Re: Packaging third-party applications (Was: Meta Toolchain Release (2008 May))

[Rod Whitby]

Pranav Desai wrote:

My attempt is to make existing opensource apps available for Openmoko.
If I have rewrite or wrap around the existing build process just to
fit the OE model then it seems a bit discouraging.


Contrast this with Debian, Ubuntu, Gentoo, etc ... you're effectively 
saying that Debian repositories with GPG signed packages are not 
required - you are saying that people should just download random 
binaries from random sites and take the consequences, or they should be 
smart enough to build the packages themselves.



Maybe its trivial to include apps into OE, but currently it seems more
than that to me.  And especially, with the toolchain out, it seems even
more easier to build apps as-is and just putting out the binaries for
people to use.


And therein starts the decline of Openmoko application availability into 
the poor practices of the windows world ... encouraging people to 
download and install binary applications with no means of determining 
exactly how they were built (and therefore not able to verify that 
something bad hasn't been included in them).


I thought the Openmoko developer community would want to better than 
that ...


-- Rod



___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Meta Toolchain Release (2008 May)

[Pranav Desai]

On 5/29/08, John Lee <[EMAIL PROTECTED]> wrote:
> On Thu, May 29, 2008 at 06:44:42PM -0700, Pranav Desai wrote:
>> On Wed, May 28, 2008 at 9:34 PM, John Lee <[EMAIL PROTECTED]> wrote:
>> >
>> > libjana isn't packaged in toolchain by default.  the default
>> > compopents are:
>> >
>> > libmokogsmd2, mokoui2, mokopanelui2, mokojournal2 and dependencies.
>> >
>> > we cannot package everything or it's going to be a huge toolchain.
>> > however, there are 2 interesting aliases in environment-setup:
>> >
>>
>> I understand, and you don't have to include all the libraries, but atleast
>> the prerequisite for building a standard openmoko app should be included.
>>
>
> That is exactly the problem.  You see, if we put april software update
> into consideration (these are standard openmoko apps as well), the
> prerequisite will not be small for a toolchain.  I'm still thinking

Not sure what you mean, is it that the dependencies for the April
software are different and hence the toolchains needs both April and
May libs ?

I have just looked at one app (dialer), so you guys know the
dependencies a lot better than I do and maybe they do make the
toolchain too big, but just for the dialer the toolchain was pretty
reasonable size.

> about this out of my personal interest.
>
>
> Regards,
> John
>

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Packaging third-party applications (Was: Meta Toolchain Release (2008 May))

[Pranav Desai]

On 5/29/08, Rod Whitby <[EMAIL PROTECTED]> wrote:
> Pranav Desai wrote:
>> On Wed, May 28, 2008 at 5:49 PM, Rod Whitby <[EMAIL PROTECTED]
>> > wrote:
>> The usual way is to add the package to OpenEmbedded, and then add
>> it's name to the task-openmoko-feed.bb
>>  recipe so that it automatically gets
>> built, packaged and deployed to the official download site.
>>
>> But wouldn't that mean writing a recipe for all packages that we want to
>> add?
>
> That is correct.
>
>> Many third party apps already have a makefile setup, why do you
>> want to change that ?
>
> Writing a recipe does not involve changing the existing Makefile. If the
> existing Makefile is written properly, then the recipe should be about 5
> lines long.
>
> But the major reason to do this is the one I gave below, which you
> didn't comment on.  Security and trust of third-party apps should be a
> very significant consideration for the Openmoko community.
>
>> Also, I am much more likely to trust a package recipe that I can
>> build and install myself using OpenEmbedded (or download from an
>> official site where the package has been built from source by a
>> trusted autobuild system), rather than downloading some unknown,
>> possibly virus-tainted binary package from some random site ...

I agree with that issue, but that choice should be left to the user,
if they don't feel comfortable downloading a binary pkg, they are more
than welcome to get the src and build it themselves.

My attempt is to make existing opensource apps available for Openmoko.
If I have rewrite or wrap around the existing build process just to
fit the OE model then it seems a bit discouraging.

Maybe its trivial to include apps into OE, but currently it seems more
than that to me. And especially, with the toolchain out, it seems even
more easier to build apps as-is and just putting out the binaries for
people to use.

-- Pranav

>
> -- Rod
>

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Freerunner Sale Price in India - Official partner with Openmoko

[Rahul Joshi]

Thanks a heap for this info, it'll certainly help. And its comforting to
know I'm not too far behind this almost-like-a-revolution.
I did get an email from THE man himself mentioning other details pertaining
to purchase of FreeRunner in India. (excited!)

As for the discount, well... congratulations! 15.4K is OMG!! Like they say,
early bird gets the umm.. Neo! :)

Rahul J


On Thu, May 29, 2008 at 11:30 PM, Ganesha Krishna <[EMAIL PROTECTED]>
wrote:

> -- Forwarded message --
>> From: "Rahul Joshi" <[EMAIL PROTECTED]>
>> To: community@lists.openmoko.org
>> Date: Thu, 29 May 2008 17:32:43 +0530
>> Subject: Re: Freerunner Sale Price in India - Official partner with
>> Openmoko
>> Hi everyone,
>>
>> I'm a little (actually quite) late in this scene so pardon my ignorance on
>> any goofy questions. I am a LAMP (and uhm ... Java) developer and had always
>> wanted/waited for a "dream box" like this. Fortunately (or unfortunately) I
>> came across openmoko 1 year late. But better late than never... yeah.
>>
>
> Welcome, I dont think you are very late at all, there were only a few
> neo1973s, the first generation free phones (swathanthra phone as some one
> pointed out)  sold in India (according to my very casual research of
> orkut/blog sphere ) and one had to order them directly from openmoko, no
> reseller. The new hardware (Freerunner) is much more juicier(sic)
> considering the variety in Indian phone market.
>
>>
>> I'll be honest when I say I came pretty close to buying an iPhone when I
>> went to US recently but the whole "Out of Stock" thing drove me mad. In fact
>> I started harboring anti-Apple feelings ($20 for s/w upgrades,
>> overtly-closed-and-controlled UI environment etc. etc.) that I decided to
>> stick with my iMate SP5m until something really gets to me. And here I am.
>
>
> of course, its the 'Handheld of the free and Phone of the brave'.
>
>
>> So, my first few questions (and I'm thrilled to know a thread on Neo
>> Indian retailing exists!!) -
>>
>> 1. Where in India can I purchase Neo FreeRunner and its related
>> accessories once it gets released. I know Rakshat is the man here but can we
>> get some more info on this please? Can I pre-order one (just in case it runs
>> out of stock!)
>>
>
> As far as I know (and I have followed this mailing list since its
> inception) Rakshat is THE Man. you can pre-order at his website.
> These are his mails form the archive
>
> http://openmoko.markmail.org/message/4gek2hynltrysxtw?q=rakshat
> http://openmoko.markmail.org/message/3ytp6yfuxoy6bwfj?q=rakshat
>
>
>
>> 2. Whats this talk of 15.4K discount Mr. Krishna is talking about. How can
>> I be entitled to it as well? :D :P
>
> The first few phones were offered for a discount, I was one of the early
> birds. unfortunately he has announced that all the slots are taken as of now
> barring last minute cancellations.
>
> Dig thru here to for the whole scoop.
> http://openmoko.markmail.org/search/?q=rakshat
>
>
>>
>> Thanks!
>>
>> Rahul J aka sector7
>>
>
> -GK
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>
>
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Meta Toolchain Release (2008 May)

[John Lee]

On Thu, May 29, 2008 at 06:44:42PM -0700, Pranav Desai wrote:
> On Wed, May 28, 2008 at 9:34 PM, John Lee <[EMAIL PROTECTED]> wrote:
> >
> > libjana isn't packaged in toolchain by default.  the default
> > compopents are:
> >
> > libmokogsmd2, mokoui2, mokopanelui2, mokojournal2 and dependencies.
> >
> > we cannot package everything or it's going to be a huge toolchain.
> > however, there are 2 interesting aliases in environment-setup:
> >
> 
> I understand, and you don't have to include all the libraries, but atleast
> the prerequisite for building a standard openmoko app should be included.
> 

That is exactly the problem.  You see, if we put april software update
into consideration (these are standard openmoko apps as well), the
prerequisite will not be small for a toolchain.  I'm still thinking
about this out of my personal interest.


Regards,
John

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Packaging third-party applications (Was: Meta Toolchain Release (2008 May))

[Rod Whitby]

Pranav Desai wrote:
On Wed, May 28, 2008 at 5:49 PM, Rod Whitby <[EMAIL PROTECTED] 
> wrote:

The usual way is to add the package to OpenEmbedded, and then add
it's name to the task-openmoko-feed.bb
 recipe so that it automatically gets
built, packaged and deployed to the official download site.

But wouldn't that mean writing a recipe for all packages that we want to 
add?


That is correct.

Many third party apps already have a makefile setup, why do you 
want to change that ?


Writing a recipe does not involve changing the existing Makefile. If the 
existing Makefile is written properly, then the recipe should be about 5 
lines long.


But the major reason to do this is the one I gave below, which you 
didn't comment on.  Security and trust of third-party apps should be a 
very significant consideration for the Openmoko community.



Also, I am much more likely to trust a package recipe that I can
build and install myself using OpenEmbedded (or download from an
official site where the package has been built from source by a
trusted autobuild system), rather than downloading some unknown,
possibly virus-tainted binary package from some random site ...  


-- Rod

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

gsmd question

[Bin Chen]

Hi,

I am a mobile phone developer and I am looking at the GSMd application
on OpenMoko. Seems all the AT command are sent to UART without waiting
the OK or ERROR response.
Because the intrinsic of AT command, if you don't wait for the
response of OK or ERROR, but send successive (quick enough) AT command
to GSM chip, you will get multiple OK or ERROR and you can't
distinguish which one is correspond to the response, because there is
no ID information in the OK or ERROR.

Please correct me if I am wrong.
Thanks.

Bin

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Packaging third-party applications (Was: Meta Toolchain Release (2008 May))

[Pranav Desai]

On Wed, May 28, 2008 at 5:49 PM, Rod Whitby <[EMAIL PROTECTED]> wrote:

> Pranav Desai wrote:
>
>> But that brings another question (which probably needs another thread),
>>> where do we store/host the ported apps if we have some?
>>> * Can we put it somewhere on downloads.openmoko.org <
>>> http://downloads.openmoko.org>
>>> * Should we create another project on project.openmoko.org <
>>> http://project.openmoko.org>.
>>> * Or do we put the responsibility on whoever ported it, to host the app.
>>>
>>
> The usual way is to add the package to OpenEmbedded, and then add it's name
> to the task-openmoko-feed.bb recipe so that it automatically gets built,
> packaged and deployed to the official download site.
>

But wouldn't that mean writing a recipe for all packages that we want to
add? Many third party apps already have a makefile setup, why do you want to
change that ? Especially, if they can be build easily enough with the
toolchain.


> Packaging and publishing random applications here, there and everywhere is
> just going to fragment the set of applications available to users.


Thats not going to stop anyway, but if we have a central place where people
can put their packages, maybe it can be reduced.


