[freenet-dev] Content Management System

[Florent Daignière]

* Ian Clarke  [2008-05-12 23:04:59]:

> On Mon, May 12, 2008 at 9:32 PM, Florent Daigni?re
>  wrote:
> >  I personnaly won't work on it for now. The main point against our
> >  current website is that it's not community-friendly... Let's see if the
> >  community feels involved and will contribute to the presumably
> >  community-friendly sandbox.
> 
> Not sure that this makes much sense.  Seems like you are saying that
> unless the "community" builds up the new site so that its better than
> the current site, then the "community" doesn't deserve a better
> website.  Hmm.  To paraphrase Margaret Thatcher, "There is no such
> thing as community", its all just individuals.
> 

Heh I'm trying to make things move forward here... sorry if that sounded
rude or inappropriate but yeah... I think we need external help to make
any progress here.

Most of the current contributors are coders and don't have good
web-designing skills ... or don't want to do it.

> Freenet needs a website worthy of a respectable and reasonably well
> funded free software project in 2008, unfortunately it doesn't have
> one right now.

We all agree that the current website sucks.

> Maybe Drupal is the answer, maybe not.  Frankly I'm
> not convinced that migrating to a CMS really solves the big problems
> with the website at all.
> 

Me neither... but Michael suggested it and no one strongly objected this
time. Last time someone did suggest it I was the one who did object...
and let's face it, the website hasn't evolved much since then... so
presumably I might have been wrong.

> Either way, its up to us to come up with a website that newbies feel
> comfortable with.  We've been very lucky with funding from large
> donors over the years, but sooner or later that luck is going to run
> out.  We can't assume that when the Google donation runs out in 6
> months, we will find someone else willing to donate $10k+.  We have 6
> months to build up enough of a userbase that we can get >$3k/month in
> donations reliably, or we can wave goodbye to Matthew's full-time
> attention.  Call me a pessimist, but this project is in big trouble if
> we lose Matthew.
> 
> Having an appealing website that efficiently turns newbies into
> dedicated Freenet users (and donors) isn't just some nice thing to
> make us all feel better, its a requirement for the survival of the
> project.

Well what's the solution then? To make Matthew work on the website? to
send a call for help on @announce (possibly a better phrased than mine)?

Shall I forget about the drupal vhost right-now and delete it?
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 

[freenet-dev] Content Management System

[Florent Daignière]

* Michael T?nzer  [2008-05-08 17:41:55]:

> Florent Daigni?re schrieb:
> > * Michael T?nzer  [2008-05-08 05:04:07]:
> It's probably not possible to migrate in two days but it seems that now
> is a good point to start the process, as Ian mentioned he wanted to
> change the website significantly (this also includes the texts). We
> probably should migrate in a soft way and try it in a test environment
> first. The Website would be a good point to start with because it has
> not so much content on it.

Okay, let's try it. I've set a drupal vhost up on emu
(http://wwwtest.freenetproject.org/).

Here is the deal: if it reaches the point where it's actually better
than the actual website then we switch permanently... Otherwise I'll
just delete it. I will grant admin access to whoever wants it... and
can install modules/themes/whatever on request. We need to put some kind
of deadline otherwise nothing will ever be done: what about one month?

I personnaly won't work on it for now. The main point against our
current website is that it's not community-friendly... Let's see if the
community feels involved and will contribute to the presumably
community-friendly sandbox.

> The other things could be done step by step,
> or never if we want to keep them (e.g. I'm not quite convinced about
> drupals bug tracker, but there are definitely better wiki engines than
> wikkawiki).
> 
> > Can a CMS have some level of history ? All the tools we use have
> > native versioning; that's a feature we don't want to loose.
> 
> Drupal has native versioning, I think that's one of the core features
> which made it one of the favourite CMSs for OpenSource projects.

It seems like it's a per-page versioning... that's not what we want, is
it ?
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 

[freenet-dev] Content Management System

[Ian Clarke]

On Mon, May 12, 2008 at 11:21 PM, Florent Daigni?re
 wrote:
>  Well what's the solution then? To make Matthew work on the website? to
>  send a call for help on @announce (possibly a better phrased than mine)?
>
>  Shall I forget about the drupal vhost right-now and delete it?

Definitely not, its a good experiment, and may yield good results.

I think you are right that none of us are good web developers, and
frankly its going to be hard to find some web development genius to
give his time to re-architecting the website.

I think the key is to take advantage of open source, to find a good
design that is released under a free license (perhaps GPL, perhaps
creative commons, maybe something else), and use it, perhaps with a
few minor modifications (logo, color scheme, etc).

I've suggested looking at the Mozilla project, because they release
their websites under creative commons, and they have some pretty good
web-designers.  Of course we should look elsewhere too.

We need to find a way to have a professional looking website, without
a) having to build it from scratch ourselves and b) having to spend
any of our precious donations on building it.  Taking advantage of
open source HTML and CSS code seems like the natural answer to this.

Ian.

-- 
Email: ian at uprizer.com
Cell: +1 512 422 3588
Skype: sanity

[freenet-dev] Content Management System

[Ian Clarke]

On Mon, May 12, 2008 at 9:32 PM, Florent Daigni?re
 wrote:
>  I personnaly won't work on it for now. The main point against our
>  current website is that it's not community-friendly... Let's see if the
>  community feels involved and will contribute to the presumably
>  community-friendly sandbox.

Not sure that this makes much sense.  Seems like you are saying that
unless the "community" builds up the new site so that its better than
the current site, then the "community" doesn't deserve a better
website.  Hmm.  To paraphrase Margaret Thatcher, "There is no such
thing as community", its all just individuals.

Freenet needs a website worthy of a respectable and reasonably well
funded free software project in 2008, unfortunately it doesn't have
one right now.  Maybe Drupal is the answer, maybe not.  Frankly I'm
not convinced that migrating to a CMS really solves the big problems
with the website at all.

