Re: ``Freerunner reliable and stable''?!

2008-10-26 Thread Chris Wright
2008/10/26 Matthias Apitz [EMAIL PROTECTED]:
 I'm a bit tired of all those (useless) threads like this. I DO USE the FR
 with Om2008.9 for everyday use, I do not even own any other cellphone.
 We should improve what we have and stop useless discussions, as well
 about Google's trick of Android which has nothing todo with free
 software.

Same here. The stability is sufficient for daily use. It is not as
stable as my previous or antepenultimate phones, but those were quite
simple, unable to do more than make calls.

The only issues I have with the Freerunner are battery life and an
occasional (one call per week, roughly) lack of sound. Neither is
sufficient to cause me to switch back to my previous phone.

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


Re: New Rasterman Image...

2008-10-06 Thread Chris Wright
2008/10/1 The Rasterman Carsten Haitzler [EMAIL PROTECTED]:
 eg:

 Press | Guess+dist
 e   e+0 w+1 r+2 d+2 s+1
 r   r+0 t+1 e+2 f+1 g+2 d+3
 k   k+0 l+1 o+2 i+3 j+3
 d   d+0 f+1 s+1 e+1 c+1 r+2 w+2

I assume this takes into account the keyboard layout, so if I use
dvorak or some random layout, it'll do the right thing. Right?

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


Re: ASU - out of memory?

2008-08-22 Thread Chris Wright
2008/8/22 Tilman Baumann [EMAIL PROTECTED]:
 Pardon.
 I don't care for the warm and fuzzy feeling you get by having malloc
 fail on you.

 It does not give you a bit more system stability! The one app
 receiving malloc errors is just not app of many. They all have a
 problem then.
 Imagine, the browser catches a failed malloc, because some other
 stupid app has eaten almost all ram.
 What is the benefit of telling the browser about low mem? It could
 only safe itself from crashing. Well done.

Imagine that malloc() returns null rather than overcommitting.
Most affected apps will receive a SIGSEGV when malloc returns null
because there's no error handling for that situation.
Some will notice that malloc() returned null and either forgo some
optional stuff or abort a single operation or simply just shut down
gracefully.

Imagine that malloc() returns an invalid pointer when it overcommits.
All affected apps will receive a SIGSEGV in this case, without some
unusual error checking.

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


Re: ASU - out of memory?

2008-08-22 Thread Chris Wright
2008/8/22 Chris Wright [EMAIL PROTECTED]:
 2008/8/22 Tilman Baumann [EMAIL PROTECTED]:
 Pardon.
 I don't care for the warm and fuzzy feeling you get by having malloc
 fail on you.

 It does not give you a bit more system stability! The one app
 receiving malloc errors is just not app of many. They all have a
 problem then.
 Imagine, the browser catches a failed malloc, because some other
 stupid app has eaten almost all ram.
 What is the benefit of telling the browser about low mem? It could
 only safe itself from crashing. Well done.

 Imagine that malloc() returns null rather than overcommitting.
 Most affected apps will receive a SIGSEGV when malloc returns null
 because there's no error handling for that situation.
 Some will notice that malloc() returned null and either forgo some
 optional stuff or abort a single operation or simply just shut down
 gracefully.

On the other hand, let's say your process allocates some memory and
doesn't use it for a while. In the meantime, some memory is freed.
This doesn't help if malloc() returned null, but it does help if the
kernel overcommitted memory instead.

I don't think that's as useful. But you could instead define a malloc:
void* _malloc(size_t length)
{
   void* pointer = malloc(length);
   if (!mlock(pointer, length)) return null; // or abort
   return pointer;
}

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


Re: Emergency call at boot time functionality (was Re: Windows CE on freerunner)

2008-08-22 Thread Chris Wright
2008/8/22 Lothar Behrens [EMAIL PROTECTED]:
 You must have a lock, since otherwise you may accidantly issue an
 emergency call.

 Lothar

With my current phone, you can make emergency calls on it, even if the
phone is locked. It happened to me once. I got a call a few minutes
later from the call center asking if everything was okay. I'd rather
not repeat that.

Locks that do not allow emergency calls are good. Especially
considering that you need to take various other steps to place an
emergency call with the Freerunner (go to an application menu, choose
an icon, dial).

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


Re: atd: how to start a job every 10 minutes ?