>
> Also, I am much more likely to trust a package recipe that I can build and
> install myself using OpenEmbedded (or download from an official site where
> the package has been built from source by a trusted autobuild system),
> rather than downloading some unknown, possibly virus-tainted binary package
> from some random site ...
>

> -- Rod
>
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Meta Toolchain Release (2008 May)

[Pranav Desai]

On Wed, May 28, 2008 at 9:34 PM, John Lee <[EMAIL PROTECTED]> wrote:

> On Sun, May 25, 2008 at 04:19:37PM -0700, Pranav Desai wrote:
> >
> > I am still having trouble building openmoko-dialer2 with the toolchain. I
> > can build openmoko-sample2 just fine.
> >
> > The toolchain seems to be missing libpulse. Initially, om-conf
> > openmok-dialer2 gave both libebook and libpulse as missing. I found that
> > libebook is present in the toolchain, but was missing the .pc file. After
> > adding that it went ahead and only complained about libpulse.
> >
> > So I started building pulseaudio-0.9.10 with the toolchain. After
> resolving
> > some dependencies like liboil and libsndfile, I was able to build
> libpulse.
> > But now I stuck at libjana.
> >
> > checking pkg-config is at least version 0.9.0... yes
> > checking for DIALER... yes
> > checking for JANA... configure: error: Package requirements (libjana
> > libjana-ecal libjana-gtk) were not met:
> >
> > No package 'libjana' found
> > No package 'libjana-ecal' found
> > No package 'libjana-gtk' found
> >
> > Any ideas, suggestions or can anyone provide with the required libraries
> ?
> >
> > Thanks
> >
> > -- Pranav
>
> libjana isn't packaged in toolchain by default.  the default
> compopents are:
>
> libmokogsmd2, mokoui2, mokopanelui2, mokojournal2 and dependencies.
>
> we cannot package everything or it's going to be a huge toolchain.
> however, there are 2 interesting aliases in environment-setup:
>

I understand, and you don't have to include all the libraries, but atleast
the prerequisite for building a standard openmoko app should be included.


> alias opkg='LD_LIBRARY_PATH=/usr/local/openmoko/arm/lib
> /usr/local/openmoko/arm/bin/opkg-cl -f
> /usr/local/openmoko/arm/etc/opkg-sdk.conf -o /usr/local/openmoko/arm'
>
> alias opkg-target='LD_LIBRARY_PATH=/usr/local/openmoko/arm/lib
> /usr/local/openmoko/arm/bin/opkg-cl -f
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi/etc/opkg.conf -o
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi'
>
> which means by opkg-target we can add the packages we need into the
> toolchain directory.
>
> at the moment the content of
>
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi/etc/opkg.conf
>
> is incorrectly pointed to local directories in build host.  if we
> replace it by the corresponding URIs such as
>
> http://buildhost.openmoko.org/daily-feed/
> http://buildhost.openmoko.org/daily-feed/all
> http://buildhost.openmoko.org/daily-feed/armv4t
> http://buildhost.openmoko.org/daily-feed/neo1973
> http://buildhost.openmoko.org/daily-feed/om-gta01
>
> then _theoretically_ we can just do
>
> opkg-target update
> opkg-target install libjana-dev
>
> I haven't tried this approach before because OE suits me just fine.
> however, for those who just want to make apps I think it's an right way
> to start.
>
>
> Regards,
> John
>
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Can update freerunner CPU to ARM926EJ-S

[luther]

I saw the Neo fr PCB diagram,it same can't do. also they have different pin 
number,and size. :-( 

- Original Message - 
From: "Ian Darwin" <[EMAIL PROTECTED]>
To: "List for Openmoko community discussion" 
Sent: Friday, May 30, 2008 3:35 AM
Subject: Re: Can update freerunner CPU to ARM926EJ-S


>>> The freerunner CPU is ARM920T, can't support android system img, I 
>>> think it can update to ARM926EJ-S ?
> 
> The Neo is not a desktop computer, so I really doubt that the CPU is 
> socketed :-)  So unless you're good at soldering SMT components, I think 
> you can forget the CPU upgrade (even if you are, I have no way of 
> telling whether the upgrade would work or not). So it's a good thing that:
> 
>> AFAIK when Google Android code will be available it will be possible to 
>> port it to work with ARMv4 CPUs.
> 
> 
> 
> 
>___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Marco Trevisan (Treviño)]

Joerg Reisenweber wrote:

Am Fr  30. Mai 2008 schrieb Marco Trevisan (Treviño):
Anyway, I'll appreciate so much if another one of the lucky Freerunner 
owners could make a battery test (also just a "standby" one) in a place 
with low GSM coverage...


Standby doesn't vary with signal-strength. Just receiver is active (except 
T3210 every few hours), so it should be no difference.


Ah... Thanks for the info. I thought this since my actual mobile battery 
really dies so slowly when I'm on places where there's good GSM 
coverage. But maybe it's due to something else!


--
Treviño's World - Life and Linux
http://www.3v1n0.net/


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Michael Shiloh]

Valid idea. Thanks for mentioning this.

Michael

David Samblas Martinez wrote:
  If anyone is really interested in make this phone an AP to whatever 
use  and  if a  external power is  an option, why not use a USB 1.1 
wireless adapter able to  do that even it only do it  in b instead of g 
if the usb 1.1 doen't allow  the 54 Mb/s.it can be very probable that 
can even use monitor mode with that dongle ;) even without external 
power it can be doable for a shorts periods of time (well battery will 
decide the timeframe)



--- El *jue, 29/5/08, Michael Shiloh /<[EMAIL PROTECTED]>/* escribió:

De: Michael Shiloh <[EMAIL PROTECTED]>
Asunto: Re: Neo as cellular modem?
Para: "List for Openmoko community discussion"

Fecha: jueves, 29 mayo, 2008 6:00

Alexey Feldgendler wrote:
> On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]>
wrote:
> 
>> We're helpless unless Atheros decided to implement Master mode in

their
>> closed firmware.  Unfortunately the power advantages of having the
bulk
>> of the ieee80211 actions managed in the firmware are pretty compelling
>> so I don't know how we get out of that bind.
> 
> The very knowledge that we can't do something that the hardware would 
> technically be capable of is annoying, 

Strongly agree. In some cases this was a result of the agreement we were 
able to reach with the chip manufacturer in order to open source the 
driver. We don't view this as a perfect solution, but rather a good 
start.
 Hopefully in the future the success of Openmoko will encourage 
chip manufacturers to become more open.



> but I don't really see why we would need to implement a true AP in the
phone. 

Strongly disagree. Innovation is stifled whenever choices are limited 
simply because we can't think of why someone would want to make that 
choice. We should always strive to make such choices available.





For any reasonable use
> case I can think of, ad-hoc mode should be enough. The only usability 
> advantage of being an AP would be that it can send beacon packets that 
> allow other devices to detect an available network, but sending beacon 
> would be a battery drain anyway.


Assuming the use case made sense, the Freerunner could be powered 
externally.


Michael

___
Openmoko community mailing
 list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community



Enviado desde Correo Yahoo! 


La bandeja de entrada más inteligente.




___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Joerg Reisenweber]

Am Fr  30. Mai 2008 schrieb Marco Trevisan (Treviño):
> ian douglas wrote:
> > I think that knowing a "best case scenario" (where you stay in the same 
> > location), you get about 6 hours of talk time, is still helpful. Cell 
> > phone manufacturers typically report a "best case scenario" when 
> > reporting talk time and standby time, with the legalese and fine print 
> > stating that "your results may vary" from their data.
> > 
> > Thoughts?
> 
> I agree, and we can't ask you more than testing like you're doing.
> 
> Unfortunately I don't live always in places where's there's full GSM 
> signal strength (there are mountains, here! :P), so I already thought to 
> this issue, but I didn't hope in test in this scenario...
> 
> The only thing I'm asking to you, Ian, is to report the GSM signal 
> strength in your testing zone, just to complete the informations you've 
> already given.
> 
> Anyway, I'll appreciate so much if another one of the lucky Freerunner 
> owners could make a battery test (also just a "standby" one) in a place 
> with low GSM coverage...

Standby doesn't vary with signal-strength. Just receiver is active (except 
T3210 every few hours), so it should be no difference.
Talktime goes down with distance^2.5 I guess. 
/j


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Brad Midgley]

David

> If anyone is really interested in make this phone an AP to whatever use
> and  if a  external power is  an option, why not use a USB 1.1 wireless
> adapter able to  do that even it only do it  in b instead of g if the usb
> 1.1 doen't allow  the 54 Mb/s

I've not had much luck finding a solid linux driver for usb-based wifi
adapter as an AP.

RE speed... get it working first, then try to optimize. :)

-- 
Brad

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Joerg Reisenweber]

Am Fr  30. Mai 2008 schrieb AVee:
> This test might not even be 'best case'. A better test would be having the 
Neo 
> really close to the cell tower for optimal conditions. I guess the 
difference 
> between testing far away from the cell tower and testing close to the tower 
> might be pretty big. There probably also is a difference between GSM900 and 
> GSM1800 (iirc 1800 has a lower range which needs to be compensated by higher 
> transmission power).

Nope, 1800 has half the max output, which is compensated by closer grid of BS.
Anyway I guess for "best case" it's no difference.
/j


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[ian douglas]

Thanks for your thoughts, Marco.


Marco Trevisan (Treviño) wrote:

I agree, and we can't ask you more than testing like you're doing.


As of today I have 165 minutes remaining of my 1,000 minute TMobile SIM 
card, so I'll have to restock it to do any more tests. I think TMobile 
still has their "1000-minutes for $100" special going on, so I'll stock 
up later this evening.


The only thing I'm asking to you, Ian, is to report the GSM signal 
strength in your testing zone, just to complete the informations you've 
already given.


Fair enough. I'll report on that when I get home as my office building 
is probably more shielded than my apartment. I live near the beach, no 
mountains to really speak of, and my AT&T phone, which was always 
showing EDGE connectivity at my old apartment always shows 
full-bandwidth "3G" at the new apartment, which is partly why I wanted 
to do some testing around my old apartment.


Anyway, I'll appreciate so much if another one of the lucky Freerunner 
owners could make a battery test (also just a "standby" one) in a place 
with low GSM coverage...


As soon as my Freerunner is back to full-charge, I'll try to emulate 
Einstein's standby tests with all 4 components (GSM, WiFi, Bluetooth, 
GPS) turned on, then off, and report my own findings. I'll try to do it 
with both SIM cards, if that makes any difference, and without any SIM 
at all -- I'm curious if having no SIM card will emulate "low GSM 
coverage" with the phone trying to constantly register on a network.


Since I have 5,000 evening/weekend minutes with AT&T, once I fill up my 
TMobile card again, I'll take my Freerunner and Blackjack 2 on errands 
this weekend, see if I can spend more than 6 hours connecting the two 
phones again in varying locations around Los Angeles.


If anyone else has suggestions on how to test battery life, I'd love to 
hear them. Especially from Michael and Steve, as they can probably 
direct us a little better on what kinds of real-life scenario testing 
they'd like us to help in.


I'm going to see if I can write a CPU-intensive script to run on the 
Freerunner too, see if I can time how long the battery will last with 
the CPU running 100% ... again, just a curiosity.


