Re: How do I project an XO

[M. Edward (Ed) Borasky]

Kurt Gramlich wrote:
> * Steve Lewis <[EMAIL PROTECTED]> [080516 07:55]:
> 
>> I am doing a talk in front of a large audience and would like to show 
>> the XO's screen on a projector. I have a laptop which
>> can be projected and ideally would like to show the XO's screen on the 
>> laptop. Other solutions (not a camera on the XO's screen) would be 
>> considered.
>> Any suggestions.
> 
> perhaps our XO-LiveCd helps?
> 
> http://dev.laptop.org/pub/livebackupcd
> 
> Viele Gruesse/Regards
> Kurt
> http://dev.laptop.org/pub/livebackupcd

There is a wiki page which lists the alternatives. I have successfully used

http://wiki.laptop.org/go/Remote_display#Cloning_a_current_Sugar_session_.28using_VNC.29

with both Linux and Windows running a VNC viewer on the laptop. Note 
that if you want the VNC viewer mouse and keyboard to work on the XO, 
you need to do

http://wiki.laptop.org/go/Remote_display#Enabling_Keyboard_and_Mouse

as well.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [Its.an.education.project] Sugar on the EEE PC

[M. Edward (Ed) Borasky]

Bernie Innocenti wrote:
>> a serious problem in the most underserved areas --- the price trend  
>> is for the second
>> generation of the "low cost" laptops to head back to $500.
>> The Asus 900 has a suggested list of $550 ?
> 
> That's weird marketing... ASUS and Intel know they will have to
> beat OLPC's offer head to head.

Not in the adult-use case USA market!

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Walter leaving and shift to XP.

[M. Edward (Ed) Borasky]

Mitch Bradley wrote:
> But in the steady 
> state, the
> web is the high-order bit, sufficient to qualify as education in and of 
> itself.

Well ... it *was* at one time -- a university library made up of 
electrons. But in my mind, that was long ago in a galaxy far away. Oh, 
sure, you can still spend time on the web in the university library. But 
now there's a red light district, shopping malls, gambling casinos, 
stalkers, bullying, con games, electronic gangs and thugs.

I've got my two G1G1 units and there are presumably two children 
somewhere who have the two Give One units. Don't get me wrong -- I think 
the *XO* is a positive force in the world. And I think it's *more* 
positive than what the WWW has become.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Update.1 RC3: candidate-703 Published!

[M. Edward (Ed) Borasky]

Chris Ball wrote:
> Hi Ed,
> 
>> OK ... I installed it on my G1G1 XO (via the GamePad / USB
>> method). It boots up into Sugar, and I can see the frame, but not
>> icons to start up any activities. Is there something I need to do??
> 
> Yes:
> 
>http://wiki.laptop.org/go/Testing_Update.1
>http://wiki.laptop.org/go/G1G1_Activity_Pack
> 
> .. or you can use Bert's script:
> 
>wget dev.laptop.org/~bert/update-activities.py
>python update-activities.py
> 
> - Chris.

OK ... got it working now.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Update.1 RC3: candidate-703 Published!

[M. Edward (Ed) Borasky]

Michael Stone wrote:
> Friends,
> 
> At long last, we have a new Update.1 Release Candidate, signed and
> waiting for your attention at
> 
>   http://download.laptop.org/xo-1/os/candidate/703/jffs2/
> 
> Release notes continue to develop at 
> 
>   http://wiki.laptop.org/go/OLPC_Update.1_Software_Release_Notes
> 
> with more technical commentary being stored at 
> 
>   http://wiki.laptop.org/go/OLPC_SW-ECO_4
> 
> Enjoy,
> 
> Michael
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

OK ... I installed it on my G1G1 XO (via the GamePad / USB method). It 
boots up into Sugar, and I can see the frame, but not icons to start up 
any activities. Is there something I need to do??
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OLPC seeks a CEO -- who was your favorite CEO elsewhere?

[M. Edward (Ed) Borasky]

John Gilmore wrote:
>   ==>  Who's the best manager or CEO you ever worked for?

Hands down, C. Norman (Norm) Winningstad!

[snip]

> OLPC has already changed the world in a small
> way, by teaching us that there's a vibrant world market for low cost,
> high function portable computers, and reminding us how much leverage
> there is in third world educational improvement.  OLPC still has a
> chance to change the world in a big way, by satisfying that market,
> rather than leaving it to commercial companies to half-assedly pick up
> the pieces.

One *big* challenge OLPC faces is that those commercial companies aren't 
half-assed at all -- they are *directly* competing with OLPC by all 
legal means!

> Steering OLPC back on to the rails before it crashes and
> burns will be a job your favorite CEO or manager will never forget.

I'm not sure your presuppositions are accurate here. First of all, I 
don't think OLPC is "off the rails" or headed towards a "crash and burn" 
fate. Second, there are very real constraints OLPC has accepted:

1. Non-profit status
2. Open-source licensing
3. Depending on volunteer software engineering.

Finally, I don't think your problem is going to be finding *applicants*. 
Your problem is going to be finding the *right* applicant.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: I/O scheduling (pdflush) on the XO

[M. Edward (Ed) Borasky]

Mikus Grinbergs wrote:
> For the fun of it, started a resource-intensive task (100% CPU, 
> sporadic floods of disk writes) on my G1G1 XO.  Eventually, it 
> failed with a "severe (38)" fortran error trying to write its 
> checkpoint file {mikus note: many quick small write operations}.

Why? This device has a 433 MHz 32-bit processor, 256 MB of RAM and a 1 
GB jffs2 flash disk. What do you expect?
> 
> That situation may be a pathological one -- but what I found 
> interesting was that in again doing something similar (different 
> worktask), 'top' showed 'pdflush' requiring lots of CPU (up to 
> 52%!!). [And 'jffs2_gcd_mtd0' keeps showing up on 'top', too.]
> 
> I take it that my XO is *struggling* when presented with heavy 
> "disk I/O".

Again, where's the surprise here? An XO is *not* a scientific 
workstation! It's a computer for elementary school students!

If you want to run stuff like this, do it on a "school server" class 
machine and write an XO client.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Setting up Fedora 7 on a ex-Windows machine (Ottawa)

[M. Edward (Ed) Borasky]

James wrote:
> Hello OLPC people!
> 
> I am working on a Snakes and Ladders game for the XO, to help young  
> children learn to count.  You can find my first draft of the game  
> here: .
> 
> 
> I'm looking for help in getting Fedora 7 to run on a Sony Vaio PCG- 
> GRT796HP laptop that used to run Windows.  It's a Pentium 4, running  
> at 2.67 GHz, with 512 MB of RAM.  I've spent several hours trying  
> various approaches and distributions, without success.
> 
> This is my first excursion into Linux territory, and I'm still finding  
> my feet with Python.  I'm more at ease with development on Macintosh,  
> and have only scraped the surface of using the Terminal.  Please don't  
> hesitate to spoonfeed me in all things Linux and Python.
> 
> What I can do
> -
> I'd almost given up hope of getting the Vaio to run Fedora when I  
> tried using the XO LiveCD from  livebackupcd>.  This worked perfectly, which encourages me to believe  
> that the issue is not with the machine but with what I am doing to it.
> 
> Where I get stuck
> -
> I've downloaded the F-7-i386-DVD.iso file from 
>   >, and burnt it to a DVD-ROM. The initial menu screen appears.  If I  
> choose the default (graphic) installation, eventually the screen  
> starts to display vibrant pulsing graphics which I do not believe are  
> intended.  If I choose the text mode for installation, and step  
> through the various screens, I eventually run into a bug in the  
> installer script.
> 
> Rodney Smith entered a description of the bug into the RedHat bugbase  
> on 2007-07-08, but there seems to have been no movement on it since  
> then.  This leads me to believe that there must be an obvious  
> workaround, so others have just side-stepped the bug and moved on.
> 
> The original bug report was marked as NEEDINFO, so I supplied that  
> info on 2008-01-21.  You can read the complete report here:
> 
> https://bugzilla.redhat.com/show_bug.cgi?id=247399
> 
> What I'm hoping to do
> -
> My aim is to install a version of Linux as close to the XO version as  
> possible.  This will make it easier for me to get into the correct  
> mindset and best practices for developing for the XO.  I'm not married  
> to the idea of getting Fedora 7 to run if the line of least resistance  
> is to install something similar.
> 
> In his bug report, Rodney Smith notes that "System previously had fc5  
> that was installed using a dvd and the graphical interface without a  
> hitch and that ran fine."
> 
> I've looked for a downloadable version of Fedora Core 5 or 6 for a x86  
> machine, but all the links that I have found end up at the Get Fedora  
> page, which now limits itself to downloads of Fedora 7 and 8 
>   >.
> 
> I get a similar bug when I try installing Fedora 8.  I've also tried  
> installing Ubuntu 6, but run into the graphic-interface-shows-vibrant- 
> pulsing-graphics issue.
>  >
> If it hadn't been for XO-LiveCD_080130.iso performing perfectly on the  
> machine, I'd have written off my Sony Vaio as being incompatible with  
> Linux.
> 
> 
> If anyone can help me get some version of Linux installed on the  
> machine, I'd be most grateful.  If there are any Python developers on  
> this list in the Ottawa area, I'd be interested to hear from them too.
> 
> Thanks in advance,
> 
> James
> 
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

There are two major Linux community distros now -- Fedora and Ubuntu. 
You've tried both of them and they've croaked. A couple of things you 
can try:

1. In general, more *recent* distros have a better shot at finding and 
dealing with unusual hardware than the older ones. So rather than 
dropping back to Fedora 5 or 6, you're better off trying to get 8 or 
pre-release 9 to work.

2. The major distros all have forums where people who are having 
problems like yours can get help.

3. When you boot a Fedora install DVD, you have an opportunity to do a 
media check to see if the download and burn was correct. If you didn't 
do that, do it now, and if you have a bad DVD, you'll need to download 
again, burn again, and media check again until you have a good one! I 
think you can do this with Ubuntu as well, but I haven't tried it recently.

4. If the graphic *installer* doesn't work, there is a text-based 
installer that might work. You'll have to set up your X and desktop 
later, but it's worth a try if the other things fail.

5. If you can't get Fedora or Ubuntu to work, there are other distros 
you can try. CentOS 5 and Debian "Etch" are solid, stable distros. They 
are probably carrying older packages than what you'd like in the ideal 
case, but if t

Re: [OLPC library] MATLAB for OLPC?

[M. Edward (Ed) Borasky]

Ian Bicking wrote:
> I'm not sure his summary here is true.  You can do efficient operations 
> over sets of data in Python (actually due to some small tweaks to the 
> language requested by NumPy/Numeric users back around the time of Python 
> 2.1).  So if you do something like "array * 6", it actually does the 
> multiplication of items in the array in C.  They bend Python's "magic 
> methods" quite a bit in NumPy, so things like array access, slicing, and 
> multiplication all avoid actually iterating over the arrays or matrixes 
> in Python.