Either way, its up to us to come up with a website that newbies feel
comfortable with.  We've been very lucky with funding from large
donors over the years, but sooner or later that luck is going to run
out.  We can't assume that when the Google donation runs out in 6
months, we will find someone else willing to donate $10k+.  We have 6
months to build up enough of a userbase that we can get >$3k/month in
donations reliably, or we can wave goodbye to Matthew's full-time
attention.  Call me a pessimist, but this project is in big trouble if
we lose Matthew.

Having an appealing website that efficiently turns newbies into
dedicated Freenet users (and donors) isn't just some nice thing to
make us all feel better, its a requirement for the survival of the
project.

Ian.

-- 
Email: ian at uprizer.com
Cell: +1 512 422 3588
Skype: sanity

Re: [freenet-dev] Content Management System

[Ian Clarke]

On Mon, May 12, 2008 at 11:21 PM, Florent Daignière
<[EMAIL PROTECTED]> wrote:
>  Well what's the solution then? To make Matthew work on the website? to
>  send a call for help on @announce (possibly a better phrased than mine)?
>
>  Shall I forget about the drupal vhost right-now and delete it?

Definitely not, its a good experiment, and may yield good results.

I think you are right that none of us are good web developers, and
frankly its going to be hard to find some web development genius to
give his time to re-architecting the website.

I think the key is to take advantage of open source, to find a good
design that is released under a free license (perhaps GPL, perhaps
creative commons, maybe something else), and use it, perhaps with a
few minor modifications (logo, color scheme, etc).

I've suggested looking at the Mozilla project, because they release
their websites under creative commons, and they have some pretty good
web-designers.  Of course we should look elsewhere too.

We need to find a way to have a professional looking website, without
a) having to build it from scratch ourselves and b) having to spend
any of our precious donations on building it.  Taking advantage of
open source HTML and CSS code seems like the natural answer to this.

Ian.

-- 
Email: [EMAIL PROTECTED]
Cell: +1 512 422 3588
Skype: sanity
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] Content Management System

[Florent Daignière]

* Ian Clarke <[EMAIL PROTECTED]> [2008-05-12 23:04:59]:

> On Mon, May 12, 2008 at 9:32 PM, Florent Daignière
> <[EMAIL PROTECTED]> wrote:
> >  I personnaly won't work on it for now. The main point against our
> >  current website is that it's not community-friendly... Let's see if the
> >  community feels involved and will contribute to the presumably
> >  community-friendly sandbox.
> 
> Not sure that this makes much sense.  Seems like you are saying that
> unless the "community" builds up the new site so that its better than
> the current site, then the "community" doesn't deserve a better
> website.  Hmm.  To paraphrase Margaret Thatcher, "There is no such
> thing as community", its all just individuals.
> 

Heh I'm trying to make things move forward here... sorry if that sounded
rude or inappropriate but yeah... I think we need external help to make
any progress here.

Most of the current contributors are coders and don't have good
web-designing skills ... or don't want to do it.

> Freenet needs a website worthy of a respectable and reasonably well
> funded free software project in 2008, unfortunately it doesn't have
> one right now.

We all agree that the current website sucks.

> Maybe Drupal is the answer, maybe not.  Frankly I'm
> not convinced that migrating to a CMS really solves the big problems
> with the website at all.
> 

Me neither... but Michael suggested it and no one strongly objected this
time. Last time someone did suggest it I was the one who did object...
and let's face it, the website hasn't evolved much since then... so
presumably I might have been wrong.

> Either way, its up to us to come up with a website that newbies feel
> comfortable with.  We've been very lucky with funding from large
> donors over the years, but sooner or later that luck is going to run
> out.  We can't assume that when the Google donation runs out in 6
> months, we will find someone else willing to donate $10k+.  We have 6
> months to build up enough of a userbase that we can get >$3k/month in
> donations reliably, or we can wave goodbye to Matthew's full-time
> attention.  Call me a pessimist, but this project is in big trouble if
> we lose Matthew.
> 
> Having an appealing website that efficiently turns newbies into
> dedicated Freenet users (and donors) isn't just some nice thing to
> make us all feel better, its a requirement for the survival of the
> project.

Well what's the solution then? To make Matthew work on the website? to
send a call for help on @announce (possibly a better phrased than mine)?

Shall I forget about the drupal vhost right-now and delete it?


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

Re: [freenet-dev] Content Management System

[Ian Clarke]

On Mon, May 12, 2008 at 9:32 PM, Florent Daignière
<[EMAIL PROTECTED]> wrote:
>  I personnaly won't work on it for now. The main point against our
>  current website is that it's not community-friendly... Let's see if the
>  community feels involved and will contribute to the presumably
>  community-friendly sandbox.

Not sure that this makes much sense.  Seems like you are saying that
unless the "community" builds up the new site so that its better than
the current site, then the "community" doesn't deserve a better
website.  Hmm.  To paraphrase Margaret Thatcher, "There is no such
thing as community", its all just individuals.

Freenet needs a website worthy of a respectable and reasonably well
funded free software project in 2008, unfortunately it doesn't have
one right now.  Maybe Drupal is the answer, maybe not.  Frankly I'm
not convinced that migrating to a CMS really solves the big problems
with the website at all.

Either way, its up to us to come up with a website that newbies feel
comfortable with.  We've been very lucky with funding from large
donors over the years, but sooner or later that luck is going to run
out.  We can't assume that when the Google donation runs out in 6
months, we will find someone else willing to donate $10k+.  We have 6
months to build up enough of a userbase that we can get >$3k/month in
donations reliably, or we can wave goodbye to Matthew's full-time
attention.  Call me a pessimist, but this project is in big trouble if
we lose Matthew.

Having an appealing website that efficiently turns newbies into
dedicated Freenet users (and donors) isn't just some nice thing to
make us all feel better, its a requirement for the survival of the
project.

Ian.