-id


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[David Samblas Martinez]

If anyone is really interested in make this phone an AP to whatever use  
and  if a  external power is  an option, why not use a USB 1.1 
wireless adapter able to  do that even it only do it  in b instead of 
g if the usb 1.1 doen't allow  the 54 Mb/s.it can be very probable that 
can even use monitor mode with that dongle ;) even without external power it 
can be doable for a shorts periods of time (well battery will decide the 
timeframe)


--- El jue, 29/5/08, Michael Shiloh <[EMAIL PROTECTED]> escribió:
De: Michael Shiloh <[EMAIL PROTECTED]>
Asunto: Re: Neo as cellular modem?
Para: "List for Openmoko community discussion" 

Fecha: jueves, 29 mayo, 2008 6:00

Alexey Feldgendler wrote:
> On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]>
wrote:
> 
>> We're helpless unless Atheros decided to implement Master mode in
their
>> closed firmware.  Unfortunately the power advantages of having the
bulk
>> of the ieee80211 actions managed in the firmware are pretty compelling
>> so I don't know how we get out of that bind.
> 
> The very knowledge that we can't do something that the hardware would 
> technically be capable of is annoying, 

Strongly agree. In some cases this was a result of the agreement we were 
able to reach with the chip manufacturer in order to open source the 
driver. We don't view this as a perfect solution, but rather a good 
start. Hopefully in the future the success of Openmoko will encourage 
chip manufacturers to become more open.


> but I don't really see why we would need to implement a true AP in the
phone. 

Strongly disagree. Innovation is stifled whenever choices are limited 
simply because we can't think of why someone would want to make that 
choice. We should always strive to make such choices available.




For any reasonable use
> case I can think of, ad-hoc mode should be enough. The only usability 
> advantage of being an AP would be that it can send beacon packets that 
> allow other devices to detect an available network, but sending beacon 
> would be a battery drain anyway.

Assuming the use case made sense, the Freerunner could be powered 
externally.

Michael

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community


  __ 
Enviado desde Correo Yahoo! La bandeja de entrada más inteligente.___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: screen protector

[Flemming Richter Mikkelsen]

On 5/29/08, "Marco Trevisan (Treviño)" <[EMAIL PROTECTED]> wrote:
> Ian Darwin wrote:
> > If anybody feels the need to use a screen protector, I just tried one
> called NavProtector that I got for a few bucks on eBay, and it fits nicely
> (cut about 1mm shorter than the screen, so check before you stick! and maybe
> center it).
> >
>
> Good. I've already asked on this list [1] about the real screen size of the
> Freerunner (better if in mm) to get a Brando one (I've heard they're the
> best ones), but I got no answer. :(
>
> Anyway I don't exactly know the size of the Freerunner LCD; I've heard it's
> 46mm x 61mm, right?
>
> Please you or any FR owner could help me? I'd like to buy one before than
> the phone itself :P

I got this info from the wiki [1]:

LCD Module (LCM)
Resistance type touch panel.
2.8" diagonal (1.7" x 2.27" - 43mm x 58mm)
Outline Dimension [2]: 43.2mm x 63.3mm x 1.9mm

[1] http://wiki.openmoko.org/wiki/Neo_FreeRunner_GTA02_Hardware
[2] http://www.tpo.biz/ENG/business-eng/Activer-Matrix-VGA.htm

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Marco Trevisan (Treviño)]

ian douglas wrote:
I think that knowing a "best case scenario" (where you stay in the same 
location), you get about 6 hours of talk time, is still helpful. Cell 
phone manufacturers typically report a "best case scenario" when 
reporting talk time and standby time, with the legalese and fine print 
stating that "your results may vary" from their data.


Thoughts?


I agree, and we can't ask you more than testing like you're doing.

Unfortunately I don't live always in places where's there's full GSM 
signal strength (there are mountains, here! :P), so I already thought to 
this issue, but I didn't hope in test in this scenario...


The only thing I'm asking to you, Ian, is to report the GSM signal 
strength in your testing zone, just to complete the informations you've 
already given.


Anyway, I'll appreciate so much if another one of the lucky Freerunner 
owners could make a battery test (also just a "standby" one) in a place 
with low GSM coverage...


Thanks again!

--
Treviño's World - Life and Linux
http://www.3v1n0.net/


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[AVee]

On Thursday 29 May 2008 21:58, ian douglas wrote:
> Andy Green wrote:
> > | Taking Lasse's advice, I set up a new test last night:
> >
> > Just a little point about these tests, AIUI the GPS stuff acts radically
> > differently in terms of current consumption depending on the distance
> > from the base station.
...
> I think that knowing a "best case scenario" (where you stay in the same
> location), you get about 6 hours of talk time, is still helpful. Cell
> phone manufacturers typically report a "best case scenario" when
> reporting talk time and standby time, with the legalese and fine print
> stating that "your results may vary" from their data.

This test might not even be 'best case'. A better test would be having the Neo 
really close to the cell tower for optimal conditions. I guess the difference 
between testing far away from the cell tower and testing close to the tower 
might be pretty big. There probably also is a difference between GSM900 and 
GSM1800 (iirc 1800 has a lower range which needs to be compensated by higher 
transmission power).

AVee

-- 
When a man sits with a pretty girl for an hour, it seems like a minute.
But let him sit on a hot stove for a minute -- and it's longer than any hour.
That's relativity.
  -- Albert Einstein

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: on screen keyboard enhaced

[George Brooke]

On Thu, 29 May 2008 22:42:41 +0200
christooss <[EMAIL PROTECTED]> wrote:

> One of selected projects in google android is Writing pad which
> enables writing one word per stroke with on screen keyboard
> 
> http://www.youtube.com/watch?v=sBOyGp25sSg
> 
> What do you think. Could this be useful on OpenMoko?
> 
> I think its really inovative usage of on screen keyboard. I could try
> to implement it on desktop machine and than when I get Freerunner
> port it to OpenMoko.
> 

How about this (http://www.strout.net/info/ideas/hexinput.html) looks
like a good idea.

solar.george

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb ian douglas:
> I think that knowing a "best case scenario" (where you stay in the same 
> location), you get about 6 hours of talk time, is still helpful. Cell 
> phone manufacturers typically report a "best case scenario" when 
> reporting talk time and standby time, with the legalese and fine print 
> stating that "your results may vary" from their data.
> 
> Thoughts?

I think *not* moving for all tests (and different types of cellphones to 
compare) is near a "best case" scenario for standby time - anyway place some 
sensitive radio or the like near the phone, to hear the typical interference 
noise when it is sending, just to make sure you don't sit on a "bad spot" 
where the phone changes cell every few minutes.

For "best case" talktime scenarios the distance to basestation is much more 
important. Here you should check for *very* good RF-signal, means very near 
to BS and thus allowing the phone-transmitter to power down to lowest level.

Also note that GSM without simcard is constantly reselecting cells, so energy 
consumption is really bad. Switch off GSM when not registering to a network.
/jOERG


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

on screen keyboard enhaced

[christooss]

One of selected projects in google android is Writing pad which enables 
writing one word per stroke with on screen keyboard


http://www.youtube.com/watch?v=sBOyGp25sSg

What do you think. Could this be useful on OpenMoko?

I think its really inovative usage of on screen keyboard. I could try to 
implement it on desktop machine and than when I get Freerunner port it 
to OpenMoko.


--
Please don't send me Word or PowerPoint attachments. 
See http://www.gnu.org/philosophy/no-word-attachments.html



___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Private data protection.

[Christoph Fink]

Ilja O. wrote:

1) Auth using PIN number (this requires encrypted image presence in
phone file system by it's boot time end -- not reallyl convenient if
SD card is used).
  
IMO encrypting Data with the PIN Number is not such a good thing, 
because the possibilities of different keys are definately not high (max 
8 digits, only numbers). A better solution would be to save the PIN on 
the encrypted storage and automatically read it.

2) Auth using key file accessible on network (when phone is connected
to your computer or local network). This means that auth can be
performed only in your place (home, work...).
  
Not such a good solution for a Phone IMO (if you're in holidays without 
laptop for example)

3) Auth using presence of another bluetooth or WiFi device (the MAC
address of this device is used as key). This means that phone fully
unlocks when your bluetooth mouse or router are around. ;)

  

Same here...

The most secure Thing IMO would be to ask a passprase while booting 
(maybe hack uboot for that, if it doesn't know that) or put data on 
another device/partition then the OS and asking later for the passphrase 
(maybe easier to implement)


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[ian douglas]

Andy Green wrote:

| Taking Lasse's advice, I set up a new test last night:

Just a little point about these tests, AIUI the GPS stuff acts radically
differently in terms of current consumption depending on the distance
from the base station.


I'm assuming you meant "GSM" not "GPS", so as I understand your point, a 
better test would be actually having the Neo move around geographically 
to hop to/from different cell towers, and that a test like this will 
give more realistic battery usage statistics?


In theory, it sounds very reasonable if it will draw varying amount of 
current from the battery. Thanks for the idea.


Given the cost of gasoline these days, though much lower than in Europe, 
I'm not sure spending two tanks of gas to drive around for 4-6 hours to 
achieve more usage stats will be feasible.


I think that knowing a "best case scenario" (where you stay in the same 
location), you get about 6 hours of talk time, is still helpful. Cell 
phone manufacturers typically report a "best case scenario" when 
reporting talk time and standby time, with the legalese and fine print 
stating that "your results may vary" from their data.


Thoughts?

-id

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb Andy Green:
> Somebody in the thread at some point said:
> | Lasse Poulsen wrote:
> |> Also it would be nice to see how long call time you have if you talk
> |> continually (might i suggest an audio-book or to). If you don't i
> |> properly will (haven't got the device yet!)
> |
> |
> | Taking Lasse's advice, I set up a new test last night:
> 
> Just a little point about these tests, AIUI the GPS stuff acts radically
s/GPS/GSM/ ;)

> differently in terms of current consumption depending on the distance
> from the base station.  All we can reasonably do is compare same-tester
> results for their different phones from the same physical location.

Yep! exactly, due to tx-power calibration, cell-handover etc. 

/jOERG


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Can update freerunner CPU to ARM926EJ-S

[Ian Darwin]

The freerunner CPU is ARM920T, can't support android system img, I 
think it can update to ARM926EJ-S ?


The Neo is not a desktop computer, so I really doubt that the CPU is 
socketed :-)  So unless you're good at soldering SMT components, I think 
you can forget the CPU upgrade (even if you are, I have no way of 
telling whether the upgrade would work or not). So it's a good thing that:


AFAIK when Google Android code will be available it will be possible to 
port it to work with ARMv4 CPUs.




___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[Marco Trevisan (Treviño)]

Bastian Muck wrote:
I have a little idea. Imagine, you put your finger somewhere on the 
right side of the screen. The "average" point is also somewhere at the 
right side. When you put another finger e.g. at the left side the 
"average" point jumps somewhere to the middle. This "jump" is where we 
could guess, that at least two fingers touch the screen. You can't use  
features  like zooming on iPhone, but it should be possible to switch 
scrolling into something else. So with one finger you could scroll the 
screen an with two fingers you could move something.


Ok... Good that's the same I get in my pc touchpad: I can scroll also 
with two fingers (MacBook like).


Can this be grabbed? What about multiple-fingers-tapping?

--
Treviño's World - Life and Linux
http://www.3v1n0.net/


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Can update freerunner CPU to ARM926EJ-S

[Marco Trevisan (Treviño)]

luther ha scritto:
The freerunner CPU is ARM920T ,can't support android system img,I think 
it can update to ARM926EJ-S ?


AFAIK when Google Android code will be available it will be possible to 
port it to work with ARMv4 CPUs.


--
Treviño's World - Life and Linux
http://www.3v1n0.net/


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: screen protector

[Marco Trevisan (Treviño)]

Ian Darwin wrote:
If anybody feels the need to use a screen protector, I just tried one 
called NavProtector that I got for a few bucks on eBay, and it fits 
nicely (cut about 1mm shorter than the screen, so check before you 
stick! and maybe center it).