2008-08-15 Thread Chris Wright
2008/8/15 Harald Koenig [EMAIL PROTECTED]:
 On Aug 15, Roland Mas wrote:

 You may want ot look at cron instead of at.  at is for one-time jobs,
 cron is for recurring jobs.

 will cron work while FR being suspended ?

 haven't checked yet, but so far I've read this only from atd,
 and atd exclusively opens /dev/rtc (see hwclock problems...)

 bute maybe there is a OM cron version which interfaces with atd ?!
 I'll check, thanks for the pointer...

If not, then you can write a script that executes whatever you want
and then adds itself to atd.

--
Those who do not understand recursion are doomed.

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


Re: FSO: is zhone here to stay?

2008-08-08 Thread Chris Wright
2008/8/7 Stefan Schmidt [EMAIL PROTECTED]:
 On Thu, 2008-08-07 at 12:39, Craig B. Allen wrote:
 If zhone is going to be around for a while, how can I make a request
 for a new feature?

 Well, that depends a lot on what kind of feature request this is. As I wrote
 above, zhone is mostly to test our APIs and implementations. It's not targeted
 to fullfill all user needs yet. Basic phone functionality is the goal.

 If it is something in this area feel free to submit a ticket for it:

 http://trac.freesmartphone.org

I take it you'd accept patches to add features to zhone, even if
they're unrelated to the framework, no?

 regards
 Stefan Schmidt

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


Re: any update on GSM interference issue

2008-08-04 Thread Chris Wright
http://gadgets.boingboing.net/2008/08/02/kill-gsm-radio-buzz.html
Maybe that could provide a less invasive means to getting rid of the problem?

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


Re: Openmoko on Design

2008-07-29 Thread Chris Wright
2008/7/29 Marek Lindner [EMAIL PROTECTED]:

 Hi,

 At the same time we heard comments from a key developer who indicated
 that the decision was made above him by unnamed individuals with whom
 the community has no obvious means of communication, and who apparently
 don't even listen to the reasonable technical arguments of key
 developers.

 Openmoko always avoided all kind of formal structures. Thus we don't have such
 a thing as key or core developer - a developer would be better.

But you do have a design team, according to Rasterman.

 This also seemed to reveal something about the internals of
 Openmoko that weren't expected: development decisions are not entirely
 made by the developers, but instead they answer to some people who the
 community cannot readily identify and who the community doesn't know how
 to interact with or if they even can interact with these decision-makers.

 May be it revealed that Openmoko itself is diverse as well. That some
 developers have different opinions than others.

Out of curiosity, how many of these developers use an Openmoko phone
as their primary phone? Do these differences of opinion tend to fall
on the same boundaries?

Still, nobody has mentioned why the design team can't be contacted or
identified.

 Openmoko has to trust those members of the community, who prove themselves
 through actual contributions, to be worthy to give input on larger design
 issues as well.

 You got the point !

Strange, I read this as Openmoko has not been, but should in the
future, trust those members

I haven't been here long enough to determine which is the case. Maybe
the company hasn't, either.

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


Re: Openmoko on Design

2008-07-29 Thread Chris Wright
2008/7/29 Marek Lindner [EMAIL PROTECTED]:
 On Tuesday, 29. July 2008 20:17:00 Chris Wright wrote:
 But you do have a design team, according to Rasterman.

 Of course we have. How do you think we are trying to get to a device that is
 ready for end user ? And this is just the beginning. We will work with more
 designers for the UI, the housing, etc to continue the direction towards the
 mass market.
 But that does not mean all that is perfect right from the beginning or that we
 don't listen to constructive criticism. Please help us making it better by
 _demonstrating_ a superior solution not by sending more emails.

Where I work, the design team is the same as the development team.

 Out of curiosity, how many of these developers use an Openmoko phone
 as their primary phone?

 What do you mean with these developers ? Everybody certainly has a different
 opinion than others on something.

Then it's vacuous to say so.

 Do these differences of opinion tend to fall on the same boundaries?

 I don't get what you trying to say here.

Something as simple as a keyboard button -- well, users were
complaining about its lack very quickly. If the design team were also
users, then they would have insisted that the error be fixed.

 Still, nobody has mentioned why the design team can't be contacted or
 identified.

 Sorry but this is just not true. I don't get why you follow this childish
 witch-hunt game.

Because you were responding to this quote:
 This also seemed to reveal something about the internals of
 Openmoko that weren't expected: development decisions are not entirely
 made by the developers, but instead they answer to some people who the
 community cannot readily identify and who the community doesn't know how
 to interact with or if they even can interact with these decision-makers.

And your response was in no way related. I was merely pointing that out.