-- 
Email: [EMAIL PROTECTED]
Cell: +1 512 422 3588
Skype: sanity
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] Content Management System

[Florent Daignière]

* Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 17:41:55]:

> Florent Daignière schrieb:
> > * Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 05:04:07]:
> It's probably not possible to migrate in two days but it seems that now
> is a good point to start the process, as Ian mentioned he wanted to
> change the website significantly (this also includes the texts). We
> probably should migrate in a soft way and try it in a test environment
> first. The Website would be a good point to start with because it has
> not so much content on it.

Okay, let's try it. I've set a drupal vhost up on emu
(http://wwwtest.freenetproject.org/).

Here is the deal: if it reaches the point where it's actually better
than the actual website then we switch permanently... Otherwise I'll
just delete it. I will grant admin access to whoever wants it... and
can install modules/themes/whatever on request. We need to put some kind
of deadline otherwise nothing will ever be done: what about one month?

I personnaly won't work on it for now. The main point against our
current website is that it's not community-friendly... Let's see if the
community feels involved and will contribute to the presumably
community-friendly sandbox.

> The other things could be done step by step,
> or never if we want to keep them (e.g. I'm not quite convinced about
> drupals bug tracker, but there are definitely better wiki engines than
> wikkawiki).
> 
> > Can a CMS have some level of history ? All the tools we use have
> > native versioning; that's a feature we don't want to loose.
> 
> Drupal has native versioning, I think that's one of the core features
> which made it one of the favourite CMSs for OpenSource projects.

It seems like it's a per-page versioning... that's not what we want, is
it ?


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

[freenet-dev] Content Management System

[Florent Daignière]

* Michael T?nzer  [2008-05-08 22:47:28]:

> Ian Clarke schrieb:
> > You could argue that all of those things have it easy, because most
> > people understand what those things do, they don't need an elaborate
> > explanation.  But look at Gnome's website:
> > 
> >   http://www.gnome.org/
> > 
> > It is clean, simple, yet if you need to you can quickly dig down to a
> > vast wealth of information.
> > 
> 
> Gnome uses Plone as it's CMS. This may be not the best choice for us
> though, as Plone is Phyton based, but that's something nextgens might
> know better than me.
> 

I don't have any problem with using python; Trac is in python too
iirc... But I don't think that Plone is what we need, it's a framework
more than a CMS.

> > Either way, since we are Java hackers for the most part, not web
> > designers, I strongly recommend that we borrow as much as we can from
> > elsewhere, even so far as using free or creative commons HTML and CSS
> > verbatim, perhaps with only a few minor changes.
> > 

What about java-based CMSes then ? :p
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 

Re: [freenet-dev] Content Management System

[Florent Daignière]

* Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 22:47:28]:

> Ian Clarke schrieb:
> > You could argue that all of those things have it easy, because most
> > people understand what those things do, they don't need an elaborate
> > explanation.  But look at Gnome's website:
> > 
> >   http://www.gnome.org/
> > 
> > It is clean, simple, yet if you need to you can quickly dig down to a
> > vast wealth of information.
> > 
> 
> Gnome uses Plone as it's CMS. This may be not the best choice for us
> though, as Plone is Phyton based, but that's something nextgens might
> know better than me.
> 

I don't have any problem with using python; Trac is in python too
iirc... But I don't think that Plone is what we need, it's a framework
more than a CMS.

> > Either way, since we are Java hackers for the most part, not web
> > designers, I strongly recommend that we borrow as much as we can from
> > elsewhere, even so far as using free or creative commons HTML and CSS
> > verbatim, perhaps with only a few minor changes.
> > 

What about java-based CMSes then ? :p


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

[freenet-dev] Content Management System

[Michael Tänzer]

Ian Clarke schrieb:
> On Thu, May 8, 2008 at 12:49 PM, Victor Denisov  
> wrote:
>> -BEGIN PGP SIGNED MESSAGE-
>>  Hash: SHA1
>>  | consider a migration seriously. If it was up to me, we would use Trac
>>  | and only Trac (for the website, wikki and bug-tracker).
>>
>>  A quick vote of confidence for Trac. It's a *very* good piece of
>>  software, and its Wiki<->Tickets<->SVN integration is amazing. We've
>>  used it for three major projects now and had nothing but *very* positive
>>  experience.
> 
> Trac isn't a bad bugtracker, probably better than Mantis, although has
> some conspicuous limitations, such as no dependencies between bugs.
> 
> That being said, I don't think Trac is designed to be a CMS, and
> frankly I don't think its appearance is enticing as a "user facing"
> website, its even worse than the current Freenet website.  If we did
> use it, it would require some major re-theming.
> 

ACK

> I think we should look to commercial and well-funded open source
> projects for inspiration about how to make our website enticing for
> first-time visitors, while still providing the depth of information we
> need to convey.
> 
> http://getfirefox.com/ is good because its colorful, inviting, and the
> "call to action" is very clear, you don't have to spend much time
> looking for that download link!  Now, its tone may be a little too
> in-your-face for Freenet, but there are things we can learn from it.
> 

getfirefox is drupal driven

> I'm a big fan of David Watanabe's work, both the software he writes,
> and the websites he designs for them.  I'd recommend looking at:
> 
>   http://xtorrent.com/
>   http://www.inquisitorx.com/safari/
>   http://www.acquisitionx.com/
> 

The design is cool, but it's a little bit too trendy in my opinion (it's
ok for stylish apple addons, but not suitable for freenet, as it's
supposed to be secure software, not another design wonder). Also some
users who don't have the bandwith would really be annoyed by the large
pictures.
Apart from that there's not much documentation on the websites, no
navigation menu (we would definitely need one, as we have more than
three pages to offer).
I don't think we could apply a similar design to freenet.

> You could argue that all of those things have it easy, because most
> people understand what those things do, they don't need an elaborate
> explanation.  But look at Gnome's website:
> 
>   http://www.gnome.org/
> 
> It is clean, simple, yet if you need to you can quickly dig down to a
> vast wealth of information.
> 

Gnome uses Plone as it's CMS. This may be not the best choice for us
though, as Plone is Phyton based, but that's something nextgens might
know better than me.