Good. I've already asked on this list [1] about the real screen size of 
the Freerunner (better if in mm) to get a Brando one (I've heard they're 
the best ones), but I got no answer. :(


Anyway I don't exactly know the size of the Freerunner LCD; I've heard 
it's 46mm x 61mm, right?


Please you or any FR owner could help me? I'd like to buy one before 
than the phone itself :P


Thanks!

[1] http://tinyurl.com/59fay8

--
Treviño's World - Life and Linux
http://www.3v1n0.net/


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Freerunner CPU-scaling

[JW]

2008/5/29 Dennis Andersen <[EMAIL PROTECTED]>:

> Anybody who knows, if the current software support CPU-scaling?
>
> And also if so, are the steps predefined or is it possible to specify a
> precise freq?


cesarb (see the wiki) did a big effort on this for GTA01
from what little i picked up its tricky as you have many cores on this
device each of which have a limited set of freqs which will work (esp the
LCM)
runnable freqs are therefore a very small subset

ah here is the wiki page

http://wiki.openmoko.org/wiki/Neo1973_GTA01_Power_Management
http://wiki.openmoko.org/wiki/User:CesarB/cpufreq

JW
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Lasse Poulsen wrote:
|> Also it would be nice to see how long call time you have if you talk
|> continually (might i suggest an audio-book or to). If you don't i
|> properly will (haven't got the device yet!)
|
|
| Taking Lasse's advice, I set up a new test last night:

Just a little point about these tests, AIUI the GPS stuff acts radically
differently in terms of current consumption depending on the distance
from the base station.  All we can reasonably do is compare same-tester
results for their different phones from the same physical location.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+9EcACgkQOjLpvpq7dMpslACdH6bGOPNwScfU0UgjoiR/kUDh
bgAAn2Vtt/O8ZhkgpiT8X6g3iEgEhwSa
=ctuM
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Michael Shiloh]

Joerg Reisenweber wrote:

Am Do  29. Mai 2008 schrieb Michael Shiloh:

Alexey Feldgendler wrote:

On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]> wrote:


We're helpless unless Atheros decided to implement Master mode in their
closed firmware.  Unfortunately the power advantages of having the bulk
of the ieee80211 actions managed in the firmware are pretty compelling
so I don't know how we get out of that bind.
The very knowledge that we can't do something that the hardware would 
technically be capable of is annoying, 
Strongly agree. In some cases this was a result of the agreement we were 
able to reach with the chip manufacturer in order to open source the 
driver. We don't view this as a perfect solution, but rather a good 
start. Hopefully in the future the success of Openmoko will encourage 
chip manufacturers to become more open.



but I don't really see why we would need to implement a true AP in the 
phone. 
Strongly disagree. Innovation is stifled whenever choices are limited 
simply because we can't think of why someone would want to make that 
choice. We should always strive to make such choices available.





For any reasonable use
case I can think of, ad-hoc mode should be enough. The only usability 
advantage of being an AP would be that it can send beacon packets that 
allow other devices to detect an available network, but sending beacon 
would be a battery drain anyway.
Assuming the use case made sense, the Freerunner could be powered 
externally.


Michael


Not if you don't find a "3 word slogan" to sell this feature to your granny.
SCNR, still got a "no Joerg!" trauma ;-)


I believe there is a legal expression, something like "the grandfather 
clause". Has Steve now invented a similar marketing expression, the 
granny test?


:-)

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Freerunner Sale Price in India - Official partner with Openmoko

[Ganesha Krishna]

>
> -- Forwarded message --
> From: "Rahul Joshi" <[EMAIL PROTECTED]>
> To: community@lists.openmoko.org
> Date: Thu, 29 May 2008 17:32:43 +0530
> Subject: Re: Freerunner Sale Price in India - Official partner with
> Openmoko
> Hi everyone,
>
> I'm a little (actually quite) late in this scene so pardon my ignorance on
> any goofy questions. I am a LAMP (and uhm ... Java) developer and had always
> wanted/waited for a "dream box" like this. Fortunately (or unfortunately) I
> came across openmoko 1 year late. But better late than never... yeah.


Welcome, I dont think you are very late at all, there were only a few
neo1973s, the first generation free phones (swathanthra phone as some one
pointed out)  sold in India (according to my very casual research of
orkut/blog sphere ) and one had to order them directly from openmoko, no
reseller. The new hardware (Freerunner) is much more juicier(sic)
considering the variety in Indian phone market.

>
> I'll be honest when I say I came pretty close to buying an iPhone when I
> went to US recently but the whole "Out of Stock" thing drove me mad. In fact
> I started harboring anti-Apple feelings ($20 for s/w upgrades,
> overtly-closed-and-controlled UI environment etc. etc.) that I decided to
> stick with my iMate SP5m until something really gets to me. And here I am.


of course, its the 'Handheld of the free and Phone of the brave'.


> So, my first few questions (and I'm thrilled to know a thread on Neo Indian
> retailing exists!!) -
>
> 1. Where in India can I purchase Neo FreeRunner and its related accessories
> once it gets released. I know Rakshat is the man here but can we get some
> more info on this please? Can I pre-order one (just in case it runs out of
> stock!)
>

As far as I know (and I have followed this mailing list since its inception)
Rakshat is THE Man. you can pre-order at his website.
These are his mails form the archive

http://openmoko.markmail.org/message/4gek2hynltrysxtw?q=rakshat
http://openmoko.markmail.org/message/3ytp6yfuxoy6bwfj?q=rakshat



> 2. Whats this talk of 15.4K discount Mr. Krishna is talking about. How can
> I be entitled to it as well? :D :P

The first few phones were offered for a discount, I was one of the early
birds. unfortunately he has announced that all the slots are taken as of now
barring last minute cancellations.

Dig thru here to for the whole scoop.
http://openmoko.markmail.org/search/?q=rakshat


>
> Thanks!
>
> Rahul J aka sector7
>

-GK
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb Michael Shiloh:
> 
> Alexey Feldgendler wrote:
> > On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]> wrote:
> > 
> >> We're helpless unless Atheros decided to implement Master mode in their
> >> closed firmware.  Unfortunately the power advantages of having the bulk
> >> of the ieee80211 actions managed in the firmware are pretty compelling
> >> so I don't know how we get out of that bind.
> > 
> > The very knowledge that we can't do something that the hardware would 
> > technically be capable of is annoying, 
> 
> Strongly agree. In some cases this was a result of the agreement we were 
> able to reach with the chip manufacturer in order to open source the 
> driver. We don't view this as a perfect solution, but rather a good 
> start. Hopefully in the future the success of Openmoko will encourage 
> chip manufacturers to become more open.
> 
> 
> > but I don't really see why we would need to implement a true AP in the 
phone. 
> 
> Strongly disagree. Innovation is stifled whenever choices are limited 
> simply because we can't think of why someone would want to make that 
> choice. We should always strive to make such choices available.
> 
> 
> 
> 
> For any reasonable use
> > case I can think of, ad-hoc mode should be enough. The only usability 
> > advantage of being an AP would be that it can send beacon packets that 
> > allow other devices to detect an available network, but sending beacon 
> > would be a battery drain anyway.
> 
> Assuming the use case made sense, the Freerunner could be powered 
> externally.
> 
> Michael

Not if you don't find a "3 word slogan" to sell this feature to your granny.
SCNR, still got a "no Joerg!" trauma ;-)

/j


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: early experince Freerunners w/ ASU load, vs. thousnads of Freeruners

[Federico Lorenzi]

On Thu, May 29, 2008 at 3:37 AM, Joerg Reisenweber <[EMAIL PROTECTED]> wrote:
> Am Do  29. Mai 2008 schrieb Marco Trevisan (Treviño):
>> In this Linuxtag report [1] (google translated here [2]) there are not
>> so good news about production :/
>> Can you share with us something more Steve?
>>
>> Bye
>>
>> [1] http://tinyurl.com/6rs6j2
>> [2] http://tinyurl.com/643y64
>>
>> --
>> Treviño's World - Life and Linux
>> http://www.3v1n0.net/
>
> I *REALLY* dislike those "tinyurl" - never give them a try. :-/
> nearly as bad as those html-only (dunno from whom) postings i use to ignore.
> just wanted to let U know
Maybe everyone should use TinyURL with preview?

Cheers,
Federico

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[ian douglas]

Lasse Poulsen wrote:

Also it would be nice to see how long call time you have if you talk
continually (might i suggest an audio-book or to). If you don't i
properly will (haven't got the device yet!)



Taking Lasse's advice, I set up a new test last night:

I fully charged my Freerunner, inserted the TMobile SIM card, and set it 
beside my computer speakers with Amarok looping some of my more rockin' 
tunes. I made sure the only thing left 'on' on the Freerunner was the 
GSM modem (wifi which is on by default, was turned off). I also enabled 
power-saving (dim, no lock -- but it locks anyway)


I called it from my AT&T phone, which I plugged in next to my TV and 
watched two full movies from Netflix and 3 episodes of The Dead Zone.


In that span, where both phones would have audio to play back and forth 
to one another, my AT&T phone beeped after 3 hours, 51 minutes and 11 
seconds that the connection was broken. I immediately redialed the 
Freerunner, and since it was already at about 90 minutes into the second 
call, I went to bed.


I got up this morning to see that my AT&T phone ended the call after 2 
hours, 11 minutes and 21 seconds. I figured this went one of two ways -- 
the Freerunner ran out of battery power, or I ran out of minutes from 
TMobile.


My TMobile SIM card was a 1,000-minute pay-as-you-go SIM, which is about 
16.7 hours of talk time, which I've only ever used for my Freerunner 
tests ... so having only three 3:51:00 phone calls (give or take a few 
minutes) and one 2:11:21 call that I still have plenty of minutes left 
on my SIM card, and sure enough -- the battery on the Freerunner was 
completely dead.


(as a side note, it's nice to see that a deeply discharged Freerunner 
doesn't have the same issue as the GTA01 where you have to charge it for 
an hour before anything shows on the screen)


Since I have TMobile minutes left, I'll finally get around to testing 
the phone from various alternate locations this coming weekend with the 
TMobile SIM, and I'll swap it out for the AT&T SIM if needed.


To recap:
- GSM turned on
- wifi turned off
- bluetooth turned off
- gps turned off
- power saving enabled (dim, no lock)
- total talk time of 6:02:32 with constant audio on both sides of the 
'conversation'


Hope this helps.

More tests to come, stay tuned!

And thanks to those who sent messages (and Steve confirming on the list) 
that my Freerunner is NOT running the ASU stack. I've watched the 
YouTube video of the ASU stack and it's completely different. My bad.


Ian Douglas
(not Ian Darwin)

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-touch?

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb Bastian Muck:
> I have a little idea. Imagine, you put your finger somewhere on the
> right side of the screen. The "average" point is also somewhere at the
> right side. When you put another finger e.g. at the left side the
> "average" point jumps somewhere to the middle. This "jump" is where we
> could guess, that at least two fingers touch the screen. You can't use

It's the same as sliding with the one finger from right to middle. Remember 
the pressure is also important for the "average", so if you hit really hard 
with the left finger, the "avg" still does no "jump" but a fast move over the 
middle some way left and then back to middle. still a single point gesture. 
Allegedly there is (or has been) sth like "pressure detection" in the 
ts-driver, that probably meassures the R *between* the 2 foils. IIRC that 
couldn't be done with GTA01, but is maybe feasible with GTA02. It would give 
some additional info to maybe distinguish multitouch from gesture. I already 
said I'll have a look at it some day...
/jOERG




signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[Mikko Rauhala]

On to, 2008-05-29 at 17:20 +0200, Bastian Muck wrote:
> I have a little idea. Imagine, you put your finger somewhere on the 
> right side of the screen. The "average" point is also somewhere at the 
> right side. When you put another finger e.g. at the left side the 
> "average" point jumps somewhere to the middle. This "jump" is where we 
> could guess, that at least two fingers touch the screen.