1. Python's array handling is the subject of an excellent chapter in 
"Beautiful Code". :)

2. I may be wrong, but IIRC NumPy is already installed on the XO by 
default! I forget what depends on it, but I think it's there as a 
dependency.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [OLPC library] OLPC+MATLAB+Greene DNA Chip = Disease Tricorder for developing world

[M. Edward (Ed) Borasky]

M. Edward (Ed) Borasky wrote:
> R has several server options, although I've never used them. 

http://www.rforge.net/Rserve/

http://cran.r-project.org/src/contrib/Descriptions/R.rsp.html

and if you absolutely positively *must* use Windows,

http://cran.r-project.org/contrib/extra/dcom/RSrv135.html


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [OLPC library] OLPC+MATLAB+Greene DNA Chip = Disease Tricorder for developing world

[M. Edward (Ed) Borasky]

Kate Davis wrote:
> At Yale, we run Matlab gridWorks (??), which is, as I understand, a 
> distributed client-server type environment.
> 
> Perhaps a slim Sugar client with a Matlab server might be a good option? 
> There are some S+ c/s options, and I can check in to R.

R has several server options, although I've never used them. By the way, 
there is also an interface between R and Python, called "RSPython". I've 
never used it, being Python-illiterate. It lives at

http://www.omegahat.org/RSPython/


There is some documentation on how to "port" Octave or Matlab code to R at

http://cran.r-project.org/doc/contrib/R-and-octave.txt

There is an R package to interface with Matlab at 
http://cran.r-project.org/src/contrib/Descriptions/R.matlab.html

and another at

http://www.omegahat.org/RMatlab/

Oh, yeah ... as long as we're talking R and BioConductor, a couple of 
years ago the BioConductor team offered an advanced R programming 
course. It's coming up again in a couple of weeks ... see

https://secure.bioconductor.org/SeattleFeb08/index.php

> 
> Any computational biology projects are well beyond both the cpu and 
> memory capabilities of the XO, but the XO might be an extremely useful 
> remote data collection client, and analytic server client, in the right 
> setup.

Getting this stuff to run on a school server is easy -- the question is 
simply how big a school server one could deploy in the developing world. 
Let me know if you want some help tweaking the R compiles. ;)


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [OLPC library] MATLAB for OLPC?

[M. Edward (Ed) Borasky]

Rafael Enrique Ortiz Guerrero wrote:
> 
> 
> 2008/1/28 C. Scott Ananian <[EMAIL PROTECTED] >:
> 
> On Jan 28, 2008 5:24 PM, Ivan Krstić
> <[EMAIL PROTECTED]
> > wrote:
>  > On Jan 28, 2008, at 8:04 PM, Cleve Moler wrote:
>  > > (I doubt that MATLAB runs in the OLPC, but I'm not sure.)
> 
> There are a number of open-source replacements for MATLAB, including
> GNU Octave ( http://www.gnu.org/software/octave/ ) and Maxima (
> http://maxima.sourceforge.net/ ).
>  --scott
> 
>  
> 
> 
> --
> 
> Another interesting open source math project also pointed as a 
> replacement of matlab is Sage
> *
> http://www.sagemath.org/**   *

My turn:

1. Both Maxima/XMaxima/wxMaxima and R run on my XO out of the box 
courtesy of "yum". With the Maximae, you get your choice of Lisp run 
times. I've successfully used both the clisp and SBCL runtimes. They do 
have a lot of dependencies, however, so watch your flash space.

2. Maxima is a Computer Algebra System and R is a graphical and 
statistical/numeric package. Both will do "number crunching", but 
they're two different beasts, and both fundamentally different beasts 
from Matlab.

3. There are two and a half "free Matlab clones". Someone mentioned 
Octave, but there is also Freemat, and a "half-free" package called 
SciLab. I call SciLab "half-free" because I don't know its exact 
license. You can download it freely, but I'm not sure all of the GPL 
freedoms are in place on it. I have used exactly none of these -- I 
learned R and don't see the need for another number cruncher.

4. On to Sage -- Sage is a wonderful package. It is written in Python 
and wraps many specialized and more general math packages. Its goal is 
to replace Mathematica, Maple, and some other less-well-known math 
packages. However -- it's huge. And it installs everything independently 
of whether you have the same package already as part of your distro.

I loaded it once, but there were only two or three rather specialized 
packages in Sage that weren't in my Gentoo repositories already. I think 
it's modular -- you don't have to load the whole enchilada. I might load 
the base on my virtual XO just to see how much space the core takes, 
because it's really an excellent collection.

If you can only load *one* math package, I highly recommend wxMaxima 
with the clisp run time. That's going to give you the most bang for your 
flash space. You don't really need XMaxima -- wxMaxima is a much better 
UI. By the way, wxMaxima also runs on Windows!!

Well ... so does R. In fact, the Windows UI for R is better than the 
core Linux UI. :)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: MIDI does support non-Western music (was: Why can't i access /dev/dsp or /dev/snd on my XO)

[M. Edward (Ed) Borasky]

imm wrote:
> Ed Borasky was suggesting scala as a useful thing to have working in  
> this regard...
> 
> Now, that got me wondering - I know it's freeware, but I'd never seen  
> the source anywhere. Have I missed it somewhere? That is, is it open  
> source?
> I'd always (somehow) assumed it was closed source.
> And if it is closed source (albeit freeware) does that put it "out of  
> bounds" for this programme?
> 
> 
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

It is free but I believe the author has placed a non-commercial 
restriction on it, which makes it incompatible with a number of open 
licenses. It is written in Ada, and if you ask the author for the 
source, he will usually send it to you. Once I get it running on my 
AMD64, I'm planning to see if he'd be willing to go with a more 
compatible license. The web site is 
http://www.xs4all.nl/~huygensf/scala/ if you want to write to him.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: MIDI does support non-Western music (was: Why can't i access /dev/dsp or /dev/snd on my XO)

[M. Edward (Ed) Borasky]

Albert Cahalan wrote:
> imm ian writes:
> On 22 Jan 2008, at 4:11, Albert Cahalan wrote:
> 
>>> You don't need to abuse pitch bends. MIDI lets you
>>> redefine the pitches of the notes. You can redefine
>>> middle C to be 1234 Hz if you like.
>> Mmm, well, yes, but...
> 
> No "but". You can redefine at will, for individual notes.
> 
> If you need a player, try timidity. If you have obsolete
> equipment that can only do pitch bends, you can use Scalia
> to convert a MIDI file. Scalia can also convert back.
> 
>> It's not so much the pitches that are the issue, it's the
>> intervals, and MIDI kind of constrains what you can do about
>> that, so you do kind of end up abusing pitch bend...
> 
> Nope. (not that abusing pitch bend is a tragedy though)
> 
> Since 1996, the MIDI tuning specification has allowed you to
> set the pitch to within 1/16384 of a semitone.
> 
> Since 1999, the MIDI tuning extensions have made this a bit
> more efficient.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

I'm still trying to get scala (not scalia -- he is or was a Supreme 
Court justice) to run on the XO. It requires some Ada run-time libraries 
and the GTK Ada bindings. It requires "gtkada" 2.8. Is that compatible 
with what's on the XO?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Why can't i access /dev/dsp or /dev/snd on my XO

[M. Edward (Ed) Borasky]

imm wrote:
> On 22 Jan 2008, at 3:43, M. Edward (Ed) Borasky wrote:
> 
>> By the way -- as far as microtonal and xentonal and "world music  
>> scales"
>> are concerned, MIDI's pitch bends are an awkward hack. Serious
>> *microtonal* algocompsynth practitioners either have to spend time
>> working around MIDI or use something else.
> 
> My worry is that a lot of the communities these machines will go to,  
> will want to perform microtonal and xentonal music, but are a long  
> way from being algocompsynth practitioners, and the MIDI tweaking  
> involved is, well, lets say "non-trivial"... I saw Csound (presumably  
> wrapped by TamTam or *something*) as the way to enable that facility  
> transparently. Perhaps I am too optimistic?
> 
> 
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Let's get our own mailing list -- or appoint a "curator of world music" 
on the Wiki. I am on the "[EMAIL PROTECTED]" mailing list, which 
has the experts on xen/microtonality, world scales, etc. But as far as I 
know I'm the only one there with an XO.

Let me see if I can get "scala" (the musical scale analysis program, not 
the programming language) up and running on my XO. That's pretty much 
required if you want to deal with world scales.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Hershey Felder, Zulu Musical Instruments, Essential To Develop Musical Traditions In Africa

[M. Edward (Ed) Borasky]

william romsay wrote:
> Hershey Felder, Zulu Musical Instruments, Essential To Develop Musical 
> Traditions In Africa
> 
> 
> African music is the music of Africans who live in a large region of 50 
> nations, each with a special culture, history and language, South of 
> Sahara. Zulu musical instruments are part of this multilingual culture. 
> African music has some distinct characteristics: the use of repetition 
> is one of them. Another important characteristic is the polyphony; this 
> is the combination of different musical parts played simultaneously.
> 
> 
> 
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel

1. Interesting ... can you post some links?

2. I think it's time the OLPC project had a list specific to audio on 
the XO and the "world music" aspects of it. Does someone on the project 
want to create such a list in the main MailMan area, or should I go 
ahead and start a Google group on the subject?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Why can't i access /dev/dsp or /dev/snd on my XO

[M. Edward (Ed) Borasky]

Albert Cahalan wrote:
> On Jan 21, 2008 12:27 PM, Walter Bender <[EMAIL PROTECTED]> wrote:
>> (b) as has been pointed out repeatedly, CSound is an open standard
>> (which incidentally predates the MIDI standard).
> 
> It may be open, but it isn't much of a standard.
> I've only found one implementation, csound itself.
> There are no hardware implementations.
> 
> Pushing this kind of thing is **wrong**.
> 
>> (c) Victor gave some very compelling reasons as to why CSound is a
>> better choice, especially for a program that is reaching out to
>> non-Western musical sensibilities.
> 
> MIDI does non-Western stuff, including unusual
> tuning systems.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Well ... I can't speak for the children of the developing world, nor do 
I have much experience with Tam Tam as wrapped around CSound. But as 
someone who has devoted a fair amount of his spare time over the past 
few decades in pursuit of algorithmic composition and synthesis 
(henceforth abbreviated algocompsynth) of music, I think I'm qualified 
to lay down some opinions here.

1. MIDI is limited but more or less universally spoken. Serious 
algocompsynth *must* involve support of MIDI. CSound recognized this 
years ago.

2. There have been numerous attempts to improve on CSound, but nothing 
else has come forth that's as comprehensive. The orchestras read like 
assembler code and the scores like a 1960s FORTRAN input card deck, but 
just about every working algocompsynth practitioner has it and knows it. 
So, serious algocompsynth *must* involve support of CSound.

3. There are a number of specialized Linux distros for audio. The three 
that I know the most about are Studio64, Jack Audio Distribution (JAD) 
and dyne:bolic. Almost all of them have a patched low-latency kernel, 
and all of them use something called the Jack Audio Connection Kit. They 
may still have to support both OSS and ALSA, but as I noted in another 
post, ALSA had support years ago for sound cards that weren't supported 
by free-as-in-freedom OSS drivers. So, serious algocompsynth on Linux 
*must* have a low-latency patched kernel, ALSA, and the Jack Audio 
Connection Kit.

4. Finally, there are three languages now in common use in 
algocompsynth. In historical order, they are Lisp/Scheme, Java and 
Python. Forth was prominent at one time as well, but the major Forth 
algocompsynth codes have mostly been ported to one of the other 
languages. So, serious algocompsynth *must* provide Python, Lisp/Scheme 
and Java support.

So the question in my mind is, "Should the XO be a platform for serious 
algocompsynth, or should it be what the project says it is -- an 
educational project for children to explore and discover?" Do children 
need MIDI, CSound, low-latency kernels, Jack, Lisp and Java? I don't 
really think so. The fact that Tam Tam has CSound and Python "under the 
hood" is only a convenience for the implementers. There are many fewer 
wheels that need to be re-invented as a result.

