[freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hi !

Feedback from devs ... they are ok to work for FProxy using jetty/Wickets.
We are ready to help devs from a git rep. We just need a first stub.

GWT would be another project especially for download management (but this
is not the top priority)

So ok for Wickets+jetty.
Templating would be the responsability of Wickets ?

Rgds
ps: Pou .. are you ok with that point of view, you lead the dev or nextgens
?
pss: 2 or 3 devs would be ok for work on that project in our side.

- Nicolas Hernandez
a-n - aleph-networks
*CEO*
http://www.aleph-networks.com




On Wed, Apr 18, 2012 at 5:51 PM, Ian Clarke  wrote:

> On Tue, Apr 17, 2012 at 3:21 AM, Nicolas Hernandez <
> nicolas.hernandez at aleph-networks.com> wrote:
>
>> i don't know ... if we can forget Velocity i am happy :-). I don't know
>> if for Freenet devs, Velocity is still a Pre-requisies.
>>
>
> I don't think Velocity was ever a prerequisite.  The only prerequisites
> are that:
>
>- We get a UI that looks good and is easy to use
>- The code for the UI is easy to maintain and extend
>- The UI can do everything that the current UI can do
>
> I am willing to trust the people that are willing to do the work to pick
> good tools, within reasonable limits.
>
> Ian.
>
> --
> Ian Clarke
> Founder, The Freenet Project
> Email: ian at freenetproject.org
>
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hi !

Feedback from devs ... they are ok to work for FProxy using jetty/Wickets.
We are ready to help devs from a git rep. We just need a first stub.

GWT would be another project especially for download management (but this
is not the top priority)

So ok for Wickets+jetty.
Templating would be the responsability of Wickets ?

Rgds
ps: Pou .. are you ok with that point of view, you lead the dev or nextgens
?
pss: 2 or 3 devs would be ok for work on that project in our side.

- Nicolas Hernandez
a-n - aleph-networks
*CEO*
http://www.aleph-networks.com




On Wed, Apr 18, 2012 at 5:51 PM, Ian Clarke i...@freenetproject.org wrote:

 On Tue, Apr 17, 2012 at 3:21 AM, Nicolas Hernandez 
 nicolas.hernan...@aleph-networks.com wrote:

 i don't know ... if we can forget Velocity i am happy :-). I don't know
 if for Freenet devs, Velocity is still a Pre-requisies.


 I don't think Velocity was ever a prerequisite.  The only prerequisites
 are that:

- We get a UI that looks good and is easy to use
- The code for the UI is easy to maintain and extend
- The UI can do everything that the current UI can do

 I am willing to trust the people that are willing to do the work to pick
 good tools, within reasonable limits.

 Ian.

 --
 Ian Clarke
 Founder, The Freenet Project
 Email: i...@freenetproject.org

___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Ian Clarke]

On Tue, Apr 17, 2012 at 3:21 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> i don't know ... if we can forget Velocity i am happy :-). I don't know if
> for Freenet devs, Velocity is still a Pre-requisies.
>

I don't think Velocity was ever a prerequisite.  The only prerequisites are
that:

   - We get a UI that looks good and is easy to use
   - The code for the UI is easy to maintain and extend
   - The UI can do everything that the current UI can do

I am willing to trust the people that are willing to do the work to pick
good tools, within reasonable limits.

Ian.

-- 
Ian Clarke
Founder, The Freenet Project
Email: ian at freenetproject.org
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Tue, Apr 17, 2012 at 3:21 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 i don't know ... if we can forget Velocity i am happy :-). I don't know if
 for Freenet devs, Velocity is still a Pre-requisies.


I don't think Velocity was ever a prerequisite.  The only prerequisites are
that:

   - We get a UI that looks good and is easy to use
   - The code for the UI is easy to maintain and extend
   - The UI can do everything that the current UI can do

I am willing to trust the people that are willing to do the work to pick
good tools, within reasonable limits.

Ian.

-- 
Ian Clarke
Founder, The Freenet Project
Email: i...@freenetproject.org
___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

i don't know ... if we can forget Velocity i am happy :-). I don't know if
for Freenet devs, Velocity is still a Pre-requisies.

we have started a POC about GWT components. The goal is to demonstrate the
use of GWT as component
- in a standard jsp/Servlet environment
- with wickets ?!
- with non-js Renderer


The main goal is to offer these components to the freenet comunity for a
use in the new FProxy architecture.
Templating can be done using wickets (if confirmed) or Velocity.

Rgds
- Nicolas Hernandez
a-n - aleph-networks
*CEO*
http://www.aleph-networks.com




On Fri, Apr 13, 2012 at 10:32 PM, Ian Clarke  wrote:

> On Thu, Apr 12, 2012 at 6:01 AM, Pouyan Zachar wrote:
>
>>  Wicket + Velocity is a excellent solution
>>
>>
> Doesn't Wicket include its own templating engine, which is just an
> extension of XHTML?
>
> Ian.
>
> --
> Ian Clarke
> Founder, The Freenet Project
> Email: ian at freenetproject.org
>
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

i don't know ... if we can forget Velocity i am happy :-). I don't know if
for Freenet devs, Velocity is still a Pre-requisies.

we have started a POC about GWT components. The goal is to demonstrate the
use of GWT as component
- in a standard jsp/Servlet environment
- with wickets ?!
- with non-js Renderer


The main goal is to offer these components to the freenet comunity for a
use in the new FProxy architecture.
Templating can be done using wickets (if confirmed) or Velocity.

Rgds
- Nicolas Hernandez
a-n - aleph-networks
*CEO*
http://www.aleph-networks.com




On Fri, Apr 13, 2012 at 10:32 PM, Ian Clarke i...@freenetproject.org wrote:

 On Thu, Apr 12, 2012 at 6:01 AM, Pouyan Zachar pouyans...@gmail.comwrote:

  Wicket + Velocity is a excellent solution


 Doesn't Wicket include its own templating engine, which is just an
 extension of XHTML?

 Ian.

 --
 Ian Clarke
 Founder, The Freenet Project
 Email: i...@freenetproject.org

___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Ian Clarke]

On Thu, Apr 12, 2012 at 6:01 AM, Pouyan Zachar  wrote:

>  Wicket + Velocity is a excellent solution
>
>
Doesn't Wicket include its own templating engine, which is just an
extension of XHTML?

Ian.

-- 
Ian Clarke
Founder, The Freenet Project
Email: ian at freenetproject.org
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Donnerstag, 12. April 2012, 14:48:08 schrieb Nicolas Hernandez:
> Yes  i know. But we are working about accessibility   for
> developpers too !
> 
> 
>- Velocity for templating
>- html+jsp for site rendering
>- gwt+nonjs for components and components rendering

How does having to know 3 different techniques help accessibility for 
developers? Especially maintainability seems to be harmed by that (I can 
already see library update nightmares: 3 times as many breakages, because some 
API changed).

Best wishes,
Arne
--
Konstruktive Kritik: 

- http://draketo.de/licht/krude-ideen/konstruktive-kritik

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Donnerstag, 12. April 2012, 14:48:08 schrieb Nicolas Hernandez:
 Yes  i know. But we are working about accessibility   for
 developpers too !
 
 
- Velocity for templating
- html+jsp for site rendering
- gwt+nonjs for components and components rendering

How does having to know 3 different techniques help accessibility for 
developers? Especially maintainability seems to be harmed by that (I can 
already see library update nightmares: 3 times as many breakages, because some 
API changed).

Best wishes,
Arne
--
Konstruktive Kritik: 

- http://draketo.de/licht/krude-ideen/konstruktive-kritik



signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Thu, Apr 12, 2012 at 6:01 AM, Pouyan Zachar pouyans...@gmail.com wrote:

  Wicket + Velocity is a excellent solution


Doesn't Wicket include its own templating engine, which is just an
extension of XHTML?

Ian.

-- 
Ian Clarke
Founder, The Freenet Project
Email: i...@freenetproject.org
___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

- Nicolas Hernandez

> *
>> Concerning GWT*:
>> *If all devs and the leader of http://goo.gl/QhPVD are ok to use GWT *we
>> can consider rewriting FProxy using GWT components.
>> We can use GWT as a pool of components only.
>> -> Using Velocity for templating (if needed).
>> -> using jsp like pages
>>
>
> I'm not sure what you mean by "using JSP-like pages"? do you mean to
> imitate JSP-Behavior by utilizing Velocity? or to use VelocityStruts (
> https://velocity.apache.org/tools/devel/struts/) to support JSP pages?
>

Nope, a minimal standard application like jetty supports jsp pages.
Velocity can be use for site templating instead og GWT one.


>
> -> developping gwt components simply embeded into jsp pages (qwt with dual
>> mode js/non-js)
>>
>
> Is there anyway to avoid JavaScript when using GWT? If I'm not mistaken,
> the GWT compiler (or better said: translator) compiles Java to JavaScript (
> http://css.dzone.com/news/understanding-gwt-compiler). Are you going to
> another framework for non-JS support?
>
> Nope, je just have to extend GWT and override it if needed ... just
standard Object Coding :-)



>
>> ps: jsp pages are a good solution for designers (as wickets)
>>
>
> In deed! However there was a discussion last year regarding using JSP (
> http://www.mail-archive.com/devl at freenetproject.org/msg25805.html).
> Matthew was more or less against using JSP and Servlets, because of the
> extra features (and size!) which we don't need. That's why we moved to
> Velocity: simplicity, fast and small in size.
>
>>
>>
Yes  i know. But we are working about accessibility :-)  for
developpers too ! ;-)


   - Velocity for templating
   - html+jsp for site rendering
   - gwt+nonjs for components and components rendering

We can use struts+gwt components if you want, or other, struts is really
too complex in my opinion for fproxy.




> *
>> *
>> - Nicolas Hernandez
>> a-n - aleph-networks
>> *CEO*
>> http://www.aleph-networks.com
>>
>>
>>
>>
>>
>> On Wed, Apr 11, 2012 at 2:32 PM, Pouyan Zachar > gmail.com>wrote:
>>
>>> Dear Nicolas,
>>>
>>> as you may know, I have proposed to rewrite the Freenet GUI in context
>>> of GSoC 2012 (see http://goo.gl/kOEAU and http://goo.gl/QhPVD)  In
>>> order to avoid duplicate implementations, I would really like to get more
>>> insight about your current approach and implementation using GWT.
>>>
>>> Regards
>>> Pouyan
>>>
>>
>>
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Pouyan Zachar]

Hi!

On Thu, Apr 12, 2012 at 9:40 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> Hello,
> *First of all*;
>
>- http://goo.gl/kOEAU
>- http://goo.gl/QhPVD
>
> Are ok for me, the only reason i prefer going to gwt is the team. It is
> easier to find dev using GWT than Wickets.
> Wicket + Velocity is a excellent solution, if there is a team of minimum 5
> devs. We can help about these 2 projects (code review, architecture design,
> and some devs later)
>

Great to hear!

*
> Concerning GWT*:
> *If all devs and the leader of http://goo.gl/QhPVD are ok to use GWT *we
> can consider rewriting FProxy using GWT components.
> We can use GWT as a pool of components only.
> -> Using Velocity for templating (if needed).
> -> using jsp like pages
>

I'm not sure what you mean by "using JSP-like pages"? do you mean to
imitate JSP-Behavior by utilizing Velocity? or to use VelocityStruts (
https://velocity.apache.org/tools/devel/struts/) to support JSP pages?

-> developping gwt components simply embeded into jsp pages (qwt with dual
> mode js/non-js)
>

Is there anyway to avoid JavaScript when using GWT? If I'm not mistaken,
the GWT compiler (or better said: translator) compiles Java to JavaScript (
http://css.dzone.com/news/understanding-gwt-compiler). Are you going to
another framework for non-JS support?


> ps: jsp pages are a good solution for designers (as wickets)
>

In deed! However there was a discussion last year regarding using JSP (
http://www.mail-archive.com/devl at freenetproject.org/msg25805.html). Matthew
was more or less against using JSP and Servlets, because of the extra
features (and size!) which we don't need. That's why we moved to Velocity:
simplicity, fast and small in size.

>
>
> *Remember, we have solutions, but we are only followers. In all cases
> FProxy rewriting needs an afficial and hitorical dev leader (like nextgen
> :-) )*


Indeed!


> *
> *
> - Nicolas Hernandez
> a-n - aleph-networks
> *CEO*
> http://www.aleph-networks.com
>
>
>
>
>
> On Wed, Apr 11, 2012 at 2:32 PM, Pouyan Zachar wrote:
>
>> Dear Nicolas,
>>
>> as you may know, I have proposed to rewrite the Freenet GUI in context of
>> GSoC 2012 (see http://goo.gl/kOEAU and http://goo.gl/QhPVD)  In order to
>> avoid duplicate implementations, I would really like to get more insight
>> about your current approach and implementation using GWT.
>>
>> Regards
>> Pouyan
>>
>
>
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Pouyan Zachar]

Hi!

On Thu, Apr 12, 2012 at 9:40 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 Hello,
 *First of all*;

- http://goo.gl/kOEAU
- http://goo.gl/QhPVD

 Are ok for me, the only reason i prefer going to gwt is the team. It is
 easier to find dev using GWT than Wickets.
 Wicket + Velocity is a excellent solution, if there is a team of minimum 5
 devs. We can help about these 2 projects (code review, architecture design,
 and some devs later)


Great to hear!

*
 Concerning GWT*:
 *If all devs and the leader of http://goo.gl/QhPVD are ok to use GWT *we
 can consider rewriting FProxy using GWT components.
 We can use GWT as a pool of components only.
 - Using Velocity for templating (if needed).
 - using jsp like pages


I'm not sure what you mean by using JSP-like pages? do you mean to
imitate JSP-Behavior by utilizing Velocity? or to use VelocityStruts (
https://velocity.apache.org/tools/devel/struts/) to support JSP pages?

- developping gwt components simply embeded into jsp pages (qwt with dual
 mode js/non-js)


Is there anyway to avoid JavaScript when using GWT? If I'm not mistaken,
the GWT compiler (or better said: translator) compiles Java to JavaScript (
http://css.dzone.com/news/understanding-gwt-compiler). Are you going to
another framework for non-JS support?


 ps: jsp pages are a good solution for designers (as wickets)


In deed! However there was a discussion last year regarding using JSP (
http://www.mail-archive.com/devl@freenetproject.org/msg25805.html). Matthew
was more or less against using JSP and Servlets, because of the extra
features (and size!) which we don't need. That's why we moved to Velocity:
simplicity, fast and small in size.



 *Remember, we have solutions, but we are only followers. In all cases
 FProxy rewriting needs an afficial and hitorical dev leader (like nextgen
 :-) )*


Indeed!


 *
 *
 - Nicolas Hernandez
 a-n - aleph-networks
 *CEO*
 http://www.aleph-networks.com





 On Wed, Apr 11, 2012 at 2:32 PM, Pouyan Zachar pouyans...@gmail.comwrote:

 Dear Nicolas,

 as you may know, I have proposed to rewrite the Freenet GUI in context of
 GSoC 2012 (see http://goo.gl/kOEAU and http://goo.gl/QhPVD)  In order to
 avoid duplicate implementations, I would really like to get more insight
 about your current approach and implementation using GWT.