Yes, this idea has been bounced around on the IRC at least (not to
belittle you coming up with it also :] ). One could indeed do limited
multi-touch emulation, recognizing quick drags and extrapolating the
position of the second finger to be at the point twice the distance from
the drag startpoint as the drag endpoint. AFAIK nobody is actually doing
that and it has its limits as you indicate (no way to reliably know
which finger is moving after the dual-tap and such); however, the way is
clear for somebody interested to experiment...

-- 
Mikko Rauhala <[EMAIL PROTECTED]>
University of Helsinki


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Freerunner CPU-scaling

[Dennis Andersen]

Anybody who knows, if the current software support CPU-scaling?

And also if so, are the steps predefined or is it possible to specify a
precise freq?
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Available Encryption algorithms

[Onno van Eijk (DT)]

> see http://en.wikipedia.org/wiki/A5/2
> as well as http://en.wikipedia.org/wiki/A5/1
> and http://en.wikipedia.org/wiki/A5/3
> 
> seems one doesn't want A5/2 anyways and its deprecated, thus eventually
> the network doesn't allow using it anymore?
That's true,

> just guessing.. would need to work through a pile of calypso docs for
> more details, and still not know whats implemented in that specific
> revision of the ti-libs anyhow.. but after reading that i think a5/1 and
> 5/3 will need to be enough till we need 3g anyhow. (when gsm gets phased
> out, somewhere in the future or THC is successful ;)

Reading the wikipedia pages you suggested, I noticed A5/3 is a 3G. It is
quite odd and 2G phone/chip would support an 3G encryption algorithm

Anyway, thanks for you input. I do not suspect you to work through a
pile of calypso docs, I guess you've got better ways to spend you're
time :-)

grz,
Onno


signature.asc
Description: This is a digitally signed message part
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

screen protector

[Ian Darwin]

If anybody feels the need to use a screen protector, I just tried one 
called NavProtector that I got for a few bucks on eBay, and it fits 
nicely (cut about 1mm shorter than the screen, so check before you 
stick! and maybe center it).


The touch screen still works! :-)

Ian Darwin

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Fwd: [Routing] http://www.openrouteservice.org

[Kyle Gordon]

This may be relevant to those folks interested in developing routing apps on 
OpenMoko.

Kyle

--  Forwarded Message  --

Subject: [Routing] http://www.openrouteservice.org
Date: Wednesday 28 May 2008 17:34
From: Pascal Neis <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]

Hi,
I want to tell that I develop a new OSM-Routing application as a Web-service
 and website.

A first version for complete Germany (over 1,4 million street segements) just
 went online in a
very first prototype version at: http://www.openrouteservice.org

The interesting thing is that it uses both free and open OSM street data and
 also an "open" international standard for the routing API. It is the Open
 Location Services Route Service Specification of the Open Geospatial
 Consortium (OGC).

It also supports "avoidareas" which are regions that shall be avoided by the
 routing algorithm.
Possible usage scenarios include disasters, traffic jams, construction works
 etc.

At the moment the first version of the interactive Website is online, but the
 idea is to publish also the
URL for the web-service itself, so that it can be used as OGC Web Service
 using a standardised
OpenLS XML-Request. This would allow to use the service easily - and in an
 interoperable way
 in your own applcaitions. Some more information on this idea can be found in
 our publications: see
at http://www.openrouteservice.org -> Info.

Of course we plan to integrate other countries, too, but first we need to
 improve the application,
adding data then is easy. As we have implemented a range of further OGC
 OpenLS core services.
in the next days further functionalities and services will be added, such as
 a OGC Geocoder
(OpenLS Location Utility Service Specification) using OSM Data ...

best whishes
   pascal

http://www.geographie.uni-bonn.de/karto/


___
Routing mailing list
[EMAIL PROTECTED]
http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/routing

---

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Qtopia Contacts issue

[Ian Darwin]

I've never used QTopia but how about something like "by groups"?  A lot 
of phones offer the option of grouping contacts (work, family, and even 
user defined groups).  It would be nice to extend this and then be able 
to filter by group?   Perhaps QTopia already has that feature?


Yes, Qtopia has groups. But with 800 contacts, even that is unwieldy, 
and I'd still like to have them shown by a sensible (to me :-)) order.


Or maybe I should prune my contacts, but even if I got it down to 400, 
I'd still have the same issues.


It is confounded by the fact that a useful QTopia feature (a feature of 
most phones now, in fact) doesn't work yet. In stock Qtopia, if you type 
a few characters of a contact name, the displayed list is filtered. This 
doesn't work now, because (I'm guessing here!) the Qwerty widget doesn't 
know where to send the characters, or, the Contacts program doesn't know 
how to select the input, or something. Hopefully Carsten will now chime 
in and say that this, too, like every single thing I've complained to 
him (except my one feature request) is already fixed in the source, and 
is working its way out to the builds.  BTW, this is another place that 
you simply must have (and this is my feature request) the ability to 
send letters one at a time to the application.


Cheers
Ian Darwin

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Michael Shiloh]

Alexey Feldgendler wrote:

On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]> wrote:


We're helpless unless Atheros decided to implement Master mode in their
closed firmware.  Unfortunately the power advantages of having the bulk
of the ieee80211 actions managed in the firmware are pretty compelling
so I don't know how we get out of that bind.


The very knowledge that we can't do something that the hardware would 
technically be capable of is annoying, 


Strongly agree. In some cases this was a result of the agreement we were 
able to reach with the chip manufacturer in order to open source the 
driver. We don't view this as a perfect solution, but rather a good 
start. Hopefully in the future the success of Openmoko will encourage 
chip manufacturers to become more open.



but I don't really see why we would need to implement a true AP in the phone. 


Strongly disagree. Innovation is stifled whenever choices are limited 
simply because we can't think of why someone would want to make that 
choice. We should always strive to make such choices available.





For any reasonable use
case I can think of, ad-hoc mode should be enough. The only usability 
advantage of being an AP would be that it can send beacon packets that 
allow other devices to detect an available network, but sending beacon 
would be a battery drain anyway.


Assuming the use case made sense, the Freerunner could be powered 
externally.


Michael

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: A disturbance in the force?

[Bastian Muck]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I guess, that it is a driver issue. The open drivers have to be done by 
reverse engeneering (at least until somwhere round february) and the 
proprietary drivers have many bugs. I hope things will get better since 
the documentation has been made open by AMD.


Greetings Bastian

Marc-Olivier Barre schrieb:
| On Wed, May 28, 2008 at 6:30 PM, Ian Darwin <[EMAIL PROTECTED]> wrote:
|> Once today, while my FreeRunner was plugged in on USB, I tapped the 
screen

|> to wake it up, and it came up, but all the pixels in text were jiggly, as
|> though the screen were being refreshed at the wrong rate. When I 
called up

|> the qwerty keyboard, it appeared quite scrambled.
|> Sadly I didn't have a video camera trained on it at the time, and by the
|> time I whipped out my pocket video camera, it had gone into suspend mode;
|> when I hit the power button to awaken it, the problem was gone!
|>
|> I know it's not my eyesight :-) This really happened. But only once.
|>
|> I'll make sure I have the video camera handy in case it comes back,
|> although, recording a 640x480 video of the freerunner will be pretty 
grainy
|> and probably won't accurately show the problem (due to things like 
refresh

|> rates)... Sigh.
|
| I think I see what you mean.
|
| the same thing happens on my Asus laptop (ATI X700 graphic card) when
| I switch from the console to X. Sometimes the screen appears like what
| you'll get when using the wrong refresh rate on a CRT screen.
| switching again to and from a console makes it go away. Never knew
| what was causing it though.
|
| Driver issue or X issue ?...
|
| Regards,
| __
| Marc-Olivier Barre,
| MarcO'Chapeau.
|
| ___
| Openmoko community mailing list
| community@lists.openmoko.org
| http://lists.openmoko.org/mailman/listinfo/community
|

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.7 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIPs56lYiDScJJ+7QRArUzAKDB7VdDcUBeA5b03Iwn2NrUC4WcEwCg586N
JLW0pnDHr9oUVp+m+MmIlTs=
=o+5Z
-END PGP SIGNATURE-


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[Bastian Muck]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I have a little idea. Imagine, you put your finger somewhere on the 
right side of the screen. The "average" point is also somewhere at the 
right side. When you put another finger e.g. at the left side the 
"average" point jumps somewhere to the middle. This "jump" is where we 
could guess, that at least two fingers touch the screen. You can't use  
features  like zooming on iPhone, but it should be possible to switch 
scrolling into something else. So with one finger you could scroll the 
screen an with two fingers you could move something.

The only problem ist the accuracy of it.

Greetings Bastian

Joerg Reisenweber schrieb:
| Am Do  29. Mai 2008 schrieb [EMAIL PROTECTED]:
|> Hi Joerg,
|> as it seems that my mail did not reach the community list (I am not a
| subscriber, just monitoring the archives) may I ask you directly the 
question

| regarding "multi tutch" :-) . See below:
|>
|> Hi all,
|> if the geometric average point is returned if multiple points are touched
| wouldn't it be possible to have a fake multi touch, e.g., by doing the
| following:
|> Press point A
|> -> Vector A is returned
|> Additionally press desired point B
|> -> X = (A+B)/2 is returned
|> -> calculate B = 2*X-A
|>
|> Questions:
|> Would this be managable?
|> What time difference would be required between first and second (and
| third ...) touch, to recognize such a multi-touch action and reliable
| distinguish it from a single touch at position X?
|> And one more comment: If that would work you could also detect 
multi-touch

| gestures (even if limited), e.g., use the first point statically (maybe a
| focus centre) and handle the second point dynamically (maybe for 
zooming in

| and out).
|>
|> Boris
|
|
| there are two resistors in the touchpad, one for X and one for Y. We 
only see

| them changing on a touch, thus giving us the coords. A concurrent second
| touch just changes the R values even more, but we get no info about 
whether

| we touched a second point or we moved the finger. So what you suggest is a
| gesture recognition. Things become nasty because even the "geometric 
middle"

| isn't true but depends on pressure of one to pressure of other touchpoint
| etc.
| You really get too few info out of the device to do anything 
reasonable with

| it beyond singletouch (at least that's the way it is now. I plan to see
| whether we can exploit dynamic pulse response of this design to get some
| additional info. Same way you're testing cat5-cable for breaks. Don't hold
| your breath though, chances are bad. And it's not on top of my todo-list)
|
| cheers
| jOERG
|
| -
|
| ___
| Openmoko community mailing list
| community@lists.openmoko.org
| http://lists.openmoko.org/mailman/listinfo/community

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.7 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFIPsnJlYiDScJJ+7QRApJnAKDQuEB1lymxNvCpsIqHHkeq96nuiQCePHTW
7ValrHjGuUuLzk8n5P6PuP4=
=0K1/
-END PGP SIGNATURE-


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[Piotr Duda]

also they used to claim that windows is best operating system ever :-)
so be careful ;-)

Piotr

ramsesoriginal pisze:

More or less I knew that, but, as you can see in the video/read in the
news, microsoft stated tat the Windows 7 Multitouch works on "standard
touchscreens"..




