Adopt a KDE Geek

sultanoslack writes "In an effort to bring together KDE hackers that are students, unemployed or by other means lacking in hardware and capital with users in that have spare goodies, Adopt-a-Geek has been launched. More details are available on how to help out. Been wondering what you can do to help out? Here's your chance!"

Comments

Do I get a framed picture of my geek?

[Harald Paulsen]

Do I get some pictures I can put in my wallet, and a certificate telling me where he lives, where he goes to school etc. Then I could flash his picture whenever people tell me how they have adopted some poor kid in the 3rd world.

Aaaaw, look at that.. not-so-cute geek!

Re:Do I get a framed picture of my geek?

[pizza_milkshake]

yeah, i think i'd skip the picture... ;)

Re:Do I get a framed picture of my geek?

[boaworm]

Do I get some pictures I can put in my wallet, and a certificate telling me where he lives

I'd be much more interrested in knowing where she lives. Let me adopt a lady and i'll consider it =)

Re:Do I get a framed picture of my geek?

[Pilferer]

Do I get some pictures I can put in my wallet, and a certificate telling me where he lives, where he goes to school etc.

Him? What about Her? Are there any female KDE Geeks to adopt? I'd like a 16 year old asian girl, please!

*cough*

Re:Do I get a framed picture of my geek?

[Anonymous Coward]

Woody Allen, is that you?

Re:Do I get a framed picture of my geek?

[CEPi]

You fucking sick pervert!!!!

Better place sto donate

[JonathanBoyd]

It's a nice idea, but aren't there better places to donate, like poorly funded school? The Geeks in question already have some skills and computers available, so how abotu we try and do the same for those who have neither?

Re:Better place sto donate

[Anonymous Coward]

You want them to donate a grammar or spelling checker, perhaps?

Re:Better place sto donate

[MikeFM]

Donate to the opensource programmers today and children of tomorrow won't have to throw their educational dollars away on constant computer upgrades and expensive commercial programs. I've been an out-of-work programmer and it's great to spend some of that free time giving back to the community but it's hard when you can't pay rent let alone buy the hardware you need to test so and so feature against. Now that I'm working I'm certainly not rich but I try to give a little here and there towards projects I like.

Re:Better place sto donate

[Proc6]

Dear Troll,

I hate when some bleeding heart socialist steps in and says money could be better spent "on the needy" in cases like this. Almost everyone reading slashdot has some kind of discretionary income. For some it's $5 a month, for other's its $500 a day. Either way, part of enjoying life is spending what you have (cash, time, knowledge) you things you enjoy. Are there other people out there who "need" things. Yes. Does that mean we should give every spare dollar to them? No.

Unless you live in a grass hut you made with your own 2 hands, dress in recycled fig leaves, give back to the land more than you consume, and produce more food personally than you consume, shut the hell up. If someone wants to spend money on the development of open source software, they should have that right without being accosted by some hippocrite. Now take the PC you used to post on Slashdot offline, sell it on Ebay, and give the money to the "needy".

Love, Proc6

Re:Better place sto donate

[JonathanBoyd]

I guess a programmer is going to see it from the point of view of the Geek and a student will see it from the point of view of the schools :^)

I'm not sure how giving computers to out-of work programmers is going to help schools maintain computers which they don't already have. Sure, if every school had computers, then getting resource-light programs to run on them would be worthwhile, but if some schools don't have computers, surely we're better off giving the school one, rather than giving it to someone to write programs for the computer they don't have?

And how many KDE programmers are there working on educational software for schools? I'm sure there are quite a few working on stuff useful at university level, but what about below that? A fully MS compatible version of KOffice would be nice for students/business, but not a 10 year old kid.

I'm going to look really stupid if all Dorling Kindersley software turns out to be written by people running KDE.

Re:Better place sto donate

[JonathanBoyd]

Excuse me, but how was I accosting anyone? There are two groups in need in here - the unemployed programmers and the school children. I personally think that if you're giving away an old computer, it would be better off going to a school. You disagree. Fine. Maybe you're right, maybe not, but why do you see me having a different opinion to you as a threat?

I never advocated coming along and taking your PC away from you or forcing you to give money to some people and not to others. You're free to do with it as you will. People have the right to spend money on developing open source software. I never said otherwise.

Neither did I say you had to give away every dollar/pound/euro/whatever you own. If you read the article, you will see that this is about old PCs you have and are no longer using. My post was in the context of someone who is already giving something way. I was not debating how much to give, but rather where was in most need. Surely that's an important question to ask when donating to people? And keeping a single computer for my own personal use is not hypocrisy.

Re:Better place sto donate

[melonman]

Donate to the opensource programmers today and children of tomorrow won't have to throw their educational dollars away on constant computer upgrades and expensive commercial programs.

It's a nice idea but, as you say:

I've been an out-of-work programmer and it's great to spend some of that free time giving back to the community but it's hard when you can't pay rent let alone buy the hardware you need to test so and so feature against.

Which is surely a good summary of the problem with the open source model. It relies on someone paying the programmers for the love of open source. Now there may be enough university departments and software manual publishers to feed the likes of Larry Wall, which is great, but I can't see this model ever scaling to the point where it 'employs' anything like the number of people currently working on commercial coding projects. You need some way to collect the money, on the basis of what work is the most useful. And the conventional way to do this is called a company in a free market.

Cf science, which started off as a hobby of the upper classes, was then patronised by the upper classes, and is now mainly funded either by business or the public sector.

I like KDE. It helps me to earn a living. I already pay for it, in the sense that I buy boxed distros. I wouldn't be averse to paying more, so that some of the money went to the people doing the coding. But I doubt if my French accountant would let me pay the programmers in hardware...

Re:Better place sto donate

[ThundaGaiden]

From everything I hear both places would be viable these days being a guru with computers (programming or otherwise) doesn't gaurantee you a job unless you've trained in 100 buzzwords , like RUP , XML , blah , blah , blah. All of which you could learn in a day or two of proper studying , but don't feel like lying on your interview form

Re:Better place sto donate

[Anonymous Coward]

Nice try but this kind of idealistic shit won't happen in reality.

Re:Better place sto donate

[carlos_benj]

But I doubt if my French accountant would let me pay the programmers in hardware...

Darn those French accountants! Try switching to a Lithuanian, I understand that they're much more flexible about things like this....

Re:Better place sto donate

[sultanoslack]

I just feel like I should note that in addition to being the author of the article and a general KDE hacker, I am active in KDE Edu which currently has 16 applications that will be part of the KDE 3.1 release this week, about half of them useful for the age range that you mention.

Giving to schools is a fine thing and needn't be exclusive to helping out KDE. In fact if every active KDE core developer were given a new computer, this at most might be enough for one school. In this case, there are probably about 20-ish (or less) core developers that could use upgrades. We're not talking about big numbers here.

