Slashdot Mirror
Debian On Compaq's iPaq Handheld
ArsonSmith writes: "It seems that at primates.helixcode.com
they have been able to get Debian GNU/Linux up and running on the Compaq iPaq PocketPC. This seems to be the first main distribution to be running on a handheld." The Debian Way may sometimes seem ponderous and conservative, but obviously that's not always the case. How'd you like to be able to apt-get new apps on your PDA? This is a detailed description which should interest anyone lucky enough to have one of these cuties.
(Gawd. I wonder what writing perl would be like on a Palm, using Graffiti, in an airplane that was in turbulence.)
:-)
You would write yourself a readable perl program
Where's my pictures?
I want Linux on a Palm. *blink* Nothing like trying to type in command line commands using Graffiti.
(Gawd. I wonder what writing perl would be like on a Palm, using Graffiti, in an airplane that was in turbulence.)
-- Talonius
My reality check bounced.
All I want to ask is why? A desktop distribution (or server) made to run on a PocketPC. Requires constant network connection, and another machine. Seems similar to, oh, WinCE 1.0 and 2.0. Copy Windows 9x interface to PDA, expect it to be useful.
Compair this to the handhelds.org project to get PDA like features running on Linux on the iPaq PocketPC. And without a network.
And I'm assuming section 1.2.2 will be how to install into main memory so that a network card could be installed instead of the CF card...
Have a look at RIM's "Developer Zone" and specifically their FAQs. The neat thing about the Blackberry is that it runs Windows .DLLs. RIM provides a simulator that runs on Windows, and provides the same API that the pager does. Instead of interfacing with actual hardware, though, the simulator's API interfaces with a virtual LCD, a virtual keypad, etc. The exact same code runs natively on the development machine and on the pager itself.
Enjoy your job, make lots of money, work within the law. Choose any two.
Is this of any practical value? Well, a DPS-6 is likely about as powerful as an 80386, and it won't likely fit in your closet, so it's fair to say that it's not a terribly practical proposal.
Likewise, an iPaq has a pretty small screen, small memory, small secondary storage, and no keyboard, thus meaning that it's not a vastly practical computing platform on which to run Debian.
However. Consider two things:
Which means that while it's not spectacularly useful now, if some people scramble to work on it now, they may start having useful software and configuration tools (and boot tools, and...) by the time the hardware is ready to be really useful.
Nobody is going to power up a DPS-6 because they want to run payroll on it, or write TECO macros to calculate Pi because that's a particularly efficient way to do it; they'll do it because they can.
I think that in another year, there may well be enough hardware on PDAs to usefully run Linux. More power to those experimenting with it now that will make it deployable by then.
If you're not part of the solution, you're part of the precipitate.
My first guess would be the heat issue. Every Intel CPU I have ever come into contact with (including the 486 and the P120 beside me, which are the really cheap, low power-style chips like you've suggested) put off a lot of heat. I'm sure it's too much to be squished in with the rest of the tiny handheld circuitry to avoid damage.
-- sudo.ca
I think that given the size of the screen and the input options for these palmtops, not to mention the cpu/mem constraints dont seem to fit the whole Unix/X way of doing things
To some extent true. But keep in mind this little device packs more power then many desktops 5-10 years ago. While input may need to be reworked to be useable with a stylus, the power is exactly why I left my Palm V and got the iPaq. I decided I wanted something more then a PDA, and the name PocketPC is very approiate. I can sit and watch streaming video from anywhere in my house via an 802.11 card in the device. I can carry the device in the same way I did my Palm V, and have so much more potential in my portable gear. All because the engineers finally shrank my computer of years ago into a small form factor with decent battery life and a color screen viewable outside and in. True, storage is a bit low, but nothing a 340 or 1gb microdrive + CF to PCMCIA adaptor + PCMCIA expansion can't solve. (PCMCIA expansion used due to it's own battery to not let the Microdrive suck the device dry quickly).
In 5-10 more years, just imagine the power of a handheld device, and what it will do to complement your existance with computers.
The Intel style chipset is versitile enough to work in a handheld easily enough, and Intel componets are certainly cheap enough to use.
.doc files.