___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: QTopia Contacts issue

[Lowell Higley]

On Wed, May 28, 2008 at 9:35 AM, Ian Darwin <[EMAIL PROTECTED]> wrote:

> Now that I have so many contacts, it would be nice if you had more control
> over the list format. The only choices I see (under Options->Display
> Settings) are "First Last or Company" and
> "Last,First or Company". I'd like to see an option that allowed company
> names and person names to be intermixed, so I had one big alphabetical list.
>
> Should this kind of thing go to the QTopia mailing list or forums, or
> should we still post general QTopia issues found under OM to this list?
>
>
I've never used QTopia but how about something like "by groups"?  A lot of
phones offer the option of grouping contacts (work, family, and even user
defined groups).  It would be nice to extend this and then be able to filter
by group?   Perhaps QTopia already has that feature?

I can't answer the original question but I thought I would throw that tidbit
in.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Will GTK be used in Openmoko?

[Arthur Marsh]

Steven Kurylo wrote, on 20/05/08 10:44:

On Mon, May 19, 2008 at 6:03 PM, Rod Whitby <[EMAIL PROTECTED]> wrote:

Marco Trevisan (Treviño) wrote:

Well, I much appreciate your work and your openness with community,
unfortunately I can't say the same about Openmoko in this occasion since
this should be an Open company and so I'd have appreciated it more if the
decision would have been debated before with developers and active part of
community (= people writing code) in public lists.

I guess the question is whether Openmoko Inc. ever promised that the
contents of the rootfs for the phone that they sell will be determined by
some sort of community consensus.

If I remember correctly, they only promised to provide an open software
platform upon which *you* can create your own personalised rootfs.  By
including both toolkits, they have not changed anything you experience with
respect to this promise.


And you'd end up arguing about the colour of the bike shed none stop.
Some decisions openmoko just needs make to deliver us a phone.




For those who didn't pick the reference, the "executive bicycle shed" is 
 a project listed in the book "Parkinson's Law" that takes forever 
while more important issues are decided immediately without forming 
commitees.


Arthur.


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: early experince Freerunners w/ ASU load, vs. thousnads of Freeruners

[Ortwin Regel]

As far as I can tell everything in the article was already mentioned
on the list. Nothing new to comment on...

Ortwin

On 5/29/08, "Marco Trevisan (Treviño)" <[EMAIL PROTECTED]> wrote:
> Marco Trevisan (Treviño) wrote:
>> steve wrote:
>>> Yes mass pro will start soon. There is another batch of a few hundred
>>> or so phones being built tomorrow.
>>> I'll  Know more after tomorrows build.
>>
>> Thanks for the news Steve!
>> I'm waiting for a better ones tomorrow... :P
>
> In this Linuxtag report [1] (google translated here [2]) there are not
> so good news about production :/
> Can you share with us something more Steve?
>
> Bye
>
> [1] http://tinyurl.com/6rs6j2
> [2] http://tinyurl.com/643y64
>
> --
> Treviño's World - Life and Linux
> http://www.3v1n0.net/
>
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: A disturbance in the force?

[Marc-Olivier Barre]

On Wed, May 28, 2008 at 6:30 PM, Ian Darwin <[EMAIL PROTECTED]> wrote:
> Once today, while my FreeRunner was plugged in on USB, I tapped the screen
> to wake it up, and it came up, but all the pixels in text were jiggly, as
> though the screen were being refreshed at the wrong rate. When I called up
> the qwerty keyboard, it appeared quite scrambled.
> Sadly I didn't have a video camera trained on it at the time, and by the
> time I whipped out my pocket video camera, it had gone into suspend mode;
> when I hit the power button to awaken it, the problem was gone!
>
> I know it's not my eyesight :-) This really happened. But only once.
>
> I'll make sure I have the video camera handy in case it comes back,
> although, recording a 640x480 video of the freerunner will be pretty grainy
> and probably won't accurately show the problem (due to things like refresh
> rates)... Sigh.

I think I see what you mean.

the same thing happens on my Asus laptop (ATI X700 graphic card) when
I switch from the console to X. Sometimes the screen appears like what
you'll get when using the wrong refresh rate on a CRT screen.
switching again to and from a console makes it go away. Never knew
what was causing it though.

Driver issue or X issue ?...

Regards,
__
Marc-Olivier Barre,
MarcO'Chapeau.

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My second battery life test with the freerunner in standby

[Flemming Richter Mikkelsen]

On 5/28/08, einstein <[EMAIL PROTECTED]> wrote:
> Hi,
>
> the day is over an the freerunner turns off after 16hours. I charged it
> over night, 7 hours long. So i only turn on the gsm. I don't use the
> Freerunner often. A short gps session of 15min an some sms, one short
> call, one missed call! And in the afternoon i wake up the freerunner
> every 30min to check the state of the device.
>
> I am supriesed of this. yesterday i think 14 hours are possible.
>
> Well, i think all the developers do a great job, and there are many
> opertunitys for more standby time.
>
> So now i go to bed, in 7 hours my night is over.
>

Great job.

Can you say what the reasons are for the improved time?
Was the system running at 400 MHz at all time?
Which image did you use?

Keep up the good work:)

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Freerunner Sale Price in India - Official partner with Openmoko

[Rahul Joshi]

Hi everyone,

I'm a little (actually quite) late in this scene so pardon my ignorance on
any goofy questions. I am a LAMP (and uhm ... Java) developer and had always
wanted/waited for a "dream box" like this. Fortunately (or unfortunately) I
came across openmoko 1 year late. But better late than never... yeah.
I'll be honest when I say I came pretty close to buying an iPhone when I
went to US recently but the whole "Out of Stock" thing drove me mad. In fact
I started harboring anti-Apple feelings ($20 for s/w upgrades,
overtly-closed-and-controlled UI environment etc. etc.) that I decided to
stick with my iMate SP5m until something really gets to me. And here I am.
So, my first few questions (and I'm thrilled to know a thread on Neo Indian
retailing exists!!) -

1. Where in India can I purchase Neo FreeRunner and its related accessories
once it gets released. I know Rakshat is the man here but can we get some
more info on this please? Can I pre-order one (just in case it runs out of
stock!)

2. Whats this talk of 15.4K discount Mr. Krishna is talking about. How can I
be entitled to it as well? :D :P

Thanks!

Rahul J aka sector7


*Ganesha Krishna* ganesh.krishna at gmail.com

*Tue May 6 11:45:30 CEST 2008*
> Hi,
> First off, my 'thank you' s are due..
> 1. To Rakshat for setting up the Indian distribution. You obviously
have
> the know how of open device and open source which is great. This shop has
> greatly relieved my anxiety of shipping and customs which tends to become
a
> hassle.
>2. To Harry, for the official announcement on Indian
distribution.Always
> good to know that you are buying from the official shop.

>  On the price point:
>  I feel 20,000 for Free runner is very decent,  (BTW I am eligible
> for 15.4K discount so that is very very decent :-) ). Neo has a potential
to
> become the "office PDA phone". Conference calls to the US and other
oversea
> destinations are extremely common in Indian IT industry. neo +wifi+skype =
> ridiculously reduced cost.

>   A GSM +GPS (GTA01 like ) box would be great for the rural India if
priced
> around 10,000. Imagine a post man plotting the shortest path to cover all
> the villages that he has to deliver mail to
> (
>
http://news.bbc.co.uk/1/shared/spl/hi/picture_gallery/06/south_asia_indian_postman/html/1.stm
)
> with the right features the government might want to buy the neo for
> them.
>   A VGA screen with GPRS means the farmer can consult the market price
> of his crops, and the wealthy ones can lookup Stock market. (Oh, yes.
> Commodity trading and stock trading are huge in my native village!! but
> there are only few brokerage houses that are miles away in the town
center,
> one needs to call them frequently to be in touch with the market. The
> village itself has GSM+GPRS coverage)

>  Students and universities:
>   Rakshat, have you considered talking to Universities/engineering
> colleges. There are mind boggling number of engineering colleges in
> Karnataka(my state) and all of them run some kind of telecommunication
> course. a couple of neos to each college with a smartly packaged SDK and
> presentation on getting hands on GSM experience would help you sell many
> phones. I know I would want to work on such a device If I were a student
> now.

> Regards,
> -GK



> >
> >
> > -- Forwarded message --
> > From: "rakshat hooja" 
> > To: community at lists.openmoko.org
> > Date: Tue, 6 May 2008 12:57:48 +0530
> > Subject: Re: Freerunner Sale Price in India - Official partner with
> > Openmoko
> >
> > > Shakthi,
> >
> >
> > I have been using  GTA01 (Neo 1973)   my main phone for some time and
felt
> > that at its price point (under INR 12000 for GSM and GPS and VGA
display) it
> > would make a great phone for the Indian market. Unfortunately with its
power
> > draining problems (I carry 2 spare batteries in my pocket) it can not be
> > mass produced (My initial email to Micheal was about reselling the GTA01
in
> > India).
> >
> > Anyway the good part is that the Openmoko and Qtopia software is GPL and
> > porting these to a GSM only hardware should be considerably easier than
> > writing a new GSM stack! We have plans get it to run on the CompuLab
X270em
> > board but thats kind of stalled at the moment. If that works out one can
> > bring out GSM only, GSM+GPS, GSM+WiFi and GSM+GPS+WiFi versions of a
phone
> > running Openmoko.
> >
> > Maybe after the success of the Freerunner (I am quite confident on that
as
> > I have seen the software really improving over the last 6 months and the
> > price point is pretty decent too) Openmoko themselves may consider
bringing
> > out a lower spec cheaper version.
> >
> > Right now I think it would be best to support the Openmoko endevour and
> > try and make the Freerunner a great success, both at the market and as a
> > philosphy.
> >
> > Rakshat
> >
> >
> >
> > >
> > > -- Forwarded message --
> > > From: "Shakthi Kannan" 
> > > To: "List for Op

Re: silence (re mass production status) is deafening

[Steven Le Roux]

Just read mails...

Steve explained that there is a last preMP and the process by the
manufacturer, test process etc...

2 days ago max...