By the way, one other note here. A number of the more advanced synthesis 
algorithms edge upon some rather well-defended patents by Yamaha and 
others. It's a similar issue to those around media codecs -- even though 
technically such things are "software" or "formulas" that can't be 
patented, once they get embodied in pieces of gear that you can buy in a 
music store, things change.

By the way -- as far as microtonal and xentonal and "world music scales" 
are concerned, MIDI's pitch bends are an awkward hack. Serious 
*microtonal* algocompsynth practitioners either have to spend time 
working around MIDI or use something else.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Why can't i access /dev/dsp or /dev/snd on my XO

[M. Edward (Ed) Borasky]

Ivo Emanuel Gonçalves wrote:
[snip]

I gave up on OSS years ago, when I discovered that there were dozens of 
high-quality sound cards without free OSS drivers! Alsa was release < 1 
back then, and there was very little documentation. That's been fixed, 
and I am not going back to OSS!
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: OLPC and GLX

[M. Edward (Ed) Borasky]

Bryan Duff wrote:
> After compiling in mesa source to get GLX working on the FC7-based
> builds.  Running `glxgears -fullscreen` I get ~25 fps.  Compared to
> Ubuntu which gets 65 fps, this is rather poor.
> 
> I think the Ubuntu performance shows that 3D (albeit simple 3D) is very
> possible and worthwhile.
> 
> I've tried the amd and fbdev drivers - they make no performance
> difference.  Neither does i386 versus i586 (w/ mmx and 3dnow! included
> in cflags).
> 
> I haven't tried DRI.  I don't understand how it would help if I
> understand DRI correctly, but perhaps I'm missing something.  I have no
> idea if processes that are re-niced automatically, or  have some sort of
> resource limitation.
> 
> I'm using the default xorg.conf.  I hope the problem is something simple
> that I'm not seeing.
> 
> Thanks.
> 
> -Bryan Duff
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Have you profiled any of the code? Compiled it with gcc flags for the 
architecture (mmx and 3dnow)? Then again, maybe the AMD libraries have 
some assembler kernels for this.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Why can't i access /dev/dsp or /dev/snd on my XO

[M. Edward (Ed) Borasky]

Albert Cahalan wrote:
> On Jan 19, 2008 4:33 PM, victor <[EMAIL PROTECTED]> wrote:
> 
>> I can't speak for TamTam because I am not involved in their
>> design details, but I can say this, Csound's standard score
>> preceeds MIDI by at least a decade (or two if you consider where
>> it came from). It is much more flexible to convey musical data
>> than MIDI. There are MIDI to csound score converters, but
>> that is beside the point, because Csound can play MIDI files
>> directly, receive realtime MIDI data and even output it.
>> There is no problem whatsoever, with the proper instruments,
>> Csound will be a MIDI synthesizer like any other. The main
>> thing is, that it is not limited to it (thank goodness...).
> 
> How about showing some support for standards by
> dropping the non-standard stuff? You can #ifdef it.
> Maybe you can even save a few bytes.
> 
> If you really must, you can embed the non-standard
> stuff into a MIDI file. It's better to avoid non-standard
> stuff entirely of course, and any extended MIDI file
> had better play decently on a standard MIDI player.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

One of the main reasons I got an XO was because it has CSound. It's a
ghastly API, but it's been around for years and there are thousands of
working instruments! There's a huge book on it, and I doubt very
seriously if anyone will ever come up with a digital sound analysis and
synthesis tool set as comprehensive without investing a lot of effort
re-inventing a bunch of wheels, levers, inclined planes and such.

By the way -- I've been meaning to check to see if this is in Trac, but
the csound-manual and csound-tutorial RPMs in the repository appear to
be empty. I can install them, but there isn't anything on the machine
after I do.

I'm also attempting to get some of the Planet CCRMA software loaded on
the system. At this point, all I really want is Common Music -- I don't
need another synthesizer since I have CSound, and I don't need a music
notation program. If anyone else has already done this, I'd love to hear
about it.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: disabling root and olpc passwords

[M. Edward (Ed) Borasky]

Mikus Grinbergs wrote:
> The 2008-1-12 OLPC News says "... so that we can finally disable the 
> root and olpc passwords".
> 
> The way I have my G1G1 system set up (I have no wireless) I *need* 
> to ftp in.  For that, I have set a password for olpc.  It would be 
> ok with me to set up a different user+password for ftp, but would 
> *not* be ok for password support to be "disabled".
> 
> Also, I don't believe in the "political correctness" of not using 
> root.  I do need to install/remove/change things as root, and 
> *strongly* prefer not to use 'sudo' for that -- I log in as root, 
> and am willing to take the risk of committing a disastrous mistake. 
>   Here, too, having a password seems "natural" to me.
> 
> I agree with the aim of making the OLPC simple to use, but please 
> don't take passwords away entirely.
> 
> mikus
> 
> 
> p.s.  I presume the existing 'passwd' command was taken from Fedora. 
>   It is too paranoid, forbidding too_short passwords, 
> too_homogeneous passwords, too_similar passwords, etc., etc., etc. 
> Such rules may be needed for a datacenter - but for a schoolroom?
> 
> 
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Typical Linux practice is the following:

1. One *never* allows remote shell login as "root" -- *ever* -- even 
behind a firewall. One allows only *one* user in the "wheel" group to 
log in to a shell account, and then *only* via "ssh".

2. When root access is needed, "sudo" is used, with the least permissive 
mode possible.

3. "ftp" is done using "sftp" and/or "scp". For Windows clients, there's 
PuTTY.

Anything less than this level of security is a bad habit -- a *very* bad 
habit. Please don't encourage such habits, or ask the open source 
community to cater to them.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fwd: Dailymotion for XO laptop

[M. Edward (Ed) Borasky]

[EMAIL PROTECTED] wrote:

> The other option is to write implementations of the codecs that avoid
> the patents. Whether that is possible depends on the exact wording of
> the patent, and sometimes it takes a few weeks working with a good
> patent attorney to work out exactly what the patent really
> says. Sometimes it just isn't possible.
> 
> We really need a open project to do patent analysis of this kind and
> determine which of these key patents (not just codecs, but also other
> important blocking patents) can be avoided, and which ones are too
> tied to the format to avoid. Perhaps the OLPC project would provide a
> good bit of motivation for people to do this type of work?

1. There are two kinds of "good patent attorneys". One kind works "pro 
bono" for free software and the other gets paid big bucks by patent 
holders. It's not just a question of quality or attorney or length of 
time spent. It's a question of two competing sides of a specific and 
very detailed technical and legal argument being thrashed out in the 
press and in the courts. If you head to the Groklaw web site, you can 
see this sort of thing (from the free software side). This process does 
not take a few weeks but *decades*.

2. I personally don't think the OLPC project has the bandwidth or the 
energy to get involved in such a struggle. As the recent events between 
OLPC and Intel have showed, when you wrestle with a pig, both of you get 
mud all over but only the pig enjoys it. :)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: GLX available?

[M. Edward (Ed) Borasky]

NoiseEHC wrote:
> Try to contact him, he ported TinyGL successfully:
> http://lists.laptop.org/pipermail/devel/2007-June/005375.html
> 
> 
> 
>> As a data point, Doom runs quite fast even at full resolution on a  
>> B4. Have not heard reports on getting Quake running. But I suspect  
>> that a software renderer hand-optimized towards the XO could be made  
>> to work at decent frame rate.
>>
>>   
> The databook of the LX is so low quality that hand optimization requires 
> testing everything (including if the published clock counts are matching 
> reality) and it has not been done yet. Also the hight cache hit penalty 
> and the in-order nature of the Geode means that a different algorithm 
> could be required than naively fetching texels while texture mapping. 
> Quake requires a preprocessed/prelit BSP tree so unless you want to 
> write an FPS it is not a perfect fit. Preprocessing also could take 
> hours on an XO (if you want children to design levels).
> 
> Just a quick calculation (400x300, 30 fps):
> 400x300x30 = 3 600 000 pixels/sec filled (with no overdraw)
> 433 000 000/3 600 000 = 120 clock/pixel (if you do not do anything else 
> in the code, but just divide it by 4 because of game logic/geometry 
> transform/overdraw)
> So you have ~30 clock/pixel and a cache miss takes 25 clocks (if the 
> documentation is ok but not tested)...
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

1. While it does not cover the Geode (or any AMD chips other than
Athlon64), there is a wonderful reference on C/C++ and assembler
tweaking, including some assembly language tools for figuring out what
your chips are up to, at http://www.agner.org/optimize.

2. The Vector Pascal compiler
(http://www.dcs.gla.ac.uk/~wpc/reports/compilers/compilerindex/x25.html)
and the book about it (_SIMD Programming Manual for Linux and Windows_)
have excellent coverage of the MMX and 3dNow! instruction sets. The
compiler and IDE are written in Java, so you probably want to do
cross-compiles/links on a Fedora box.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: A jabber hosting offer...

[M. Edward (Ed) Borasky]

Dave Belfer-Shevett wrote:
> I've just received my, er,  my son's XO, and he's ecstatic with it, 
> enjoying fiddling with Python programs and other tidbits.  I've heard 
> that the Jabber 'chat' functions are disabled on the US XO's, mostly 
> because the existing jabber hosts can't really take the load of all 
> these machines going out.
> 
> I have machinery and bandwidth available for setting up a dedicated, 
> fairly powerful machine specifically to run Jabber for the OLPC 
> community.  I have no problems building, configuring, installing, and 
> maintaining the machine in a colo facility in Bedford, MA.  I'll donate 
> the hardware and time to make it work, if it'll benefit the project.
> 
> I'm looking at a dual-Xeon 2.8 gig Ubuntu Gutsy box (1U) with a pair of 
> mirrored drives.  The facility has multiple peered connections (it 
> supports a series of VOIP servers), and is well managed.
> 
> I have experience running jabber servers, and sysadminning.  Would this 
> be of benefit to the community?
> 
> Please let me know.  I can be reached on jabber at 
> [EMAIL PROTECTED] , or email at [EMAIL PROTECTED] is fine too. 
> I'm also (obviously) on the devel list :)
> 
> Thanks.  I do want to contribute to the project in any way I can...
> 
>   -dbs
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Some other people have suggested local Jabber servers, and I agree with
them. What I think would be really wonderful would be if someone could
make a LiveCD that would boot up as a Jabber server for XOs! It's a real
bear to configure one, and a LiveCD would make it feasible for
communities to do it.

I've been on the xochat.com server from time to time, and there are
sometimes so many people you can't really "meet" anyone.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: shutting off wireless for air travel

[M. Edward (Ed) Borasky]

Mitch Bradley wrote:
> This is all documented on the wiki.  See:
> 
> http://wiki.laptop.org/go/Airplane_mode
> 
> and
> 
> http://wiki.laptop.org/go/Support_FAQ#How_do_I_disable_wireless_when_flying.3F
> 
> If you don't want the wireless to restart automatically after a reboot, 
> renaming /lib/firmware/usb8388.bin works.