 Regards
 Pouyan



___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

- Nicolas Hernandez

 *
 Concerning GWT*:
 *If all devs and the leader of http://goo.gl/QhPVD are ok to use GWT *we
 can consider rewriting FProxy using GWT components.
 We can use GWT as a pool of components only.
 - Using Velocity for templating (if needed).
 - using jsp like pages


 I'm not sure what you mean by using JSP-like pages? do you mean to
 imitate JSP-Behavior by utilizing Velocity? or to use VelocityStruts (
 https://velocity.apache.org/tools/devel/struts/) to support JSP pages?


Nope, a minimal standard application like jetty supports jsp pages.
Velocity can be use for site templating instead og GWT one.



 - developping gwt components simply embeded into jsp pages (qwt with dual
 mode js/non-js)


 Is there anyway to avoid JavaScript when using GWT? If I'm not mistaken,
 the GWT compiler (or better said: translator) compiles Java to JavaScript (
 http://css.dzone.com/news/understanding-gwt-compiler). Are you going to
 another framework for non-JS support?

 Nope, je just have to extend GWT and override it if needed ... just
standard Object Coding :-)




 ps: jsp pages are a good solution for designers (as wickets)


 In deed! However there was a discussion last year regarding using JSP (
 http://www.mail-archive.com/devl@freenetproject.org/msg25805.html).
 Matthew was more or less against using JSP and Servlets, because of the
 extra features (and size!) which we don't need. That's why we moved to
 Velocity: simplicity, fast and small in size.



Yes  i know. But we are working about accessibility :-)  for
developpers too ! ;-)


   - Velocity for templating
   - html+jsp for site rendering
   - gwt+nonjs for components and components rendering

We can use struts+gwt components if you want, or other, struts is really
too complex in my opinion for fproxy.




 *
 *
 - Nicolas Hernandez
 a-n - aleph-networks
 *CEO*
 http://www.aleph-networks.com





 On Wed, Apr 11, 2012 at 2:32 PM, Pouyan Zachar pouyans...@gmail.comwrote:

 Dear Nicolas,

 as you may know, I have proposed to rewrite the Freenet GUI in context
 of GSoC 2012 (see http://goo.gl/kOEAU and http://goo.gl/QhPVD)  In
 order to avoid duplicate implementations, I would really like to get more
 insight about your current approach and implementation using GWT.

 Regards
 Pouyan




___
Devl mailing list
Devl@freenetproject.org
https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Tuesday 13 Mar 2012 02:54:47 Steve Dougherty wrote:
> I very much like this idea! Presenting a single page with choices,
> especially when the final button is "Finish and connect now" is a
> superb way to speed up the first run setup. The current use of
> multiple pages without even a progress bar puts more pressure on each
> decision and feels very slow.
> 
> Hopefully we could add arrows next to the options which can be clicked
> to expand with additional explanation and information instead of
> presenting a wall of text.
> 
> Does someone else want to implement this? I'd love to, but I won't
> have time for almost two months.

I'd be willing to implement it if nobody else does (I'm sure Ian would be 
delighted with it!). We need to discuss the details first but it's looking 
promising.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Tuesday 13 Mar 2012 02:54:47 Steve Dougherty wrote:
 I very much like this idea! Presenting a single page with choices,
 especially when the final button is Finish and connect now is a
 superb way to speed up the first run setup. The current use of
 multiple pages without even a progress bar puts more pressure on each
 decision and feels very slow.
 
 Hopefully we could add arrows next to the options which can be clicked
 to expand with additional explanation and information instead of
 presenting a wall of text.
 
 Does someone else want to implement this? I'd love to, but I won't
 have time for almost two months.

I'd be willing to implement it if nobody else does (I'm sure Ian would be 
delighted with it!). We need to discuss the details first but it's looking 
promising.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Ian Clarke]

On Fri, Mar 9, 2012 at 11:45 AM, Matthew Toseland  wrote:

> Last time I asked, just about everyone actively using FMS who commented on
> the matter was of the view that for Freenet to *require* javascript would
> be utterly unacceptable.
>

Yes, when you ask people "how would you feel about being forced to use X",
the answer will always be negative.  If you ask people "would you like
Freenet to have a user interface of comparable quality to Google Docs or
Gmail?", I suspect they'd say "yes".  You can get the answer you want by
phrasing the question carefully.


> Aggravating your core userbase is usually a bad strategy, even if you hope
> that you might gain casual users to replace them eventually. IMHO it is
> important - not necessarily of primary importance, but an important
> consideration - not to alienate what is at present, and probably in future,
> an important demographic - paranoid (justifiably or not) semi-techie users.
>

We should accomodate justified concerns, but we should not accomodate
unjustified concerns - and concerns about Javascript fall into the latter
category.  Many projects die because they are held hostage by a vocal
minority of their existing user base who resist any change.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Fri, Mar 9, 2012 at 11:45 AM, Matthew Toseland t...@amphibian.dyndns.org
 wrote:

 Last time I asked, just about everyone actively using FMS who commented on
 the matter was of the view that for Freenet to *require* javascript would
 be utterly unacceptable.


Yes, when you ask people how would you feel about being forced to use X,
the answer will always be negative.  If you ask people would you like
Freenet to have a user interface of comparable quality to Google Docs or
Gmail?, I suspect they'd say yes.  You can get the answer you want by
phrasing the question carefully.


 Aggravating your core userbase is usually a bad strategy, even if you hope
 that you might gain casual users to replace them eventually. IMHO it is
 important - not necessarily of primary importance, but an important
 consideration - not to alienate what is at present, and probably in future,
 an important demographic - paranoid (justifiably or not) semi-techie users.


We should accomodate justified concerns, but we should not accomodate
unjustified concerns - and concerns about Javascript fall into the latter
category.  Many projects die because they are held hostage by a vocal
minority of their existing user base who resist any change.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Steve Dougherty]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I very much like this idea! Presenting a single page with choices,
especially when the final button is "Finish and connect now" is a
superb way to speed up the first run setup. The current use of
multiple pages without even a progress bar puts more pressure on each
decision and feels very slow.

Hopefully we could add arrows next to the options which can be clicked
to expand with additional explanation and information instead of
presenting a wall of text.

Does someone else want to implement this? I'd love to, but I won't
have time for almost two months.

- -operhiem1

On 03/11/2012 08:53 AM, Arne Babenhauserheide wrote:
> Am Freitag, 9. M?rz 2012, 21:52:22 schrieb Steve Dougherty:
>> Should we get rid of more of the steps in the first-run setup?
> 
> I think yes - if people choose the default installation, they
> should not be forced to take additional decisions.
> 
> Maybe it would even be possible to just show a pane ?Basic Config?
> which shows preselected configuration values, each with an
> edit-button (?change?).
> 
> (I use [button] for buttons with the text ?button?)
> 
> ### Basic Config ###
> 
> Please check the default configuration. Change values which don?t
> fit:
> 
> * Encrypted Datastore Size: 2 GiB [change] * Assigned Bandwidth: 10
> KiB/s (26 GiB per month) [change] * ?
> 
> [Finish and connect now]
> 
> (4 values, one for each pane of the current wizard)
> 
>> I2P works very nicely with something that would work without
>> Javascript, but works much better with: it shows for a given
>> speed setting how much it will likely use per month. That a way
>> better way to deal with monthly bandwidth limits than asking
>> directly.
> 
> I would like that.
> 
> Best wishes, Arne
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=0ueI
-END PGP SIGNATURE-

Re: [freenet-dev] webui and accessibility

[Steve Dougherty]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I very much like this idea! Presenting a single page with choices,
especially when the final button is Finish and connect now is a
superb way to speed up the first run setup. The current use of
multiple pages without even a progress bar puts more pressure on each
decision and feels very slow.

Hopefully we could add arrows next to the options which can be clicked
to expand with additional explanation and information instead of
presenting a wall of text.

Does someone else want to implement this? I'd love to, but I won't
have time for almost two months.

- -operhiem1

On 03/11/2012 08:53 AM, Arne Babenhauserheide wrote:
 Am Freitag, 9. März 2012, 21:52:22 schrieb Steve Dougherty:
 Should we get rid of more of the steps in the first-run setup?
 
 I think yes - if people choose the default installation, they
 should not be forced to take additional decisions.
 
 Maybe it would even be possible to just show a pane “Basic Config”
 which shows preselected configuration values, each with an
 edit-button (“change”).
 
 (I use [button] for buttons with the text “button”)
 
 ### Basic Config ###
 
 Please check the default configuration. Change values which don’t
 fit:
 
 * Encrypted Datastore Size: 2 GiB [change] * Assigned Bandwidth: 10
 KiB/s (26 GiB per month) [change] * …
 
 [Finish and connect now]
 
 (4 values, one for each pane of the current wizard)
 
 I2P works very nicely with something that would work without
 Javascript, but works much better with: it shows for a given
 speed setting how much it will likely use per month. That a way
 better way to deal with monthly bandwidth limits than asking
 directly.
 
 I would like that.
 
 Best wishes, Arne
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=0ueI
-END PGP SIGNATURE-
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. M?rz 2012, 21:52:22 schrieb Steve Dougherty:
> Should we get rid of more of the steps in the first-run setup? 

I think yes - if people choose the default installation, they should not be 
forced to take additional decisions.

Maybe it would even be possible to just show a pane ?Basic Config? which shows 
preselected configuration values, each with an edit-button (?change?).

(I use [button] for buttons with the text ?button?)

### Basic Config ###

Please check the default configuration. Change values which don?t fit:

* Encrypted Datastore Size: 2 GiB [change]
* Assigned Bandwidth: 10 KiB/s (26 GiB per month) [change]
* ?

[Finish and connect now]

(4 values, one for each pane of the current wizard)

> I2P works very nicely with
> something that would work without Javascript, but works much better
> with: it shows for a given speed setting how much it will likely use per
> month. That a way better way to deal with monthly bandwidth limits than
> asking directly.

I would like that.

Best wishes,
Arne
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. März 2012, 21:52:22 schrieb Steve Dougherty:
 Should we get rid of more of the steps in the first-run setup?

I think yes - if people choose the default installation, they should not be
forced to take additional decisions.

Maybe it would even be possible to just show a pane “Basic Config” which shows
preselected configuration values, each with an edit-button (“change”).

(I use [button] for buttons with the text “button”)

### Basic Config ###

Please check the default configuration. Change values which don’t fit:

* Encrypted Datastore Size: 2 GiB [change]
* Assigned Bandwidth: 10 KiB/s (26 GiB per month) [change]
* …

[Finish and connect now]

(4 values, one for each pane of the current wizard)

 I2P works very nicely with
 something that would work without Javascript, but works much better
 with: it shows for a given speed setting how much it will likely use per
 month. That a way better way to deal with monthly bandwidth limits than
 asking directly.

I would like that.

Best wishes,
Arne

signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. M?rz 2012, 23:59:30 schrieb Arne Babenhauserheide:
> Am Mittwoch, 7. M?rz 2012, 15:29:45 schrieb Ian Clarke:
> > > So what is the experience of new users? Install freenet, see automaticly
> > > shipped wot+freetalk, try them, see their issues, maybe browse some
> > > sites and uninstall freenet again, since things either do rarely work or
> > > there is no content. A cool web-UI wont change those points, so from my
> > > perspective, a different interface may be nice, but it wont solve the
> > > bigger issues with freenet.
> > 
> > So again, you think that just because a new UI won't solve every problem
> > with Freenet, then we shouldn't do it?  That's completely illogical.
> 
> If it creates more problems than it solves, it is a problem.

I just saw a message on Sone, which shows one quite extreme position on 
Javascript, and I thought you should know about it before deciding whether to 
make Freenet dependant on Javascript:

jacg:
Achtung, rant ahead. JS, serverside scripting, cookies, web "apps", canvases, 
media players, clientside SQL databases, the recent brain damage of "web 
sockets", are all cancer and the faster they all die, the better. 

A big part of why I really enjoy reading Freesites is that all of this brain 
damage is disallowed, pointless, and/or technically impossible to implement, 
and thus the sites have only two things left to focus on: template and 
content; the latter being the only one of any importance. 

Web pages are documents. If you don't see how far wrong the things have gone, 
try to imagine yourself having to open a Word document each time you want to 
play minesweeper. Or opening a Word document to run an image editor inside of 
it. Or opening a Word document to open a web browser to go to a website that 
allows you to write Word documents. 

It's been 20 years of idiots trying to turn web pages into applications, and 
the result is several incompatible implementations, shitty development tools, 
ever-increasing abuse of user privacy, millions of developers' man-hours 
wasted on dealing with legacy crap, retrofitting protocols and formats to do 
things they were never designed to do, and so on and on. 

The web is a fucking frankenstein and it's getting worse with every iteration. 
Please keep this turd off my lawn.

> Best wishes,
> Arne
--
Ein W?rfel System - einfach saubere Regeln: 

- http://1w6.org

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Saturday 10 Mar 2012 02:54:15 Steve Dougherty wrote:
> Should we get rid of more of the steps in the first-run setup? Asking
> about datastore size is a must, but perhaps we can get away with
> something like what I2P does where it assumes a very low bandwidth limit
> and encourages users to raise it afterwards. This could work well
> alongside requiring a 5KiB/s minimum. 

No, unfortunately Freenet won't work well with less than about 16KB/sec :(

And a lot of people who want or even need to use it have less than that 
available.

Usually we can detect the bandwidth from the router. We can't skip it even in 
that case because of the next item ...

> I2P works very nicely with
> something that would work without Javascript, but works much better
> with: it shows for a given speed setting how much it will likely use per
> month. That a way better way to deal with monthly bandwidth limits than
> asking directly.

More advanced users will definitely want to know the exact figure per second, 
but maybe we need an advanced mode - that's an extra step but it could be a 
checkbox or something.

I agree 100% that we need to show traffic per month on the bandwidth selector. 
This is planned but not implemented IIRC. Although I'd need to see a mockup - 
it's a lot of information, doing it separately as we do now may be easier.

Also more generally, fewer but more complex steps isn't necessarily more user 
friendly.
> 
> -operhiem1
> 
> On 03/09/2012 07:50 PM, Arne Babenhauserheide wrote:
> > Am Freitag, 9. M?rz 2012, 23:14:04 schrieb Matthew Toseland:
> >> On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
> >>> Am Mittwoch, 7. M?rz 2012, 22:14:39 schrieb Thomas Sachau:
>  So what is the experience of new users? Install freenet, see automaticly
>  shipped wot+freetalk, try them, see their issues, maybe browse some
>  sites and uninstall freenet again, since things either do rarely work or
>  there is no content.
> >>>
> >>> I think it goes even deeper: The installer fails for many.
> >>
> >> Does it still? If so we need to determine why.
> > 
> > Last week someone in IRC had problems again.
> > 
> > And I think the installer has to be as simple and straightforward as 
> > possible. 
> > What I saw in the chinese freenet installation youtube movie was much too 
> > complex. 6 steps in the installer (or so) and then 4 configuration steps 
> > (do 
> > we really need 2 steps with only 2 buttons on the site?).
> > 
> > And then there are firewall and port-forwarding issues.
> > 
> >>> Also Freetalk is not integrated with the other parts of freenet. There
> >>> should be attachment options, where the post is only sent once the file
> >>> finished uploading. And lots of other ?small? stuff on the UI side.
> >>
> >> Agreed, as does p0s, but first it has to WORK.
> > 
> > Jupp. Any plans about Sone?
> > 
> > Best wishes,
> > Arne
> > 
> > 
> > 
> > ___
> > Devl mailing list
> > Devl at freenetproject.org
> > http://freenetproject.org/cgi-bin/mailman/listinfo/devl
> 
> 
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. M?rz 2012, 23:14:04 schrieb Matthew Toseland:
> On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
> > Am Mittwoch, 7. M?rz 2012, 22:14:39 schrieb Thomas Sachau:
> > > So what is the experience of new users? Install freenet, see automaticly
> > > shipped wot+freetalk, try them, see their issues, maybe browse some
> > > sites and uninstall freenet again, since things either do rarely work or
> > > there is no content.
> > 
> > I think it goes even deeper: The installer fails for many.
> 
> Does it still? If so we need to determine why.