> Either way, since we are Java hackers for the most part, not web
> designers, I strongly recommend that we borrow as much as we can from
> elsewhere, even so far as using free or creative commons HTML and CSS
> verbatim, perhaps with only a few minor changes.
> 
> Ian.
> 

Neo at NHNG
-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: 

[freenet-dev] Content Management System

[Victor Denisov]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

| consider a migration seriously. If it was up to me, we would use Trac
| and only Trac (for the website, wikki and bug-tracker).

A quick vote of confidence for Trac. It's a *very* good piece of
software, and its Wiki<->Tickets<->SVN integration is amazing. We've
used it for three major projects now and had nothing but *very* positive
experience.

Regards,
Victor Denisov.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIIz0rS81Mh9/iCDgRAsoSAKDSzT3tW+DaA/FkOefLAU4MUuVkfACgz8sx
QflbOaD1swFHdHMWj7Uzago=
=QlxK
-END PGP SIGNATURE-

[freenet-dev] Content Management System

[Florent Daignière]

* Michael T?nzer  [2008-05-08 17:41:55]:

> Florent Daigni?re schrieb:
> > * Michael T?nzer  [2008-05-08 05:04:07]:
> > 
> >> In the last few weeks I've done some work on the website. While
> >> translating it, there were some things that struck me so I changed them.
> >> But our site is still far from perfect. It lacks a attractive design and
> >> some features that would be quite handy (e.g. select the language by
> >> hand, RSS-Feeds, a search) but are a little bit difficult to implement
> >> (at least if we want to do it in a safe and efficient way) or at least I
> >> don't have the time and skills to do it.
> > 
> > Select language by hand is trivial to implement and we can delegate the 
> > search
> > to google so that's trivial too... okay RSS would require some work
> > 
> 
> I know it's not that hard to do but someone actually has to do it.

No one has bothered that's why it hasn't been done.

[snip.]

> > At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
> > home-maid website and *loads* of custom scripting for almost
> > everything... How do you plan to migrate existing content ?
> > 
> 
> The fully custom made site is one of the problems, as we are not experts
> in some of the things we did. I saw that you fixed some security issues
> in our php code today, some issues that dealt with character escaping
> and such things.

The broken code wasn't mine! I have already fixed the exact same bug 3
years ago and someone reintroduced it since then!

We should really have regression tests; even for the website.

> I'm no PHP expert but I think these are things which
> are obvious to a professional php-developer but can completely break our
> security, which means if some  here> guy used this issue to hack into our server and replace the
> binaries we provide, then this could be rather dangerous for our users.
> 

I'm not a fan of security by obscurity but let's face it: we have fixed
only a few security related bugs in the last few years... Drupal had many
more (and that's logical given that it's a gaz plant compared to our
requirements). Their last release was on the 9th of April and guess
what? It's a security bugfix!

> What I want to say: If you're not absolutely sure about what you're
> doing, leave it to the pros, they know how to deal with it, and we can
> concentrate on what we do best: provide our users with tools to give
> them true freedom of speech.
> 

Go on with that logic... and we end up being dependant on a 3rd party
entity. We left SourceForge because their service wasn't up to our
expectations anymore and at the time there was no good alternative.

> It's probably not possible to migrate in two days but it seems that now
> is a good point to start the process, as Ian mentioned he wanted to
> change the website significantly (this also includes the texts). We
> probably should migrate in a soft way and try it in a test environment
> first. The Website would be a good point to start with because it has
> not so much content on it. The other things could be done step by step,
> or never if we want to keep them (e.g. I'm not quite convinced about
> drupals bug tracker, but there are definitely better wiki engines than
> wikkawiki).

I don't share your views here. Either we switch to a CMS and use it for
everything or we don't.

They are good and bad reasons to switch to a CMS: I don't think that
security is a good one. As you've highlighted, our website doesn't
evolve much and has a long history; that's why it's pretty secure
overall. On the other hand, integration of services into the CMS
is a good reason to make the switch. Find a CMS which has a good
integration with mantis or can import its tickets and then we can
consider a migration seriously. If it was up to me, we would use Trac
and only Trac (for the website, wikki and bug-tracker).

A few weeks ago someone asked me to set a blog engine up (Wordpress), I
did and so far no one used it... We obviously don't want the same thing
to happen with a Drupal, do we ?
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 

[freenet-dev] Content Management System

[Michael Tänzer]

Florent Daigni?re schrieb:
> * Michael T?nzer  [2008-05-08 05:04:07]:
> 
>> In the last few weeks I've done some work on the website. While
>> translating it, there were some things that struck me so I changed them.
>> But our site is still far from perfect. It lacks a attractive design and
>> some features that would be quite handy (e.g. select the language by
>> hand, RSS-Feeds, a search) but are a little bit difficult to implement
>> (at least if we want to do it in a safe and efficient way) or at least I
>> don't have the time and skills to do it.
> 
> Select language by hand is trivial to implement and we can delegate the search
> to google so that's trivial too... okay RSS would require some work
> 

I know it's not that hard to do but someone actually has to do it. And
if there is an already existent solution, why not do it with that one
instead of inventing the wheel yet another time.

>> I also noticed, that the format
>> we use to save the content (it's just a php file containing HTML which
>> is included in some kind of very simple template) leaves room for
>> optimization (for both, the author who needs to write valid HTML and
>> know about the things he can do with it (not all of us do know how to
>> write clean and valid HTML (I do not exclude myself from this
>> statement)), and the user, who might get malformed HTML or ugly pages
>> because the browser has some bugs the author didn't know of). We also
>> have the problem, that our site consists of many different components:
>> there's the homepage, the wiki, emu, SVN, which looks very fragmented.
>>
>> We could address most of this problems by using a CMS (content
>> management system).
> 
> It's not the first time it's being debated...
> 
>> Of course a CMS is not a Swiss Army knife for
>> everything and it does raise several issues: is it fast enough to
>> survive a slashdot, can we use our already existent database, how can we
>> migrate, is it safe?
> 
> Don't worry about performances for now...
> 

I don't but Ian did when I had a discussion with him about the some web
design issues recently where I mentioned CMSs.