The Compaq iPaq Blackberry wireless e-mail PDA does use an i386 chip inside of it. I definitly have to agree to some extent that it would make running programs on the PDA easier if it had an Intel x86 chip inside, but one problem. The device will be years behind current processing power, and storage will be an issue. Once you dig up Word 6.0, it would run, but Word 6.0 (And other programs) are designed for a keyboard and mouse interface with a decent resolution. Work would still have to be done to change the UI to make it run on a small screen, and by then, it would have been easier to instead make PocketWord, and throw in the benefit of compatibility with 95, 97, and 2000
I don't see a problem with all handhelds running non Intel x86 chips for now. Once we start all using virtual monitors on these, then it will be useful.
Actually if the goal is just to get linux onto another device, you're right...no justification. But the reality in this case is a little bit different from my perspective...
First off they are developing for a PocketPC (aka WinCE3.0) not Palm. Big ifference in hardware etc.
I have a Casio E105. Its a really nice little WinCE unit that unfortunately I can not upgrade to PocketPC. So basically no further OS & little application support & development going forward. Now when I invested in this unit (to the tune of $400+) I knew WinCE was going to have a finite life, but knowing that the OS and in rom apps were on a nice daughterboard I figured I would be able to upgrade for a reasonable amount. Woops missed the window, Casio says too bad shell out another $500. (Last Casio I buy)
I love this PDA, light, powerful, good memory/battery life etc. WinCE2.1 has some bugs and the UI has some issues. Most of the Apps I have loaded (or that came from MS/Casio) are well designed. I do not at all regret getting the Casio. It has served me very well already...but now I have this issue, already WinCe2.1 compatible SW is getting harder to find, and I have no upgrade options available for PocketPC...
Along comes linux for the VR series of chips...LinuxCE... familiar OS, fairly well developed, Open Source cross compiler, etc etc etc
Right now I have a fully functional Linux kernel booted up on it with Video, networking, serial, touch panel, sound,CF support etc...running a 2.3.9 kernel. Linux is entirely on a CF card, and I still have WinCE available with a reset. There are a lot of issues with Linux useability on the PDA right now, but they are not what you might think. Software keyboard drivers/apps (like the jot etc) are developing nicely...and userland is starting to be brought over. So it's not there but it is being actively developed and it is using tools I am more familiar with so if I need something or want to give a hand, I dont have to learn as much.
Now as to power/performance of the CE handhelds...here's part of the boot transcript:
and the
Now those BogoMIPS aren't anything to scream about by todays desktop measures but I think it's probably quite a bit more capable than you may have given them credit for. One of Linux strong features is it's ability to scale down and run on lower end HW quite well. It may not be there yet, but at least I have a hope of continued development for this platform with linux. The same most definately CAN NOT be said with Windows...and I don't relish the idea of replacing a PDA every year to two years... no thanks. If Linux can at all extend the usefull life of my Casio then it is worth every moment of my time and effort. The fact that it is Open Source etc is just a very nice bonus (for me at least.)
Why does Linux have to run on *everything*?
I know this will just be marked -1 troll, but really folks, we complain about Billy G making windows into things it was never meant to be, yet here we have people trying to cram a linux kernel in to a PDA. Now I have to ask why.
Sure, it's cool. But that's it. No PDA to date has had a CLI, nor do I believe one would be benificial. In the case of a PDA, pointing and clicking is many times faster than *ANY* kind of text input developed so far.
1) Now, let's look at Linux. No graphical interface, natively anyway. To get the GUI, you need an extra layer of programming. Now, look at the Palm. GUI through and though, no extra layer.
2) This is a good thing. Why? Because PDA's are obviously *NOT* desktop machine, or servers! They have a very limited capacity by todays standards. Every extra layer on something that small means more CPU, more memory, and more waiting.
3) Which gets me to my point. I love linux, I run it at home, at work, and everywhere I can reasonably in between. I like it to develop code. I don't use it because it's user friendly, I don't use it because it's got all the killer games. I use it to be productive, mainly on a CLI. Now take away that CLI. What do you have? Well, you have Mac, Windows, BeOS, whatever pretty WM you can run to emulate any of the above or do about anything you want. But consider this, wouldn't a kernel that natively used graphics fair a bit better?
Now like I said, mod me down all you want, I use Linux, I love Linux, and it's a great alternative to any other OS. But I have to say, I think the Linux crowd here has gotten out the hammer and decided this problem looks like a nail. I think maybe it looks more like a push pin.
Free Online Woodworking Resources Directory