Last week someone in IRC had problems again.

And I think the installer has to be as simple and straightforward as possible. 
What I saw in the chinese freenet installation youtube movie was much too 
complex. 6 steps in the installer (or so) and then 4 configuration steps (do 
we really need 2 steps with only 2 buttons on the site?).

And then there are firewall and port-forwarding issues.

> > Also Freetalk is not integrated with the other parts of freenet. There
> > should be attachment options, where the post is only sent once the file
> > finished uploading. And lots of other ?small? stuff on the UI side.
> 
> Agreed, as does p0s, but first it has to WORK.

Jupp. Any plans about Sone?

Best wishes,
Arne
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. M?rz 2012, 22:14:39 schrieb Thomas Sachau:
> So what is the experience of new users? Install freenet, see automaticly
> shipped wot+freetalk, try them, see their issues, maybe browse some
> sites and uninstall freenet again, since things either do rarely work or
> there is no content.

I think it goes even deeper: The installer fails for many. 
Also Freetalk is not integrated with the other parts of freenet. There should 
be attachment options, where the post is only sent once the file finished 
uploading. And lots of other ?small? stuff on the UI side.

Essentially polish.

I think being really polished and getting rid of all the small rough edges 
would be a much stronger usability boost than using new tech. (except if that 
new tech does not add more rough corners and makes is much easier to fix 
existing problems)

Best wishes,
Arne
-- 
1w6 sie zu achten,
sie alle zu finden,
in Spiele zu leiten
und sacht zu verbinden.
? http://1w6.org

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Saturday 10 Mar 2012 02:54:15 Steve Dougherty wrote:
 Should we get rid of more of the steps in the first-run setup? Asking
 about datastore size is a must, but perhaps we can get away with
 something like what I2P does where it assumes a very low bandwidth limit
 and encourages users to raise it afterwards. This could work well
 alongside requiring a 5KiB/s minimum. 

No, unfortunately Freenet won't work well with less than about 16KB/sec :(

And a lot of people who want or even need to use it have less than that 
available.

Usually we can detect the bandwidth from the router. We can't skip it even in 
that case because of the next item ...

 I2P works very nicely with
 something that would work without Javascript, but works much better
 with: it shows for a given speed setting how much it will likely use per
 month. That a way better way to deal with monthly bandwidth limits than
 asking directly.

More advanced users will definitely want to know the exact figure per second, 
but maybe we need an advanced mode - that's an extra step but it could be a 
checkbox or something.

I agree 100% that we need to show traffic per month on the bandwidth selector. 
This is planned but not implemented IIRC. Although I'd need to see a mockup - 
it's a lot of information, doing it separately as we do now may be easier.

Also more generally, fewer but more complex steps isn't necessarily more user 
friendly.
 
 -operhiem1
 
 On 03/09/2012 07:50 PM, Arne Babenhauserheide wrote:
  Am Freitag, 9. März 2012, 23:14:04 schrieb Matthew Toseland:
  On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
  Am Mittwoch, 7. März 2012, 22:14:39 schrieb Thomas Sachau:
  So what is the experience of new users? Install freenet, see automaticly
  shipped wot+freetalk, try them, see their issues, maybe browse some
  sites and uninstall freenet again, since things either do rarely work or
  there is no content.
 
  I think it goes even deeper: The installer fails for many.
 
  Does it still? If so we need to determine why.
  
  Last week someone in IRC had problems again.
  
  And I think the installer has to be as simple and straightforward as 
  possible. 
  What I saw in the chinese freenet installation youtube movie was much too 
  complex. 6 steps in the installer (or so) and then 4 configuration steps 
  (do 
  we really need 2 steps with only 2 buttons on the site?).
  
  And then there are firewall and port-forwarding issues.
  
  Also Freetalk is not integrated with the other parts of freenet. There
  should be attachment options, where the post is only sent once the file
  finished uploading. And lots of other “small” stuff on the UI side.
 
  Agreed, as does p0s, but first it has to WORK.
  
  Jupp. Any plans about Sone?
  
  Best wishes,
  Arne
  
  
  
  ___
  Devl mailing list
  Devl@freenetproject.org
  http://freenetproject.org/cgi-bin/mailman/listinfo/devl
 
 


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. März 2012, 23:59:30 schrieb Arne Babenhauserheide:
 Am Mittwoch, 7. März 2012, 15:29:45 schrieb Ian Clarke:
   So what is the experience of new users? Install freenet, see automaticly
   shipped wot+freetalk, try them, see their issues, maybe browse some
   sites and uninstall freenet again, since things either do rarely work or
   there is no content. A cool web-UI wont change those points, so from my
   perspective, a different interface may be nice, but it wont solve the
   bigger issues with freenet.
 
  So again, you think that just because a new UI won't solve every problem
  with Freenet, then we shouldn't do it?  That's completely illogical.

 If it creates more problems than it solves, it is a problem.

I just saw a message on Sone, which shows one quite extreme position on
Javascript, and I thought you should know about it before deciding whether to
make Freenet dependant on Javascript:

jacg:
Achtung, rant ahead. JS, serverside scripting, cookies, web apps, canvases,
media players, clientside SQL databases, the recent brain damage of web
sockets, are all cancer and the faster they all die, the better.

A big part of why I really enjoy reading Freesites is that all of this brain
damage is disallowed, pointless, and/or technically impossible to implement,
and thus the sites have only two things left to focus on: template and
content; the latter being the only one of any importance.

Web pages are documents. If you don't see how far wrong the things have gone,
try to imagine yourself having to open a Word document each time you want to
play minesweeper. Or opening a Word document to run an image editor inside of
it. Or opening a Word document to open a web browser to go to a website that
allows you to write Word documents.

It's been 20 years of idiots trying to turn web pages into applications, and
the result is several incompatible implementations, shitty development tools,
ever-increasing abuse of user privacy, millions of developers' man-hours
wasted on dealing with legacy crap, retrofitting protocols and formats to do
things they were never designed to do, and so on and on.

The web is a fucking frankenstein and it's getting worse with every iteration.
Please keep this turd off my lawn.

 Best wishes,
 Arne
--
Ein Würfel System - einfach saubere Regeln:

- http://1w6.org



signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. M?rz 2012, 15:29:45 schrieb Ian Clarke:
> > So what is the experience of new users? Install freenet, see automaticly
> > shipped wot+freetalk, try them, see their issues, maybe browse some
> > sites and uninstall freenet again, since things either do rarely work or
> > there is no content. A cool web-UI wont change those points, so from my
> > perspective, a different interface may be nice, but it wont solve the
> > bigger issues with freenet.
> 
> So again, you think that just because a new UI won't solve every problem
> with Freenet, then we shouldn't do it?  That's completely illogical.

If it creates more problems than it solves, it is a problem.

Not being able to use freenet without Javascript would be a problem. Dillo has 
no Javascript support. Neither have w3m, lynx, and a whole bunch of other low-
profile browsers. Mainly used by geeky people - who are a natural target 
audience for freenet.

I don?t mind a good Javascript UI. But Sone was actually the first really 
useful example for that which I ever saw. Yahoo is a really good counter-
example: they even reimplemented tabs in Javascript?

Javascript has to be complementary, though: The UI has to work without it. And 
it can. After all, a reload of a local page is blazingly fast (when there are 
no other bottlenecks).

Also there are many points on the UI side which could be improved that don?t 
need JS. Better download-queue, integrated WoT and Sone (well, that benefits 
from JS), and so on.

On the other hand, drag-and-drop uploading with image previews and all that is 
quite a usability booster. So I am not against JS (anymore). I am just against 
requiring it.

Best wishes,
Arne

-- 
Unpolitisch sein
hei?t politisch sein, 
ohne es zu merken. 
- Arne (http://draketo.de)


-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
> Am Mittwoch, 7. M?rz 2012, 22:14:39 schrieb Thomas Sachau:
> > So what is the experience of new users? Install freenet, see automaticly
> > shipped wot+freetalk, try them, see their issues, maybe browse some
> > sites and uninstall freenet again, since things either do rarely work or
> > there is no content.
> 
> I think it goes even deeper: The installer fails for many. 

Does it still? If so we need to determine why.

> Also Freetalk is not integrated with the other parts of freenet. There should 
> be attachment options, where the post is only sent once the file finished 
> uploading. And lots of other ?small? stuff on the UI side.

Agreed, as does p0s, but first it has to WORK.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. M?rz 2012, 13:40:56 schrieb Ian Clarke:
> Yes, the vast number of people that refuse to use anything but Lynx?  There
> is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
> so we don't have to worry about him.

I regularly browse freenet with a raw text interface. ssh to my box and lynx 
or w3m-mode in emacs.

It?s much faster and much more convenient than starting a firefox on the 300 
Mhz machine I use as laptop.

Best wishes,
Arne
-- 
1w6 sie zu achten,
sie alle zu finden,
in Spiele zu leiten
und sacht zu verbinden.
? http://1w6.org

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 316 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Steve Dougherty]

Should we get rid of more of the steps in the first-run setup? Asking
about datastore size is a must, but perhaps we can get away with
something like what I2P does where it assumes a very low bandwidth limit
and encourages users to raise it afterwards. This could work well
alongside requiring a 5KiB/s minimum. I2P works very nicely with
something that would work without Javascript, but works much better
with: it shows for a given speed setting how much it will likely use per
month. That a way better way to deal with monthly bandwidth limits than
asking directly.

-operhiem1

On 03/09/2012 07:50 PM, Arne Babenhauserheide wrote:
> Am Freitag, 9. M?rz 2012, 23:14:04 schrieb Matthew Toseland:
>> On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
>>> Am Mittwoch, 7. M?rz 2012, 22:14:39 schrieb Thomas Sachau:
 So what is the experience of new users? Install freenet, see automaticly
 shipped wot+freetalk, try them, see their issues, maybe browse some
 sites and uninstall freenet again, since things either do rarely work or
 there is no content.
>>>
>>> I think it goes even deeper: The installer fails for many.
>>
>> Does it still? If so we need to determine why.
> 
> Last week someone in IRC had problems again.
> 
> And I think the installer has to be as simple and straightforward as 
> possible. 
> What I saw in the chinese freenet installation youtube movie was much too 
> complex. 6 steps in the installer (or so) and then 4 configuration steps (do 
> we really need 2 steps with only 2 buttons on the site?).
> 
> And then there are firewall and port-forwarding issues.
> 
>>> Also Freetalk is not integrated with the other parts of freenet. There
>>> should be attachment options, where the post is only sent once the file
>>> finished uploading. And lots of other ?small? stuff on the UI side.
>>
>> Agreed, as does p0s, but first it has to WORK.
> 
> Jupp. Any plans about Sone?
> 
> Best wishes,
> Arne
> 
> 
> 
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 16:23:49 Nicolas Hernandez wrote:
> Hello Ian,
> 
> We are still looking for the best choice between running in jetty
> a. Accessibility for devs (nice eclipse plugin for example)

Important but not absolutely vital. E.g. a fair bit happens with the current UI 
in spite of it not being very dev friendly.

> b. Accessibility for user

Vitally important but has a lot of meanings. E.g. totally blind people may use 
text readers or audio readers with limited or no javascript; paranoid users may 
turn off javascript; some users would like to be able to build custom themes 
(but this isn't important). More generally, users want the UI to be obvious, 
logical, responsive and pretty, in more or less that order: If it's easy to 
find what you are trying to do, it's a better UI than if it looks great but is 
completely incomprehensible, so the key thing is actually the design *from a 
usability point of view* i.e. have people thought about how the user will 
actually use it? This is mostly independant of toolkit (e.g. too much text = 
confusion; too little text = confusion). On the other hand, responding quickly 
and obviously to what the user is doing can help a lot, and this does ideally 
need javascript. And a few parts of the UI are just vastly better with 
javascript - notably stuff to do with chat, compare gmail to phpbb (Freetalk is 
much more like phpbb).

Also, some stuff is clearly controller side - a classic Freenet bug, fixed 
recently, is that when you ask Freenet to load a plugin, the browser would go 
off into limbo for minutes or more, rather than immediately returning to the 
plugins page showing that it is trying to load the plugin.

The web-pushing branch originally included code to use javascript and 
long-polling to auto-update a lot of the UI. This was never put back because of 
the bugs in the GWT web-pushing code, but for some elements it is useful, e.g. 
alerts, status pages etc.

> c. Light weight

Important. We don't want to have to add 20MB to the 10MB download just for the 
UI framework.

> d. Performance

Very important IMHO.
> 
> In our side the priority are like that a>b>d>c (+eclipse plugin)
> Four some freenet devs it loks like c>d>a>b (+velocity for templating)

I guess I'd say B D C A for me?
> 
> Not so easy to make the good choice. We have three mains ideas
> A- using Apache Wickets http://wicket.apache.org/
> B- gwt
> C- struts+extjs
> *D-  freenet devs ideas*

I'm afraid I don't have detailed knowledge of frameworks ...
> 
> This frameworks have to;
> 1. be active and old enough
> 2. including web n.m fionctionnalities
> 3. replace the Toad controller :-)
> 4. using jetty
> 5. be fully toolled for Eclipse
6. Ideally it should be possible to fall back to non-javascript, without 
actually having two complete UIs except where it is absolutely vital (IMHO 
maintaining two separate UIs would result in one of them being unmaintained in 
the long run and cause problems, as well as causing more work when we change 
something).
7. Ideally it should be possible to implement live updating using long-polling 
or similar, like the web-pushing branch, and without using one connection per 
window/tab. This is not needed immediately but it should be possible. It's not 
vital though, as we can use the existing GWT code if necessary.
> 
> We hope to have freenet devs ideas on it before starting anything :-)
> 
> Rgds
> 
> - Nicolas Hernandez
> a-n - aleph-networks
> *associ?*
> http://www.aleph-networks.com
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 21:19:38 Nicolas Hernandez wrote:
> rendering:
> it is important to differentiate the Redenderer and the Controller.
> GWT can be used:
> -  as a Controler
> -  as Renderer

Please explain briefly controller vs renderer? Is this Model/View/Controller 
you are talking about, or something else?
> 
> as Renderer gwt can offer a
> - MinimalHtmlRenderer Interface for admin and simple actions
> - Web2Renderer
> - AndroidRender ?
> 
> In this case the interface templates could be in xml and i am not sure than
> using velocity is a good idea
> 
> the beast:
> if GWT is the BEAST, ... there is some patterns against this.
> 
> gentoo:
> i love gentoo ... sorry about the gwt epic fail. Scripting if for us the
> main goal. Packaging is really important for accesibility too, as lynx for
> some admin stuffs.
> 
> So, the controler ? the application server ?
> 
> Its really a pleasure to be there :-)
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 19:40:56 Ian Clarke wrote:
> On Wed, Mar 7, 2012 at 1:21 PM, Matthew Toseland
> wrote:
> 
> > On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
> > > Unless those freenet devs are willing to build the GUI themselves, I
> > > recommend that you do not allow them to prevent you from proceeding.
> > >  Someone willing to make things happen should not be prevented from doing
> > > so by someone with an opinion, but who isn't willing to do the work.
> >
> > Even if it means breaking existing code for a large minority of users?
> > (The whole no-javascript lobby)?
> 
> Yes, the vast number of people that refuse to use anything but Lynx?  There
> is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
> so we don't have to worry about him.
> 
> But seriously, it would be insanity to hold up development of a decent user
> interface just because of the griping of a few people who, for no logical
> reason, refuse to enable JavaScript in their browsers.