> At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
> home-maid website and *loads* of custom scripting for almost
> everything... How do you plan to migrate existing content ?
> 

The fully custom made site is one of the problems, as we are not experts
in some of the things we did. I saw that you fixed some security issues
in our php code today, some issues that dealt with character escaping
and such things. I'm no PHP expert but I think these are things which
are obvious to a professional php-developer but can completely break our
security, which means if some  guy used this issue to hack into our server and replace the
binaries we provide, then this could be rather dangerous for our users.

What I want to say: If you're not absolutely sure about what you're
doing, leave it to the pros, they know how to deal with it, and we can
concentrate on what we do best: provide our users with tools to give
them true freedom of speech.

It's probably not possible to migrate in two days but it seems that now
is a good point to start the process, as Ian mentioned he wanted to
change the website significantly (this also includes the texts). We
probably should migrate in a soft way and try it in a test environment
first. The Website would be a good point to start with because it has
not so much content on it. The other things could be done step by step,
or never if we want to keep them (e.g. I'm not quite convinced about
drupals bug tracker, but there are definitely better wiki engines than
wikkawiki).

> Can a CMS have some level of history ? All the tools we use have
> native versioning; that's a feature we don't want to loose.

Drupal has native versioning, I think that's one of the core features
which made it one of the favourite CMSs for OpenSource projects.

Neo at NHNG

-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: 

Re: [freenet-dev] Content Management System

[Michael Tänzer]

Ian Clarke schrieb:
> On Thu, May 8, 2008 at 12:49 PM, Victor Denisov <[EMAIL PROTECTED]> wrote:
>> -BEGIN PGP SIGNED MESSAGE-
>>  Hash: SHA1
>>  | consider a migration seriously. If it was up to me, we would use Trac
>>  | and only Trac (for the website, wikki and bug-tracker).
>>
>>  A quick vote of confidence for Trac. It's a *very* good piece of
>>  software, and its Wiki<->Tickets<->SVN integration is amazing. We've
>>  used it for three major projects now and had nothing but *very* positive
>>  experience.
> 
> Trac isn't a bad bugtracker, probably better than Mantis, although has
> some conspicuous limitations, such as no dependencies between bugs.
> 
> That being said, I don't think Trac is designed to be a CMS, and
> frankly I don't think its appearance is enticing as a "user facing"
> website, its even worse than the current Freenet website.  If we did
> use it, it would require some major re-theming.
> 

ACK

> I think we should look to commercial and well-funded open source
> projects for inspiration about how to make our website enticing for
> first-time visitors, while still providing the depth of information we
> need to convey.
> 
> http://getfirefox.com/ is good because its colorful, inviting, and the
> "call to action" is very clear, you don't have to spend much time
> looking for that download link!  Now, its tone may be a little too
> in-your-face for Freenet, but there are things we can learn from it.
> 

getfirefox is drupal driven

> I'm a big fan of David Watanabe's work, both the software he writes,
> and the websites he designs for them.  I'd recommend looking at:
> 
>   http://xtorrent.com/
>   http://www.inquisitorx.com/safari/
>   http://www.acquisitionx.com/
> 

The design is cool, but it's a little bit too trendy in my opinion (it's
ok for stylish apple addons, but not suitable for freenet, as it's
supposed to be secure software, not another design wonder). Also some
users who don't have the bandwith would really be annoyed by the large
pictures.
Apart from that there's not much documentation on the websites, no
navigation menu (we would definitely need one, as we have more than
three pages to offer).
I don't think we could apply a similar design to freenet.

> You could argue that all of those things have it easy, because most
> people understand what those things do, they don't need an elaborate
> explanation.  But look at Gnome's website:
> 
>   http://www.gnome.org/
> 
> It is clean, simple, yet if you need to you can quickly dig down to a
> vast wealth of information.
> 

Gnome uses Plone as it's CMS. This may be not the best choice for us
though, as Plone is Phyton based, but that's something nextgens might
know better than me.

> Either way, since we are Java hackers for the most part, not web
> designers, I strongly recommend that we borrow as much as we can from
> elsewhere, even so far as using free or creative commons HTML and CSS
> verbatim, perhaps with only a few minor changes.
> 
> Ian.
> 

[EMAIL PROTECTED]
-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/



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

[freenet-dev] Content Management System

[Florent Daignière]

* Michael T?nzer  [2008-05-08 05:04:07]:

> In the last few weeks I've done some work on the website. While
> translating it, there were some things that struck me so I changed them.
> But our site is still far from perfect. It lacks a attractive design and
> some features that would be quite handy (e.g. select the language by
> hand, RSS-Feeds, a search) but are a little bit difficult to implement
> (at least if we want to do it in a safe and efficient way) or at least I
> don't have the time and skills to do it.

Select language by hand is trivial to implement and we can delegate the search
to google so that's trivial too... okay RSS would require some work

> I also noticed, that the format
> we use to save the content (it's just a php file containing HTML which
> is included in some kind of very simple template) leaves room for
> optimization (for both, the author who needs to write valid HTML and
> know about the things he can do with it (not all of us do know how to
> write clean and valid HTML (I do not exclude myself from this
> statement)), and the user, who might get malformed HTML or ugly pages
> because the browser has some bugs the author didn't know of). We also
> have the problem, that our site consists of many different components:
> there's the homepage, the wiki, emu, SVN, which looks very fragmented.
> 
> We could address most of this problems by using a CMS (content
> management system).

It's not the first time it's being debated...

> Of course a CMS is not a Swiss Army knife for
> everything and it does raise several issues: is it fast enough to
> survive a slashdot, can we use our already existent database, how can we
> migrate, is it safe?

Don't worry about performances for now...