Yeah ... when you take off, your XO is powered off and stowed. Then when
they allow you to operate it, you'll boot it up and the wireless will
come on until you can get to a terminal and type the
"sugar-control-panel" command. So ... I think this hack is necessary if
you're going to operate the XO in flight.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: XO as a scientific platform: wiki page

[M. Edward (Ed) Borasky]

David W Hogg wrote:
> FWIW, I started a wiki page (on my research group wiki) about setting
> up my G1G1 XO for scientific writing on the road (and, eventually,
> research, but right now my job is to write a grant proposal on the XO
> as I travel around this weekend).
> 
> http://howdy.physics.nyu.edu/index.php/Setting_up_an_XO
> 
> Sorry for the spam.  Hogg
> 

Yeah ... I'm going in similar directions. A couple of other notes:

1. Both LyX and TeXmacs are in the repositories, and I've installed both
successfully. I forget what the dependencies are, etc. I prefer LyX for
ease of use, but TeXmacs will allow you to insert "sessions" from many
other packages painlessly.

2. If you do number crunching, the XO comes with a fair amount of
(Python) numerical software. It also has the Atlas linear algebra
libraries, although the ones loaded on the box are only tuned to i386
level. More highly optimized versions are available via "yum".

3. wxMaxima costs about 28 MB IIRC, if you install the "clisp" run time,
and you get clisp as part of the bargain. The default Maxima uses gcl
and is about the same size. The SBCL version is also available, but it's
much larger. If you have TeXmacs and wxMaxima, you can get a Maxima
session directly into a document.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Uniting the community's infrastructure

[M. Edward (Ed) Borasky]

Chris Hager wrote:
> Hi all!
> 
>  From now on, the channel #olpc-groups is open with the ambition to 
> connect local communities from everywhere! I can imagine a lot of 
> potential for collaborations, projects, problem solving and 
> not-reinventing-the-wheel :) !
> 
> SJ and I have talked about the current status of the irc-infrastructure 
> for the communities, and had some ideas for the future. Basically,
> 
> 1. many of the local channels are inactive
> 2. we have had no channel to cross-connect communities
> 3. the channel #olpc-europe developed from the ccc
> 4. the whole olpc project is gaining momentum
> 
> =>
> 
> 5. the logical next step would be a united communities-channel to focus 
> our common effort!
> 
> We could basically connect via #olpc-groups on a global basis, and use 
> #olpc-europe/asia/africa/... for regional discussions.
> 
> The same idea could apply to some of the mailing-lists. We can basically 
> use the grassroots list (http://lists.laptop.org/listinfo/grassroots)! 
> Please post feedback to [EMAIL PROTECTED]
> 
> I wish you all an interesting and successful year 2008!
> 
> - chris
> (olpc austria)

Well, like a lot of the Internet and life in general, things tend to
"self-organize". I've started a Google group for Portland Oregon XO
"owners", and I'm personally on quite a few of the other mailing lists.
In any event, I'm not sure it's time yet for any "consolidation", either
on mailing lists/Google Groups or in IRC. I think the best we should
hope for at this point is to ask everyone on IRC to use the characters
"olpc" in naming channels. That way, the IRC client channel listers can
find them.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Updates API documentation for everything.

[M. Edward (Ed) Borasky]

Edward Cherlin wrote:
> Does anybody know of a documentation tool for Open Firmware, or for
> FORTH more generally? Exploring using 'words' and 'see'

Are you looking for automated documentation generation, or FORTH coding
and documentation standards? I don't know about the former, but there is
a well-established set of the latter, and given adherence, I'm sure the
former is eminently possible.

"The FORTH community" is fairly small (relative to, say, Python), and as
a result, most FORTH programmers don't have much trouble reading the
code of other FORTH programmers. But I don't know about outsiders coming
to FORTH from "more traditional" languages. :)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Kernel configuration options

[M. Edward (Ed) Borasky]

Mitch Bradley wrote:
>  From a security standpoint, there is an advantage to building in 
> everything.  The main kernel is verified with a crypto signature before 
> it is executed.  Loading a module without first verifying a 
> similarly-strong signature weakens the security.
> 
> Modules are a good idea for kernels that are intended to run on a wide 
> variety of hardware.  I am in favor of treating XO like an appliance and 
> making the kernel as monolithic as possible.

I'm not familiar with the security stuff in general or this case in
particular. But I think the "trend" in the Linux community has been
towards more flexibility, moving stuff from kernel space to user space, etc.

Then again, since the *hardware* is soldered onto the mainboard and
can't be easily expanded, why shouldn't the *kernel* be just as
inflexible? ;)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Kernel configuration options

[M. Edward (Ed) Borasky]

John Richard Moser wrote:
> I'm also noticing some things like KALLSYMS and BUG(), BSD process 
> accounting, and the like.  KALLSYMS, BUG(), and printk() are useful; on 
> a true embedded device I'd say remove 'em but I can't justify it here... 
> BSD process accounting and auditd support though?

Yeah ... I run stuff like that in *my* kernels, but I'm a kernel geek.
Does that stuff need to be in a machine in a village school in Rwanda?

P.S.: I do know a Red Hat geek in Nigeria. :)

> 
> In the same line, a lot of debugging options are in use.  I'm using 
> Build 653, I guess it may be a developer's build and thus there's a lot 
> of debugging stuff; but in the final should things like CONFIG_PROFILING 
> , CONFIG_KPROBES, CONFIG_DEBUG_FS, CONFIG_UNUSED_SYMBOLS, 
> CONFIG_SCHEDSTATS, CONFIG_TIMER_STATS, CONFIG_DEBUG_PREEMPT, 
> CONFIG_DEBUG_SPINLOCK ... be removed?

I think I agree with you here with the possible exception of
CONFIG_PROFILING aka "oprofile support".

> 
> Finally, I'm noticing a lot of stuff can be built as modules, but is 
> built in.  Networking in its entirety; sound drivers; mouse; and USB 
> (the mouse looks like it's PS2) can be loaded by HAL and UDEV, but this 
> will increase boot time (then again, HAL apparently needs to be fixed 
> anyway[2], so this could be encouraging?).  More interesting to me 
> however is that EXT2, EXT3, Joliet, ZISOFS, RAMFS, NFS, and possibly 
> PROMFS because I don't think JFFS2 depends on it BUT I'm not sure if 
> it's used at some point before it can be loaded as a module.

I'd keep as much as possible that's XO-specific and *always* going to be
loaded built in. You're always going to need sound, mouse and keyboard,
wireless and video, and I'd keep the core USB stuff built in as well,
since that's really the only way to talk to the machine besides
wireless. But the rest of it should be built as modules.

The filesystems should be there (as modules). The XO is going to be
talking to school servers, and there's no point in ruling out NFS,
Samba, plugging in an external USB DVD reader/writer, etc. Again, the
question is, "would a village in Rwanda format a USB stick ext3?" :)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Is the "yum" repository down?

[M. Edward (Ed) Borasky]

I just built a VMware virtual XO (build 655) and I can't seem to access
the "yum" repositories:

-bash-3.2# yum search chess
http://mock.laptop.org/repos/local.ship2/repodata/repomd.xml: [Errno 12]
Timeout: 
Trying other mirror.
Error: Cannot retrieve repository metadata (repomd.xml) for repository:
olpc_local-ship2. Please verify its path and try again
-bash-3.2#


I haven't put my real XO on the network today -- too busy crawling
around under my work table dealing with useless KVM switches. ;) But I
can't think of any reason why this shouldn't work.

P.S.: "wget" on the host is also not happy:

$ wget http://mock.laptop.org/repos/local.ship2/repodata/repomd.xml
--18:30:46--  http://mock.laptop.org/repos/local.ship2/repodata/repomd.xml
   => `repomd.xml'
Resolving mock.laptop.org... 18.85.46.52
Connecting to mock.laptop.org|18.85.46.52|:80... failed: Connection refused.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: I got a developer key -- now what? :)

[M. Edward (Ed) Borasky]

Chris Ball wrote:
> Hi,
> 
>> I signed up for a developer key, so I have one now. But what can I
>> do with it?
>
> You can do anything that you'd expect to do with a standard laptop;
> install any operating system, and flash a new BIOS.
> 
>> How can I be sure I'm not going to nuke the XO beyond all recovery?
>> Is there some kind of documentation on what's risky and what's
>> safe?
> 
> You're safe no matter what you do to the NAND, because the firmware can
> flash a new NAND image from USB.  If you want to be sure of not bricking
> it, you should avoid flashing firmware that isn't signed by OLPC.
> 
> - Chris.

Ah ... OK. I have the procedure for flashing the NAND, but I haven't
seen one for flashing the firmware. Is that documented somewhere?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

I got a developer key -- now what? :)

[M. Edward (Ed) Borasky]

I signed up for a developer key, so I have one now. But what can I do
with it? How can I be sure I'm not going to nuke the XO beyond all
recovery? Is there some kind of documentation on what's risky and what's
safe?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 3rd Fedora disk curdle

[M. Edward (Ed) Borasky]

Gerard J. Cerchio wrote:
> M. Edward (Ed) Borasky wrote:
>> 1. Define "curdle their ext3 disk".
>>   
> I use the word curdle to describe a disk with lost inodes, sectors that
> are multiply  allocated, and other such problems that fsck valiantly
> tries to correct but winds up with a non-working system. In the last
> case I got a clean boot and clean subsequent fsck, however many
> applications just seg faulted when I tried to run them.
>> 2. What version of VMware are you using?
>>   
> VMWare server 1.04 (latest freebe)
>> 3. What's your host version?
>>   
> Windows XP 64 SP2. Booting off an Nvidia RAID5 4 disk 1 TB.
>> 4. Are you using the default settings on your virtual disks or are you
>> changing them in some way?
>>   
> The first two I just took all the defaults. This new one I put on the
> IDE controller. I don't have a whole bunch of faith in the snapshot
> system either so  I just don't use it.
> 
> Now here is something interesting, I just did a NTFS chkdsk on the RAID
> and it moved my vmdk's into a copy of the the VM's directory while
> reporting 2 lost files. These are the first errors I have ever seen on
> the NVidia RAID after running it for a year. This has me nervous enough
> to move the VM's off the RAID and onto an IDE backup drive in the
> system. Maybe I am looking at some VMare/XP64 interaction. Anyway I'll
> report any new problems on this thread.

Yeah, VMware can't deal very well with host-side platform problems. :)

How sure are you that VMware Server 1.04 works on Windows XP 64 SP2? I'm
running VMware Workstation *6* on my AMD64 box -- I haven't messed with
the free server in a while and I've never tried it on a 64-bit machine.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: 3rd Fedora disk curdle

[M. Edward (Ed) Borasky]

Gerard J. Cerchio wrote:
> In the short time I have been working with olpc I have had my Fedora 
> VMware machines curdle their ext3 disks 3 times.
> 
> I have been running 2.4 and 2.6 Redhats and Debians for over a year with 
> no such problems. Once the first Fedora 7 machine broke its disk I have 
> been very careful to shutdown every time. I cannot point to any 
> particular activity that has corrupted the disk image. Things will start 
> to go "wrong" and a subsequent reboot with fschk will yield a disk 
> hopelessly in trouble.
> 
> Does the jhbuild emulator do any exotic direct to disk IO that may be 
> causing this?
> Does Fedora aggressively modify its ext3, vfs or SCSI drivers?
> 
> I have built a third Fedora 8 VM this time using IDE disk IO in hopes 
> that this problem will go away.
> 
> Has anyone else seen this kind of problem?

1. Define "curdle their ext3 disk".
2. What version of VMware are you using?
3. What's your host version?
4. Are you using the default settings on your virtual disks or are you
changing them in some way?