Last time I asked, just about everyone actively using FMS who commented on the 
matter was of the view that for Freenet to *require* javascript would be 
utterly unacceptable.

Aggravating your core userbase is usually a bad strategy, even if you hope that 
you might gain casual users to replace them eventually. IMHO it is important - 
not necessarily of primary importance, but an important consideration - not to 
alienate what is at present, and probably in future, an important demographic - 
paranoid (justifiably or not) semi-techie users.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 20:09:51 Ian Clarke wrote:
> On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau  wrote:
> 
> > One reason to not use pre-compiled stuff is, that you can patch or
> > adjust the code before you compile it. In addition, only those patches
> > have a chance to go upstream, probably noone will accept a patch against
> > the generated javascript code.
> 
> How likely is that in practice?  In the entire history of Freenet has any
> third party felt the need to patch it?  

There have been a number of patches. The max-peers patch for example is very 
popular. There were various others in the past.

> This seems like an extreme
> edge-case to me, and that it shouldn't dictate key architectural decisions.
> 
> 
> > Since it is Gentoo policy to give the user this choice and ability,
> > shipping the precompiled code is not really an option.
> 
> Then perhaps the real problem here is Gentoo's policy?  In any case, I
> don't think Gentoo's policy should tie our hands one way or another - our
> policy is that we need a decent user interface :-)

Gentoo's policy is exactly the same as Debian's policy AFAIK and there are good 
reasons for it. However, this does not prevent us providing our own packages. 
Having an official gentoo package is a debatable point, it certainly is of some 
value (it isn't possible for other distros and many people who may be 
interested in freenet do run gentoo), but it's not THAT many potential users.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 21:29:45 Ian Clarke wrote:
> On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau  wrote:
> 
> > Ian Clarke schrieb:
> > Do you really think, that having some nice javascript or otherwise
> > adjusted interface would magically raise the numbers of users?
> 
> I think having a nice, well defined user interface will raise the number of
> users, yes.  Javascript is just a means to this end, since it's well suited
> to rich web UIs, which Freenet requires.

Do you have an example in mind?
> 
> > Some eye-candy
> 
> I'm not talking about "eye-candy", I'm talking about good user interface
> design.

Javascript can be used for both eye-candy and good user interface design, but 
the latter doesn't require javascript.
> 
> > wont keep new users nor will it get you many new users,
> > it is way more important to have a reliable working freenet with some
> > nice additional apps (which themselves should also be reliable).
> > The only reliable working additional apps seem to be FMS and maybe Frost
> > (where the later has of course issues with spam). So due to the spam,
> > the only app left is in the end FMS. Now instead of supporting that,
> > something different is started and suggested (with wot and freetalk),
> > which still have many issues, are not reliable and where until now
> > nobody was able to even tell me, how wot itself works.

Ian is correct: We have a volunteer here interested in UIs, let's give them all 
the help we can.

However, I agree in part that functionality and performance are often more 
important than small improvements to the UI, which we have had third party 
reviews praising as at least somewhat usable. (See e.g. some press a while 
back).

WoT and Freetalk are not bundled, they are installed if you ask it to. We could 
have WoT auto-installed if we wanted to - it won't do anything if you don't 
create an identity, although the next step would be to create one optionally on 
setup. WoT is a lot more mature than Freetalk. On the other hand, architectural 
changes are still likely ...

A lot of people say Sone works. It's not even on the plugin list because:
1) nobody's read the code except Bombe and
2) it has severe scalability problems which AFAIK have still not been 
addressed, and which will probably result in major changes to functionality 
eventually IMHO.
> 
> > So what is the experience of new users? Install freenet, see automaticly
> > shipped wot+freetalk, try them, see their issues, maybe browse some
> > sites and uninstall freenet again, since things either do rarely work or
> > there is no content. A cool web-UI wont change those points, so from my
> > perspective, a different interface may be nice, but it wont solve the
> > bigger issues with freenet.

Content is a chicken and egg problem. But it's also a matter of making it 
really easy to upload content with better tools, and that includes better UIs. 
E.g. the blog wizard we have is rather geeky.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 21:29:45 Ian Clarke wrote:
 On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau to...@gentoo.org wrote:
 
  Ian Clarke schrieb:
  Do you really think, that having some nice javascript or otherwise
  adjusted interface would magically raise the numbers of users?
 
 I think having a nice, well defined user interface will raise the number of
 users, yes.  Javascript is just a means to this end, since it's well suited
 to rich web UIs, which Freenet requires.

Do you have an example in mind?
 
  Some eye-candy
 
 I'm not talking about eye-candy, I'm talking about good user interface
 design.

Javascript can be used for both eye-candy and good user interface design, but 
the latter doesn't require javascript.
 
  wont keep new users nor will it get you many new users,
  it is way more important to have a reliable working freenet with some
  nice additional apps (which themselves should also be reliable).
  The only reliable working additional apps seem to be FMS and maybe Frost
  (where the later has of course issues with spam). So due to the spam,
  the only app left is in the end FMS. Now instead of supporting that,
  something different is started and suggested (with wot and freetalk),
  which still have many issues, are not reliable and where until now
  nobody was able to even tell me, how wot itself works.

Ian is correct: We have a volunteer here interested in UIs, let's give them all 
the help we can.

However, I agree in part that functionality and performance are often more 
important than small improvements to the UI, which we have had third party 
reviews praising as at least somewhat usable. (See e.g. some press a while 
back).

WoT and Freetalk are not bundled, they are installed if you ask it to. We could 
have WoT auto-installed if we wanted to - it won't do anything if you don't 
create an identity, although the next step would be to create one optionally on 
setup. WoT is a lot more mature than Freetalk. On the other hand, architectural 
changes are still likely ...

A lot of people say Sone works. It's not even on the plugin list because:
1) nobody's read the code except Bombe and
2) it has severe scalability problems which AFAIK have still not been 
addressed, and which will probably result in major changes to functionality 
eventually IMHO.
 
  So what is the experience of new users? Install freenet, see automaticly
  shipped wot+freetalk, try them, see their issues, maybe browse some
  sites and uninstall freenet again, since things either do rarely work or
  there is no content. A cool web-UI wont change those points, so from my
  perspective, a different interface may be nice, but it wont solve the
  bigger issues with freenet.

Content is a chicken and egg problem. But it's also a matter of making it 
really easy to upload content with better tools, and that includes better UIs. 
E.g. the blog wizard we have is rather geeky.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 20:09:51 Ian Clarke wrote:
 On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau to...@gentoo.org wrote:
 
  One reason to not use pre-compiled stuff is, that you can patch or
  adjust the code before you compile it. In addition, only those patches
  have a chance to go upstream, probably noone will accept a patch against
  the generated javascript code.
 
 How likely is that in practice?  In the entire history of Freenet has any
 third party felt the need to patch it?  

There have been a number of patches. The max-peers patch for example is very 
popular. There were various others in the past.

 This seems like an extreme
 edge-case to me, and that it shouldn't dictate key architectural decisions.
 
 
  Since it is Gentoo policy to give the user this choice and ability,
  shipping the precompiled code is not really an option.
 
 Then perhaps the real problem here is Gentoo's policy?  In any case, I
 don't think Gentoo's policy should tie our hands one way or another - our
 policy is that we need a decent user interface :-)

Gentoo's policy is exactly the same as Debian's policy AFAIK and there are good 
reasons for it. However, this does not prevent us providing our own packages. 
Having an official gentoo package is a debatable point, it certainly is of some 
value (it isn't possible for other distros and many people who may be 
interested in freenet do run gentoo), but it's not THAT many potential users.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 19:40:56 Ian Clarke wrote:
 On Wed, Mar 7, 2012 at 1:21 PM, Matthew Toseland
 t...@amphibian.dyndns.orgwrote:
 
  On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
   Unless those freenet devs are willing to build the GUI themselves, I
   recommend that you do not allow them to prevent you from proceeding.
Someone willing to make things happen should not be prevented from doing
   so by someone with an opinion, but who isn't willing to do the work.
 
  Even if it means breaking existing code for a large minority of users?
  (The whole no-javascript lobby)?
 
 Yes, the vast number of people that refuse to use anything but Lynx?  There
 is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
 so we don't have to worry about him.
 
 But seriously, it would be insanity to hold up development of a decent user
 interface just because of the griping of a few people who, for no logical
 reason, refuse to enable JavaScript in their browsers.

Last time I asked, just about everyone actively using FMS who commented on the 
matter was of the view that for Freenet to *require* javascript would be 
utterly unacceptable.

Aggravating your core userbase is usually a bad strategy, even if you hope that 
you might gain casual users to replace them eventually. IMHO it is important - 
not necessarily of primary importance, but an important consideration - not to 
alienate what is at present, and probably in future, an important demographic - 
paranoid (justifiably or not) semi-techie users.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 21:19:38 Nicolas Hernandez wrote:
 rendering:
 it is important to differentiate the Redenderer and the Controller.
 GWT can be used:
 -  as a Controler
 -  as Renderer

Please explain briefly controller vs renderer? Is this Model/View/Controller 
you are talking about, or something else?
 
 as Renderer gwt can offer a
 - MinimalHtmlRenderer Interface for admin and simple actions
 - Web2Renderer
 - AndroidRender ?
 
 In this case the interface templates could be in xml and i am not sure than
 using velocity is a good idea
 
 the beast:
 if GWT is the BEAST, ... there is some patterns against this.
 
 gentoo:
 i love gentoo ... sorry about the gwt epic fail. Scripting if for us the
 main goal. Packaging is really important for accesibility too, as lynx for
 some admin stuffs.
 
 So, the controler ? the application server ?
 
 Its really a pleasure to be there :-)


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 16:23:49 Nicolas Hernandez wrote:
 Hello Ian,
 
 We are still looking for the best choice between running in jetty
 a. Accessibility for devs (nice eclipse plugin for example)

Important but not absolutely vital. E.g. a fair bit happens with the current UI 
in spite of it not being very dev friendly.

 b. Accessibility for user

Vitally important but has a lot of meanings. E.g. totally blind people may use 
text readers or audio readers with limited or no javascript; paranoid users may 
turn off javascript; some users would like to be able to build custom themes 
(but this isn't important). More generally, users want the UI to be obvious, 
logical, responsive and pretty, in more or less that order: If it's easy to 
find what you are trying to do, it's a better UI than if it looks great but is 
completely incomprehensible, so the key thing is actually the design *from a 
usability point of view* i.e. have people thought about how the user will 
actually use it? This is mostly independant of toolkit (e.g. too much text = 
confusion; too little text = confusion). On the other hand, responding quickly 
and obviously to what the user is doing can help a lot, and this does ideally 
need javascript. And a few parts of the UI are just vastly better with 
javascript - notably stuff to do with chat, compare gmail to phpbb (Freetalk is 
much more like phpbb).

Also, some stuff is clearly controller side - a classic Freenet bug, fixed 
recently, is that when you ask Freenet to load a plugin, the browser would go 
off into limbo for minutes or more, rather than immediately returning to the 
plugins page showing that it is trying to load the plugin.

The web-pushing branch originally included code to use javascript and 
long-polling to auto-update a lot of the UI. This was never put back because of 
the bugs in the GWT web-pushing code, but for some elements it is useful, e.g. 
alerts, status pages etc.

 c. Light weight

Important. We don't want to have to add 20MB to the 10MB download just for the 
UI framework.

 d. Performance

Very important IMHO.
 
 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)

I guess I'd say B D C A for me?
 
 Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/
 B- gwt
 C- struts+extjs
 *D-  freenet devs ideas*

I'm afraid I don't have detailed knowledge of frameworks ...
 
 This frameworks have to;
 1. be active and old enough
 2. including web n.m fionctionnalities
 3. replace the Toad controller :-)
 4. using jetty
 5. be fully toolled for Eclipse
6. Ideally it should be possible to fall back to non-javascript, without 
actually having two complete UIs except where it is absolutely vital (IMHO 
maintaining two separate UIs would result in one of them being unmaintained in 
the long run and cause problems, as well as causing more work when we change 
something).
7. Ideally it should be possible to implement live updating using long-polling 
or similar, like the web-pushing branch, and without using one connection per 
window/tab. This is not needed immediately but it should be possible. It's not 
vital though, as we can use the existing GWT code if necessary.
 
 We hope to have freenet devs ideas on it before starting anything :-)
 
 Rgds
 
 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. März 2012, 13:40:56 schrieb Ian Clarke:
 Yes, the vast number of people that refuse to use anything but Lynx?  There
 is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
 so we don't have to worry about him.

I regularly browse freenet with a raw text interface. ssh to my box and lynx
or w3m-mode in emacs.

It’s much faster and much more convenient than starting a firefox on the 300
Mhz machine I use as laptop.

Best wishes,
Arne
--
1w6 sie zu achten,
sie alle zu finden,
in Spiele zu leiten
und sacht zu verbinden.
→ http://1w6.org



signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. März 2012, 15:29:45 schrieb Ian Clarke:
  So what is the experience of new users? Install freenet, see automaticly
  shipped wot+freetalk, try them, see their issues, maybe browse some
  sites and uninstall freenet again, since things either do rarely work or
  there is no content. A cool web-UI wont change those points, so from my
  perspective, a different interface may be nice, but it wont solve the
  bigger issues with freenet.

 So again, you think that just because a new UI won't solve every problem
 with Freenet, then we shouldn't do it?  That's completely illogical.

If it creates more problems than it solves, it is a problem.

Not being able to use freenet without Javascript would be a problem. Dillo has
no Javascript support. Neither have w3m, lynx, and a whole bunch of other low-
profile browsers. Mainly used by geeky people - who are a natural target
audience for freenet.

I don’t mind a good Javascript UI. But Sone was actually the first really
useful example for that which I ever saw. Yahoo is a really good counter-
example: they even reimplemented tabs in Javascript…

Javascript has to be complementary, though: The UI has to work without it. And
it can. After all, a reload of a local page is blazingly fast (when there are
no other bottlenecks).

Also there are many points on the UI side which could be improved that don’t
need JS. Better download-queue, integrated WoT and Sone (well, that benefits
from JS), and so on.