On Thu, May 29, 2008 at 6:08 AM, Steven Kurylo <[EMAIL PROTECTED]>
wrote:

> On Wed, May 28, 2008 at 8:08 PM, Ron K. Jeffries <[EMAIL PROTECTED]>
> wrote:
> > (I am not complaining, just sayin'...)
> >
> > We're not hearing much about the status of production
> > for Freerunner. From available bits of info, I deduce:
>
> Only a week ago there was a thread "Whats up with the freerunner mass
> production?".  I'd hardly call that a lack of communication.
>
> --
> Steven Kurylo
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>



-- 
Steven Le Roux
Jabber-ID : [EMAIL PROTECTED]
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb [EMAIL PROTECTED]:
> Hi Joerg,
> as it seems that my mail did not reach the community list (I am not a 
subscriber, just monitoring the archives) may I ask you directly the question 
regarding "multi tutch" :-) . See below:
> 
> 
> Hi all,
> if the geometric average point is returned if multiple points are touched 
wouldn't it be possible to have a fake multi touch, e.g., by doing the 
following:
> 
> Press point A
> -> Vector A is returned
> Additionally press desired point B
> -> X = (A+B)/2 is returned
> -> calculate B = 2*X-A
> 
> Questions:
> Would this be managable?
> What time difference would be required between first and second (and 
third ...) touch, to recognize such a multi-touch action and reliable 
distinguish it from a single touch at position X?
> 
> And one more comment: If that would work you could also detect multi-touch 
gestures (even if limited), e.g., use the first point statically (maybe a 
focus centre) and handle the second point dynamically (maybe for zooming in 
and out).
> 
> 
> Boris


there are two resistors in the touchpad, one for X and one for Y. We only see 
them changing on a touch, thus giving us the coords. A concurrent second 
touch just changes the R values even more, but we get no info about whether 
we touched a second point or we moved the finger. So what you suggest is a 
gesture recognition. Things become nasty because even the "geometric middle" 
isn't true but depends on pressure of one to pressure of other touchpoint 
etc.
You really get too few info out of the device to do anything reasonable with 
it beyond singletouch (at least that's the way it is now. I plan to see 
whether we can exploit dynamic pulse response of this design to get some 
additional info. Same way you're testing cat5-cable for breaks. Don't hold 
your breath though, chances are bad. And it's not on top of my todo-list)

cheers
jOERG


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Joerg Reisenweber]

Am Do  29. Mai 2008 schrieb Andy Green:
> Somebody in the thread at some point said:
> 
> |> Please someone correct me if I'm wrong, but I thought the WiFi chip in
> |> fact _can_ do AP mode, but that mode is not allowed in the open source
> |> driver.
> |
> | its a firmware. the wifi module has its own firmware and does the 802.11
> | handling there autonomous.
> | that concept is called hardmac and was there earlier, e.g on the old
> | 'orinoco silver' aka hermes pcmcia cards.
> 
> or "fullmac".  It just doesn't support Master mode.
> 
> | that firmware can currently do client mode and ad-hoc.
> | _in theory_ every wifi radio can do ap-mode, its just a question if you
> | can send packets at a low-enough layer in the right format.
> | this is controlled by firmware on the wifi module in this specific case.
> 
> It's also a matter of receiving bulk packets efficiently and they don't
> give us monitor mode either.
> 
> We're helpless unless Atheros decided to implement Master mode in their
> closed firmware.  Unfortunately the power advantages of having the bulk
> of the ieee80211 actions managed in the firmware are pretty compelling
> so I don't know how we get out of that bind.
> 

Well it's some time ago since I read about Soft|Hard|Free|FullMAC for prism 
chipsets. Dunno whether there was a way to run even fullmac cards with a 
softmac stack.
A much more interesting point: where is the firmware to download to our 
Atheros-chip. (NO, no flamewar on free firmware|no firmware again!). Just a 
simple question: do we have any way to reflash the FW?

/jOERG


signature.asc
Description: This is a digitally signed message part.
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: multi-tutch?

[ramsesoriginal]

More or less I knew that, but, as you can see in the video/read in the
news, microsoft stated tat the Windows 7 Multitouch works on "standard
touchscreens"..

On Thu, May 29, 2008 at 7:37 AM, Federico Lorenzi <[EMAIL PROTECTED]> wrote:
> On Thu, May 29, 2008 at 7:33 AM, Federico Lorenzi <[EMAIL PROTECTED]> wrote:
>> Not really, normal touch screen tablets have a capacitive touch
>> screen, which is why they usually only work with a special pen, and
>> you can rest your hand on them.
> Correcting myself - from Wikipedia:
> Passive tablets, most notably those by Wacom, make use of
> electromagnetic induction technology, where the horizontal and
> vertical wires of the tablet operate as both transmitting and
> receiving coils (as opposed to the wires of the RAND Tablet which only
> transmit). The tablet generates an electromagnetic signal, which is
> received by the LC circuit in the pen. The wires in the tablet then
> change to a receiving mode and read the signal generated by the pen.
> Modern arrangements also provide pressure sensitivity and one or more
> switches (similar to the buttons on a mouse), with the electronics for
> this information present in the pen itself, not the tablet. On older
> tablets, changing the pressure on the pen nub or pressing a switch
> changed the properties of the LC circuit, affecting the signal
> generated by the pen, which modern ones often encode a digital data
> stream onto the signal. By using electromagnetic signals, the tablet
> is able to sense the stylus position without the stylus having to even
> touch the surface, and powering the pen with this signal means that
> devices used with the tablet never need batteries. Wacom's patents
> don't permit their competitors to employ such techniques.
>
> Cheers,
> Federico
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>



-- 
My corner of the web: http://blog.ramsesoriginal.org

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Ivo Anjo]

I think, for example, you can't do WPA in ad-hoc mode (only WPA2), so
there's a reason. There are some others like this one.

Ivo

On Thu, May 29, 2008 at 10:31 AM, Alexey Feldgendler <[EMAIL PROTECTED]>
wrote:

> On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]> wrote:
>
>  We're helpless unless Atheros decided to implement Master mode in their
>> closed firmware.  Unfortunately the power advantages of having the bulk
>> of the ieee80211 actions managed in the firmware are pretty compelling
>> so I don't know how we get out of that bind.
>>
>
> The very knowledge that we can't do something that the hardware would
> technically be capable of is annoying, but I don't really see why we would
> need to implement a true AP in the phone. For any reasonable use case I can
> think of, ad-hoc mode should be enough. The only usability advantage of
> being an AP would be that it can send beacon packets that allow other
> devices to detect an available network, but sending beacon would be a
> battery drain anyway.
>
>
> --
> Alexey Feldgendler <[EMAIL PROTECTED]>
> [ICQ: 115226275] http://feldgendler.livejournal.com
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community
>
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Alexey Feldgendler]

On Thu, 29 May 2008 08:59:10 +0200, Andy Green <[EMAIL PROTECTED]> wrote:


We're helpless unless Atheros decided to implement Master mode in their
closed firmware.  Unfortunately the power advantages of having the bulk
of the ieee80211 actions managed in the firmware are pretty compelling
so I don't know how we get out of that bind.


The very knowledge that we can't do something that the hardware would  
technically be capable of is annoying, but I don't really see why we would  
need to implement a true AP in the phone. For any reasonable use case I  
can think of, ad-hoc mode should be enough. The only usability advantage  
of being an AP would be that it can send beacon packets that allow other  
devices to detect an available network, but sending beacon would be a  
battery drain anyway.



--
Alexey Feldgendler <[EMAIL PROTECTED]>
[ICQ: 115226275] http://feldgendler.livejournal.com

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Available Encryption algorithms

[Joachim Steiger]

just as a sidenote:

see http://en.wikipedia.org/wiki/A5/2

as well as http://en.wikipedia.org/wiki/A5/1
and http://en.wikipedia.org/wiki/A5/3

seems one doesn't want A5/2 anyways and its deprecated, thus eventually
the network doesn't allow using it anymore?

just guessing.. would need to work through a pile of calypso docs for
more details, and still not know whats implemented in that specific
revision of the ti-libs anyhow.. but after reading that i think a5/1 and
5/3 will need to be enough till we need 3g anyhow. (when gsm gets phased
out, somewhere in the future or THC is successful ;)

-- 

Joachim Steiger
Openmoko Central Services

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My second battery life test with the freerunner in standby

[Alexander Frøyseth]

How about charge the mobile every night?
And the powersavingapp isen't ready yet. After what I know.

Alexander Frøyseth

Eildert Groeneveld skrev:

Hei Folks

can someone enlighten me on this issue of battery life: it seems that the 
freerunner stayed alive for 16 hours and everyone seems exited. My little 
crappy Motorala stays alive for days if it is not used to make calls.


How does this relate? Are we waiting for suspend to work? or shall we need a 
stack of batteries to swap?


greetings

Eildert

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community
  



___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: [Fwd: u-blox binary protocol boilerplate code]

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| If you look at the ATR0635 datasheet directly from Atmel:
|
| http://www.atmel.com/dyn/resources/prod_documents/doc4928.pdf
|
| you'll read on page 14:
| "RAW (RAW message support requires an additional license)"
|
| U-blox (Atmel's software partner) only sells some of its products with
RAW
| message support (notably the LEA-4T)
| So even if it's a supported feature we can not really be sure if the
firmware
| on the chip allows us access to the RAW messages :(
|
| I guess we'll have to wait until somebody interested in this topic
receives
| his/her own Freerunner...

Here is the reset header that device spews:

$GPTXT,01,01,02,u-blox ag - www.u-blox.com*50
$GPTXT,01,01,02,ANTARIS ATR062x HW 80040001*26
$GPTXT,01,01,02,ROM CORE   5.00Jan 09 2006 12:00:00*76
$GPTXT,01,01,02,LIC 1EBF-BD07-E83D-6BE1-0F7A*50

It has some kind of LICense, good, bad or indifferent I dunno.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+XnYACgkQOjLpvpq7dMpS+gCfeFsg2+PtRlx4T5/aA7QefwUE
CbQAnAm/a8HcP6tSdZpY3BzUk+GmyEcP
=I0S3
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: IGES & STEP CAD file issues

[Philip Stubbs]

2008/5/22 Jeremiah Flerchinger <[EMAIL PROTECTED]>:
> I'm trying to install a 30 day trial version of Pro/E Wildfire 4.0 right
> now.  If I'm lucky I can run it in Wine and I can see if it can export
> models that work any better for me.

Pro/E Wildfire 3.0 and 2.0 were available as native Linux
applications. If you could get a trial of one of those, I am sure you
will have much more success.
-- 
Philip Stubbs

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My second battery life test with the freerunner in standby

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:
| Hei Folks
|
| can someone enlighten me on this issue of battery life: it seems that the
| freerunner stayed alive for 16 hours and everyone seems exited. My little
| crappy Motorala stays alive for days if it is not used to make calls.
|
| How does this relate? Are we waiting for suspend to work? or shall we
need a
| stack of batteries to swap?

It's not bad because the CPU was up at 400MHz all that time.  Yes, when
Linux suspend is used you will get radically longer survival times off a
charge.  So the number is interesting for the suspend implication as
much as for what it actually tested.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+WdUACgkQOjLpvpq7dMo7aQCfcfePrSm2ghuiNqC/stOIuAtL
n0UAn2i2j54CfHXlS9WKbB1lG/xxsOQE
=NisE
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My experience with the Freerunner

[The Rasterman]

On Wed, 28 May 2008 22:40:27 -0700 Matt Mets <[EMAIL PROTECTED]> babbled:

> > to correct - just backspace! :) (left slide). the magnifying thing is
> > possible
> > - but somehow i saw it as superfluous as chances are u press and release
> > very fast like a keypress on a normal keyboard and then notice the mistake.
> > even so
> > - the dictionary lookup will be correcting if it's in the dictionary and not
> > too far of a typo (press too far away from intended key). admittedly the
> > dictionary we ship has only 5000 words - but hey. it's a simple text
> > file. :) 
> Ok, keeping that in mind, pressing in the general area and using the 
> word lookup feature seems to work pretty well with a finger.  At least i 
> have been able to put in some simple phrases quite easily.

that's the main idea - that it is pretty usable even with fingers (within
touchscreen limitations - ie - don't press hard enough and it won't register a
press). it's meant to have a wide margin of error so it can correct your
mis-hits to be what you intended. as with all things - it's a guess. it's never
perfect, but humans definitely are not perfect (if we were we'd be able to hit
the touchscreen with pixel-level accuracy! :)).