My answers to 2 - 4:

2. VMware Workstation 6
3. Gentoo Linux on an AMD64 (Athlon64 X2 dual-core with 4 GB of RAM)
4. No ... I turn off the snapshots and make my virtual disks independent
and persistent, and I turn on write caching.

I have built virtual XOs, virtual Fedora 7 and 8 systems (32-bit!!), and
jhbuild inside a virtual Fedora 8 system, all without any data
corruption. The XOs I build with IDE drives because they won't boot from
a SCSI drive -- "hda" is hard-wired somewhere that I don't recall off
the top of my head -- but all the others use SCSI drives with no problem.

So the first thing I recommend is to check your hard disk settings in
VMware. The defaults *should* work without incident, but the snapshot
logic may be doing something to you.
> 
> -Gerard
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Error: Missing Dependency: libpoppler-glib.so.1 is needed by package evince-olpc

[M. Edward (Ed) Borasky]

Jim Oser wrote:
> I am trying to make a developer system with the source code.
> 
> I downloaded and installed ship2-olpc-653.vmdk and ship2-olpc-653.vmx
> 
> This is working fine in my VMWare Fusion setup.
> 

Where are the prebuilt vmdk and vmx on the Internet? I've been
downloading the ".img" files and making my own with "qemu" and VMware
Workstation!
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [olpc-help] setting correct time

[M. Edward (Ed) Borasky]

Chih-yu Chao wrote:
> Hi,
> 
> Please refer to
> http://wiki.laptop.org/go/Support_FAQ#How_do_I_set_the_date_and_time_on_my_laptop.3F
> 
> The OLPC team is currently working on that wiki section, so you might
> want to check back later for more updates.
> 
> Thanks a lot!
> Chih-yu
> 
> 
> On Dec 26, 2007 12:56 AM, Anonymous <[EMAIL PROTECTED]> wrote:
>> hello, I just received this neat OLPC and am enjoying it immensely. Playing 
>> a bit at the OS level, I determined that the system clock is about 13 hrs 
>> fast. It's already tomorrow  :D I tried a conventional linux date -s command 
>> to reset the time, but it replied with an error that I could not do that. Is 
>> there another way?  TIA
>>
>> dw
>>
>>
>>
>>
>>
>> ___
>> community-support mailing list
>> [EMAIL PROTECTED]
>> http://lists.laptop.org/listinfo/community-support
>>
> ___
> community-support mailing list
> [EMAIL PROTECTED]
> http://lists.laptop.org/listinfo/community-support
> 

There are two steps required. First, you have to set the time zone. Open
up the terminal activity and type "sugar-control-panel -h
timezone|more". This will page you through the list of all the possible
time zones. The space bar goes to the next page, and a carriage
return/enter goes to the next line.

Once you find the one you want, do a "q" to get out of "more". Then type
"su" to become root. Do *not* type "su -" -- this will mess up the path
settings. Then type "sugar-control-panel -s timezone ". To check it, type "sugar-control-panel -g timezone".

OK ... *now* type "su -". Then type "ntpdate pool.ntp.org". The machine
will sync to the time standard server. That's it.

Question for the developers -- some Linux systems will automatically
sync to the time server every time the network starts up. When you
install Fedora, for example, you get the option to do that. How
difficult would it be to enable that option in an XO?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Give One Get One laptop for software development

[M. Edward (Ed) Borasky]

Edward Cherlin wrote:
> I would want to work out algorithms in APL and translate to something
> else after they were working. I may be able to provide an APL for the
> XO next month. It's 29K. See Executable Math on the Wiki.

Very cool!! I've always wanted to learn APL but never had an excuse. I
think I installed APlus when I was running Debian, but when I switched
to Gentoo, I couldn't find the source.

>> 2. "Native" develop and test on the XO itself.
>>
>> The XO as currently shipping does *not* contain a development tool set
>> other than Python and Etoys, which is a version of Squeak, which is a
>> Smalltalk IDE. In particular, there is no Perl interpreter on the XO,
>> and there is not GCC or any of the conventional Linux development tool set.
> 
> The XO is memory and storage constrained. You can plug in a multi-G
> flash drive or an SD card for more storage, but you can' get more
> memory. There is no swap partition. You could plug in an external hard
> drive and set up a swap there.
> 
> There are instructions on the OLPC Wiki for dual-booting to Debian.
> You could adapt them to create a command-line development environment
> using a text-mode programming editor.

1. I've experimented with some other tools, and just about all of them
have a Perl dependency. So it might not be all that big a deal to just
bite the bullet and install Perl. When my main computer was a
Hewlett-Packard HP100-LX, I did on-box development in Perl 4 and Forth
and everything worked just fine.

2. I don't understand what advantages "dual-booting to Debian" offers
for the command-line developer over the modified Fedora 7 that's already
on the XO.

3. I suspect swap to a SD card would be slow. If the USB slots have the
power to drive one, you might be better off with a micro-hard drive
instead of flash. You don't need all that much swap space anyhow -- if
you've overloaded RAM to the point where you've filled up 256 MB of RAM
and another 256 MB - 512 MB of swap, your performance is going to be
horrible anyhow. You're much better off designing an efficient explicit
"out of core" algorithm than letting the swapper try to guess what
should be in RAM and what should be on "backing store".

>> On top of that, there is very little room on the XO's base flash hard
>> drive (1 GB) for any additional software at the system level. As a
>> consequence, I think you'd want to either cross-develop using the
>> standard GNU/Linux tools, or use a more compact development tool set on
>> the XO itself.
> 
> Yum is provided, so you can install a few hundred megabytes of
> software with no difficulty. Large amounts would require a bit of
> surgery, to install software other than in default locations and
> expand the path or symlink everything.

Yeah ... that's probably what I'll end up doing if I fill up the flash
drive. But right now it doesn't look like I'm going to need to do that.
I'm probably stuck with Perl and Tcl/Tk/Tkinter, but other than that and
gForth I don't think I need much more of an on-system toolchain.
> 
>> I recommend Forth for the "compact development tool set on the XO
>> itself". The stock Fedora "gforth" package should work, although I've
>> only tested it on an emulated XO -- my physical unit has not arrived
>> yet. And there is a version of Forth that is more or less "XO-native" --
>> it's a port of the boot firmware to run in the Linux environment.
> 
> Power on the XO. During the boot countdown, press the Esc key (top
> left, x in a black circle). This will put you at the FORTH 'ok'
> prompt. There are FORTH lessons on the Wiki.

Only if you have a developer key, I'm told, is the boot firmware Forth
available. Of course, I've applied for a developer key, but I'm somewhat
used to gForth already, so I've loaded it. It's under a megabyte and has
all the ANS floating point words. I'm not sure the boot Forth does.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Playing with IDEs

[M. Edward (Ed) Borasky]

Bernardo Innocenti wrote:
> Jeffrey Kesselman wrote:
> 
>> Think VERY carefully about his.  Your opening up a world of potential
>> hurt for 2D game developers and similar kinds of apps.
> 
> I designed a few 2D arcade games myself and I've found that it
> only takes a minimal amount of thought to make them properly
> scale within a reasonable range or resolutions.
> 
> PC games had the requirement of supporting multiple resolutions
> since the MS-DOS era.  Today Cairo makes it super easy.
> 
> And even if a very peculiar 2D bitmap game couldn't be made
> to scale easily, this is not a good excuse for all the rest
> of the activities to lazily hardcode screen coordinates and
> font sizes instead of computing them based on their window size.
> 

Well ... let me put on my marketing hat for a bit here:

The current XO is designed to have a life of five years. Now if it only
takes a bright child a year or two to outgrow the *hardware* of the XO,
that means that the current XO will probably get handed down to a
younger child and *someone, not necessarily OLPC*, will have an
opportunity to produce a more advanced hardware platform for "graduates"
of the XO.

So I'm with Bernardo on this one -- unless OLPC is willing to rule out
being the provider of the "XO Plus" for older children, activity
developers should do nothing that rules out portability, growth paths,
agility, or use cases only a little bit removed from the original use
cases around which the XO was designed.

What I do think is a good idea is to specify a *minimum* screen size
which activities must support. I'd guess 1024x768 is a fair choice at
this point in time -- I don't see any reason why activity developers
should be forced to run on 800x600.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Give One Get One laptop for software development

[M. Edward (Ed) Borasky]

Yoshiki Ohshima wrote:
>   Edward,
> 
>> a. There isn't enough room for the RPM on the base 1 GB hard drive.
> 
>   This, and what you wrote to the Ruby mailing list makes me think
> that there is some differerence between your environment and a typical
> installment on XO.  A clean installation of later Update.1 gives me
> about 65% free space.  your emulation environment doesn't take the
> compression of jffs2 into account, or you have something else
> installed?
> 
> -- Yoshiki
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Yeah, I discovered that yesterday when I got my physical machine. I've
been build my emulated XOs from the ext3 images as described somewhere
on the wiki. If there's a way to build an emulated XO using a jffs2
image instead, I'll switch over to that.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: [olpc-help] clicking 'back' crashes browse

[M. Edward (Ed) Borasky]

Anonymous wrote:
> this is a weird one, but its happened now several times so i wanted to post 
> it. roughly half the time i try to go 'back' during a browse session, the 
> activity will close abruptly. i would say that this is because of a complex 
> page but it happens on simple sites like the olpc wiki (although not, so far, 
> on this forum). anyone else have a similar issue?
> 
> 
> 
> 
> 
> ___
> community-support mailing list
> [EMAIL PROTECTED]
> http://lists.laptop.org/listinfo/community-support
> 
Yep, it's happened to me too -- if I can reproduce it, I'll file a
ticket on the bug tracker. I'm pretty sure Sugar keeps the log files
around for the developers to examine, so it's probably an easy thing to
fix. I'm cross-posting this to the Developer list.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

A couple of "adult use case" questions

[M. Edward (Ed) Borasky]

1. My typical use of laptops is nearly always with the AC adapter
plugged in. I don't travel a lot, and when I do, I generally don't
operate the laptop in an aircraft. I've been told that this is harder on
batteries than allowing them to discharge and recharge.

With the XO, though, I'm planning to use it more like a laptop -- only
connect to AC when required. Is that easier on the battery, harder on
the battery, or about the same as what I've done with my older laptops?

2. My experience with wireless in hotels and Linux is that:

   a. You usually need Windows and IE to authenticate the first time.
This isn't a problem because my other laptops have been dual-booted, but
the XO isn't and won't be.

   b. I've had a number of instances where something in the way the
Linux wireless configuration (it's a DHCP thing) can actually crash the
hotel's wireless infrastructure and require a reboot of it! This has
happened to me (and possibly other hotel guests at the same time) at
least three times. I don't know enough about the details to know exactly
what the mechanism is, but again, since I wouldn't have Windows, I
wouldn't be able to follow the support tech's instructions if I crashed it.