On the other hand, drag-and-drop uploading with image previews and all that is
quite a usability booster. So I am not against JS (anymore). I am just against
requiring it.

Best wishes,
Arne

--
Unpolitisch sein
heißt politisch sein,
ohne es zu merken.
- Arne (http://draketo.de)




signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Mittwoch, 7. März 2012, 22:14:39 schrieb Thomas Sachau:
 So what is the experience of new users? Install freenet, see automaticly
 shipped wot+freetalk, try them, see their issues, maybe browse some
 sites and uninstall freenet again, since things either do rarely work or
 there is no content.

I think it goes even deeper: The installer fails for many.
Also Freetalk is not integrated with the other parts of freenet. There should
be attachment options, where the post is only sent once the file finished
uploading. And lots of other “small” stuff on the UI side.

Essentially polish.

I think being really polished and getting rid of all the small rough edges
would be a much stronger usability boost than using new tech. (except if that
new tech does not add more rough corners and makes is much easier to fix
existing problems)

Best wishes,
Arne
--
1w6 sie zu achten,
sie alle zu finden,
in Spiele zu leiten
und sacht zu verbinden.
→ http://1w6.org



signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
 Am Mittwoch, 7. März 2012, 22:14:39 schrieb Thomas Sachau:
  So what is the experience of new users? Install freenet, see automaticly
  shipped wot+freetalk, try them, see their issues, maybe browse some
  sites and uninstall freenet again, since things either do rarely work or
  there is no content.
 
 I think it goes even deeper: The installer fails for many. 

Does it still? If so we need to determine why.

 Also Freetalk is not integrated with the other parts of freenet. There should 
 be attachment options, where the post is only sent once the file finished 
 uploading. And lots of other “small” stuff on the UI side.

Agreed, as does p0s, but first it has to WORK.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Arne Babenhauserheide]

Am Freitag, 9. März 2012, 23:14:04 schrieb Matthew Toseland:
 On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
  Am Mittwoch, 7. März 2012, 22:14:39 schrieb Thomas Sachau:
   So what is the experience of new users? Install freenet, see automaticly
   shipped wot+freetalk, try them, see their issues, maybe browse some
   sites and uninstall freenet again, since things either do rarely work or
   there is no content.
 
  I think it goes even deeper: The installer fails for many.

 Does it still? If so we need to determine why.

Last week someone in IRC had problems again.

And I think the installer has to be as simple and straightforward as possible.
What I saw in the chinese freenet installation youtube movie was much too
complex. 6 steps in the installer (or so) and then 4 configuration steps (do
we really need 2 steps with only 2 buttons on the site?).

And then there are firewall and port-forwarding issues.

  Also Freetalk is not integrated with the other parts of freenet. There
  should be attachment options, where the post is only sent once the file
  finished uploading. And lots of other “small” stuff on the UI side.

 Agreed, as does p0s, but first it has to WORK.

Jupp. Any plans about Sone?

Best wishes,
Arne

signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Steve Dougherty]

Should we get rid of more of the steps in the first-run setup? Asking
about datastore size is a must, but perhaps we can get away with
something like what I2P does where it assumes a very low bandwidth limit
and encourages users to raise it afterwards. This could work well
alongside requiring a 5KiB/s minimum. I2P works very nicely with
something that would work without Javascript, but works much better
with: it shows for a given speed setting how much it will likely use per
month. That a way better way to deal with monthly bandwidth limits than
asking directly.

-operhiem1

On 03/09/2012 07:50 PM, Arne Babenhauserheide wrote:
 Am Freitag, 9. März 2012, 23:14:04 schrieb Matthew Toseland:
 On Friday 09 Mar 2012 23:02:49 Arne Babenhauserheide wrote:
 Am Mittwoch, 7. März 2012, 22:14:39 schrieb Thomas Sachau:
 So what is the experience of new users? Install freenet, see automaticly
 shipped wot+freetalk, try them, see their issues, maybe browse some
 sites and uninstall freenet again, since things either do rarely work or
 there is no content.

 I think it goes even deeper: The installer fails for many.

 Does it still? If so we need to determine why.
 
 Last week someone in IRC had problems again.
 
 And I think the installer has to be as simple and straightforward as 
 possible. 
 What I saw in the chinese freenet installation youtube movie was much too 
 complex. 6 steps in the installer (or so) and then 4 configuration steps (do 
 we really need 2 steps with only 2 buttons on the site?).
 
 And then there are firewall and port-forwarding issues.
 
 Also Freetalk is not integrated with the other parts of freenet. There
 should be attachment options, where the post is only sent once the file
 finished uploading. And lots of other “small” stuff on the UI side.

 Agreed, as does p0s, but first it has to WORK.
 
 Jupp. Any plans about Sone?
 
 Best wishes,
 Arne
 
 
 
 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl



signature.asc
Description: OpenPGP digital signature
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Florent Daigniere]

On Wed, Mar 07, 2012 at 05:23:49PM +0100, Nicolas Hernandez wrote:
> Hello Ian,
> 
> We are still looking for the best choice between running in jetty
> a. Accessibility for devs (nice eclipse plugin for example)
> b. Accessibility for user
> c. Light weight
> d. Performance
> 
> In our side the priority are like that a>b>d>c (+eclipse plugin)
> Four some freenet devs it loks like c>d>a>b (+velocity for templating)
> 
> Not so easy to make the good choice. We have three mains ideas
> A- using Apache Wickets http://wicket.apache.org/

I am a big fan of Wicket. I might consider mentoring someone to work on it.

Florent

Re: [freenet-dev] webui and accessibility

[Florent Daigniere]

On Wed, Mar 07, 2012 at 05:23:49PM +0100, Nicolas Hernandez wrote:
 Hello Ian,
 
 We are still looking for the best choice between running in jetty
 a. Accessibility for devs (nice eclipse plugin for example)
 b. Accessibility for user
 c. Light weight
 d. Performance
 
 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)
 
 Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/

I am a big fan of Wicket. I might consider mentoring someone to work on it.

Florent
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

rendering:
it is important to differentiate the Redenderer and the Controller.
GWT can be used:
-  as a Controler
-  as Renderer

as Renderer gwt can offer a
- MinimalHtmlRenderer Interface for admin and simple actions
- Web2Renderer
- AndroidRender ?

In this case the interface templates could be in xml and i am not sure than
using velocity is a good idea

the beast:
if GWT is the BEAST, ... there is some patterns against this.

gentoo:
i love gentoo ... sorry about the gwt epic fail. Scripting if for us the
main goal. Packaging is really important for accesibility too, as lynx for
some admin stuffs.

So, the controler ? the application server ?

Its really a pleasure to be there :-)




- Nicolas Hernandez
a-n - aleph-networks
*associ?*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 9:09 PM, Ian Clarke  wrote:

> On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau  wrote:
>
>> One reason to not use pre-compiled stuff is, that you can patch or
>> adjust the code before you compile it. In addition, only those patches
>> have a chance to go upstream, probably noone will accept a patch against
>> the generated javascript code.
>>
>
> How likely is that in practice?  In the entire history of Freenet has any
> third party felt the need to patch it?  This seems like an extreme
> edge-case to me, and that it shouldn't dictate key architectural decisions.
>
>
>> Since it is Gentoo policy to give the user this choice and ability,
>> shipping the precompiled code is not really an option.
>>
>
> Then perhaps the real problem here is Gentoo's policy?  In any case, I
> don't think Gentoo's policy should tie our hands one way or another - our
> policy is that we need a decent user interface :-)
>
> Ian.
>
> --
> Ian Clarke
> Personal blog: http://blog.locut.us/
>
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Thomas Sachau]

Ian Clarke schrieb:
> On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau  wrote:
> 
>> One reason to not use pre-compiled stuff is, that you can patch or
>> adjust the code before you compile it. In addition, only those patches
>> have a chance to go upstream, probably noone will accept a patch against
>> the generated javascript code.
>>
> 
> How likely is that in practice?  In the entire history of Freenet has any
> third party felt the need to patch it?  This seems like an extreme
> edge-case to me, and that it shouldn't dictate key architectural decisions.
> 
> 
>> Since it is Gentoo policy to give the user this choice and ability,
>> shipping the precompiled code is not really an option.
>>
> 
> Then perhaps the real problem here is Gentoo's policy?  In any case, I
> don't think Gentoo's policy should tie our hands one way or another - our
> policy is that we need a decent user interface :-)

Do you really think, that having some nice javascript or otherwise
adjusted interface would magically raise the numbers of users?

Some eye-candy wont keep new users nor will it get you many new users,
it is way more important to have a reliable working freenet with some
nice additional apps (which themselves should also be reliable).
The only reliable working additional apps seem to be FMS and maybe Frost
(where the later has of course issues with spam). So due to the spam,
the only app left is in the end FMS. Now instead of supporting that,
something different is started and suggested (with wot and freetalk),
which still have many issues, are not reliable and where until now
nobody was able to even tell me, how wot itself works.

So what is the experience of new users? Install freenet, see automaticly
shipped wot+freetalk, try them, see their issues, maybe browse some
sites and uninstall freenet again, since things either do rarely work or
there is no content. A cool web-UI wont change those points, so from my
perspective, a different interface may be nice, but it wont solve the
bigger issues with freenet.



-- 

Thomas Sachau
Gentoo Linux Developer

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 380 bytes
Desc: OpenPGP digital signature
URL: 

[freenet-dev] webui and accessibility

[Thomas Sachau]

Ian Clarke schrieb:
> On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau  wrote:
> 
>> I dont contribute code as a freenet dev, but i do package it for Gentoo
>> linux and from that perspective, i have to strongly vote against gwt.
>> That thing is a big beast with many included external libs, often even
>> modified ones and a complex build system. I once tried to create a
>> package for it, it took me many hours and there was still no good
>> result. So unless this has greatly increased or someone else can provide
>> a sane package for Gentoo, requiring this framework to build freenet
>> would result in freenet being dropped from Gentoo as a package.
> 
> 
> I assume this would only be a problem if it was a requirement that the
> Java-JavaScript compilation occurs during the Gentoo build process, rather
> than just distributing the JavaScript already compiled - right?
> 
> Ian.

One reason to not use pre-compiled stuff is, that you can patch or
adjust the code before you compile it. In addition, only those patches
have a chance to go upstream, probably noone will accept a patch against
the generated javascript code.

Since it is Gentoo policy to give the user this choice and ability,
shipping the precompiled code is not really an option.

Matthew Toseland schrieb:
> Also, there is a package for GWT for Debian, I wonder if somebody has
> done one for Gentoo? A lot of stuff uses GWT ...

There is no package for GWT in the main tree of Gentoo. While it is
possible do create something for GWT, the real problem is to do it
properly with clean building from source and without included external libs.


> If it's an official package it needs to be built from scratch. And
> iirc right now it is an official package.
>

I dont know the details about the debian policy. "built from scratch"
can both mean "build the gwt code from scratch without the included
external libs" or "just build everything gwt ships from scratch". If it
is the first and if this is done for the debian package, then it might
be possible for Gentoo too. It would still mean many hours of work to
get it done.

-- 

Thomas Sachau
Gentoo Linux Developer

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 380 bytes
Desc: OpenPGP digital signature
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
> On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez <
> nicolas.hernandez at aleph-networks.com> wrote:
> 
> > In our side the priority are like that a>b>d>c (+eclipse plugin)
> > Four some freenet devs it loks like c>d>a>b (+velocity for templating)
> >
> 
> Unless those freenet devs are willing to build the GUI themselves, I
> recommend that you do not allow them to prevent you from proceeding.
>  Someone willing to make things happen should not be prevented from doing
> so by someone with an opinion, but who isn't willing to do the work.

Even if it means breaking existing code for a large minority of users? (The 
whole no-javascript lobby)?
> 
> Not so easy to make the good choice. We have three mains ideas
> > A- using Apache Wickets http://wicket.apache.org/
> 
> I've used this before (back in 2007), and it's creator, Jonathan Locke, is
> a friend of mine.  It does degrade nicely if there is no JavaScript, but
> does have a significant learning curve.
> 
> > B- gwt
> 
> I have only played with GWT, but I like the idea of it, I like the fact
> that Google is behind it, and my wife has used it on a project and she said
> that she'd be very happy to use it again.

We have some GWT code in Freenet for web-pushing (live updating of image 
loading etc). It's buggy, but that's probably not GWT's fault. I'm not wedded 
to it; it's turned off by default due to the bugs; but using GWT in its full 
capabilities would certainly prevent us from having a no-javascript option.
> 
> > C- struts+extjs
> 
> I know very little about this.
> 
> > We hope to have freenet devs ideas on it before starting anything :-)
> 
> Yes, but please don't let one stubborn person prevent you from making
> progress.  The opinion that really matters is the opinion of the person
> that is willing to do the work.  You should consider their opinions, but
> don't let them block progress.  All other opinions are secondary (including
> mine!), your opinion is what really matters if you are willing to do the
> actual work.

Unless it breaks the code for a large number of actively contributing users, 
and therefore costs us HUGELY. Which it will if there is no non-javascript 
support.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 19:17:42 Ian Clarke wrote:
> On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau  wrote:
> 
> > I dont contribute code as a freenet dev, but i do package it for Gentoo
> > linux and from that perspective, i have to strongly vote against gwt.
> > That thing is a big beast with many included external libs, often even
> > modified ones and a complex build system. I once tried to create a
> > package for it, it took me many hours and there was still no good
> > result. So unless this has greatly increased or someone else can provide
> > a sane package for Gentoo, requiring this framework to build freenet
> > would result in freenet being dropped from Gentoo as a package.
> 
> I assume this would only be a problem if it was a requirement that the
> Java-JavaScript compilation occurs during the Gentoo build process, rather
> than just distributing the JavaScript already compiled - right?

If it's an official package it needs to be built from scratch. And iirc right 
now it is an official package.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 17:55:57 Thomas Sachau wrote:
> Nicolas Hernandez schrieb:
> > Hello Ian,
> > 
> > We are still looking for the best choice between running in jetty
> > a. Accessibility for devs (nice eclipse plugin for example)
> > b. Accessibility for user
> > c. Light weight
> > d. Performance
> > 
> > In our side the priority are like that a>b>d>c (+eclipse plugin)
> > Four some freenet devs it loks like c>d>a>b (+velocity for templating)
> > 
> > Not so easy to make the good choice. We have three mains ideas
> > A- using Apache Wickets http://wicket.apache.org/
> > B- gwt
> 
> I dont contribute code as a freenet dev, but i do package it for Gentoo
> linux and from that perspective, i have to strongly vote against gwt.
> That thing is a big beast with many included external libs, often even
> modified ones and a complex build system. I once tried to create a
> package for it, it took me many hours and there was still no good
> result. So unless this has greatly increased or someone else can provide
> a sane package for Gentoo, requiring this framework to build freenet
> would result in freenet being dropped from Gentoo as a package.

Generally we don't WANT packages, because they get frozen in time for years. 
But gentoo is an exception.

Also, there is a package for GWT for Debian, I wonder if somebody has done one 
for Gentoo? A lot of stuff uses GWT ...