 Strange, I read this as Openmoko has not been, but should in the
 future, trust those members

 I haven't been here long enough to determine which is the case. Maybe
 the company hasn't, either.

 Meant was: Openmoko pays much more attention to installable solutions and
 listens to hackers that provide those instead to people that complain all day
 long but don't get their hands dirty.

Okay.

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


Re: Alternative stylus - Group Buy

2008-07-21 Thread Chris Wright
2008/7/19 Steven ** [EMAIL PROTECTED]:
 Anyone interested in a different group buy?
 I want one of these:
 http://www.jetmall.net/index.asp?PageAction=VIEWPRODProdID=303

I'm with you, too.

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


Re: Rules based policy engine

2008-07-20 Thread Chris Wright
2008/7/20 Ryan Meador [EMAIL PROTECTED]:
 I think it's important that we use an existing general-purpose platform such 
 as
 Prolog (at least, it's about as general purpose as logic programming gets...).

I would favor Rhino.DSL, simply because I know it integrates well with
a language with dbus bindings. And because it can probably result in
very readable syntax. And it's based on Boo, which is very nearly
Python, and thus more accessible to most programmers than Prolog.

Does dbus allow you to specify your priority when listening to an
event, and to prevent it from being published to other listeners? If
not, then the first step is to separate the relevant dbus events and
come up with an application that merely translates unconditionally
between the two. And that allows you to insert any rules engine you
want.

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


Re: openmoko availability from the online store, set opening time if not date?

2008-07-02 Thread Chris Wright
2008/7/2 Crane, Matthew [EMAIL PROTECTED]:

 Maybe you guys could set a time that the web store will open, on
 whatever day it does open?  E.g. if it's not open by 10am, it's not
 going to open that day.

 That could give everybody a fair shot of getting their orders asap.

 Matt

Or at least don't bother checking for the next ten hours/three
days/whatever. Information ish good.

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


Re: QVGA V/s VGA for GTA03 (was something about yummy CPU-GPU combos!)

2008-06-13 Thread Chris Wright
2008/6/10 The Rasterman Carsten Haitzler [EMAIL PROTECTED]:
 On Tue, 10 Jun 2008 16:40:08 +0100 Stroller [EMAIL PROTECTED] babbled:


 On 10 Jun 2008, at 02:17, Carsten Haitzler (The Rasterman) wrote:
  browsing full web pages scrammed into a 2.8 screen as many have
  suggested, is
  really... pushing such a tiny screen far beyond its usefulness. web
  pages are
  designed for 14 or 17 screens or so. squeezing them down into
  2.8 is nigh
  madness. it's possible - but vga vs qvga there isn't the factor
  (imho) :)

 I'm sorry, Carsten, but this just makes me think you're nuts. Um, I
 mean, eccentric.

 I mean, I know you know loads more about this sort of thing than I
 do, but mobile phone web-browsers are absolutely standard these days.

 never said otherwise but the screen is physically small. very. put it at a
 normal usage distance and it covers a small fraction of the field of view a
 standard desktop screen does. web pages are normally designed for the field of
 view of a desktop screen. either you play zooming games to squeeze it down, or
 scrolling, or play re-formatting games. however you look at it - you won't get
 close to the same experience.

True...someone mentioned a month view for a calendar with a meaningful
amount of content as a use case for a higher resolution screen, but on
a three-inch display, you're not getting any significant amount of
data across without a magnifying glass.

Using a terminal emulator would be far more pleasant with the higher
resolution screen, but you're not going to get an 80x25 window in
there; with a 640x480 display plus an on-screen keyboard, you're going
to have either an 80x30 or so window in portrait mode and an
unreadable font, or you're going to get about 80x15 in landscape.

Video's going to scale anyway (which says more about processor / gpu
issues than about acceptable quality). Really, viewing photos is the
only thing that will suffer significantly. This isn't as much of an
issue since the Freerunner doesn't have a camera. But the GTA04
probably will, and the QVGA might be somewhat annoying.

Since the alternative to using QVGA is using VGA with a faster GPU and
processor, and this is a phone that runs on batteries, I'm inclined to
encourage the use of QVGA on future Openmoko phones. I have a phone
from 2002 with a battery that used to last a week; it annoys me to
deal with much less than that.

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


Re: QVGA V/s VGA for GTA03 (was something about yummy CPU-GPU combos!)

2008-06-13 Thread Chris Wright
2008/6/13 enaut [EMAIL PROTECTED]:
 I'm feeling reminded of those creditcard sized poket calendars we have
 in germany. on those you have got the calendar information of a whole
 year + the Information of hollidays and other special days. land
 everything is readable and pretty clear to read. so the size is not the
 problem. but rather the resollution as well as the clearness of the display.