Now lets say that one school switches to KDE / Open Source from MS desktops. The cost savings in that alone outweigh the cost of diverting machines which might have gone to schools to KDE developers. In fact there have been a good handful of schools switch to KDE based desktops -- dragging an Open Source envirionment with all of its Free tools and such behind it.

Remember the ideal is for people pushing technology in schools to keep in mind both hardware and software concerns; this is a partnership, not a competition. When you send in your hardware donation I'll even be glad to direct it to a KDE Edu developer. :-)

Re:Better place sto donate

[rseuhs]

If that poorly funded school would use KDE/Linux instead of Windows and would stop forcing its students to buy/pirate Office and Windows this would helt the students and their budget much more.

Re:Better place sto donate

[JonathanBoyd]

There's still the problem of schools getting the hardware in the first place. Free software is all well and good, but if you have nothing to run it on, then it's not much use. That said, it's good to hear that education software is being developed. As long as you take share away from Microsoft and not Apple :^)

Given that quite a lot of developers seem to be after newer faster machines, so complies are quicker, what happens to their old computers if they get a new one? If their old ones were passed on to schools then I see both groups benefitting. Doing and mentionning that could encourage more people to donate to you. I'd certainly approve of the scheme in that case. Keep everybody happy.

Re:Better place sto donate

[JonathanBoyd]

Students are not to my knowledge forced to buy/pirate those. There are plenty of programs compatible to the level required of school work that are free or cheap. Unless you're having to run Access

Re:Better place sto donate

[swillden]

Which is surely a good summary of the problem with the open source model.

The first problem with the open source "model" is that there isn't one, at least not in the sense of a standard, documented business model, or even any realistic idea about how "all" software could be produced by it. In other words, there are a thousand open source models, and none of them are really complete. That means you have to define precisely what you mean by "open source model" before you criticize it, else you're just attacking a strawman, and one whose details are known only to you.

I can't see this model ever scaling to the point where it 'employs' anything like the number of people currently working on commercial coding projects.

Aside from the question of what model you mean, exactly, why would you expect open source to 'employ' that many people? More on that below.

You need some way to collect the money, on the basis of what work is the most useful. And the conventional way to do this is called a company in a free market.

The problem with these statements is that you're assuming your conclusion. How? You're thinking only about the world of off-the-shelf packaged software as the way in which software is developed. In fact, only a small fraction of the world's software is developed that way. The vast majority of the software that is written is in-house, custom software. And the majority of software developers are employed in writing this sort of application.

Here's a more realistic view (in my opinion, at least): There will always be certain categories of software that will be produced primarily in the for-profit-software-company style. Some people, particularly those who run or work for for-profit-software-companies would like *all* software to be developed that way, but there's a great deal of software that can never be written that way, because there's just no market for it; it's too application-specific. Finally, there's a goodly chunk of software that can benefit tremendously from an ("an", not "the") open souce approach. Most of this software falls into the category of "infrastructure" -- operating systems and their components, development tools and libraries, basic office and business applications, etc.

Now, all of this software, from all categories, is needed. People want it, and they're willing to "pay" for it. Some of this "paying" is in the form of donated labor, equipment, etc., some of it is in the form of a P.O., and there are other forms as well. A fundamental rule of economics: Where there is adeqate demand, a supplier will step forward. People want the software, ergo, it will get written, one way or another. And only programmers can write it. And programmers must eat. Therefore, programmers will get paid for writing software, one way or another.

It *is* possible that open source software will reduce the number of programmers employed in writing software, but if it does it will be because of the greater efficiencies provided by open source. All of those programmers hacking out in-house, custom apps will have this massive base of tools and almost-right applications that they can use, so they can do the job in less time, with less effort, less people and at less cost to their employer, freeing up that capital to be employed elsewhere.

And that, my friend, is unarguably a *good* thing. Sure, it may mean that the world needs fewer programmers, but that's also a good thing, since it frees up all of those smart people to apply their effort and intellect to other, more valuable tasks. The only way that all of this could be bad would be if it got us into a situation where the software we needed could not be produced, maintained or supported, and economics would not seem to permit such a situation to exist.

Re:Better place sto donate

[Anonymous Coward]

Good and kind sir,

Almost everyone reading slashdot has some kind of discretionary income. For some it's $5 a month, for other's its $500 a day.

How, about this... When I was in college, I had to actually work for the money I spent. I spent that money on food, textbooks, and yes even a computer I used to code on. I didn't earn $5 a month, nor $500 a day.

If someone wants to spend money on the development of open source software, they should have that right without being accosted by some hippocrite.

I'm not critisizing that anyone wants to spend their money on open source development, or providing people with means to do open source development. People who wish to help by spending money or used hardware can do so, it's their stuff.

What I am going to critisize is the fact that some people have the nerve to ask for damn good hardware. Don't send me 32Mb of RAM, we can't use that. It has to be at least 64. Don't send me a celeron 600, it has to be at least an AMD 800, and don't be cheap, no Durons either.

If you're looking for help, accept what you get, and don't get picky. Start looking for secondhand PCs in the newspaper, or find a cheap one on Ebay if you want something specific, but don't ask for help and tell people that you don't want crappy hardware.

Some people code for a living, and don't even have the kind of machine these people describe at their day jobs. If your compiling takes too long, grow some patience, and while you're waiting for that compile to finish, perhaps you could go out and work for that extra cash needed for your computer, instead of begging and demanding superior hardware.

Oh, I'm envious too when my neighbour shows off his new Ghz machine, while he only uses it to type his letters and surfs to the CNN and BBC news websites. I'd have thoughts like: "If I'd switch the motherboards he wouldn't even notice". But the truth is, he worked for his machine, just as much or even more, than I did for mine.

And this last comment is something that pisses me off too:

I hate when some bleeding heart socialist steps in and says money could be better spent "on the needy" in cases like this.--snip-- Unless you live in a grass hut you made with your own 2 hands, dress in recycled fig leaves, give back to the land more than you consume, and produce more food personally than you consume, shut the hell up.

I love open source software, and I try to contribute as much as I can, by effort but nothing material. But you cannot deny that there are people who would benefit more from any kind of donation than a minimum of hardware. The people asking the hardware have the ability to get an education (as most of them are college students), while children in Africa are starving and in Eastern Europe and some parts of Asia are being forced into prostitution.

You tell people that if they don't give to good causes, they should shut up. Part of the taxes I pay every year are given to humanitary aid, and in my country a social security fund that takes care of the sick and unemployed (for as long as it will still exist). I drop a coin in the beggars stereotypical cup every once in a while, and I don't care how he spends it. I give money to good causes when I feel they are good causes that deserve money.

In my opinion (and this is not to convince the rest of the world), these are just whining students, tired of that old 'n slow machine they can't compile fast enough, too lazy to work a little for what they want. I'd donate a machine to the local student community right now, if they hang out a poster "any machine will do". It'd be a pentium 2 that's just sitting here, but when there's a minimum limit, I'd rather sell it, and spend the money on myself or perhaps a cause that didn't have minimum requirements.