However I am skeptical about GWT because of the javascript vs non-javascript 
issue. Maybe that's less now but there was a very strong party on FMS etc for 
providing a non-JS option last time we discussed it.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Thomas Sachau]

Nicolas Hernandez schrieb:
> Hello Ian,
> 
> We are still looking for the best choice between running in jetty
> a. Accessibility for devs (nice eclipse plugin for example)
> b. Accessibility for user
> c. Light weight
> d. Performance
> 
> In our side the priority are like that a>b>d>c (+eclipse plugin)
> Four some freenet devs it loks like c>d>a>b (+velocity for templating)
> 
> Not so easy to make the good choice. We have three mains ideas
> A- using Apache Wickets http://wicket.apache.org/
> B- gwt

I dont contribute code as a freenet dev, but i do package it for Gentoo
linux and from that perspective, i have to strongly vote against gwt.
That thing is a big beast with many included external libs, often even
modified ones and a complex build system. I once tried to create a
package for it, it took me many hours and there was still no good
result. So unless this has greatly increased or someone else can provide
a sane package for Gentoo, requiring this framework to build freenet
would result in freenet being dropped from Gentoo as a package.

-- 

Thomas Sachau
Gentoo Linux Developer

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 380 bytes
Desc: OpenPGP digital signature
URL: 

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

Thanks a lot Ian,

We try to federate the maximum of devs ... but the dev start (the core
project) in 2 months. Just the time to take the decision and write the
firsts docs.  And we don't have time to wait a lot about that ... so this
project starts anyway :-) ... and hope for a finished soon project.

Feel free to expose your opinon (for the devs) !

Rgds
- Nicolas Hernandez
a-n - aleph-networks
*associ?*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 6:18 PM, Ian Clarke  wrote:

> On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez <
> nicolas.hernandez at aleph-networks.com> wrote:
>
>> In our side the priority are like that a>b>d>c (+eclipse plugin)
>> Four some freenet devs it loks like c>d>a>b (+velocity for templating)
>>
>
> Unless those freenet devs are willing to build the GUI themselves, I
> recommend that you do not allow them to prevent you from proceeding.
>  Someone willing to make things happen should not be prevented from doing
> so by someone with an opinion, but who isn't willing to do the work.
>
> Not so easy to make the good choice. We have three mains ideas
>> A- using Apache Wickets http://wicket.apache.org/
>>
>
> I've used this before (back in 2007), and it's creator, Jonathan Locke, is
> a friend of mine.  It does degrade nicely if there is no JavaScript, but
> does have a significant learning curve.
>
>
>> B- gwt
>>
>
> I have only played with GWT, but I like the idea of it, I like the fact
> that Google is behind it, and my wife has used it on a project and she said
> that she'd be very happy to use it again.
>
>
>> C- struts+extjs
>>
>
> I know very little about this.
>
>
>> We hope to have freenet devs ideas on it before starting anything :-)
>>
>
> Yes, but please don't let one stubborn person prevent you from making
> progress.  The opinion that really matters is the opinion of the person
> that is willing to do the work.  You should consider their opinions, but
> don't let them block progress.  All other opinions are secondary (including
> mine!), your opinion is what really matters if you are willing to do the
> actual work.
>
> Ian.
>
> --
> Ian Clarke
> Personal blog: http://blog.locut.us/
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Zlatin Balevsky]

Just want to add JRuby to the list.  Rails has lots of fans and third-party
tools and it's under active development and supposedly on java 1.7 it runs
faster than the C implementation.

On Wed, Mar 7, 2012 at 11:23 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> Hello Ian,
>
> We are still looking for the best choice between running in jetty
> a. Accessibility for devs (nice eclipse plugin for example)
> b. Accessibility for user
> c. Light weight
> d. Performance
>
> In our side the priority are like that a>b>d>c (+eclipse plugin)
> Four some freenet devs it loks like c>d>a>b (+velocity for templating)
>
> Not so easy to make the good choice. We have three mains ideas
> A- using Apache Wickets http://wicket.apache.org/
> B- gwt
> C- struts+extjs
> *D-  freenet devs ideas*
>
> This frameworks have to;
> 1. be active and old enough
> 2. including web n.m fionctionnalities
> 3. replace the Toad controller :-)
> 4. using jetty
> 5. be fully toolled for Eclipse
>
> We hope to have freenet devs ideas on it before starting anything :-)
>
> Rgds
>
> - Nicolas Hernandez
> a-n - aleph-networks
> *associ?*
> http://www.aleph-networks.com
>
>
>
>
> On Wed, Mar 7, 2012 at 3:33 PM, Ian Clarke  wrote:
>
>> Hi Nicolas,
>>
>> Did you ever make any progress on this?
>>
>> Ian.
>>
>> On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez <
>> nicolas.hernandez at aleph-networks.com> wrote:
>>
>>> Hello,
>>>
>>> This is a sort of personnal introduction:
>>>
>>> We want to work on freenet accessibility. The first step could be the
>>> webui. In the case of a total rework of the webui. Wich framework could be
>>> the 'good' one in the freenet philosophy ? Wich software architecture ?
>>>
>>> Is GWT and his model a good idea ? Could it be possible to have help
>>> from GSoc 2012 - (i can be the man with the Umbrella) ?
>>>
>>> ... lots of questions ...
>>>
>>> Rgds
>>>
>>> - Nicolas Hernandez
>>> a-n - aleph-networks
>>> *associ?*
>>> http://www.aleph-networks.com
>>>
>>>
>>>
>>> ___
>>> Devl mailing list
>>> Devl at freenetproject.org
>>> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>>>
>>
>>
>>
>> --
>> Ian Clarke
>> Personal blog: http://blog.locut.us/
>>
>
>
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hello Ian,

We are still looking for the best choice between running in jetty
a. Accessibility for devs (nice eclipse plugin for example)
b. Accessibility for user
c. Light weight
d. Performance

In our side the priority are like that a>b>d>c (+eclipse plugin)
Four some freenet devs it loks like c>d>a>b (+velocity for templating)

Not so easy to make the good choice. We have three mains ideas
A- using Apache Wickets http://wicket.apache.org/
B- gwt
C- struts+extjs
*D-  freenet devs ideas*

This frameworks have to;
1. be active and old enough
2. including web n.m fionctionnalities
3. replace the Toad controller :-)
4. using jetty
5. be fully toolled for Eclipse

We hope to have freenet devs ideas on it before starting anything :-)

Rgds

- Nicolas Hernandez
a-n - aleph-networks
*associ?*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 3:33 PM, Ian Clarke  wrote:

> Hi Nicolas,
>
> Did you ever make any progress on this?
>
> Ian.
>
> On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez <
> nicolas.hernandez at aleph-networks.com> wrote:
>
>> Hello,
>>
>> This is a sort of personnal introduction:
>>
>> We want to work on freenet accessibility. The first step could be the
>> webui. In the case of a total rework of the webui. Wich framework could be
>> the 'good' one in the freenet philosophy ? Wich software architecture ?
>>
>> Is GWT and his model a good idea ? Could it be possible to have help from
>> GSoc 2012 - (i can be the man with the Umbrella) ?
>>
>> ... lots of questions ...
>>
>> Rgds
>>
>> - Nicolas Hernandez
>> a-n - aleph-networks
>> *associ?*
>> http://www.aleph-networks.com
>>
>>
>>
>> ___
>> Devl mailing list
>> Devl at freenetproject.org
>> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>>
>
>
>
> --
> Ian Clarke
> Personal blog: http://blog.locut.us/
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau  wrote:

> Ian Clarke schrieb:
> Do you really think, that having some nice javascript or otherwise
> adjusted interface would magically raise the numbers of users?
>

I think having a nice, well defined user interface will raise the number of
users, yes.  Javascript is just a means to this end, since it's well suited
to rich web UIs, which Freenet requires.


> Some eye-candy


I'm not talking about "eye-candy", I'm talking about good user interface
design.


> wont keep new users nor will it get you many new users,
> it is way more important to have a reliable working freenet with some
> nice additional apps (which themselves should also be reliable).
> The only reliable working additional apps seem to be FMS and maybe Frost
> (where the later has of course issues with spam). So due to the spam,
> the only app left is in the end FMS. Now instead of supporting that,
> something different is started and suggested (with wot and freetalk),
> which still have many issues, are not reliable and where until now
> nobody was able to even tell me, how wot itself works.
>

So what is your point?  The fact that Freetalk isn't perfect means that if
someone is willing to work on a different aspect of Freenet we should tell
them to go away?


> So what is the experience of new users? Install freenet, see automaticly
> shipped wot+freetalk, try them, see their issues, maybe browse some
> sites and uninstall freenet again, since things either do rarely work or
> there is no content. A cool web-UI wont change those points, so from my
> perspective, a different interface may be nice, but it wont solve the
> bigger issues with freenet.
>

So again, you think that just because a new UI won't solve *every* problem
with Freenet, then we shouldn't do it?  That's completely illogical.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau  wrote:

> One reason to not use pre-compiled stuff is, that you can patch or
> adjust the code before you compile it. In addition, only those patches
> have a chance to go upstream, probably noone will accept a patch against
> the generated javascript code.
>

How likely is that in practice?  In the entire history of Freenet has any
third party felt the need to patch it?  This seems like an extreme
edge-case to me, and that it shouldn't dictate key architectural decisions.


> Since it is Gentoo policy to give the user this choice and ability,
> shipping the precompiled code is not really an option.
>

Then perhaps the real problem here is Gentoo's policy?  In any case, I
don't think Gentoo's policy should tie our hands one way or another - our
policy is that we need a decent user interface :-)

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Tom Elovi Spruce]

User experience and user interfaces go hand and hand. I believe Thomas is 
addressing issues he sees with the user experience of Freenet. 

If you're updating the interface to improve the user experience, it's a 
worthwhile investment. However, mainly putting effort into updating the visuals 
of an interface instead of improved parts of the experience of the software, 
isn't worthwhile. You can still have a 90s look, but if the experience of using 
the software is pleasant, it can go a longs way. 

---
i?computers

On Mar 7, 2012, at 1:29 PM, Ian Clarke wrote:

> On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau  wrote:
> Ian Clarke schrieb:
> Do you really think, that having some nice javascript or otherwise
> adjusted interface would magically raise the numbers of users?
> 
> I think having a nice, well defined user interface will raise the number of 
> users, yes.  Javascript is just a means to this end, since it's well suited 
> to rich web UIs, which Freenet requires.
>  
> Some eye-candy
> 
> I'm not talking about "eye-candy", I'm talking about good user interface 
> design.
>  
> wont keep new users nor will it get you many new users,
> it is way more important to have a reliable working freenet with some
> nice additional apps (which themselves should also be reliable).
> The only reliable working additional apps seem to be FMS and maybe Frost
> (where the later has of course issues with spam). So due to the spam,
> the only app left is in the end FMS. Now instead of supporting that,
> something different is started and suggested (with wot and freetalk),
> which still have many issues, are not reliable and where until now
> nobody was able to even tell me, how wot itself works.
> 
> So what is your point?  The fact that Freetalk isn't perfect means that if 
> someone is willing to work on a different aspect of Freenet we should tell 
> them to go away?
>  
> So what is the experience of new users? Install freenet, see automaticly
> shipped wot+freetalk, try them, see their issues, maybe browse some
> sites and uninstall freenet again, since things either do rarely work or
> there is no content. A cool web-UI wont change those points, so from my
> perspective, a different interface may be nice, but it wont solve the
> bigger issues with freenet.
> 
> So again, you think that just because a new UI won't solve every problem with 
> Freenet, then we shouldn't do it?  That's completely illogical.
> 
> Ian.
> 
> -- 
> Ian Clarke
> Personal blog: http://blog.locut.us/
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl

-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 1:21 PM, Matthew Toseland
wrote:

> On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
> > Unless those freenet devs are willing to build the GUI themselves, I
> > recommend that you do not allow them to prevent you from proceeding.
> >  Someone willing to make things happen should not be prevented from doing
> > so by someone with an opinion, but who isn't willing to do the work.
>
> Even if it means breaking existing code for a large minority of users?
> (The whole no-javascript lobby)?


Yes, the vast number of people that refuse to use anything but Lynx?  There
is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
so we don't have to worry about him.

But seriously, it would be insanity to hold up development of a decent user
interface just because of the griping of a few people who, for no logical
reason, refuse to enable JavaScript in their browsers.

If they care so much about it nobody is preventing them from creating their
own non-JavaScript UI, but they shouldn't be able to hold up progress for
the vast majority of people that don't have an irrational fear of
JavaScript.


> > Yes, but please don't let one stubborn person prevent you from making
> > progress.  The opinion that really matters is the opinion of the person
> > that is willing to do the work.  You should consider their opinions, but
> > don't let them block progress.  All other opinions are secondary
> (including
> > mine!), your opinion is what really matters if you are willing to do the
> > actual work.
>
> Unless it breaks the code for a large number of actively contributing
> users, and therefore costs us HUGELY. Which it will if there is no
> non-javascript support.
>

If this is such a concern to them then they should be willing to implement
their own non-JavaScript UI.  They should not be permitted to stand in the
way of people who are actually willing to do real work to improve the UI.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau  wrote:

> I dont contribute code as a freenet dev, but i do package it for Gentoo
> linux and from that perspective, i have to strongly vote against gwt.
> That thing is a big beast with many included external libs, often even
> modified ones and a complex build system. I once tried to create a
> package for it, it took me many hours and there was still no good
> result. So unless this has greatly increased or someone else can provide
> a sane package for Gentoo, requiring this framework to build freenet
> would result in freenet being dropped from Gentoo as a package.


I assume this would only be a problem if it was a requirement that the
Java-JavaScript compilation occurs during the Gentoo build process, rather
than just distributing the JavaScript already compiled - right?

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> In our side the priority are like that a>b>d>c (+eclipse plugin)
> Four some freenet devs it loks like c>d>a>b (+velocity for templating)
>

Unless those freenet devs are willing to build the GUI themselves, I
recommend that you do not allow them to prevent you from proceeding.
 Someone willing to make things happen should not be prevented from doing
so by someone with an opinion, but who isn't willing to do the work.

Not so easy to make the good choice. We have three mains ideas
> A- using Apache Wickets http://wicket.apache.org/
>

I've used this before (back in 2007), and it's creator, Jonathan Locke, is
a friend of mine.  It does degrade nicely if there is no JavaScript, but
does have a significant learning curve.


> B- gwt
>

I have only played with GWT, but I like the idea of it, I like the fact
that Google is behind it, and my wife has used it on a project and she said
that she'd be very happy to use it again.


> C- struts+extjs
>

I know very little about this.


> We hope to have freenet devs ideas on it before starting anything :-)
>

Yes, but please don't let one stubborn person prevent you from making
progress.  The opinion that really matters is the opinion of the person
that is willing to do the work.  You should consider their opinions, but
don't let them block progress.  All other opinions are secondary (including
mine!), your opinion is what really matters if you are willing to do the
actual work.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

Hi Nicolas,

Did you ever make any progress on this?

Ian.

On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> Hello,
>
> This is a sort of personnal introduction:
>
> We want to work on freenet accessibility. The first step could be the
> webui. In the case of a total rework of the webui. Wich framework could be
> the 'good' one in the freenet philosophy ? Wich software architecture ?
>
> Is GWT and his model a good idea ? Could it be possible to have help from
> GSoc 2012 - (i can be the man with the Umbrella) ?
>
> ... lots of questions ...
>
> Rgds
>
> - Nicolas Hernandez
> a-n - aleph-networks
> *associ?*
> http://www.aleph-networks.com
>
>
>
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>



-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

Hi Nicolas,

Did you ever make any progress on this?