Of course, this use case is 180 degrees away from the reason the machine
was designed. But still, if I do go on the road, say, to a G1G1
conference somewhere, are things like this going to be a problem?
Conversely, if there *are* going to be G1G1 conferences, might I suggest
that they be held on a university campus rather than in a hotel? :)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Gen2 pointing device

[M. Edward (Ed) Borasky]

Albert Cahalan wrote:
> It's time to think outside the box again. These office-oriented
> pointing devices aren't working very well. Kids are commonly
> coated with stuff. That includes abrasive grit, conductive liquid,
> and opaque liquid. Mud and food are particularly common.
> 
> The touchpad fails with conductive liquid. Remember that people
> do not have equally sweaty hands, even in an equal environment.
> What works for you, or in a cold office, won't work for everybody.
> Now add mud and food.
> 
> Using a stylus unsurprisingly requires a stylus. This is not good.
> If the stylus were to be a primary input device, it would need to
> be in relative mode. Anything pressure-sensitive is going to
> encourage heavy pressure with sharp objects, leading to damage.
> On the positive side, at least it works fine while filthy.
> 
> The touchscreen is first of all just plain inadequate. It lacks the
> ability to hover, which is used throughout sugar. It is inaccurate.
> Depending on mechanism, it has either the touchpad or stylus problems.
> The really terrible thing is screen damage. Screens will not last long
> if kids are rubbing them with filthy fingers. Mud contains grit.
> 
> This could work:  http://en.wikipedia.org/wiki/Pointing_stick
> It might be possible to mold it right into the keyboard, preventing
> entry of liquids or grit while cutting costs.
> 
> This could work: http://en.wikipedia.org/wiki/Analog_stick
> http://en.wikipedia.org/wiki/Image:GameCube_Analog_Stick.jpg
> It's especially good for tablet mode, assuming that gets kept.
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 
What about a second camera and tracking head movements? Sonar? Can't I
just "point with my nose?"
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Playing with IDEs

[M. Edward (Ed) Borasky]

Jeffrey Kesselman wrote:
> Figured it was time for a new thread for this
> 
> Idle is actually included on the olpc in /usr/lib/python2.5/idlelib
> 
> However trying to invoke idle.py gives this error...
> 
> ** IDLE can't import Tkinter.  Your python may not be configured for Tk.
> 
> Question for those more familiar with python on linux, Is there
> something I can yum
> or otherwise download and install that would fix this?
> 
> JK
> 
I think the Tkinter RPM isn't loaded by default ... I think "yum install
 tkinter" will bring it in. If I knew how to do anything in Idle, I'd
try it on my emulated XO. :)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Give One Get One laptop for software development

[M. Edward (Ed) Borasky]

Rob Savoye wrote:

>> 2. "Native" develop and test on the XO itself.
>   GCC and G++ are both used with the XO.

Used with an XO, sure. But not *installed on* the XO by default. I run
virtual XOs in VMware Workstation 6. You can't install VMware Tools on
one because:

a. There isn't enough room for the RPM on the base 1 GB hard drive.
b. If you build a virtual XO with a larger hard drive, you can install
the RPM, but the configure script doesn't work. It needs Perl, which
isn't installed. If you install Perl, you find out that it also needs
"gcc" and "make", and if you install those, you find out that the kernel
headers or some other kernel package doesn't match the running kernel.

> I believe the geode optimized GCC and Glibc are included now. If not,
> here's instructions on building your own:
> http://wiki.gnashdev.org/wiki/index.php/Building_OLPC_Tools

I may check these out -- my base distro is Gentoo, so I may just skip
all this. :)

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Give One Get One laptop for software development

[M. Edward (Ed) Borasky]

Jeffrey Kesselman wrote:
> The discussion thread rather looks like it suffered from a 'too many
> cooks" problem.
> 
> Aren't there any existing small ide's written in pyton for python?

The two main IDEs for Python are "Eric" and "Idle" (bonus points for
knowing who Eric Idle is and what *other* group besides Monty Python he
belonged to.) :)

That said, IIRC a lot of the Python numeric code is bundled into the XO
base, and a lot of *that* is already optimized. For example, the Atlas
linear algebra library is on the XO (although it's the i386 version --
it's not yet optimized for the Geode). So I am guessing the original
task -- elliptic curve cryptography -- could well be done efficiently on
the XO in Python using "pippy" as the IDE.

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Give One Get One laptop for software development

[M. Edward (Ed) Borasky]

Danilo Câmara wrote:
> I'm a student at State University of Campinas, Brazil. I'm researching
> efficient implementation of Elliptic Curve Cryptography in constrained
> environments. I'm working with an ARM XScale PXA270 platform but would
> like also to work with a x86-based constrained platform. I think the
> OLPC laptop is an interesting option for many reasons.
> 
> I'd like to know if one of those laptops of the Give One Get One program
> are suitable for software development? I guess so, but would like to be
> sure.

How are you going to get an XO? The Give One Get One program is only in
the USA and Canada, and the other units go to young children.

Assuming that you can obtain one, there are two ways to develop for the XO:

1. Cross-develop on a more powerful platform, download the software to
the XO, and test it.

2. "Native" develop and test on the XO itself.

The XO as currently shipping does *not* contain a development tool set
other than Python and Etoys, which is a version of Squeak, which is a
Smalltalk IDE. In particular, there is no Perl interpreter on the XO,
and there is not GCC or any of the conventional Linux development tool set.

On top of that, there is very little room on the XO's base flash hard
drive (1 GB) for any additional software at the system level. As a
consequence, I think you'd want to either cross-develop using the
standard GNU/Linux tools, or use a more compact development tool set on
the XO itself.

I recommend Forth for the "compact development tool set on the XO
itself". The stock Fedora "gforth" package should work, although I've
only tested it on an emulated XO -- my physical unit has not arrived
yet. And there is a version of Forth that is more or less "XO-native" --
it's a port of the boot firmware to run in the Linux environment.

> Do I need any special hardware or cable to connect to the OLPC laptop
> from my desktop? A telnet or SSH connection is all I need.

Yes ... once you set the "root" and "olpc" passwords on the unit itself,
you can "ssh" in over either the wireless network or via a USB standard
Ethernet connection. I don't believe the "telnet" server is there --
nobody I know uses telnet any more because it's insecure.
> 
> I want real timings, so I think an emulated solution would not be
> suitable.

Well ... I guess that depends on how good your emulator is. But the
Geode is more like an Athlon than anything else -- I think it has MMX
and 3DNOW! but not SSE or 3DNOWEXT or any later SSE instruction sets.
That's one of the reasons I think you might want to look at Forth --
it's a complete IDE, it has an assembler built in, and it's pretty easy
to profile Forth code at the chip level. That's tough to do in C because
of the edit-compile-link-run cycle. BTW, gForth works just fine on the
XScale. :)

Oh, yeah -- one other question -- how are you doing elliptic curve
crypto on the XSCale? Isn't it floating-point intensive?

___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Our Stories: Commercialization?

[M. Edward (Ed) Borasky]

Tom Boonsiri wrote:
> After reading the Bender update, I checked out Anna's recent effort
> (http://wiki.laptop.org/go/User:Anna_B) which seemed to be similar to
> Ian Daniher's Telehealth module effort and I noticed the following goal.
> 
> -Van on commercializing Our Stories: would consumers pay for something
> like a virtual safari tour? A micro-franchise idea
> 
> Are we actually avoiding the commercialization of the technologies we
> develop under OLPC?

Well ...

1. A recent news item stated that some countries that are heavily into
hosting on line gambling casinos are asking the World Trade Organization
to fine the USA because we have legislation to protect our citizens from
on line gambling casinos.

2. Any talk of "commercialization" of anything needs to be gone over
with a fine-toothed comb by dozens of accountants and attorneys here is
the USA.

In short, there are always going to be grinches trying to steal our
Christmases. :) But I'm personally a lot more interested in the stories
of the children than I am in a "virtual safari". And yes, I'd pay money
for the literature, art and music that comes out of the project.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fooling with Java

[M. Edward (Ed) Borasky]

Bert Freudenberg wrote:
> On Dec 22, 2007, at 2:04 , Jeffrey Kesselman wrote:
>> Okay,
>>
>> I have a JDK installed and it seems to work.  For grins i put netbeans
>>  on my USB stick and fired it up.
>>
>> It seems to be working however I get no main display. I do get pop up
>> dialogs though.
>>
>> My suspicion is that Netbeans is asking the X wm for a Window and,
>> sicne from what I cna see the "window manager" in the OLPC is
>> windowless (one fullscreen root window only) it ends u pwitha null
>> window and all the drawing goes down the bit bucket.
>>
>> Has anyone else played with this?  If so does anyone know any magic to
>> get it to use the root "window" for the app frame?
> 
> Our window manager is "Matchbox", it is not "window-less" but "full- 
> screen", that is, it resizes each top-level window to cover the whole  
> screen. In theory this should work fine with all well-behaved X11  
> apps, and in practice it seems to work for most. Can't help with Java  
> specifically.

I brought up a couple of X apps from the terminal window on my virtual
XO. Alt-Tab cycles through the windows, including the Journal.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Mounting a USB drive (windows format)

[M. Edward (Ed) Borasky]

Jeffrey Kesselman wrote:
> On Dec 21, 2007 2:36 PM, Greg KH <[EMAIL PROTECTED]> wrote:
>> On Fri, Dec 21, 2007 at 02:05:29PM -0500, Jeffrey Kesselman wrote:
>>> More Diagnosis:
>>>
>>> I bought an OLPC in the buy one/give one so i just tried the USB stick
>>> on the real machine and it works fine.
>>>
>>> So it just my emulated image that is refusing to mount it for some reason :(
>>>
>>> I'd really like to keep using VMWare.  On my machine its both more
>>> convenient and much faster then qemu even with the kqemu wedge.  (its
>>> a 64 bit dual core, but you can't run qemu in 64 bit mode with kqemu )
>>>
>>> Has anyoen else gotten emulation under vmware recognizing USB memory
>>> stick drives?
>> You need to tell vmware to do this and take away the device from the
>> underlying operating system that vmware is running on.
>>
>> See the vmware documentation for how to do this properly, it's not an
>> OLPC image issue, but rather a vmware one.
> 
> I've already done that Greg.  (You do that with the VM->removable
> Devices menu.  I'm somewhat of an old hand at VMWare.)
> 
> The VM has the USB as a device, but the OLPC emulation is not mounting it.
> 
> So it was a good guess but thats not the problem.
> 
> It deosnt seem to be mounting the CD-Rom either, which suggests to me
> that it could be an issue with the automounter configuration in the
> ship2 build.
> 
> Is that what everyone else is usign for development? I assume that
> matches whats actually on my OLPC?
> 
> 
> 
> 

Hmmm ... looks like the two of us have been going down similar paths.
How many other folks are out there who are building VMware XO emulations
as development systems?

Speaking of which, given the hacks I previously posted to get a large
base hard drive, you can *install* VMware Tools, but you can't configure
them. That requires "perl", "gcc", "make" and correct kernel headers and
kernel build tools. I'm not sure what other dependencies there are -- I
haven't crossed the kernel build barrier yet.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Mounting a USB drive (windows format)

[M. Edward (Ed) Borasky]

Jeffrey Kesselman wrote:
> Hi Guys,
> 
> im trying to get started developign but ist pretty clear that my
> development environment wont fit on the emulation image's disk.
> 
> No biggie,  I figured, I'll put it on a USB memory stick.
> 
> I can tell my emulator (VMWARE) to make the USb memory stick acessible
> to the emulated hardware no problem but I can't figure out how to get
> it to mount in the OLPC.
> 
> Help?  I'm running ship2, if thats any help.
> 
> 
> JK
> 

I've been doing a lot of VMware emulations of various OLPC images on my
Linux box. I have an AMD64 running Gentoo Linux and VMware Workstation
6. This could probably be made to work on Windows but I don't have any
largish Windows systems.

1. Download the image ext3 file. I have a "wget" script that does it.
2. You need to install "qemu". It has a utility to convert a .img to a
.vmdk. Here's the bash script that does it, assuming "qemu" is installed.