The thing with the minimum requirements proposed is that in 5 months or so, they'll have to raise these minimum requirements because the software packages have gotten even bigger by then, and they'll need faster machines because it takes them just as long to compile as it did 5 months ago. Will they be whining again in 5 months?

Re:Better place sto donate

[MikeFM]

I've been a student and have worked in several schools doing IT type work dealing with from Preschool to PhD level users. I'd still suggest donating to developers before schools because it will benefit a lot more people to get that code out there.

I'd suggest donating to schools and community centers and let the programmers have access to those machines for programming and testing tasks but somehow I can't see many schools that'd go for that. I have a big interest in community centers especially as they are open to the general public and not just students. I'd love to be able to set up and manage a community computing center. Hardware is less of an issue than having a rent/utils covered by donatations. I can put very nice, full multimedia, computers in for around $300-$400 each including monitor. It isn't hard to get a community business or organization to donate that much.

A lot of programmers are working on educational software - myself included. Edutainment type software is one of the growing hotspots of interest. Myself I develop in Python w/ Pygame or wxPython as it makes the software portable between Windows, Linux, Mac, etc which IMO is a good thing. If you have a specific interest in donating to educational software check out seul.org/edu/ for some possible recipients.

Re:Better place sto donate

[Anonymous Coward]

I was forced to at my technical degree 2 years ago btw. I made my way through the entire school year on an apple laptop vnc'ing into someones desktop when I needed windows (which was very rarely). I left after a year because most people teaching knew less than me about well, everything.

Re:Better place sto donate

[melonman]

The first problem with the open source "model" is that there isn't one

I agree with the sentence, but maybe not with the thought behind it. The multiplicity of OSS models does a lot of people's heads in. At least with the commercial model it is very simple (you may be being ripped off, but at least you can put that into your spreadsheet)...

It is possible that open source software will reduce the number of programmers employed in writing software.

That isn't my concern, if I have a 'concern' on this matter. My projection into the future goes more like this:

You can start a project on your s/h PC in your dorm, and, if it's a great idea, it may get a lot of interest. When it gets too big for one dorm room, you can network. But if you want to produce a piece of software with the breadth of, say, Windows XP, the 'doing it for love or donated equipment' model just won't do it.

IMHO, the current state of Linux/GNU (or whatever it is I have on my Redhat CD) illustrates this fairly well. The kernel is state of the art. And a lot of big companies are putting a lot of $$$ into keeping it that way, because, on a purely commercial basis, it is cheaper to do this than to keep developing and supporting bespoke OSs, especially for minority platforms. Some of the geeky applications are pretty whizzy too, because they catch the imagination of the dorm-room developers.

On the other hand, the less 'sexy' bits are often pretty bad. I use Redhat in French, except that half the applications still come up in American English. In linguistic terms, Windows runs away with the ball. Why? Because translating page after page of someone else's text prompts is not what sets your average geek's imagination on fire. Then there are drivers. Or not. I bought a Polaroid digital camera over a year ago, plugs and plays with Windows, still waiting for it to work with a standard Linux distro. The last 2 versions of RH claimed to support it: v8 has a driver, but only via the USB port, which is funny since that model never had a USB connector. What's the problem here? Producing drivers for umpteem million products is grunt work, and unless some geek somewhere happens to have my camera,it just ain't going to happen.

And the lack of development on the non-sexy bits of Linux/GNU is, IMHO, one of the main obstacles to its widespread adoption on the desktop. So I don't think we are looking at less programmers, I just think that the OSS/Commercial split is reaching dynamic equilibrium, and that the balance will not shift further towards OSS unless or until we find a sane way of financing the dull bits of infrastructure development.

Re:Better place sto donate

[swillden]

At least with the commercial model it is very simple

Only if you don't read EULAs, and don't think about the future.

In linguistic terms, Windows runs away with the ball. Why? Because translating page after page of someone else's text prompts is not what sets your average geek's imagination on fire.

This is a relatively simple problem. Non-programmers can do it, and we're just now starting to see projects recruiting non-technical people in a serious way.

I'm also not at all sure that Windows does, in fact, "run away with the ball" here. Sure, the apps from Microsoft are very widely internationalized, but what about third party applications? I'm sure that the packages you find on the shelves in France are in French, but the majority of Windows software as a whole is not.

Just out of curiosity, have you tried Mandrake? Given that they're a French company, they may do a better job of supporting your preferred language.

That said, I'll certainly agree that there's a level of professional polish that is lacking in OSS projects, and will always be lacking as long as it's strictly a hobbiest thing. But it isn't just a hobbiest thing anymore, and all it takes is for a bunch of, say, French companies to decide they want to use a particular software and to spend a few thousand euros each (often cheaper than licensing a commercial version) to fix it up. This makes sense for exactly the same reason that it makes sense for IBM to enhance Linux rather than put that money into AIX. We haven't seen a lot of it, but that's because this whole OSS thing is very new and not widely understood, yet. It's coming, though.

Then there are drivers. Or not.

Do you think Microsoft writes those drivers? The state of this situation is exactly the opposite of what you're saying. Writing rogue drivers is a lot of geeky fun, particularly since it generally requires reverse engineering the protocol. However, it is very difficult and time-consuming, so we should really be amazed that Linux supports as much hardware as it does. The thing MS has going for it in regard to drivers is that the manufacturers of hardware write the Windows drivers themselves. As Linux becomes more widespread, we're beginning to see hardware manufacturers who chose to support Linux themselves, and it's reasonable to expect that trend to accelerate.

The downside is that many of the manufacturers will keep their drivers closed. In my experience, Linux has fewer drivers, but they're often higher-quality, and almost always more flexible, because of the openness.

And the lack of development on the non-sexy bits of Linux/GNU is, IMHO, one of the main obstacles to its widespread adoption on the desktop.

On the home desktop, sure. On the business desktop, the main barrier is the Office lock-in. Once companies can get past that, the cost of polishing up the rough edges that annoy them is pretty small, and there is a business model for doing that development. The really big win for the corporate world, however, is the ability to *customize* that doesn't exist with most closed software. The IT industry really hasn't caught on to that yet, but they will.

Re:Better place sto donate

[melonman]

Only if you don't read EULAs, and don't think about the future.

I have no fears on that score. Like any extorsion racket, it is in MS's interests not to kill off their victims (or hurt them so much that they decide to fight back), so, whatever the EULA might allow in a /.er's paranoid mind, it just ain't gonna happen. (And my business is 99% Linux anyway, so I don't care too much if I'm wrong on this...)

Most software isn't internationalised

But if, as you more or less say yourself, the MS apps are, and the products I can buy in the country where I live are, who cares whether I could buy a Czech version of some obscure program in a supermarket in Equador?