Ian.

On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 Hello,

 This is a sort of personnal introduction:

 We want to work on freenet accessibility. The first step could be the
 webui. In the case of a total rework of the webui. Wich framework could be
 the 'good' one in the freenet philosophy ? Wich software architecture ?

 Is GWT and his model a good idea ? Could it be possible to have help from
 GSoc 2012 - (i can be the man with the Umbrella) ?

 ... lots of questions ...

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl




-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hello Ian,

We are still looking for the best choice between running in jetty
a. Accessibility for devs (nice eclipse plugin for example)
b. Accessibility for user
c. Light weight
d. Performance

In our side the priority are like that abdc (+eclipse plugin)
Four some freenet devs it loks like cdab (+velocity for templating)

Not so easy to make the good choice. We have three mains ideas
A- using Apache Wickets http://wicket.apache.org/
B- gwt
C- struts+extjs
*D-  freenet devs ideas*

This frameworks have to;
1. be active and old enough
2. including web n.m fionctionnalities
3. replace the Toad controller :-)
4. using jetty
5. be fully toolled for Eclipse

We hope to have freenet devs ideas on it before starting anything :-)

Rgds

- Nicolas Hernandez
a-n - aleph-networks
*associé*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 3:33 PM, Ian Clarke i...@locut.us wrote:

 Hi Nicolas,

 Did you ever make any progress on this?

 Ian.

 On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez 
 nicolas.hernan...@aleph-networks.com wrote:

 Hello,

 This is a sort of personnal introduction:

 We want to work on freenet accessibility. The first step could be the
 webui. In the case of a total rework of the webui. Wich framework could be
 the 'good' one in the freenet philosophy ? Wich software architecture ?

 Is GWT and his model a good idea ? Could it be possible to have help from
 GSoc 2012 - (i can be the man with the Umbrella) ?

 ... lots of questions ...

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl




 --
 Ian Clarke
 Personal blog: http://blog.locut.us/

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)


Unless those freenet devs are willing to build the GUI themselves, I
recommend that you do not allow them to prevent you from proceeding.
 Someone willing to make things happen should not be prevented from doing
so by someone with an opinion, but who isn't willing to do the work.

Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/


I've used this before (back in 2007), and it's creator, Jonathan Locke, is
a friend of mine.  It does degrade nicely if there is no JavaScript, but
does have a significant learning curve.


 B- gwt


I have only played with GWT, but I like the idea of it, I like the fact
that Google is behind it, and my wife has used it on a project and she said
that she'd be very happy to use it again.


 C- struts+extjs


I know very little about this.


 We hope to have freenet devs ideas on it before starting anything :-)


Yes, but please don't let one stubborn person prevent you from making
progress.  The opinion that really matters is the opinion of the person
that is willing to do the work.  You should consider their opinions, but
don't let them block progress.  All other opinions are secondary (including
mine!), your opinion is what really matters if you are willing to do the
actual work.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

Thanks a lot Ian,

We try to federate the maximum of devs ... but the dev start (the core
project) in 2 months. Just the time to take the decision and write the
firsts docs.  And we don't have time to wait a lot about that ... so this
project starts anyway :-) ... and hope for a finished soon project.

Feel free to expose your opinon (for the devs) !

Rgds
- Nicolas Hernandez
a-n - aleph-networks
*associé*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 6:18 PM, Ian Clarke i...@locut.us wrote:

 On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez 
 nicolas.hernan...@aleph-networks.com wrote:

 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)


 Unless those freenet devs are willing to build the GUI themselves, I
 recommend that you do not allow them to prevent you from proceeding.
  Someone willing to make things happen should not be prevented from doing
 so by someone with an opinion, but who isn't willing to do the work.

 Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/


 I've used this before (back in 2007), and it's creator, Jonathan Locke, is
 a friend of mine.  It does degrade nicely if there is no JavaScript, but
 does have a significant learning curve.


 B- gwt


 I have only played with GWT, but I like the idea of it, I like the fact
 that Google is behind it, and my wife has used it on a project and she said
 that she'd be very happy to use it again.


 C- struts+extjs


 I know very little about this.


 We hope to have freenet devs ideas on it before starting anything :-)


 Yes, but please don't let one stubborn person prevent you from making
 progress.  The opinion that really matters is the opinion of the person
 that is willing to do the work.  You should consider their opinions, but
 don't let them block progress.  All other opinions are secondary (including
 mine!), your opinion is what really matters if you are willing to do the
 actual work.

 Ian.

 --
 Ian Clarke
 Personal blog: http://blog.locut.us/

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Thomas Sachau]

Nicolas Hernandez schrieb:
 Hello Ian,
 
 We are still looking for the best choice between running in jetty
 a. Accessibility for devs (nice eclipse plugin for example)
 b. Accessibility for user
 c. Light weight
 d. Performance
 
 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)
 
 Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/
 B- gwt

I dont contribute code as a freenet dev, but i do package it for Gentoo
linux and from that perspective, i have to strongly vote against gwt.
That thing is a big beast with many included external libs, often even
modified ones and a complex build system. I once tried to create a
package for it, it took me many hours and there was still no good
result. So unless this has greatly increased or someone else can provide
a sane package for Gentoo, requiring this framework to build freenet
would result in freenet being dropped from Gentoo as a package.

-- 

Thomas Sachau
Gentoo Linux Developer



signature.asc
Description: OpenPGP digital signature
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau to...@gentoo.org wrote:

 I dont contribute code as a freenet dev, but i do package it for Gentoo
 linux and from that perspective, i have to strongly vote against gwt.
 That thing is a big beast with many included external libs, often even
 modified ones and a complex build system. I once tried to create a
 package for it, it took me many hours and there was still no good
 result. So unless this has greatly increased or someone else can provide
 a sane package for Gentoo, requiring this framework to build freenet
 would result in freenet being dropped from Gentoo as a package.


I assume this would only be a problem if it was a requirement that the
Java-JavaScript compilation occurs during the Gentoo build process, rather
than just distributing the JavaScript already compiled - right?

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 17:55:57 Thomas Sachau wrote:
 Nicolas Hernandez schrieb:
  Hello Ian,
  
  We are still looking for the best choice between running in jetty
  a. Accessibility for devs (nice eclipse plugin for example)
  b. Accessibility for user
  c. Light weight
  d. Performance
  
  In our side the priority are like that abdc (+eclipse plugin)
  Four some freenet devs it loks like cdab (+velocity for templating)
  
  Not so easy to make the good choice. We have three mains ideas
  A- using Apache Wickets http://wicket.apache.org/
  B- gwt
 
 I dont contribute code as a freenet dev, but i do package it for Gentoo
 linux and from that perspective, i have to strongly vote against gwt.
 That thing is a big beast with many included external libs, often even
 modified ones and a complex build system. I once tried to create a
 package for it, it took me many hours and there was still no good
 result. So unless this has greatly increased or someone else can provide
 a sane package for Gentoo, requiring this framework to build freenet
 would result in freenet being dropped from Gentoo as a package.

Generally we don't WANT packages, because they get frozen in time for years. 
But gentoo is an exception.

Also, there is a package for GWT for Debian, I wonder if somebody has done one 
for Gentoo? A lot of stuff uses GWT ...

However I am skeptical about GWT because of the javascript vs non-javascript 
issue. Maybe that's less now but there was a very strong party on FMS etc for 
providing a non-JS option last time we discussed it.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 19:17:42 Ian Clarke wrote:
 On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau to...@gentoo.org wrote:
 
  I dont contribute code as a freenet dev, but i do package it for Gentoo
  linux and from that perspective, i have to strongly vote against gwt.
  That thing is a big beast with many included external libs, often even
  modified ones and a complex build system. I once tried to create a
  package for it, it took me many hours and there was still no good
  result. So unless this has greatly increased or someone else can provide
  a sane package for Gentoo, requiring this framework to build freenet
  would result in freenet being dropped from Gentoo as a package.
 
 I assume this would only be a problem if it was a requirement that the
 Java-JavaScript compilation occurs during the Gentoo build process, rather
 than just distributing the JavaScript already compiled - right?

If it's an official package it needs to be built from scratch. And iirc right 
now it is an official package.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
 On Wed, Mar 7, 2012 at 10:23 AM, Nicolas Hernandez 
 nicolas.hernan...@aleph-networks.com wrote:
 
  In our side the priority are like that abdc (+eclipse plugin)
  Four some freenet devs it loks like cdab (+velocity for templating)
 
 
 Unless those freenet devs are willing to build the GUI themselves, I
 recommend that you do not allow them to prevent you from proceeding.
  Someone willing to make things happen should not be prevented from doing
 so by someone with an opinion, but who isn't willing to do the work.

Even if it means breaking existing code for a large minority of users? (The 
whole no-javascript lobby)?
 
 Not so easy to make the good choice. We have three mains ideas
  A- using Apache Wickets http://wicket.apache.org/
 
 I've used this before (back in 2007), and it's creator, Jonathan Locke, is
 a friend of mine.  It does degrade nicely if there is no JavaScript, but
 does have a significant learning curve.
 
  B- gwt
 
 I have only played with GWT, but I like the idea of it, I like the fact
 that Google is behind it, and my wife has used it on a project and she said
 that she'd be very happy to use it again.

We have some GWT code in Freenet for web-pushing (live updating of image 
loading etc). It's buggy, but that's probably not GWT's fault. I'm not wedded 
to it; it's turned off by default due to the bugs; but using GWT in its full 
capabilities would certainly prevent us from having a no-javascript option.
 
  C- struts+extjs
 
 I know very little about this.
 
  We hope to have freenet devs ideas on it before starting anything :-)
 
 Yes, but please don't let one stubborn person prevent you from making
 progress.  The opinion that really matters is the opinion of the person
 that is willing to do the work.  You should consider their opinions, but
 don't let them block progress.  All other opinions are secondary (including
 mine!), your opinion is what really matters if you are willing to do the
 actual work.

Unless it breaks the code for a large number of actively contributing users, 
and therefore costs us HUGELY. Which it will if there is no non-javascript 
support.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 1:21 PM, Matthew Toseland
t...@amphibian.dyndns.orgwrote:

 On Wednesday 07 Mar 2012 17:18:54 Ian Clarke wrote:
  Unless those freenet devs are willing to build the GUI themselves, I
  recommend that you do not allow them to prevent you from proceeding.
   Someone willing to make things happen should not be prevented from doing
  so by someone with an opinion, but who isn't willing to do the work.

 Even if it means breaking existing code for a large minority of users?
 (The whole no-javascript lobby)?


Yes, the vast number of people that refuse to use anything but Lynx?  There
is RMS, who are the other ones?  So far as I know RMS doesn't use Freenet
so we don't have to worry about him.

But seriously, it would be insanity to hold up development of a decent user
interface just because of the griping of a few people who, for no logical
reason, refuse to enable JavaScript in their browsers.

If they care so much about it nobody is preventing them from creating their
own non-JavaScript UI, but they shouldn't be able to hold up progress for
the vast majority of people that don't have an irrational fear of
JavaScript.


  Yes, but please don't let one stubborn person prevent you from making
  progress.  The opinion that really matters is the opinion of the person
  that is willing to do the work.  You should consider their opinions, but
  don't let them block progress.  All other opinions are secondary
 (including
  mine!), your opinion is what really matters if you are willing to do the
  actual work.

 Unless it breaks the code for a large number of actively contributing
 users, and therefore costs us HUGELY. Which it will if there is no
 non-javascript support.


If this is such a concern to them then they should be willing to implement
their own non-JavaScript UI.  They should not be permitted to stand in the
way of people who are actually willing to do real work to improve the UI.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Thomas Sachau]

Ian Clarke schrieb:
 On Wed, Mar 7, 2012 at 11:55 AM, Thomas Sachau to...@gentoo.org wrote:
 
 I dont contribute code as a freenet dev, but i do package it for Gentoo
 linux and from that perspective, i have to strongly vote against gwt.
 That thing is a big beast with many included external libs, often even
 modified ones and a complex build system. I once tried to create a
 package for it, it took me many hours and there was still no good
 result. So unless this has greatly increased or someone else can provide
 a sane package for Gentoo, requiring this framework to build freenet
 would result in freenet being dropped from Gentoo as a package.
 
 
 I assume this would only be a problem if it was a requirement that the
 Java-JavaScript compilation occurs during the Gentoo build process, rather
 than just distributing the JavaScript already compiled - right?
 
 Ian.

One reason to not use pre-compiled stuff is, that you can patch or
adjust the code before you compile it. In addition, only those patches
have a chance to go upstream, probably noone will accept a patch against
the generated javascript code.

Since it is Gentoo policy to give the user this choice and ability,
shipping the precompiled code is not really an option.

Matthew Toseland schrieb:
 Also, there is a package for GWT for Debian, I wonder if somebody has
 done one for Gentoo? A lot of stuff uses GWT ...

There is no package for GWT in the main tree of Gentoo. While it is
possible do create something for GWT, the real problem is to do it
properly with clean building from source and without included external libs.


 If it's an official package it needs to be built from scratch. And
 iirc right now it is an official package.


I dont know the details about the debian policy. built from scratch
can both mean build the gwt code from scratch without the included
external libs or just build everything gwt ships from scratch. If it
is the first and if this is done for the debian package, then it might
be possible for Gentoo too. It would still mean many hours of work to
get it done.

-- 

Thomas Sachau
Gentoo Linux Developer



signature.asc
Description: OpenPGP digital signature
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau to...@gentoo.org wrote:

 One reason to not use pre-compiled stuff is, that you can patch or
 adjust the code before you compile it. In addition, only those patches
 have a chance to go upstream, probably noone will accept a patch against
 the generated javascript code.


How likely is that in practice?  In the entire history of Freenet has any
third party felt the need to patch it?  This seems like an extreme
edge-case to me, and that it shouldn't dictate key architectural decisions.


 Since it is Gentoo policy to give the user this choice and ability,
 shipping the precompiled code is not really an option.


Then perhaps the real problem here is Gentoo's policy?  In any case, I
don't think Gentoo's policy should tie our hands one way or another - our
policy is that we need a decent user interface :-)

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Thomas Sachau]

Ian Clarke schrieb:
 On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau to...@gentoo.org wrote:
 
 One reason to not use pre-compiled stuff is, that you can patch or
 adjust the code before you compile it. In addition, only those patches
 have a chance to go upstream, probably noone will accept a patch against
 the generated javascript code.

 
 How likely is that in practice?  In the entire history of Freenet has any
 third party felt the need to patch it?  This seems like an extreme
 edge-case to me, and that it shouldn't dictate key architectural decisions.
 
 
 Since it is Gentoo policy to give the user this choice and ability,
 shipping the precompiled code is not really an option.

 
 Then perhaps the real problem here is Gentoo's policy?  In any case, I
 don't think Gentoo's policy should tie our hands one way or another - our
 policy is that we need a decent user interface :-)

Do you really think, that having some nice javascript or otherwise
adjusted interface would magically raise the numbers of users?