At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
home-maid website and *loads* of custom scripting for almost
everything... How do you plan to migrate existing content ?

Can a CMS have some level of history ? All the tools we use have
native versioning; that's a feature we don't want to loose.
-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 

[freenet-dev] Content Management System

[Ian Clarke]

On Thu, May 8, 2008 at 12:49 PM, Victor Denisov  wrote:
> -BEGIN PGP SIGNED MESSAGE-
>  Hash: SHA1
>  | consider a migration seriously. If it was up to me, we would use Trac
>  | and only Trac (for the website, wikki and bug-tracker).
>
>  A quick vote of confidence for Trac. It's a *very* good piece of
>  software, and its Wiki<->Tickets<->SVN integration is amazing. We've
>  used it for three major projects now and had nothing but *very* positive
>  experience.

Trac isn't a bad bugtracker, probably better than Mantis, although has
some conspicuous limitations, such as no dependencies between bugs.

That being said, I don't think Trac is designed to be a CMS, and
frankly I don't think its appearance is enticing as a "user facing"
website, its even worse than the current Freenet website.  If we did
use it, it would require some major re-theming.

I think we should look to commercial and well-funded open source
projects for inspiration about how to make our website enticing for
first-time visitors, while still providing the depth of information we
need to convey.

http://getfirefox.com/ is good because its colorful, inviting, and the
"call to action" is very clear, you don't have to spend much time
looking for that download link!  Now, its tone may be a little too
in-your-face for Freenet, but there are things we can learn from it.

I'm a big fan of David Watanabe's work, both the software he writes,
and the websites he designs for them.  I'd recommend looking at:

  http://xtorrent.com/
  http://www.inquisitorx.com/safari/
  http://www.acquisitionx.com/

You could argue that all of those things have it easy, because most
people understand what those things do, they don't need an elaborate
explanation.  But look at Gnome's website:

  http://www.gnome.org/

It is clean, simple, yet if you need to you can quickly dig down to a
vast wealth of information.

Either way, since we are Java hackers for the most part, not web
designers, I strongly recommend that we borrow as much as we can from
elsewhere, even so far as using free or creative commons HTML and CSS
verbatim, perhaps with only a few minor changes.

Ian.

-- 
Email: ian at uprizer.com
Cell: +1 512 422 3588
Skype: sanity

Re: [freenet-dev] Content Management System

[Ian Clarke]

On Thu, May 8, 2008 at 12:49 PM, Victor Denisov <[EMAIL PROTECTED]> wrote:
> -BEGIN PGP SIGNED MESSAGE-
>  Hash: SHA1
>  | consider a migration seriously. If it was up to me, we would use Trac
>  | and only Trac (for the website, wikki and bug-tracker).
>
>  A quick vote of confidence for Trac. It's a *very* good piece of
>  software, and its Wiki<->Tickets<->SVN integration is amazing. We've
>  used it for three major projects now and had nothing but *very* positive
>  experience.

Trac isn't a bad bugtracker, probably better than Mantis, although has
some conspicuous limitations, such as no dependencies between bugs.

That being said, I don't think Trac is designed to be a CMS, and
frankly I don't think its appearance is enticing as a "user facing"
website, its even worse than the current Freenet website.  If we did
use it, it would require some major re-theming.

I think we should look to commercial and well-funded open source
projects for inspiration about how to make our website enticing for
first-time visitors, while still providing the depth of information we
need to convey.

http://getfirefox.com/ is good because its colorful, inviting, and the
"call to action" is very clear, you don't have to spend much time
looking for that download link!  Now, its tone may be a little too
in-your-face for Freenet, but there are things we can learn from it.

I'm a big fan of David Watanabe's work, both the software he writes,
and the websites he designs for them.  I'd recommend looking at:

  http://xtorrent.com/
  http://www.inquisitorx.com/safari/
  http://www.acquisitionx.com/

You could argue that all of those things have it easy, because most
people understand what those things do, they don't need an elaborate
explanation.  But look at Gnome's website:

  http://www.gnome.org/

It is clean, simple, yet if you need to you can quickly dig down to a
vast wealth of information.

Either way, since we are Java hackers for the most part, not web
designers, I strongly recommend that we borrow as much as we can from
elsewhere, even so far as using free or creative commons HTML and CSS
verbatim, perhaps with only a few minor changes.

Ian.

-- 
Email: [EMAIL PROTECTED]
Cell: +1 512 422 3588
Skype: sanity
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] Content Management System

[Victor Denisov]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

| consider a migration seriously. If it was up to me, we would use Trac
| and only Trac (for the website, wikki and bug-tracker).

A quick vote of confidence for Trac. It's a *very* good piece of
software, and its Wiki<->Tickets<->SVN integration is amazing. We've
used it for three major projects now and had nothing but *very* positive
experience.

Regards,
Victor Denisov.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.5 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIIz0rS81Mh9/iCDgRAsoSAKDSzT3tW+DaA/FkOefLAU4MUuVkfACgz8sx
QflbOaD1swFHdHMWj7Uzago=
=QlxK
-END PGP SIGNATURE-
___
Devl mailing list
Devl@freenetproject.org
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Re: [freenet-dev] Content Management System

[Florent Daignière]

* Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 17:41:55]:

> Florent Daignière schrieb:
> > * Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 05:04:07]:
> > 
> >> In the last few weeks I've done some work on the website. While
> >> translating it, there were some things that struck me so I changed them.
> >> But our site is still far from perfect. It lacks a attractive design and
> >> some features that would be quite handy (e.g. select the language by
> >> hand, RSS-Feeds, a search) but are a little bit difficult to implement
> >> (at least if we want to do it in a safe and efficient way) or at least I
> >> don't have the time and skills to do it.
> > 
> > Select language by hand is trivial to implement and we can delegate the 
> > search
> > to google so that's trivial too... okay RSS would require some work
> > 
> 
> I know it's not that hard to do but someone actually has to do it.

No one has bothered that's why it hasn't been done.