Do you think MS writes those drivers?

No, I'll give you that one, wrong discussion...

Mandrake

It might be better in language terms, but the version I tried was extremely bad in many other ways. I'm going back a couple of years here. SUSE was better than RH language-wise, and, in general, I liked it a lot, but I just got fed up with having to fix every program I wanted to run because none of the files were in the normal (ie RH) place.

Office lock-in

Glad you mentioned that, because office software is another really bad thing with Linux. Star Office/Open Office are great office suites for people who don't need an office suite. I get to try MS Office users on Star Office in my cybercafe most days, and they all end up wanting to chew their arms off, and that is before we mention 'features' like locking the entire system up big time if you try to save a file directly onto a floppy disc (this bug has been around for at least a year, survived from SO 5.2 to OO 1, and also happens on a laptop with a different distro). Thing is, geeks are not usually Office power users...

Re:Better place sto donate

[swillden]

so, whatever the EULA might allow in a /.er's paranoid mind, it just ain't gonna happen

Given that some competition is arising, I'm sure you're right. Given complete domination, however, I have no doubt that MS would come up with stuff that would amaze even the most paranoid.

But if, as you more or less say yourself, the MS apps are, and the products I can buy in the country where I live are, who cares whether I could buy a Czech version of some obscure program in a supermarket in Equador?

My point was that this is a problem that corrects itself as the software becomes more widely used. By the time Linux has grown to a, say, 5% market share of desktops worldwide, internationalization will be far better than it was when Windows achieved that same penetration. Why? Because anyone who wants to can do it, and more and more people are wanting to.

With respect to RedHat/Mandrake/SuSe and internationalization, honestly I just have to plead ignorance. I don't use any of the above (I like Debian, 'cause I don't like reinstalling), and, frankly, my first language is English and that's what I use. I do know that I can switch KDE to Spanish and everything in KDE changes.

Glad you mentioned that, because office software is another really bad thing with Linux.

It's weak. OpenOffice is getting there pretty rapidly, although I wonder sometimes if the half-open nature isn't slowing it down. However, you're comparing a decade's worth of work by Microsoft's very large Office team to what has been done in the two to three years since people got interested in building an office suite for Linux. And, for that matter, there are areas in which the OSS offerings are vastly superior to MS Office. Automatic document generation, for example, is supremely easy using OpenOffice.

Also, I think you overestimate the problems users have with OpenOffice. I recently put it on my brother-in-law's computer (he's completely non-technical, and computer stuff is pretty hard for him) and he and his family have had no problems with it at all. That's the Windows version, of course, but it's essentially the same. His daughter saves documents to floppy to take to school all the time. I can see how MS Office users would be driven insane by the StarOffice 5.0 take-over-the-desktop approach, but have you tried it again with a newer version?

BTW, I just tested, and on my Debian system with OOo 1.0.2, I can save to floppies with no problem. Tried both FAT and EXT2 file systems, too.

A little more information

[Anonymous Coward]

From the Relevant Page:

KDE developers put their computers through a lot of work. Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours. Many developers are working on systems which cannot fully build KDE in under 24 hours, and many KDE developers do so several times a week. Profiling and debugging tools for optimizing code are very processor and memory intensive. Hardware often is a bottleneck to KDE developers' productivity.

So keep this in mind before you ask why they're requesting this. Thanks :).

Re:A little more information

[JohnFluxx]

Riiiiight.
How long do you think it takes to compile windows, with it's .. 10 million lines of code isn't it now?

Re:A little more information

[Anonymous Coward]

Maybe if KDE offered software in individual packages instead of package groups, it'd take less time to compile. :P

I can sympathize, my 1.53 GHz Athlon, 512 MB ram, took hours. And I didn't install much - the base, libs, graphics, a few other things.

I'd cringe at the thought of compiling KDE, in its entirety.

Re:A little more information

[oliverthered]

If KDE is slick on a crap PC then it's going to run like a dream on a 1.4 GHz Athlon, 512 MB RAM
(or have a div(0) if someones put a bad timing loop in there!).

Also....

"Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours. Many developers are working on systems which cannot fully build KDE in under 24 hours"

What?
1: get the pre-compiled header patched version of GCC
2: Do you always build the whole tree every time? A kernel compile takes 30mins or so my PC, but if I change a module is takes 1min because everything else is build already and I don't make clean/mrproper.

Re:A little more information

[Big Mark]

The analogy is flawed. KDE doesn't have a kernel, drivers, window server (? whatever XFree86 is) etc to compile, which Windows does. So of course Windows will take longer to compile, Windows is much, much more than KDE will ever be.

And on this machine Windows XP is more responsive than KDE for some reason. Go figure.

-Mark

Re:A little more information

[Chris Canfield]

KDE developers put their computers through a lot of work. Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours. Many developers are working on systems which cannot fully build KDE in under 24 hours, and many KDE developers do so several times a week. Profiling and debugging tools for optimizing code are very processor and memory intensive. Hardware often is a bottleneck to KDE developers' productivity.

Thinking back to the useless hours being wasted trying to crack the X-Box encryption, how much of this compiling could be distributed? Obviously it wouldn't accelerate live debugging or optimizing tools, but what if there were networks of computers who people volunteered to standby and remotely download, build, and upload code, and a linker on the initiating machine to reassemble globals, etc?

I know nothing about distributed compiling, which probably means that either A: I should go back to college (very likely) or B: compiling doesn't break down nicely into chunks.

If it is possible, a network of volunteer Open Source compilers would probably build in a significantly faster time than many of the aformentioned older systems can, assuming no major bandwidth bottlenecks, and would probably find a rather large home of OSS and Free Software supporters who don't have the time to code as much as they would like to. Such a structure would probably support the compiling of any large linux project, such as X, or Gnome, or... err... Well, Kde, X or Gnome. Any of these projects would be worthwile.

Someone with more experience, please stand up! If it were possible, many people would become that much more involved, and the community would prosper. Could you imagine teams of people competing to help out the KDE developers as much as they do the seti@home project?

Re:A little more information

[Bronster]

Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours. Many developers are working on systems which cannot fully build KDE in under 24 hours

Given that most source files don't get changed every build, the major problem is going to be disk space. We need to donate some of those old 10 Gig disks that are too small for our MP3 (sorry, Ogg) collections now.

Oh, and we should give them all a pointer to Compiler Cache while we're at it.

Must install that on my own system, it look sweeeeet.

Re:A little more information

[Big Mark]

Do you always build the whole tree every time?

They might, it depends on what they change to require the build. They're called dependencies and are specified in makefiles, they define which part of the source tree is dependent upon what other parts.

Almost all of the tree will be dependent on the core libs, so if something fundamental changes in one of them the whole tree needs to be rebuilt to take this into account - even if it's only a one-line change. If something in a higher-level library changes significantly - an entire function is reimplemented, or something - only a few things will have to be rebuilt, as the make program is smart enough to rebuild only those parts of the tree that depend upon the changed part.