Some eye-candy wont keep new users nor will it get you many new users,
it is way more important to have a reliable working freenet with some
nice additional apps (which themselves should also be reliable).
The only reliable working additional apps seem to be FMS and maybe Frost
(where the later has of course issues with spam). So due to the spam,
the only app left is in the end FMS. Now instead of supporting that,
something different is started and suggested (with wot and freetalk),
which still have many issues, are not reliable and where until now
nobody was able to even tell me, how wot itself works.

So what is the experience of new users? Install freenet, see automaticly
shipped wot+freetalk, try them, see their issues, maybe browse some
sites and uninstall freenet again, since things either do rarely work or
there is no content. A cool web-UI wont change those points, so from my
perspective, a different interface may be nice, but it wont solve the
bigger issues with freenet.



-- 

Thomas Sachau
Gentoo Linux Developer



signature.asc
Description: OpenPGP digital signature
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Nicolas Hernandez]

rendering:
it is important to differentiate the Redenderer and the Controller.
GWT can be used:
-  as a Controler
-  as Renderer

as Renderer gwt can offer a
- MinimalHtmlRenderer Interface for admin and simple actions
- Web2Renderer
- AndroidRender ?

In this case the interface templates could be in xml and i am not sure than
using velocity is a good idea

the beast:
if GWT is the BEAST, ... there is some patterns against this.

gentoo:
i love gentoo ... sorry about the gwt epic fail. Scripting if for us the
main goal. Packaging is really important for accesibility too, as lynx for
some admin stuffs.

So, the controler ? the application server ?

Its really a pleasure to be there :-)




- Nicolas Hernandez
a-n - aleph-networks
*associé*
http://www.aleph-networks.com




On Wed, Mar 7, 2012 at 9:09 PM, Ian Clarke i...@locut.us wrote:

 On Wed, Mar 7, 2012 at 1:52 PM, Thomas Sachau to...@gentoo.org wrote:

 One reason to not use pre-compiled stuff is, that you can patch or
 adjust the code before you compile it. In addition, only those patches
 have a chance to go upstream, probably noone will accept a patch against
 the generated javascript code.


 How likely is that in practice?  In the entire history of Freenet has any
 third party felt the need to patch it?  This seems like an extreme
 edge-case to me, and that it shouldn't dictate key architectural decisions.


 Since it is Gentoo policy to give the user this choice and ability,
 shipping the precompiled code is not really an option.


 Then perhaps the real problem here is Gentoo's policy?  In any case, I
 don't think Gentoo's policy should tie our hands one way or another - our
 policy is that we need a decent user interface :-)

 Ian.

 --
 Ian Clarke
 Personal blog: http://blog.locut.us/

 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau to...@gentoo.org wrote:

 Ian Clarke schrieb:
 Do you really think, that having some nice javascript or otherwise
 adjusted interface would magically raise the numbers of users?


I think having a nice, well defined user interface will raise the number of
users, yes.  Javascript is just a means to this end, since it's well suited
to rich web UIs, which Freenet requires.


 Some eye-candy


I'm not talking about eye-candy, I'm talking about good user interface
design.


 wont keep new users nor will it get you many new users,
 it is way more important to have a reliable working freenet with some
 nice additional apps (which themselves should also be reliable).
 The only reliable working additional apps seem to be FMS and maybe Frost
 (where the later has of course issues with spam). So due to the spam,
 the only app left is in the end FMS. Now instead of supporting that,
 something different is started and suggested (with wot and freetalk),
 which still have many issues, are not reliable and where until now
 nobody was able to even tell me, how wot itself works.


So what is your point?  The fact that Freetalk isn't perfect means that if
someone is willing to work on a different aspect of Freenet we should tell
them to go away?


 So what is the experience of new users? Install freenet, see automaticly
 shipped wot+freetalk, try them, see their issues, maybe browse some
 sites and uninstall freenet again, since things either do rarely work or
 there is no content. A cool web-UI wont change those points, so from my
 perspective, a different interface may be nice, but it wont solve the
 bigger issues with freenet.


So again, you think that just because a new UI won't solve *every* problem
with Freenet, then we shouldn't do it?  That's completely illogical.

Ian.

-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Tom Elovi Spruce]

User experience and user interfaces go hand and hand. I believe Thomas is 
addressing issues he sees with the user experience of Freenet. 

If you're updating the interface to improve the user experience, it's a 
worthwhile investment. However, mainly putting effort into updating the visuals 
of an interface instead of improved parts of the experience of the software, 
isn't worthwhile. You can still have a 90s look, but if the experience of using 
the software is pleasant, it can go a longs way. 

---
i❤computers

On Mar 7, 2012, at 1:29 PM, Ian Clarke wrote:

 On Wed, Mar 7, 2012 at 3:14 PM, Thomas Sachau to...@gentoo.org wrote:
 Ian Clarke schrieb:
 Do you really think, that having some nice javascript or otherwise
 adjusted interface would magically raise the numbers of users?
 
 I think having a nice, well defined user interface will raise the number of 
 users, yes.  Javascript is just a means to this end, since it's well suited 
 to rich web UIs, which Freenet requires.
  
 Some eye-candy
 
 I'm not talking about eye-candy, I'm talking about good user interface 
 design.
  
 wont keep new users nor will it get you many new users,
 it is way more important to have a reliable working freenet with some
 nice additional apps (which themselves should also be reliable).
 The only reliable working additional apps seem to be FMS and maybe Frost
 (where the later has of course issues with spam). So due to the spam,
 the only app left is in the end FMS. Now instead of supporting that,
 something different is started and suggested (with wot and freetalk),
 which still have many issues, are not reliable and where until now
 nobody was able to even tell me, how wot itself works.
 
 So what is your point?  The fact that Freetalk isn't perfect means that if 
 someone is willing to work on a different aspect of Freenet we should tell 
 them to go away?
  
 So what is the experience of new users? Install freenet, see automaticly
 shipped wot+freetalk, try them, see their issues, maybe browse some
 sites and uninstall freenet again, since things either do rarely work or
 there is no content. A cool web-UI wont change those points, so from my
 perspective, a different interface may be nice, but it wont solve the
 bigger issues with freenet.
 
 So again, you think that just because a new UI won't solve every problem with 
 Freenet, then we shouldn't do it?  That's completely illogical.
 
 Ian.
 
 -- 
 Ian Clarke
 Personal blog: http://blog.locut.us/
 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Zlatin Balevsky]

Just want to add JRuby to the list.  Rails has lots of fans and third-party
tools and it's under active development and supposedly on java 1.7 it runs
faster than the C implementation.

On Wed, Mar 7, 2012 at 11:23 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 Hello Ian,

 We are still looking for the best choice between running in jetty
 a. Accessibility for devs (nice eclipse plugin for example)
 b. Accessibility for user
 c. Light weight
 d. Performance

 In our side the priority are like that abdc (+eclipse plugin)
 Four some freenet devs it loks like cdab (+velocity for templating)

 Not so easy to make the good choice. We have three mains ideas
 A- using Apache Wickets http://wicket.apache.org/
 B- gwt
 C- struts+extjs
 *D-  freenet devs ideas*

 This frameworks have to;
 1. be active and old enough
 2. including web n.m fionctionnalities
 3. replace the Toad controller :-)
 4. using jetty
 5. be fully toolled for Eclipse

 We hope to have freenet devs ideas on it before starting anything :-)

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com




 On Wed, Mar 7, 2012 at 3:33 PM, Ian Clarke i...@locut.us wrote:

 Hi Nicolas,

 Did you ever make any progress on this?

 Ian.

 On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez 
 nicolas.hernan...@aleph-networks.com wrote:

 Hello,

 This is a sort of personnal introduction:

 We want to work on freenet accessibility. The first step could be the
 webui. In the case of a total rework of the webui. Wich framework could be
 the 'good' one in the freenet philosophy ? Wich software architecture ?

 Is GWT and his model a good idea ? Could it be possible to have help
 from GSoc 2012 - (i can be the man with the Umbrella) ?

 ... lots of questions ...

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl




 --
 Ian Clarke
 Personal blog: http://blog.locut.us/



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 22 Feb 2012 14:58:43 Ian Clarke wrote:
> Hi, that would be great!
> 
> We did consider GWT a few years back, but the concern was the lack of
> graceful degradation if the browser didn't support JavaScript.  I'm not
> sure how relevant that concern is in this day and age where browser support
> for JavaScript is assumed on many websites.
> 
> Ultimately my position is that if someone is willing to do the work, I'll
> defer to their decision on what framework will work best for the project,
> within reason.  I personally wouldn't have a problem with GWT if that was
> what you guys are most comfortable with if it means that the job gets done,
> but others might disagree.

The other difficulty is that we will need some tag-level stuff around loading 
freesites anyway. This is currently implemented by abusing GWT however.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: 

[freenet-dev] webui and accessibility

[Ian Clarke]

Hi, that would be great!

We did consider GWT a few years back, but the concern was the lack of
graceful degradation if the browser didn't support JavaScript.  I'm not
sure how relevant that concern is in this day and age where browser support
for JavaScript is assumed on many websites.

Ultimately my position is that if someone is willing to do the work, I'll
defer to their decision on what framework will work best for the project,
within reason.  I personally wouldn't have a problem with GWT if that was
what you guys are most comfortable with if it means that the job gets done,
but others might disagree.

Ian.

On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> Hello,
>
> This is a sort of personnal introduction:
>
> We want to work on freenet accessibility. The first step could be the
> webui. In the case of a total rework of the webui. Wich framework could be
> the 'good' one in the freenet philosophy ? Wich software architecture ?
>
> Is GWT and his model a good idea ? Could it be possible to have help from
> GSoc 2012 - (i can be the man with the Umbrella) ?
>
> ... lots of questions ...
>
> Rgds
>
> - Nicolas Hernandez
> a-n - aleph-networks
> *associ?*
> http://www.aleph-networks.com
>
>
>
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>



-- 
Ian Clarke
Personal blog: http://blog.locut.us/
-- next part --
An HTML attachment was scrubbed...
URL: 

Re: [freenet-dev] webui and accessibility

[Ian Clarke]

Hi, that would be great!

We did consider GWT a few years back, but the concern was the lack of
graceful degradation if the browser didn't support JavaScript.  I'm not
sure how relevant that concern is in this day and age where browser support
for JavaScript is assumed on many websites.

Ultimately my position is that if someone is willing to do the work, I'll
defer to their decision on what framework will work best for the project,
within reason.  I personally wouldn't have a problem with GWT if that was
what you guys are most comfortable with if it means that the job gets done,
but others might disagree.

Ian.

On Tue, Feb 21, 2012 at 11:23 AM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 Hello,

 This is a sort of personnal introduction:

 We want to work on freenet accessibility. The first step could be the
 webui. In the case of a total rework of the webui. Wich framework could be
 the 'good' one in the freenet philosophy ? Wich software architecture ?

 Is GWT and his model a good idea ? Could it be possible to have help from
 GSoc 2012 - (i can be the man with the Umbrella) ?

 ... lots of questions ...

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl




-- 
Ian Clarke
Personal blog: http://blog.locut.us/
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Matthew Toseland]

On Wednesday 22 Feb 2012 14:58:43 Ian Clarke wrote:
 Hi, that would be great!
 
 We did consider GWT a few years back, but the concern was the lack of
 graceful degradation if the browser didn't support JavaScript.  I'm not
 sure how relevant that concern is in this day and age where browser support
 for JavaScript is assumed on many websites.
 
 Ultimately my position is that if someone is willing to do the work, I'll
 defer to their decision on what framework will work best for the project,
 within reason.  I personally wouldn't have a problem with GWT if that was
 what you guys are most comfortable with if it means that the job gets done,
 but others might disagree.

The other difficulty is that we will need some tag-level stuff around loading 
freesites anyway. This is currently implemented by abusing GWT however.


signature.asc
Description: This is a digitally signed message part.
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

[freenet-dev] webui and accessibility

[Pouyan Zachar]

Hey,

we had a discussion on how to re-implement the webui in context of GSoC
2011. The final consensus was to use a templating engine such as Apache
velocity and externalize the whole HTML generation as template file. Main
focus was on a light-weight and fast approach.

Cheers
Pouyan

On Tue, Feb 21, 2012 at 6:23 PM, Nicolas Hernandez <
nicolas.hernandez at aleph-networks.com> wrote:

> Hello,
>
> This is a sort of personnal introduction:
>
> We want to work on freenet accessibility. The first step could be the
> webui. In the case of a total rework of the webui. Wich framework could be
> the 'good' one in the freenet philosophy ? Wich software architecture ?
>
> Is GWT and his model a good idea ? Could it be possible to have help from
> GSoc 2012 - (i can be the man with the Umbrella) ?
>
> ... lots of questions ...
>
> Rgds
>
> - Nicolas Hernandez
> a-n - aleph-networks
> *associ?*
> http://www.aleph-networks.com
>
>
>
> ___
> Devl mailing list
> Devl at freenetproject.org
> http://freenetproject.org/cgi-bin/mailman/listinfo/devl
>
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hello,

This is a sort of personnal introduction:

We want to work on freenet accessibility. The first step could be the
webui. In the case of a total rework of the webui. Wich framework could be
the 'good' one in the freenet philosophy ? Wich software architecture ?

Is GWT and his model a good idea ? Could it be possible to have help from
GSoc 2012 - (i can be the man with the Umbrella) ?

... lots of questions ...

Rgds

- Nicolas Hernandez
a-n - aleph-networks
*associ?*
http://www.aleph-networks.com
-- next part --
An HTML attachment was scrubbed...
URL: 

[freenet-dev] webui and accessibility

[Nicolas Hernandez]

Hello,

This is a sort of personnal introduction:

We want to work on freenet accessibility. The first step could be the
webui. In the case of a total rework of the webui. Wich framework could be
the 'good' one in the freenet philosophy ? Wich software architecture ?

Is GWT and his model a good idea ? Could it be possible to have help from
GSoc 2012 - (i can be the man with the Umbrella) ?

... lots of questions ...

Rgds

- Nicolas Hernandez
a-n - aleph-networks
*associé*
http://www.aleph-networks.com
___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] webui and accessibility

[Pouyan Zachar]

Hey,

we had a discussion on how to re-implement the webui in context of GSoC
2011. The final consensus was to use a templating engine such as Apache
velocity and externalize the whole HTML generation as template file. Main
focus was on a light-weight and fast approach.

Cheers
Pouyan

On Tue, Feb 21, 2012 at 6:23 PM, Nicolas Hernandez 
nicolas.hernan...@aleph-networks.com wrote:

 Hello,

 This is a sort of personnal introduction:

 We want to work on freenet accessibility. The first step could be the
 webui. In the case of a total rework of the webui. Wich framework could be
 the 'good' one in the freenet philosophy ? Wich software architecture ?

 Is GWT and his model a good idea ? Could it be possible to have help from
 GSoc 2012 - (i can be the man with the Umbrella) ?

 ... lots of questions ...

 Rgds

 - Nicolas Hernandez
 a-n - aleph-networks
 *associé*
 http://www.aleph-networks.com



 ___
 Devl mailing list
 Devl@freenetproject.org
 http://freenetproject.org/cgi-bin/mailman/listinfo/devl

___
Devl mailing list
Devl@freenetproject.org
http://freenetproject.org/cgi-bin/mailman/listinfo/devl