[snip.]

> > At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
> > home-maid website and *loads* of custom scripting for almost
> > everything... How do you plan to migrate existing content ?
> > 
> 
> The fully custom made site is one of the problems, as we are not experts
> in some of the things we did. I saw that you fixed some security issues
> in our php code today, some issues that dealt with character escaping
> and such things.

The broken code wasn't mine! I have already fixed the exact same bug 3
years ago and someone reintroduced it since then!

We should really have regression tests; even for the website.

> I'm no PHP expert but I think these are things which
> are obvious to a professional php-developer but can completely break our
> security, which means if some  here> guy used this issue to hack into our server and replace the
> binaries we provide, then this could be rather dangerous for our users.
> 

I'm not a fan of security by obscurity but let's face it: we have fixed
only a few security related bugs in the last few years... Drupal had many
more (and that's logical given that it's a gaz plant compared to our
requirements). Their last release was on the 9th of April and guess
what? It's a security bugfix!

> What I want to say: If you're not absolutely sure about what you're
> doing, leave it to the pros, they know how to deal with it, and we can
> concentrate on what we do best: provide our users with tools to give
> them true freedom of speech.
> 

Go on with that logic... and we end up being dependant on a 3rd party
entity. We left SourceForge because their service wasn't up to our
expectations anymore and at the time there was no good alternative.

> It's probably not possible to migrate in two days but it seems that now
> is a good point to start the process, as Ian mentioned he wanted to
> change the website significantly (this also includes the texts). We
> probably should migrate in a soft way and try it in a test environment
> first. The Website would be a good point to start with because it has
> not so much content on it. The other things could be done step by step,
> or never if we want to keep them (e.g. I'm not quite convinced about
> drupals bug tracker, but there are definitely better wiki engines than
> wikkawiki).

I don't share your views here. Either we switch to a CMS and use it for
everything or we don't.

They are good and bad reasons to switch to a CMS: I don't think that
security is a good one. As you've highlighted, our website doesn't
evolve much and has a long history; that's why it's pretty secure
overall. On the other hand, integration of services into the CMS
is a good reason to make the switch. Find a CMS which has a good
integration with mantis or can import its tickets and then we can
consider a migration seriously. If it was up to me, we would use Trac
and only Trac (for the website, wikki and bug-tracker).

A few weeks ago someone asked me to set a blog engine up (Wordpress), I
did and so far no one used it... We obviously don't want the same thing
to happen with a Drupal, do we ?


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

Re: [freenet-dev] Content Management System

[Michael Tänzer]

Florent Daignière schrieb:
> * Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 05:04:07]:
> 
>> In the last few weeks I've done some work on the website. While
>> translating it, there were some things that struck me so I changed them.
>> But our site is still far from perfect. It lacks a attractive design and
>> some features that would be quite handy (e.g. select the language by
>> hand, RSS-Feeds, a search) but are a little bit difficult to implement
>> (at least if we want to do it in a safe and efficient way) or at least I
>> don't have the time and skills to do it.
> 
> Select language by hand is trivial to implement and we can delegate the search
> to google so that's trivial too... okay RSS would require some work
> 

I know it's not that hard to do but someone actually has to do it. And
if there is an already existent solution, why not do it with that one
instead of inventing the wheel yet another time.

>> I also noticed, that the format
>> we use to save the content (it's just a php file containing HTML which
>> is included in some kind of very simple template) leaves room for
>> optimization (for both, the author who needs to write valid HTML and
>> know about the things he can do with it (not all of us do know how to
>> write clean and valid HTML (I do not exclude myself from this
>> statement)), and the user, who might get malformed HTML or ugly pages
>> because the browser has some bugs the author didn't know of). We also
>> have the problem, that our site consists of many different components:
>> there's the homepage, the wiki, emu, SVN, which looks very fragmented.
>>
>> We could address most of this problems by using a CMS (content
>> management system).
> 
> It's not the first time it's being debated...
> 
>> Of course a CMS is not a Swiss Army knife for
>> everything and it does raise several issues: is it fast enough to
>> survive a slashdot, can we use our already existent database, how can we
>> migrate, is it safe?
> 
> Don't worry about performances for now...
> 

I don't but Ian did when I had a discussion with him about the some web
design issues recently where I mentioned CMSs.

> At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
> home-maid website and *loads* of custom scripting for almost
> everything... How do you plan to migrate existing content ?
> 

The fully custom made site is one of the problems, as we are not experts
in some of the things we did. I saw that you fixed some security issues
in our php code today, some issues that dealt with character escaping
and such things. I'm no PHP expert but I think these are things which
are obvious to a professional php-developer but can completely break our
security, which means if some  guy used this issue to hack into our server and replace the
binaries we provide, then this could be rather dangerous for our users.

What I want to say: If you're not absolutely sure about what you're
doing, leave it to the pros, they know how to deal with it, and we can
concentrate on what we do best: provide our users with tools to give
them true freedom of speech.

It's probably not possible to migrate in two days but it seems that now
is a good point to start the process, as Ian mentioned he wanted to
change the website significantly (this also includes the texts). We
probably should migrate in a soft way and try it in a test environment
first. The Website would be a good point to start with because it has
not so much content on it. The other things could be done step by step,
or never if we want to keep them (e.g. I'm not quite convinced about
drupals bug tracker, but there are definitely better wiki engines than
wikkawiki).

> Can a CMS have some level of history ? All the tools we use have
> native versioning; that's a feature we don't want to loose.

Drupal has native versioning, I think that's one of the core features
which made it one of the favourite CMSs for OpenSource projects.