So, core KDE developers need monster PCs as they're work has a fundamental impact upon the entire project; app hackers do just as well on less cutting-edge equipment as they shouldn't really need to build everything - that's what binary distributions are for.

-Mark

Re:A little more information

[oliverthered]

Have you tried KDE 3.1 with a 2.5 kernel, it's actually slick, even with my crap unacellerated GFX card.
The kernel made more of a differnace than anything else, with kernel 2.4.19 with CK performance patchset KDE is clunky as hell, xine skips frames and isn't smooth. kernel 2.5.54 smooth KDE no missing frames in XINE.

Re:A little more information

[oliverthered]

? you may have to relink I suppose or are the core API's/interfaces getting changed on an ad-hoc bases?

I can compile core.c and is turns into core.o
I can compile fish.c that depends on core.h compiles to fish.o
I link fish.o that depends on core.o

If I change core.h (an API change) then I must recompile fish.c
If I chnage core.c then I only need to relink fish.o against core.o

If i use dynamic libraries then I don't need to relink atall.

Changes to core.h should be in the form of, 'Right lads, were changing the API, get you design and documentation heads on'

Re:A little more information

[realnowhereman]

LOS(KDE) + LOS(Linux kernel) < LOS(Windows)
QOS(KDE) + QOS(Linux kernel) > QOS(Windows)

The analogy is not implicitly flawed it's just incomplete. Also, KDE includes *lots* of applications, Windows does not. You'd better start adding in the amount of cruft in Office as well.

Comparisons like this are always going to be subjective. I can say right back at you that on this machine Windows XP is less responsive than KDE. Does it prove anything? Nope.

Re:A little more information

[Seli]

2a: You apparently never tried to compile a larger C++ project. GCC needs considerably more time to compile C++ code than C. In general, the GNU toolchain has worse support for C++ as compared to C (yes, C++ is more complicated than C, but I don't think it's _that_ much). Precompiled headers will hopefully help here.

2b:No wonder remaking kernel after changing one module is so fast. Usually nothing except the module itself depends on it, so nothing else needs to be rebuilt. But if somebody commits a change to some of the kdelibs header files, many files have to be rebuilt.

1a: Most people probably don't want even to patch and compile even their GCC. It's just one more thing to take care of.

1b: Moreover, GCC is not the only bottleneck. The linker (not ld.so, but ld, the one creating binaries) is pretty slow as well. Or you could try debugging some KDE app (with debug info compiled in) in GDB - THAT will teach you what 'slow' really means (and maybe you'll even suddenly find KDE's performance quite acceptable).

Re:A little more information

[Anonymous Coward]

Did you even read the post you are replying to?

It did mention the hardware required KDE. Other than to say that many KDE developers (i.e. people who use KDE regularly and find it usable on their hardware) are running systems so old that it takes more than 3 times a long to do a compile of the full KDE system (includeing KOffice, KDevelop, and all the KDE apps)) as it would do on a modern machine.

Re:A little more information

[oliverthered]

Cool, that gentoo build doesn't look so bad now...

I'll look at distcc too since I've more than one PC at home.

Re:A little more information

[tjansen]

Unfortunately the pre-compiled GCC version requires some heavy modifications of the sources - you can only have one pre-compiled header per file! This makes it pretty useless right now. To make use of it you had to bundle all headers into a single file. This, however, would slow down compilations for GCC versions without pre-compiled header support. So the situation is not trivial...

Re:A little more information

[JohnFluxx]

It does indeed parallise very well. The trouble is that over the internet, bandwidth is probably going to be your problem.

Imagine a server that could compile kde in 1 second. But on a dialup it takes you 12 hours to get it there and another few hours to get the results back. :)

Re:A little more information

[oliverthered]

I found this in one of the posts, looks promising.
Cached compliation, basicly like pre-compiled headers on a per make run basis. No changes required.

Re:A little more information

[CBravo]

If you are a student, without hardware but with a fat pipe this would work then eh? I seem to fit the profile.

Re:A little more information

[Xrikcus]

True, but on the other hand many students will have easy access to cs department cpu servers.

Re:A little more information

[Seli]

Hmm, ok, more precisely said, the linking itself is slow. No wonder, ld has to load the huge .o files created by gcc (huge because of all the debug info and things duplicated in every .o). For example all .o files here for libkdeui.so are together 45MiB (101 files), forming 21MiB large libkdeui.so. Creating the library needs almost 30s here (1Ghz Athlon, hdparm shows disk can do 40MiB/s). I don't know how about you, but I call that slow, regardless of what exactly is causing this.

Re:A little more information

[fault0]

In the new days, with computers that match the hardware, these desktops are still snappy as hell. If you want to run the newest version of KDE or GNOME, then get new hardware. They run quite fine on my Athlon XP 2200+. Perhaps you have slow hardware that makes things look bloated.

Windows 3.1 was slow on my 286 (and I beleive the computer could only do cga). I ran Windows 3.0 instead.

Re:A little more information

[fault0]

See.. some people hate having to compile thirty or so odd packages. Other people hate having to compile six or so packages in which they can't chose to compile some. Either way, you can't have both.

Of course, i routinely rebuild kde, mozilla, and OO several times on my box(es) each week (not each one several times), and stay fairly up-to-date on xfree-cvs too. This is brought to you by the magic of ccache, distcc, and emerge.

Re:A little more information

[Anonymous Coward]

If KDE is slick on a crap PC then it's going to run like a dream on a 1.4 GHz Athlon, 512 MB RAM

but it doesn't... yes it's faster than gnome.. but it is useless on a P-II 350... something that it should fly on.

basically, we need to force all the KDE developers to use nothing more advanced than a P-III 500 with 128 meg of ram.. make it go good on that and we hit the mark that makes it awesome... there is way too much in it that has little or no value and only serves to hog processor cycles.... and yes, I dont want my file manager to render HTML... I want it to be a file manager.

Re:A little more information

[Nadir]

Also the upcoming gcc-3.4 will support (at last) Precompiled Headers which will speed up compilation of big apps.

Re:A little more information

[tjansen]

CCache only helps when you compile exactly the same source for a second time - usually done when re-building the source. It does not help you when you compile a file for the first time, when you compile a file that has changed or when a included header has changed. Pre-compiled headers do, even in the latter case (because the header has to be re-compiled only once, not in every file).

Re:A little more information

[oliverthered]

"only helps when you compile exactly the same source for a second time",

That's not what it says on the box, though it seems a bit strange.
"If you ever run "make clean; make" then you can probably benefit from ccache."... I do make clean to remove the cache (.o) files kicking around.
"...can get exactly the same effect as "make clean; make" but much faster. ", umm... nope/

"......which constantly does clean builds of Samba on about 30 machines after each CVS commit. On some of those machines the build took over an hour. By using ccache we get the same effect as clean builds but about 6 times faster."