> > horizontal is for quick selecting the most likely matches for correction
> > (or if no matches - exactly what you typed), and if it doesn't fit u can
> > access ALL matches from the popup list.
> >   
> One other thing I noticed was that the widget for the popup window 
> covers up the leftmost horizontal match or two, making it impossible to 
> select them.

i know. spotted that. fixed already in svn a few days back. :)

> > it's possible we can do this - in svn there is even a full qwerty kbd
> > layout i initially used - with ctrl, alt, etc. for terminal junkies, BUT
> > for now correction is always-on.
> >   
> Here is a funny idea: how about replacing the word lookup with 
> bash-style command completion when in the terminal?  That could be 
> really cool!  Maybe the word lookup feature already has an interface 
> that can be hooked into. I could see it being useful for auto-completing 
> really any application-specific data as well. :-D  For example, maybe in 
> the dialer program to show your closest contacts that match the number 
> (one of my favorite features of the GTK-dialer).

already considered. not now, but later, allow the application to hint things
like:


please use dictionary name: "en" (english) or "de" (german) or "shell" (a shell
dictionary that includes all the commands in $PATH and common options and so
on) and the layout can be different for "shell" mode (request a shell layout,
not the default or numeric" etc. etc. - but of course all to be done over time.
can't do everything on day 0. but have enough there so this can be expanded.
allow for custom dictionaries to be generated from other personal data (contact
lists etc.) and later even abstract out dictionary engines maybe to a dbus
service, abstract out keyboard ui handling to modules etc. etc. - so each
element of a keybard is its own abstracted unit that can be plugged-in. right
now its a big blob of stuff - but i do intend to abstract it and make the
barrier of entry to write just part of a keyboard (and not have to write the
whole thing if you don't want to) to b4e easy. move as much into config files
as possible and not require code (eg keyboard layouts are right now just text
config files - as are dictionary files along with frequency of use info).

i'm a big fan of making as much a config option as possible.

-- 
Carsten Haitzler (The Rasterman) <[EMAIL PROTECTED]>

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Michael Shiloh]

Joerg Reisenweber wrote:

hmmm you really have a talent to ask interesting questions


I'll take that as a compliment, although most people find it irritating :-)

In this case it's actually my job :-)



dunno... yet


I understand. It's not a trivial question. I know that you won't claim 
something unless you have convinced yourself of the truth. I look 
forward to your analysis and consideration.





/j


___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Can update freerunner CPU to ARM926EJ-S

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| The freerunner CPU is ARM920T ,can't support android system img,I think
| it can update to ARM926EJ-S ?
| We can replace CPU directly ?

No, CPU is definitely not "user replaceable part".

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+VWYACgkQOjLpvpq7dMrVzACfY2zW4xTy53vub1bXwzrb7DdR
iwUAnAslrv2T9VjqjUbasEKWHw/RqD1n
=qQuy
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Neo as cellular modem?

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

|> Please someone correct me if I'm wrong, but I thought the WiFi chip in
|> fact _can_ do AP mode, but that mode is not allowed in the open source
|> driver.
|
| its a firmware. the wifi module has its own firmware and does the 802.11
| handling there autonomous.
| that concept is called hardmac and was there earlier, e.g on the old
| 'orinoco silver' aka hermes pcmcia cards.

or "fullmac".  It just doesn't support Master mode.

| that firmware can currently do client mode and ad-hoc.
| _in theory_ every wifi radio can do ap-mode, its just a question if you
| can send packets at a low-enough layer in the right format.
| this is controlled by firmware on the wifi module in this specific case.

It's also a matter of receiving bulk packets efficiently and they don't
give us monitor mode either.

We're helpless unless Atheros decided to implement Master mode in their
closed firmware.  Unfortunately the power advantages of having the bulk
of the ieee80211 actions managed in the firmware are pretty compelling
so I don't know how we get out of that bind.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+VD4ACgkQOjLpvpq7dMo4nACdFkBC+552ElF8o0cy2x+ffu0A
mfQAn2+zAX7OTDveChC1r4ulzlBjV7kG
=swEq
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Converting Blackberry contacts into QTopia

[Michael Shiloh]

Marco Trevisan (Treviño) wrote:

Ian Darwin wrote:

I won't say it was easy or pretty, but I did it.


Cool... I want do it also with my Motorola phone importing contacts, 
SMSs and events from the binary seem files... Maybe I could use some of 
your advices!




Marco, and others: I'm going to create an entry on my blog for those of 
you switching from X to Openmoko. Please add your experiences.


Michael

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: USB power direction in host mode

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| Is it possible to charge the Neo Freerunner when using the USB port in
| host mode, by ascerting LOGICAL hostmode and by NOT asserting EN_USBHOST?

I verified this a couple of months ago.  EN_USBHOST asserted is
deliberately fatal for any charging activity (we do not want to charge
ourselves off of the 5V we just generated!), but you can have logical
host mode up with EN_USBHOST down if you externally add 15K each to D+
and D-.

This is how we formulated the "Y cable" recipe, which features the 15Ks
and the charger, together with a USB A socket for your USB device.  With
this you can have charging, wallsocket power, USB host mode and a socket
for your USB device.

- -Andy
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+UyQACgkQOjLpvpq7dMon8QCfUDJyN1aKV0KjSDv+J4CQ8+qO
E/cAn0MUlzsKo7dNMIpNnRZ7pLDlGJKD
=uXf/
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Meta Toolchain Release (2008 May)

[John Lee]

I have just realized that opkg-cl isn't in toolchain.  :(

added and committed, will show up in the next toolchain build.

On Thu, May 29, 2008 at 12:34:40PM +0800, John Lee wrote:
> 
> libjana isn't packaged in toolchain by default.  the default
> compopents are:
> 
> libmokogsmd2, mokoui2, mokopanelui2, mokojournal2 and dependencies.
> 
> we cannot package everything or it's going to be a huge toolchain.
> however, there are 2 interesting aliases in environment-setup:
> 
> alias opkg='LD_LIBRARY_PATH=/usr/local/openmoko/arm/lib 
> /usr/local/openmoko/arm/bin/opkg-cl -f 
> /usr/local/openmoko/arm/etc/opkg-sdk.conf -o /usr/local/openmoko/arm'
> 
> alias opkg-target='LD_LIBRARY_PATH=/usr/local/openmoko/arm/lib 
> /usr/local/openmoko/arm/bin/opkg-cl -f 
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi/etc/opkg.conf -o 
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi'
> 
> which means by opkg-target we can add the packages we need into the
> toolchain directory.
> 
> at the moment the content of 
> 
> /usr/local/openmoko/arm/arm-angstrom-linux-gnueabi/etc/opkg.conf
> 
> is incorrectly pointed to local directories in build host.  if we
> replace it by the corresponding URIs such as
> 
> http://buildhost.openmoko.org/daily-feed/
> http://buildhost.openmoko.org/daily-feed/all
> http://buildhost.openmoko.org/daily-feed/armv4t
> http://buildhost.openmoko.org/daily-feed/neo1973
> http://buildhost.openmoko.org/daily-feed/om-gta01
> 
> then _theoretically_ we can just do
> 
> opkg-target update
> opkg-target install libjana-dev
> 
> I haven't tried this approach before because OE suits me just fine.
> however, for those who just want to make apps I think it's an right way
> to start.
> 
> 
> Regards,
> John
> 

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: 3G USB Dongle (was Re: Neo as cellular modem?)

[Andy Green]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Somebody in the thread at some point said:

| YEP. but from this sniplet I don't understand whether it also will enter
| LOGICAL hostmode like with a short to GND on ID-pin.
|
| That's the crux, do *both*

No, nothing is watching ID level right now.  You have to do it by hand
until we implement something.  But there are no mini USB OTG Host or Y
cables shipped either :-/

echo 1 > /sys/devices/platform/neo1973-pm-host.0/hostmode ; echo host >
/sys/devices/platform/s3c2410-ohci/usb_mode

to get into host mode and this:

echo 0 > /sys/devices/platform/neo1973-pm-host.0/hostmode ; echo device
| /sys/devices/platform/s3c2410-ohci/usb_mode

to get back to device.

- -Andy

-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkg+UdkACgkQOjLpvpq7dMpCGQCgh9doLt06m5qfWYOXqeCEgz0o
s0EAnRpoYeFOYwgfVY6ln8GL4roY+yAK
=C8SV
-END PGP SIGNATURE-

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: [Fwd: u-blox binary protocol boilerplate code]

[beren]

If you look at the ATR0635 datasheet directly from Atmel:

http://www.atmel.com/dyn/resources/prod_documents/doc4928.pdf

you'll read on page 14:
"RAW (RAW message support requires an additional license)"

U-blox (Atmel's software partner) only sells some of its products with RAW 
message support (notably the LEA-4T)
So even if it's a supported feature we can not really be sure if the firmware 
on the chip allows us access to the RAW messages :(

I guess we'll have to wait until somebody interested in this topic receives 
his/her own Freerunner...

-- beren

On Wednesday 28 May 2008 16:36:55 Joseph Reeves wrote:
> According to the datasheet for the chip ATR0635, DGPS is a supported
> feature:
>
> http://www.u-blox.com/products/Data_Sheets/ATR0630_35_SglChip_Data_Sheet(GP
>S.G4-X-06009).pdf
>
> If I'm reading this correct, we've also got "supersense" available,
> which should improve accuracy.
>
> I'll be the first to admit, however, that I really am not the best
> person to talk to about the technical side of such things.
>
> Joseph
>
> 2008/5/28  <[EMAIL PROTECTED]>:
> > On Wednesday 28 May 2008 11:46:45 Joseph Reeves wrote:
> >> We're looking to develop a combined DGPS and A-GPS application for the
> >> platform, for which we need the data contained within this binary
> >> protocol.
> >
> > That sounds great - I presume you need access to the RXM-RAW message
> > which to the best of my knowledge is not available/accessible on all
> > Antaris chips due to licensing/firmware reasons. Could anybody confirm
> > that it is possible to access the RXM-RAW information from the Antaris
> > chip in the Freerunner? That would be really great news because without
> > it DGPS would be impossible...
> >
> > -- beren
> >
> > ___
> > Openmoko community mailing list
> > community@lists.openmoko.org
> > http://lists.openmoko.org/mailman/listinfo/community
>
> ___
> Openmoko community mailing list
> community@lists.openmoko.org
> http://lists.openmoko.org/mailman/listinfo/community



___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: My second battery life test with the freerunner in standby

[Eildert Groeneveld]

Hei Folks

can someone enlighten me on this issue of battery life: it seems that the 
freerunner stayed alive for 16 hours and everyone seems exited. My little 
crappy Motorala stays alive for days if it is not used to make calls.

How does this relate? Are we waiting for suspend to work? or shall we need a 
stack of batteries to swap?

greetings

Eildert

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community

Re: Password safe

[Michael Shiloh]

Uncle Kridley wrote:

Ian Darwin wrote:
The only two things I really need the Blackberry for (apart from 
stable calling :-)) are the password safe (which I have written a 
replacement for, and others exist)...


I'm totally dependent on gnukeyring on my Treo, so this is good news. 
I'd been hoping that somebody with some GUI coding experience would 
build one on Openmoko, since I've never written a (non-web) GUI program.


Are you going to release your password safe?  What are the others you 
speak of?


I thought that perhaps gnome-keyring could be ported to Openmoko...



I use pwsafe, which is a text-only program. I imagine it would be easy 
to throw a GTK (or QT) wrapper around it and cross-complile for Neo.


M

___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community