[EMAIL PROTECTED]

-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/



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

[freenet-dev] Content Management System

[Michael Tänzer]

In the last few weeks I've done some work on the website. While
translating it, there were some things that struck me so I changed them.
But our site is still far from perfect. It lacks a attractive design and
some features that would be quite handy (e.g. select the language by
hand, RSS-Feeds, a search) but are a little bit difficult to implement
(at least if we want to do it in a safe and efficient way) or at least I
don't have the time and skills to do it. I also noticed, that the format
we use to save the content (it's just a php file containing HTML which
is included in some kind of very simple template) leaves room for
optimization (for both, the author who needs to write valid HTML and
know about the things he can do with it (not all of us do know how to
write clean and valid HTML (I do not exclude myself from this
statement)), and the user, who might get malformed HTML or ugly pages
because the browser has some bugs the author didn't know of). We also
have the problem, that our site consists of many different components:
there's the homepage, the wiki, emu, SVN, which looks very fragmented.

We could address most of this problems by using a CMS (content
management system). Of course a CMS is not a Swiss Army knife for
everything and it does raise several issues: is it fast enough to
survive a slashdot, can we use our already existent database, how can we
migrate, is it safe?

The three commonly used Open CMS' are:

Typo3 - the elder:
-first release in 1998, therefore probably pretty safe by now
-complicated to administer and design (has it's own template-language)
-but therefore very powerful
-according to some sites Typo3 needs a powerful server

Joomla! - the most used:
-easy to administer and design (at least the last time I used it)
-very big community
-had some security vulnerabilities in the past (hopefully this will have
more or less disappeared with the ground-up rewrite in version 1.5 - the
most vulnerabilities where in third party modules though)

Drupal - the community focused (and therefore my favourite):
-should be as easy as Joomla
-has some features which are especially interesting for communities
(like us - mozilla.org and other OpenSource projects seem to use it too)

All of them are licensed under GPL, they all provide caching techniques
to cope with high traffic, they all can use mySQL, other databases are
also supported. It's important, that the functionality we want to have
is covered with the standard modules as much as possible, third party
modules are a major security risk.

Looking forward to your comments
Neo at NHNG

P.S.: The question whether Joomla! or Drupal is the better CMS seems to
be a question of belief.
-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/

-- next part --
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
URL: 

Re: [freenet-dev] Content Management System

[Florent Daignière]

* Michael Tänzer <[EMAIL PROTECTED]> [2008-05-08 05:04:07]:

> In the last few weeks I've done some work on the website. While
> translating it, there were some things that struck me so I changed them.
> But our site is still far from perfect. It lacks a attractive design and
> some features that would be quite handy (e.g. select the language by
> hand, RSS-Feeds, a search) but are a little bit difficult to implement
> (at least if we want to do it in a safe and efficient way) or at least I
> don't have the time and skills to do it.

Select language by hand is trivial to implement and we can delegate the search
to google so that's trivial too... okay RSS would require some work

> I also noticed, that the format
> we use to save the content (it's just a php file containing HTML which
> is included in some kind of very simple template) leaves room for
> optimization (for both, the author who needs to write valid HTML and
> know about the things he can do with it (not all of us do know how to
> write clean and valid HTML (I do not exclude myself from this
> statement)), and the user, who might get malformed HTML or ugly pages
> because the browser has some bugs the author didn't know of). We also
> have the problem, that our site consists of many different components:
> there's the homepage, the wiki, emu, SVN, which looks very fragmented.
> 
> We could address most of this problems by using a CMS (content
> management system).

It's not the first time it's being debated...

> Of course a CMS is not a Swiss Army knife for
> everything and it does raise several issues: is it fast enough to
> survive a slashdot, can we use our already existent database, how can we
> migrate, is it safe?

Don't worry about performances for now...

At the moment we are using mantis as a BTS, Wikka as a wiki-engine, a
home-maid website and *loads* of custom scripting for almost
everything... How do you plan to migrate existing content ?

Can a CMS have some level of history ? All the tools we use have
native versioning; that's a feature we don't want to loose.


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

[freenet-dev] Content Management System

[Michael Tänzer]

In the last few weeks I've done some work on the website. While
translating it, there were some things that struck me so I changed them.
But our site is still far from perfect. It lacks a attractive design and
some features that would be quite handy (e.g. select the language by
hand, RSS-Feeds, a search) but are a little bit difficult to implement
(at least if we want to do it in a safe and efficient way) or at least I
don't have the time and skills to do it. I also noticed, that the format
we use to save the content (it's just a php file containing HTML which
is included in some kind of very simple template) leaves room for
optimization (for both, the author who needs to write valid HTML and
know about the things he can do with it (not all of us do know how to
write clean and valid HTML (I do not exclude myself from this
statement)), and the user, who might get malformed HTML or ugly pages
because the browser has some bugs the author didn't know of). We also
have the problem, that our site consists of many different components:
there's the homepage, the wiki, emu, SVN, which looks very fragmented.

We could address most of this problems by using a CMS (content
management system). Of course a CMS is not a Swiss Army knife for
everything and it does raise several issues: is it fast enough to
survive a slashdot, can we use our already existent database, how can we
migrate, is it safe?

The three commonly used Open CMS' are:

Typo3 - the elder:
-first release in 1998, therefore probably pretty safe by now
-complicated to administer and design (has it's own template-language)
-but therefore very powerful
-according to some sites Typo3 needs a powerful server

Joomla! - the most used:
-easy to administer and design (at least the last time I used it)
-very big community
-had some security vulnerabilities in the past (hopefully this will have
more or less disappeared with the ground-up rewrite in version 1.5 - the
most vulnerabilities where in third party modules though)

Drupal - the community focused (and therefore my favourite):
-should be as easy as Joomla
-has some features which are especially interesting for communities
(like us - mozilla.org and other OpenSource projects seem to use it too)

All of them are licensed under GPL, they all provide caching techniques
to cope with high traffic, they all can use mySQL, other databases are
also supported. It's important, that the functionality we want to have
is covered with the standard modules as much as possible, third party
modules are a major security risk.

Looking forward to your comments
[EMAIL PROTECTED]

P.S.: The question whether Joomla! or Drupal is the better CMS seems to
be a question of belief.
-- 
Follow the blue rabbit - The Freenet Project - http://freenetproject.org/



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