This is kinda what developers would be doing, though I'm still a little confused about the make clean.

Re:A little more information

[Samrobb]

KDE developers put their computers through a lot of work.

Perhaps they should consider using ccache to avoid recompiling unchanged sources and distcc to make use of the almost-inevitable collection of low-powered machines every developer seems to have?

Re:A little more information

[Anonymous Coward]

See, my Windows machine is snappy enough, without having to recompile the kernel bi-weekly.

Re:A little more information

[oliverthered]

No, but you have to install a patch Bi-weekly.

My uptimes almost a month (since I compiled the kernel).

And yes the 2.4 kernel is shit on the desktop, and windows is shit on the server side.

For the short time I spend re-compiling, fixing &co the kernel I have a mail/web/file server/NAT router running postgress, all easy to maintain and all 'free'.

I know if there's a cash I can probably fix the cause, or at least pass it directly to someone who can.

Re:A little more information

[AaronW]

As someone who regularly compiles KDE from sources for Solaris where I work, it isn't quite this simple. When running the configure script there is an option, --enable-final, which causes the build process to create a single .cpp file that includes all of the other .cpp files in a library. This has two advantages over compiling all of the .cpp files separately. 1. The total compile time is shorter, and 2. the compiler can better optimize the code through inlining.

What is even more time consuming seems to be linking. For some reason libtool takes forever before it starts the actual link process.

Granted, not using --enable-final will speed up the patches, but compiling is still a long and demanding process.

As for debugging, with all the shared libraries, gdb will easily consume 200MB of RAM just to load symbols. God forbid that you link with something like Electric Fence and try and start up a process. A couple of years ago I used this to debug a problem with konsole crashing and starting one konsole session with EF consumed something like 200MB. Loading the resulting core file in GDB took forever since the machine only had 512MB of RAM.

It takes a lot more horsepower to debug and profile code than it takes to run the final code.

-Aaron

Re:A little more information

[Ogerman]

KDE developers put their computers through a lot of work. Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours. Many developers are working on systems which cannot fully build KDE in under 24 hours, and many KDE developers do so several times a week. Profiling and debugging tools for optimizing code are very processor and memory intensive. Hardware often is a bottleneck to KDE developers' productivity.

What doesn't make sense is that hardware is so cheap these days and yet some of these developers are using old crap. Why? Are they really that dirt poor? Seriously! We're talking about like ~$50 Athlon xp 1700, ~$50 motherboard, ~$80 512 MB RAM. Lets say $200 with shipping. Is there anyone who can't afford that kinda upgrade even if they have to save a couple months? $200 is a drop in the hat even with a $30k/year income. It seems these guys either have no concept of managing their personal finance or else they're purposely living in poverty / self-pity. I would hope it's simply the former, because KDE is a really excellent project and its developers deserve a lot more than they give themselves credit for.

Just a thought: Try consulting on the side.. A handful of consistent clients is enough to support a reasonable lifestyle.

Re:A little more information

[CBravo]

We, in Delft, have FAT PIPES at home (think Mbytes/s /system). At the faculty however some still work at Indy machines from the early 90's (100MHz?).

Re:A little more information

[wrenkin]

If you actually read the article, it mentions many of these developers are students. Speaking as a student, I can understand their reluctance to forgo food, shelter, and books for computing horsepower.

Re:A little more information

[mfh]

Speaking also as a student, I cannot understand their reluctance to find a decent job so that they don't have to forgo food, shelter, and books for computing horsepower.

Re:A little more information

[Graphyx]

Well as far as compiling goes there are often many different components that are compiled seperately. But each can have a list of dependants that must be compiled before they are able to be compiled. The smallest distributed portion would have to be a mere single .c and .h file. The true question would be if it is quicker to send those out to compile and wait for them to come back before sending another module out than just compile them homeside.

Re:A little more information

[Anonymous Coward]

speaking as a serious student, I cannot understand where you fit work into the 5 free hours a week I have.

Re:A little more information

[PurpleFloyd]

What about those students who have to work decent jobs to pay for their tuition, food, shelter, and books? It is damn near impossible to work part-time and earn any disposable income, and working full-time while going to school is roughly equivalent to asking your professors to just fail you for the quarter/semester.

Re:A little more information

[wrenkin]

Whatever, I meant 'to not forgo'... don't get excited.

If only..

[Anonymous Coward]

If only somone had adopted me 2 years ago when I was dialing out to a shell from a TRS80 for internet access and coding purposes.. Ever use 'ed'? Ever use it every day for 6 months? There are many more geeks out there in need of help than just the KDE team. That 50mb IDE drive you are using as a doorstop could revolutionise somone's work. Find somone in need and help them out!

-MadCamel [EnergyMech IRC Bot - www.energymech.net]

Re:If only..

[Billly Gates]

Dam. I was stuck 3 years ago on an aging 486dx2 and I thought I had it bad.

If I had to use ed then I would never of learned programming. I admire your patience and willingness to learn and I hope you had good documentation or a great instructor to help you along. Learning vim in console mode and then fvwm after several months at trying to get X to work on my 486 was tough enough for me. I barely got anything done. But ed ?? (shudder)

I take it you must be from a third world country. Or a country in Europe in which your parents do not own such a system. If your stuck with ed I would of donated my fathers old 286 with borland c/c++ and turbo pascal. Even the dos editor had to be better.

I can easily get a pentium1 from going in the garbage at most back alleys. You should do the same if you live in a city. My pentium FreeBSD server is from the trash of a computer repair store near the WTC in New York. The sticker on it says Canter Fitzgarold which means its probably from a victim of 9-11. For those not aware, Canter Fitzgarold lost close to %90 of its empolyees on that day. They lost more then any other company because they owned the top 5 floors of both towers. I feel I can at least make some use of the tragedy by using the equipment since it was desitined for the landfill anyway and the indivual who had it is probably gone.

My point is I am sure you can find old computer equipment from various agencies. Many bussinesses in the US get tax refunds from donating to charity so they like to donate rather then pitch there old systems. Or ask anyone who has been using computers for years. They tend to have old systems they no longer use.

Re:If only..

[Anonymous Coward]

Yes, ed takes some getting used to, but it got the job done, sort of. I'm from worse than a third world country, a small town with less than 2000 people, yikes! Finding hardware there was hard. After moving to the Boston area however I have collected an adequate amount of hardware from various sources such you have mentioned. Rural areas do not make a good geek habitat it seems.

Your FreeBSD server sounds like a good find, not because of it's specs but because of where it's from, I hope you take good care of it :)

-MadCamel

My geek...

[LucidityZero]

My geek can program in C/C++, Java, Perl AND LISP.

And he's captain of the chess club!

I'm so proud of my adopted geek! :D

Re:My geek...