Those include no more text than numbers for the date. Which you can do
with qvga as well, I am certain, if it's one month. And you can also
include colors to indicate there's an event on that day, or it's a
holiday, or whatnot -- none of which requires a larger display.

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


Re: Why not use votation system?

2008-06-13 Thread Chris Wright
2008/6/12 Brad Midgley [EMAIL PROTECTED]:
 Michele

 Do you want feature X or feature Y?

 you might need a table representing the cost and business feasibility
 of each feature. We've seen a lot more transparency on the 3g decision
 recently which reveals it to be more complicated than people thought.

I have some notes here on the knapsack problem, if we go that route

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


Re: GTA03 case should incorporate stylus holder

2008-06-09 Thread Chris Wright
2008/6/6 steve [EMAIL PROTECTED]:
 Stroller you crack me up. (vi rules.)

 We need a religious war between the Digitii, the Stylii, and Quertii.

 Or three products?

 Rant away.  Your views are always welcome.

What of the Dvoraksai?

I do think that thumb keyboards are best -- tactile feedback. Finding
space for such is not terribly easy. There are clamshell designs,
though that prevents you from using the primary display without
opening the device; then there are sliding designs, which seem to have
a bad angle.

Thumb typing on a touchscreen is probably more efficient than using a
stylus. Thumb typing on a keyboard is probably much faster than using
a touchscreen.

Though if people want a stylus holder, how hard would it be to design
one that clips onto a Freerunner?

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


Re: QVGA V/s VGA for GTA03 (was something about yummy CPU-GPU combos!)

2008-06-09 Thread Chris Wright
2008/6/9 Dotan Cohen [EMAIL PROTECTED]:
 2008/6/9 The Rasterman Carsten Haitzler [EMAIL PROTECTED]:
 an 80x24 terminal is possible to make it readable @ qvga - if we allow
 scrolling. (and possible in landscape with an ultra-tiny 4-pixel wide font -
 possible (3 pixels for text, 1 for space). not very nice though.

 That is unusable. I'm sorry, I will not force myself to use that. I
 will buy something else. Using the terminal on a remote machine with a
 4 pixels font is possibile like it is possible to eat cockroaches for
 lunch everyday. There are some 'can get by' that I will not force
 myself to get by with.

The same with having to flip between right/left panes: it turns a
half-second operation into a five-second one, and it taxes the brain
at the same time.

Moreover, this has to fit a keyboard along with anything else. The
keyboard itself has very minimal needs in terms of resolution, but it
steals about a third of the screen in portrait mode, more in landscape
-- 640x480 is probably a bare minimum. You'd be getting 200 by 240
usable space with qvga.

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


OT: ajax image galleries

2008-06-04 Thread Chris Wright
2008/6/3 Marco Trevisan (Treviño) [EMAIL PROTECTED]:
 There are some pics and videos by Einstein from freeyourphone.de:
  - http://tinyurl.com/66ktzl

(URL maps to 
http://freeyourphone.de/portal_v1/gallery/menu.php?gallery=membersalbum_id=8
)

Those AJAX image widgets seem to be designed adversarially. They
prevent tabbed browsing of images, make it difficult to link to an
image, impose additional delays on the viewer, and introduce a new UI
for navigating the galleries that doesn't fit in with any existing
browser. Viewing a large image and then a small one requires
additional scrolling, unlike any non-ajax solutions I've seen. The one
that that link goes to allows you to click on the right half of the
image to go to the next image or the left half of the image to go to
the previous, with no visual clues that that is what is happening.

Do they have any recognizable merit whatsoever? Why do people use them?

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


Re: Yummy new CPU/GPU combo

2008-06-03 Thread Chris Wright
The cost is probably on the order of paraplegia. And I wonder about
the power requirements -- there haven't been significant advances in
battery technology lately.

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


Re: Open Hardware

2008-06-01 Thread Chris Wright
Openmoko phones are as cheap as they are because they use commodity
hardware, I'm given to understand. If you wanted a phone with open
hardware, you'd probably be paying thousands for all the custom
components. Plus there's testing and certification for various parts,
which probably is also expensive

It's doable, but it'd be harder by far than the OGP. And consider that
the first card released by the OGP costs $1500, which is two or three
times the cost of the more expensive commodity cards, while delivering
less performance (their target is 20fps in Quake 3 at 1280x1024).