#! /bin/bash -v
export
WHERE=http://xs-dev.laptop.org/~cscott/olpc/streams/ship.2/build653/devel
_ext3
export WHAT=olpc-redhat-stream-ship.2-devel_ext3
wget ${WHERE}/${WHAT}.img.bz2
wget ${WHERE}/${WHAT}.img.bz2.md5
md5sum -c ${WHAT}.img.bz2.md5
bunzip2 ${WHAT}.img.bz2
qemu-img convert ${WHAT}.img -O vmdk ${WHAT}.vmdk

3. OK ... now you have a vmdk image. Rather than mess with a USB stick
or anything like that, I just make a bigger virtual disk and copy the
image over to it! Here's what you have to do:

a. Create a virtual machine. Give it 256 MB and a new virtual disk. The
new virtual disk has to be an IDE and *not* the default SCSI. The
drivers are looking for an IDE drive. Then, *add* a hard drive and
select "use existing virtual disk". *That* one is the image you
downloaded and converted to vmdk format.

b. Edit the ".vmx" file from the virtual machine and add the following
line at the bottom:

  bios.forceSetupOnce = "TRUE"

This will make the machine come up in the emulated BIOS the first time
you boot it. Go over to the boot order selection menu and make sure the
CD-ROM boots before the hard drive.

c. Why do we want to boot from the CD-ROM? Because we have downloaded a
Fedora 8 rescue CD image and we are going to boot *that* to do the copy!
So ... you've got your Fedora image ".iso" file ... tell VMware to use
that for the CD-ROM and boot the virtual machine.

d. It will come up in the menu ... go to the default "rescue" option. It
will ask you for a keyboard and language. It will ask you if you want to
configure the network. Say "no". It will ask you if you want to rescue
an image. You want to go to the "Skip" option.

e. Now you're "root" in a RAM disk. The first thing you need to do is
format the new hard drive. I usually do it manually with "fdisk", but
"parted" is also on the Fedora rescue CD if you want to use that. You
want a single primary partition, partition 1, spanning the entire disk.
The disk will be called /dev/sda.

f. Now execute the following commands:

mkfs.ext3 /dev/sda1
e2label /dev/sda1 OLPCRoot
mkdir /orig
mkdir /copy
mount /dev/sdb1 /orig
mount /dev/sda1 /copy
cp -a /orig/* /copy/

g. One last step and you've got it. Type "grub" at the shell prompt.
You'll get a "grub" prompt. Type "root (hd0,0)". It should find an
"ext2" filesystem. Then type "setup (hd0)". That puts the boot loader
into the boot sector.

h. Reboot. The system will come back up in the rescue CD, but you've got
about 60 second before it will do anything. Power down the virtual
machine. Go into the settings and remove the second hard drive (the
original image). Disconnect the ".iso" of the Fedora rescue CD. Close
the virtual machine. Make a zip archive of the whole virtual machine
directory if you want to.

Now you have a virtual XO with a larger hard drive. I usually make them
2 GB, but I've made some as large as the default, 8 GB.





___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Mounting a USB drive (windows format)

[M. Edward (Ed) Borasky]

Jeffrey Kesselman wrote:
> No...  reformatting to Fat32 didnt help :(
> 
> On Dec 21, 2007 1:49 PM, Jeffrey Kesselman <[EMAIL PROTECTED]> wrote:
>> Hmm.
>>
>> It is a FAT file system.
>>
>> But it isn't automounting :(  And I can't figure out what it "name"
>> would be to manually mount it...
>>
>> Maybe its the weird U3 Cruiser software.  I'll get a second drive,
>> they're cheap now, and reformat it and see if that helps.
>>
>> JK
>>
>>
>> On Dec 21, 2007 1:39 PM, Ivo Emanuel Gonçalves <[EMAIL PROTECTED]> wrote:
>>> What kind of "windows format" do you mean?  In my experience HAL
>>> automatically mounts FAT and FAT32 USB keys and probably big drives as
>>> well.
>>>
>>> If your drive is formatted NTFS that may be the problem, as I do not
>>> reckon the XO to ship with the NTFS write-mode driver.
>>>
>>> -Ivo
>>>
>>
>>
>>
>> --
>> ~~ Microsoft help desk says: reply hazy, ask again later. ~~
>>
> 
> 
> 

It's very rare to find a USB stick formatted with anything other than
FAT32, although it does happen. I've also had Windows refuse to read USB
sticks formatted FAT32 by a Linux machine. The other way almost always
works -- format it FAT32 on a Windows machine and you'll be able to use
it anywhere.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Oprofile, swap

[M. Edward (Ed) Borasky]

Jordan Crouse wrote:
> On 18/12/07 12:39 -0500, Chris Ball wrote:
>> Hi,
>>
>>> However, you appear to be correct about the oprofile kernel.
>>
>>> $ grep OPROFILE config*
>>> config-olpc-generic:# CONFIG_OPROFILE is not set
>>
>> It is enabled in our kernel:
>>
>> -bash-3.2# grep OPROFILE /boot/config-2.6.22-20071204.2.olpc.9679b65c8c5ed6e 
>> CONFIG_OPROFILE=m
>>
>> Our kernel config lives in olpc-2.6/arch/i386/configs.
> 
> It is indeed there - but oprofile is still having issues.  It seems that 
> the sample data that is being written out is invalid (files are filled with
> only zeros, as far as  I can tell) , and opreport spits back errors, 
> consistant with badly formed sample files.
> 
> We need some people who understand oprofile to take a look at whats happening
> and diagnose it.
> 
> Jordan
> 

Uh ... you're from AMD? Can you get CodeAnalyst to work on an XO? That 
would be &$^%&$ awesome!
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: sudo, not su.

[M. Edward (Ed) Borasky]

[EMAIL PROTECTED] wrote:
>> I'd like to draw devel@'s attention to trac bug 5537, which might land
>> sometime soon:
>>   http://dev.laptop.org/ticket/5537
>> The upshot would be that, instead of logging in directly as root with
>> no password, you would log in directly as *olpc* with no password, and
>> then sudo to root (if you need root).  Please comment in the bug if
>> you have strong objections and/or be prepared to try logging in as
>> 'olpc' if 'root' seems to stop working in a new build.
> 
> +1
> 
> Yes, I think logging in directly as root is a misfeature that should go 
> away.  Most of the other unix-derived platforms have been doing their best 
> to kill it off or at least reduce its attractiveness... and being able to 
> sudo is usually pretty trivial.
> 
> --elijah
> ___
> Devel mailing list
> Devel@lists.laptop.org
> http://lists.laptop.org/listinfo/devel
> 

Yeah ... sudo is more secure than su. In fact, some systems, for 
example, the Gentoo LiveCD, scrambles the root password. So you have to do

$ sudo su -

and then set a password to ssh in as root.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Oprofile, swap

[M. Edward (Ed) Borasky]

Ivan Krstić wrote:
> On Dec 18, 2007, at 6:27 PM, John Richard Moser wrote:
>> I like to think of programs like kernels, or kernels like programs.  
>> Either way, I like to treat applications like microkernels.  In the  
>> embedded scene, this may actually be critical; maybe you should  
>> think that way for the XO, in a little part.
> 
> Please post a link to this list when you have (proof-of-concept) code  
> available.

Forth is the proof of this concept. ;)
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Oprofile, swap

[M. Edward (Ed) Borasky]

John Richard Moser wrote:
> 
> (Note:  most of this message isn't very useful probably; it's about 
> theoretical software architecture, that nobody's going to implement, 
> that I can't prove, that I'm not really 100% sure about.  Still, if you 
> WANT to read it, hey... remember, bad ideas sometimes get corrected by 
> people who are smart enough to turn them into GOOD ideas)
> 
> 
> Ivan Krstić wrote:
>> On Dec 18, 2007, at 12:27 PM, Jameson Chema Quinn wrote:
>>> Has anyone looked at Psyco on the XO?
>>
>>
>> Psyco improves performance at the cost of memory. On a 
>> memory-constrained machine, it's a tradeoff that can only be made in 
>> laser-focused, specific cases. We have not done the work -- partly for 
> 
> It would be wise to throw out the idea of laser-focusing which engine to 
> use.  Think of memory costs for running multiple versions of Python. 
> Then again, what IS wise?
> 
> Any such system needs to efficiently use memory.  I like the idea of one 
> based on Mono since it has that whole compacting garbage collector, 
> which (although being a cache destroyer by nature) at least shrinks down 
> memory usage.  Of course then you still have Mono on top of it, and CIL 
> code that's been generated, and reflected, and JIT'd, which means you 
> (again) have 2 interpreters in memory (one in CIL, one being Mono 
> itself), and one gets dynamic recompiled (the Python one in CIL), and 
> all the intermediary (CIL) code gets kept for later profiling and 
> optimization...
> 
> ... didn't I say before I hate the concept of JIT dynamic compilation? 
> Interpreters just suck by nature due to dcache problems (code becomes 
> data, effectively your instruction working set is fixed, the load 
> doesn't go onto icache and dcache both as the program gets bigger...) 
> and due to the fact that you have to do a LOT of work to decode an insn 
> in software (THINK ABOUT QEMU).  Interpreters for specific script 
> languages like Python and Perl have the advantage of not having to be a 
> general CPU emulator, so they can have instructions that are just 
> function calls that go into native code.
> 
> 
> So execution time order:
> 
> Native code <  // *1
> JIT < // *2
> Specific language interpreter < // *3
> General bytecode interpreter < // *4
> Parser script interpreter // *5
> 
> *1:  Native code.  C, obj-C, something compiled.  everything else I 
> could mention is out of date.
> 
> *2:  Technically JIT is native code, but there's also extra 
> considerations with memory use and cache pressure comes into play 
> slightly.  After the ball gets rolling it just eats more memory but 
> cache and execution speed are fine.
> 
> *3:  A specific language interpreter might call a native code strcpy() 
> function instead of have an insn for CALL that goes into a bytecode 
> implementation of strcpy(), or having an insn for CALL that goes into a 
> bytecode strcpy() that just sets up a binding and calls real native 
> strcpy().  The interpreter would head straight for native land, going 
> "function foobar() gets assigned token 0x?? and I'll know what to do 
> when I see it."
> 
> *4:  A general CPU interpreter is going to have to be a CPU emulator. 
> Java and Mono count, for Java and CIL CPUs.  These CPUs don't really 
> exist but those interpreters work that way, they even have their own 
> assembly.
> 
> *5:  Some script engines are REALLY FREAKING DUMB and actually send each 
> line through a parser every time they see it, which is megaslow.  These 
> usually don't last, or just function as proof of concept until a real 
> bytecode translator gets written to make a specific language interpreter.
> 
> 
> Maybe, MAYBE by twiddling with a JIT, you could convince it to discard 
> generated bytecode.  For example, assuming we're talking about a Python 
> implementation on top Mono, and we can modify Mono any way we want with 
> reasonably little effort:
> 
>  - Python -> internal tree (let's say Gimple, like gcc)
>  - Gimple -> optimizer (Python)
>  - Gimple (opt) -> optimizer (general)
>  - Gimple (opt) -> CIL data (for reflection)
>  - FREE:  Gimple
>  - CIL (data) -> Reflection (CIL)
>  - FREE:  CIL data (for reflection)
>  - CIL -> CIL optimizer
>  - CIL (opt) -> JIT (x86)
>  - While (not satisfied)
>- The annoying process of dynamic profiling
>- CIL (opt, profiled) -> JIT (x86)
>  - FREE:  CIL
> 
> NOTE:  at the FREE CIL data step, we are talking about the Python 
> interpreter freeing the CIL data that it has; Mono has now loaded a copy 
> as CIL code, we don't need to give it to it again, we're done with it.
> 
> At this point we should have:
> 
>  - A CIL program for a Python interpreter
>  - A CIL interpreter (Mono)
>  - x86 native code for the program
> 
> Further, you should be able to make the Python interpreter do a number 
> of things:
> 
>  - Translate any Python-written libraries via JIT on a method-for-method
>basis
>  - Translate Python bindings (Python calling C) to active CIL bindings
>