[Daleks]

My adopted geek can kick your adopted geek's ass! Well, actually, I doubt either of them could kick anyones ass, but mine does play a mean game of Scorched Earth.

Re:My geek...

[Asprin]

Umm.... how long are they supposed to sleep?

You see, I adopted my geek and brought him home (his name's Mike!) and put him in a jar by the computer so, you know, he could program when I wasn't using the computer and he could help my wife with her email quesions and so on.

Anyway, this morning I went to change the paper and feed him (pepperoni pizza - yum!) and he's *not* *moving*. I prodded him with a Playstation controller and even offered him a game of Moria, but he won't budge.

Should I try calling tech support?
Do I need to put air holes in the jar?
Does anyone know CPR?

Oh well, back to dot.kde.org!

Re:My geek...

[vofka]

1st Thought: Aww, someone wanted to adopt me (I'm a Mike, and a Geek!)
2nd Thought: Only part time programming... Insanity! Withdrawal!
3rd Thought: Help with E-Mail Problems? What do you take me for? I'm not /that/ kind of geek!
4th Thought: Pepperoni Pizza... My Faaaavourite!
5th Thought: You B****rd, You've killed me!

Go for it

[JohnFluxx]

I think it's cool.

I remember when was I younger I had to stop coding for almost year when my power supply blew and I couldn't afford another one...

It put me behind my classmates (the good ones that is) - a year of knowledge is quite a lot :)

Re:Go for it

[MikeFM]

I agree. If not for people that have helped me out along the way I'd probably have been a janitor and the world would be short several fairly cool bits of software and even hardware.

Seriously, if you run opensource software and aren't giving back code of your own, documentation, or something worthwhile then the least you could do is donate $10 or a mobo or something to one of the projects you use. Not having to spend $500 for your software should be worth that $10. :)

Re:Don't do it until you read this

[Anonymous Coward]

Yes, but KDE runs faster on my 486/25mhz :)

And the developers' old hardware goes where?

[Snover]

I would certainly consider giving it a good home here. I'm not discriminate. You can give me an 8086 and I'll be happy. All that copper will be good to keep away the evil mind-controlling radio waves.

All very good i'm sure

[Suchetha]

But if you *really* want to help out.. why not get in touch with some of the organisations that rebuild old computers to ship them to developing countries (with Linux as the running OS)..

i belive techsoup.org has a list of organisations near you

Suchetha

Re:All very good i'm sure

[Anonymous Coward]

I agree. Only in US do things like these happen. Don't people have better things to do? There are organizations like Big Brothers/Sisters foundations that provide mentoring to kids.

Re:All very good i'm sure

[Anonymous Coward]

But obviousally not running KDE.

it's sad when the OS runs well on a Pentium II but the wm and FM cant...

it tells you that the developers needs to freeze all features and rip out at least 30% of them

Re:All very good i'm sure

[fault0]

KDE 1 runs fine on really old PII's. KDE 3.x runs fine on higher end pII's (like my pII 450). Of course, it's performance could be better, but once you disable most of the visual effects, it works fine. I dont end up running KDE much on it however, as it's a router box and I dont run X on it much.

Re:All very good i'm sure

[Anonymous Coward]

why not get in touch with some of the organisations that rebuild old computers to ship them to developing countries (with Linux as the running OS)..

What are these "developing" countries? Countries that produce code? Or, backwards countries that can't get their shit together enough to feed themselves, nevermind use computers?

Is that kind of like "Hire a Hermit?"

[lingqi]

I mean... pretty much the same, partly - live a secluded life, usually very eccentric, scare away "normal people."

But Hermits can't hack out bulletproof code... hmm...

FYI: back in the old-old days, Castle owners found it "fashionable*" to gave a hermit or two living on their property to... do whatever hermits do. There ARE professional hermits.

* I can't think of another word - I mean, besides peer pressure, why else would you get a hermit? At least your geek would write you some CS homework code for some pizza (I would assume)

Re:Is that kind of like "Hire a Hermit?"

[Suchetha]

a (rich) couple in britain actually put an ad to hire a hermit in november.. one of the articles about it is here

Suchetha

TV Commericals

[johnraphone]

How long till we see the commericals? I know thier coming.... For only a stick of memory a day, you too can help a KDE geek!

From the comments on the KDE page:

[captainclever]

Warning, warning about to get slashdotted
by Yocihc on Monday 27/Jan/2003, @11:45

SLASHDOT crowd coming!!

lol :)

Yup

[Anonymous Coward]

Ok, so give me an ATi AIW Radeon 9700 Pro (not one of those suky-pesky GFFX... ;) and I'll consider it.

Re:Algerbraic is Dying

[Anonymous Coward]

My HP-48GX thinks that your are the man.

Re:Nice to see

[Anonymous Coward]

So, do you work entirely with proprietary software, or do you use opensource software?

uhh ?

[Anonymous Coward]

excuse me, but why would anyone just donate hardware to those people especially if its supposed to be pretty much top-of-the-notch hardware ? can't these people get a job or something? i can buy a 2.4ghz p4 including mainboard and memory with my student/evening job, i don't see why they cant...

Wow.

[Gyan]

Geeks and a family. Go figure.

I am active kde hacker who needs some equipment

[Billly Gates]

I did a full "hello World" application using kdevelop with 2 full buttons for "print" and "Exit".

I feel I need to upgrade to a better system to expand my programming knowledge and help society and Kde in general. I am in desperate need of newer hardware and software since my low end athlon +1800MP with a half a gig of ram just doesn't cut it. A sun workstation 2000 with 2 gigs of ram as well as the Enterpise edition of Forte for java, Borland Jbuilder Enterprise Edition, as well as the full version of Kylix is what will really help me for my quest to help man kind. To help me write great software for you a nice scalable server to help beta test my high end client/server apps would also rock.

PS, I also wouldn't need oops I mean mind a dual XEON 3ghz with the Enterprise edition of Visual Studio.NET and Adobe Photoshop to port some of my great free software to Windows that I am sure I oops I mean none of you can live without. But I can live with just the 2 sun's.

Thanks guys I appreciate your help in this since I can't afford any of these nice toys oops I mean tools. Will you please adopt me.

Re:I am active kde hacker who needs some equipment

[AsparagusChallenge]

You try, unsucessfully, being funny.

The problem to be tackled here is that some KDE core developers spend over 24 hours in a full compile. That is way off from your "Hello world", surely made from already compiled libraries.

Know how to ask...

[e8johan]

If you know how to ask you can quite easily get ahold of most hardware (except HDDs) from technology companies. As long as you can live with 1-2 years old hardware and some DIY to set things up, you can get most for free.

Re:Know how to ask...

[Anonymous Coward]

I don't know how to ask, and very much would like to know :)

Re:Know how to ask...

[moosesocks]

I'm skeptical. Where? How? Why would they want to give ME hardware?