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


Re: Library can handle MP3 id3tag encoding

2008-05-27 Thread Chris Wright
2008/5/27 Bin Chen [EMAIL PROTECTED]:
 HI,

 Is there a good library can handle MP3 id3tag encoding easily? AFAIK,
 the encoding that in the id3tag can't be decided, it maybe ASCII,
 UTF-8 and others which sometimes cause the software to decode some
 error character.
 The libid3tag can do raw reading to id3tag but not handle the encoding easily.

You need one library to read the tag's text fields as raw byte arrays
and another to determine the encoding. The latter is highly
nontrivial.

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


Re: Switch from GTK to QT (was: ASU software - pre-pre-release impressions)

2008-05-21 Thread Chris Wright
2008/5/21 Nkoli [EMAIL PROTECTED]:

 On Tue, May 20, 2008 at 1:32 PM, Carlo E. Prelz [EMAIL PROTECTED] wrote:


 My complaint is that it would be difficult for me to put my hands into
 the default apps. They are C++, QT, and expectedly using enough of
 those creepy C++-isms (possibly, even those yecchy templates or
 whereabouts). I would be comfortable with tinkering with CGTK main
 apps. On the other hand, I would find C++QT main apps closed boxes (I
 perfectly know that I could very well write C/Ruby new code on the
 OM).

 From this statement, one would think that you don't use _any_ applications
 written in C++ or Qt for the simple reason that you can't tinker with the
 code. I am sure this is not the case. You use the applications written in
 C++/Qt and play with those written in the languages you're comfortable with
 or you write your own from scratch. Right now OM needs more new apps than it
 needs shipped apps taken apart (which will probably be done by a number of
 people anyway).

In this case, openmoko is still being marketed to developers and is
still alpha software. As such, I have a reasonable expectation that I
will want to hack a fair number of the applications it ships with. So
that is a valid complaint. Of course, you could simply rewrite each
application in your preferred language, but that's a waste of effort.
Or you could not mess with the applications written in a language you
do not prefer. Or you could wait until there's a more severe need and
then learn the language in question and start working on the
application.

With desktop Linux, most things are sufficiently mature that I expect
not to have to hack them. So the language they are written in is far
less important.

Of course, one person's opinion doesn't matter much, as long as there
are plenty of people familiar with the languages that are commonly
used in openmoko. I think there are far more people who can
effectively write C collaboratively than who can effectively write C++
collaboratively.

The larger issue is extending a C++ codebase in another language.

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


Re: ASU software - pre-pre-release impressions

2008-05-20 Thread Chris Wright
2008/5/20 Ben Burdette [EMAIL PROTECTED]:
 Travis Tabbal wrote:
 C++ is difficult to link against, so that limits the languages that you can
 use for development.  For instance, D has a hard time with linking to C++
 (or did last I checked).  C, on the other hand is far simpler to link to.
  That's why many standard APIs use C interfaces - OpenGL for instance.

Recent versions of dmd can link against C++ code compiled with dmc,
but it's a bit unreliable, from what I've heard. And D has *extremely
good* support for linking with C++, compared with other languages.

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


Re: Will GTK be used in Openmoko? (was: Re: Software Status Update)

2008-05-15 Thread Chris Wright
2008/5/15 Flemming Richter Mikkelsen [EMAIL PROTECTED]:
 On 5/15/08, Geoff Ruscoe [EMAIL PROTECTED] wrote:
 I think gtk should definitely be part of the platform.  As far as I know the
 Java Swing hooks are for gtk.  Since its shipping with Java it should be
 able to use the gtk, right?

 Plus as far as I can tell the ruby and python hooks for gtk are slightly
 cleaner and easier than working with qt.  So from a developer standpoint I
 think we're better off with gtk.

 Thoughts?

 Qt and GTK is the only options. GTK has the best license _and_ is very
 easy. GTK has a lot of bindings and is very crossplatform.

Considering that Rasterman is working with openmoko, I'd be surprised
if ewt were not an option.

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


Re: Root password and ssh?

2008-05-14 Thread Chris Wright
2008/5/14 Mo Abrahams [EMAIL PROTECTED]:
 Except for if music files, images etc. on the phone are owned by root,
 in which case we wouldn't be able to access them via ssh.

On the contrary!
1. Disable root login via ssh.
2. Log in as a luser.
3. Su to root.

-- 
http://dsource.org/projects/dmocks -- Mock objects and more for the D
programming language
___
Openmoko community mailing list
community@lists.openmoko.org
http://lists.openmoko.org/mailman/listinfo/community