Re: Oprofile, swap

[M. Edward (Ed) Borasky]

John Richard Moser wrote:
> I just got my OX laptop (hopefully some kid gets the other one soon... 
> or not), and noticed it's slow and kind of buggy.  I think I'll get a 
> $25 4GB SD card for a SWAP area...
> 
> I should run oprofile too, and have it write to the SD card.  I 
> understand what an interpreted language like Python does to the CPU but 
> it shouldn't be this bad... it's only going to be like 100 times slower? 
>   An actual interpreter will...
> 
>   - Put pressure on the data cache as its code grows
>   - ... but keep the actual interpreter (code) in cache better
>   - Use a relatively large chunk of data for a look-up table
>   - ... or use some convoluted and hard to maintain code
>   - ... or optimally, a look-up table to start the decoding process, if
> like a CPU bytecode interpreter (Java, CIL) it has an insn + address
> mode + data (not QUITE optimal for Python, but maybe since simple
> addition and call happens)
>   - Wind up doing what can easily become a multi-hundred-cycle decoding
> process for each executed bytecode insn
> 
> Python rewrites to bytecode (good, interpreting text is slow!  Multiple 
> parsing!) but a lot of the main function calls in the API should be C, 
> not Python (taking some of the pressure off).  This means Python should 
> be doing a lot of logic in native space, rather than interpreting a lot 
> (unlike Java, which had its whole library written in Java...)
> 
> I suggest taking a look at PyPy for Python, which will dynamic recompile 
> Python to native code and likely give some good performance benefits.  I 
> really can't stand JIT compilation and would prefer something that takes 
> advantage of Mono's own facilities, to centralize the effort in the JIT 
> at least (Mono has nice stuff), but IronPython is Microsoft Permissive 
> License which is not OSI approved.
> 
> As for real solutions, I want to profile things and see where they're 
> hanging.  I may need a Python profiler too, to get a look inside the 
> Python code and see if some functions there are also bad; oprofile will 
> tell me if Python itself is spending an ungodly amount of time in its 
> decoder functions but that's it.
> 

I don't have my physical unit yet, but I too am interested in profiling 
and performance tuning. Unfortunately I have no Python tuning experience 
so I can't be of much help at the moment. I do have a "virtual ship2", 
running on a 2.2 GHz Athlon64 X2, but that of course is cheating. :)

Oprofile is a bit tough to work with -- it makes you install a whole 
bunch of GUI libraries just to get at the low-level profiling stuff. And 
the kernel needs to be rebuilt with the right options -- I don't know if 
the OLPC kernel does so. So for now, I think you'll probably be better 
off with lower-level command-line tools.

I know "top" is there, but as far as I'm concerned the one must-have 
package is "sysstat". "sysstat" is a work of pure genius -- it started 
out as a Linux re-implementation of "sar" and "iostat", but it is much 
more than that now. Once I get my physical unit, I'll be looking at 
things in some detail.

I'm guessing that adding swap isn't going to help you. If you're memory 
bound, the solution is to stop activities that you aren't using, not 
forcing the kernel to move stuff in and out of RAM. "top" will tell you. 
Open a terminal window and type "top". At the top of the display you'll 
see memory used, free, cached, etc. There's a keystroke that will sort 
processes by their resident set size. Type "h" to get a help menu. If I 
get a chance tonight, I'll fire up a bunch of activities in my virtual 
XO and see what it does when it runs out of RAM.


___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: KDE, QT, and the XO

[M. Edward (Ed) Borasky]

Bernardo Innocenti wrote:
> (adding devel@ and Bernhard Rosenkraenzer to Cc list)
> 
> Samuel Klein wrote:
> 
>> Two questions for you:
>>   1) is anyone working on a slimmed-down version of QT?  It looks as though 
>> it would be 30M of code and dependencies not in our current image.  (see 
> 
> It's always been quite easy to rebuild Qt with stripped down
> functionality.  I've been doing it on both embedded Linux and
> Windows environments with Qt 3.x.  The worst offenders were
> the large encoding tables for charsets.
> 
> The problem with KDE applications is that they depend on kdelibs
> too, which in turn tend to depend on the full blown Qt.
> 
> Back at the time of KDE 3.0, there used to be options to build
> stripped down versions of kdelibs on top of Qt/Embedded.
> Ultimately, you could have Konqueror running off a statically
> linked binary of 5-6MB.
> 
> The project was here, but it seems abandoned:
> 
>   http://www.konqueror.org/embedded/
> 
> 
> 
>> our latest image, emulation works cleanly : wiki.laptop.org/go/Emulation)
>>   2) is anyone working on a KDE package for the XO of any sort?
> 
> I initially thought that relocating the KDE runtime would be hard.
> Then I remembered that everything in KDE is relative to $KDEDIR
> or $QTDIR.  So, easy as pie!
> 

I gave up on KDE (and Gnome) a couple of years ago because they were so 
bloated. For about six months I ran XFCE 4.2. I did some beta testing on 
  fairly late betas of XFCE 4.4 and it was starting to look too bloated 
for me as well. So I ended up using "GNUStep/Windowmaker" after trying 
most of the lighter ones. I'm still on it, although I do run Gnome on my 
Fedora virtual machine.

I have to admit I haven't tried matchbox yet, but I have tried 
Fluxbox/Blackbox, IceWM and Enlightenment. Enlightenment is definitely 
an aquired taste. :) That said, I don't see the point in putting KDE or 
even QT on the XO. I know there's a lot of educational software 
available in and for KDE, but doesn't most of it have lighter-weight 
alternatives?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Activity depends on Fedora-packaged binary code

[M. Edward (Ed) Borasky]

Bernardo Innocenti wrote:
> Asheesh Laroia wrote:
> 
>> I suppose I'll have to include a libraptor.so.1 in my own .xo's lib/ 
>> directory.  Is the normal(ish) way to do this by just unpacking the RPM 
>> and yoinking the Fedora-compiled .so and jamming that file into my .xo?
> 
> This is my understanding.  As long as it's just one library,
> it isn't a big problem.  I have a friend who is packaging
> 5MB of Mono libraries for an activity written in C#.
> 
> This thing obviously doesn't scale and in the long term we
> may end up reinventing a full blown package manager with
> dependency tracking, plus tools like apt for downloading
> and installing them.

I've spent a fair amount of time with both "yum" and "apt" and I fail to 
see any "superiority" of apt over yum. Then again, my main distro is 
Gentoo, and I *know* Portage (and recompiling everything on an XO) isn't 
going to scale. :)

But seriously, does the XO really need two package managers? What's 
wrong with Fedora/RPM/yum? Do people really need to spend ergs on 
supporting Debian?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Upgrading to Fedora 8

[M. Edward (Ed) Borasky]

Bernardo Innocenti wrote:
> Yesterday I tried upgrading yoyride to Fedora 8, to see
> how much pain it would be.
> 
> After enabling the fedora repo in /etc/yum.repos.d/fedora.repo and
> tweaking it to make it fetch from version 8, I just did:
> 
>   init 3
>   yum -y dist-upgrade --exclude poppler
>   init 5
> 
> Going to init 3 is necessary to conserve memory, otherwise
> yum fights with sugar and cranks the system to death.
> 
> Excluding poppler precents a dependency problem in our
> custom package.
> 
> Surprisingly, everything worked perfectly after the update,
> until I rebotted.
> 
> Then I got into a minor glitch that prevents input hotplug from
> working when the X server is run by the unprivileged user olpc
> (X is a suid binary).  If I run X manually from the console,
> everything works fine.  There's probably a stricter check on
> the EUID somewhere in dbus or in hal.
> 

Does the symlink in /boot to the kernel get updated? I've tried 
smaller-scale updates (just the kernel) and it still reboots into the 
old kernel.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Status of Develop.activity?

[M. Edward (Ed) Borasky]

C. Scott Ananian wrote:
> On Dec 5, 2007 10:15 PM, Albert Cahalan <[EMAIL PROTECTED]> wrote:
>> Finally we have the problem of NO systems programming language
>> being supplied. It's less than 9 MB for the whole C development
>> environment, including a decent collection of *-devel packages.
>> You even get a second language thrown in for free, x86 assembly.
>> Pretty much everything that matters is written in C, including
>> the Python interpreter.
> 
> I'd be very interested in hearing details of your 'whole C development
> environment'.  By my casual inspection, 'rpm -qi gcc' says that gcc
> alone is over 10MB, and that *doesn't* include any of the *-devel
> packages needed to make it actually useful.
> 
> I believe there has been some work done on identifying a "lightweight"
> C development infrastructure; your assistance there would be helpful.
>  --scott
> 

Correct me if I'm wrong, but isn't the XO a platform for learning, not 
software development? And isn't it supposed to be lightweight and 
child-friendly? I would dearly love to have Ruby in there, and R, and 
SBCL, and Guile, and Maxima, and Perl, because I don't know *any* 
Python. But the philosophy of the machine is that languages other than 
Python and Squeak/Smalltalk are discouraged. I don't mind learning 
Python to be able to program the machine.

Now, if you truly want a lightweight development environment, install 
"gforth". I believe it's under 2 MB, and it's a full ANS
Forth, not the basic low-level Forth that's in the boot firmware. It 
has, like most Forths, an assembler and dis-assembler.

That said, when I get my G1G1, I'm certainly planning to load additional 
software on a Secure Digital card, but that will probably be 
cross-compiled on another system, rather than native-compiled on the XO, 
simply because I've got much bigger workstations to use for compile 
engines.
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel

Re: Fake mesh over IP.

[M. Edward (Ed) Borasky]

Morgan Collett wrote:
> Yes, the presence service operates either over the mesh, or via a Jabber
> server if you have Internet access. However due to scalability issues,
> the G1G1 software is not configured with a real jabber server by
> default. There is no way we could handle the 100,000 G1G1 laptops trying
> to connect to the current infrastructure.
> 
> We are working on server scalability issues for Update.1, at which time
> there will be a new software release and hopefully a usable server.
> 
> I do NOT recommend anybody advising G1G1 users to connect to the current
> developer jabber servers, as that will ensure that those servers become
> unusable for the development community.

Some G1G1 users are going to be sophisticated enough to search the web, 
find a server URL on the wiki and attempt to collaborate using it. :) I 
think the "real answer" is to find a corporate entity who will view this 
as a marketing opportunity and provide the infrastructure, much like 
T-Mobile have done with a year of free wireless connectivity.

Speaking of such "sophisticated G1G1 users", I've built VMware images of 
ship2, and I can do so for Joyride as well. Is there a place where I 
could post such images for people who are waiting for their G1G1 
hardware, so they can experiment with the interface?
___
Devel mailing list
Devel@lists.laptop.org
http://lists.laptop.org/listinfo/devel