Either way, the demands of the Adopt-a-geek program ask for computers way better than my own (heh...)

Re:Know how to ask...

[e8johan]

I have a friend who works as a computer builder (he builds computers according to customers specs). His employer evaluates lots of fun hardware before choosing a supplyer and stuff gets left over. Also, spare parts may go old (1-2 years). These parts (evaluate + spare parts) is what I pick up from time to time. The company saves money by not having to pay for destruction and I get cool hardware.

If you google around a bit, I think that there are quite a few sites concerning this subject. I rember reading a HOWTO about it once.

Re:Know how to ask...

[Anonymous Coward]

and why don't they just sell it on ebay?

Re:Know how to ask...

[Dead_Smiley]

I want to work for that company! They give me hell here for tossing out old P90's with 16MB. I have to make sure it's "broken" first...

Cool but...

[pigeonhk]

It's indeed not a bad idea to help developers out like this, but seriously why only KDE? Not that I hate KDE, but seriously don't use KDE at all (and before you could ask, I don't use GNOME either).

I'm hoping to see something like adopt an open source developer.

Re:Cool but...

[fault0]

Probably because the guy who started this is a KDE hacker, and thus has the closest links with other KDE developers :)

> I'm hoping to see something like adopt an open source developer.

start it then :)

Why?

[hughk]

Must you build the whole thinmg for debugging? Is it not possible to build and debug KDE at the component or subsystem level rather than to always try to go for the whole thing?

Probably the single largest thing I have needed when working on debugging/building large systems was enough diskspace. Processor power and memory were usually not a major problem because although we would be running the entire system, only part would be in debug at any one time. The rest would be instrumented, but that is all.

It seems like someone should be looking very hard at the engineering aspects if this is really a bottleneck.

Re:Why?

[Anonymous Coward]

apparently you have never heard of Big-O. compiling takes O(n^3) time.

Re:Why?

[JohnFluxx]

Where n=files or n=lines of code, or what?
Have you got a reference to this?

Re:Why?

[Anonymous Coward]

Who cares, I love it when people write the "this is an O(random-function), making it NP-moderately-hard" posts, not even giving us a hint the input. I guess it's typical of the current state of engineering education: read something in a book, refuse to verify or doubt it, but quote it knowledgeably. Big-Oh is often useless anyway, especially with complex algorithms where more than one factor is involved...

Re:Why?

[JohnFluxx]

It's worse than that.

I just had an exam (last thursday) where they asked me for the time and space complexity of a load of algorithms. I put my hand up and asked if they meant time complexity for searching or insertion or what. The guy who wrote the exam had to come all the way up the exam room and replied "oh er well use your judgement".

Jeez.

Re:Why?

[Anonymous Coward]

Should we blame Manchester for this (your home page link)? That kind of thing wouldn't surprise me there :-).

this is lame

[Anonymous Coward]

If you want hardware, go work at K0mart or mcdonalds for a few weeks, a new cpu motherboard and case isnt THAT expensive... but yes, it WILL require you to get off of your fat lazy ass, so i suppose that might be a bit too much to ask then..

this world has plenty of really helpless out there

[Anonymous Coward]

adopt a geek my ass... just cos u spend some hours of compiling some software...

jeez how rotten and technocratic have people become... adopt some poor asian or african kid, rather than adding more stuff and efforts in areas where people arent really poor compared to the huge problems in the rest of the world out there...

donate some money or goods for the real problems, and opensource and the western world will not die because they lack the latest hardware, but the kids and folks in third world WILL...

morons!

Possible alternative donation options

[Nemus]

For those of us without spare hardware lying around, here are some (possibly) acceptible alternatives:

A Case of Bawls - $29.99
Caffeniated Soap - $6.99-$14.99
Caffeine Candy Sampler, v3.0 - $19.99

And various other assorted goods and sundries.

Now, some people make think this is a joke post, but its not. Even if its not hardware, I think anyone who uses KDE should feel compelled to donate something. As someone who does a lot of Volunteer work for local charities, it always feels good when someone recognizes all the hard work you've put into a job. And since alot of these guys can't really spend alot of money on luxury items, I say give em something to make a geek's day a lil brighter.

Re:Possible alternative donation options

[asv108]

Instead of buying overpriced soda from a website, go to Sam's club and buy 5 cases of Jolt for the same price as one case of "Bawls."

We need this!

[falonaj]

The idea might seems quite funny to at first glance, but it actually makes sense.

I am involved in KDE (maintainership of one of the web sites), and I know of cases where lack of hardware has indeed prevented people from working on very interesting projects. It is not only about the speed of compilation, it is also about disc space. This is especially true for projects dealing with Gnome interoperability, as this sometimes requires to compile _two_ huge desktops from source.

Of course, lack of hardware will not stop things forever - other geeks or some distribution will step in eventually - but it has slowed down interoperability effords.

Adversity

[realnowhereman]

During my formative years as a geek I (as I'm sure many of you did) had to make do with whatever was available. Although being pampered and showered with cool gear would have been nice, my lack of up-to-the-minute equipment did not damage me - in fact, I would go as far as saying that my abilities to fix equipment in the middle of a field come directly from those early days and put me and my skills in demand today.

The reason the requirements for Windows keep increasing and increasing, every release requiring the most modern hardware is because the developers all have modern hardware and don't see it as a problem to make full use of it. (Games are even more of a culprit here, but that's a little more forgiveable)

Whatever hardware the developers have is what the hardware requirements will be in the end; if that is a gameboy and a piece of string then so much the better for the project.

Re:Adversity

[AsparagusChallenge]

Agree.

But, on the other side, the people on the optimization team can do its job better if they have decent equipment.

Re:Adversity

[Anonymous Coward]

"Games are even more of a culprit here, but that's a little more forgiveable."

Could you imagine if it weren't so forgiveable?

Look at consoles. The first games on a new console look better than anything else out there, true. Yet at the end of a console's lifespan, the games that were released much later make those early games look like utter crap.

Since console game houses aren't dealing with constantly changing hardware, they gain a massive amount of experience, and eventually know all of the little tricks and shortcuts to make a game truly take full advantage of hardware.

PC game houses don't have that advantage. The hardware is ever-changing, thanks to NVidia/ATI.

That, in itself, is not necessarily a bad thing - new technology is always good. However, I wonder how much graphic processing ability is wasted on all of our PC cards?

Re:Adversity

[stubear]

"e reason the requirements for Windows keep increasing and increasing, every release requiring the most modern hardware is because the developers all have modern hardware and don't see it as a problem to make full use of it."

I disagree. I'm running Windows XP Professional and IIS 5.1 (development webserver and it runs my own website, small few hits) on a dual Pentium II Xeon 450 with 768MB RAM and it's quite responsive. This isn't exactly new hardware and I have no compliants. Well, except that I would like to have faster hardware to run InDesign and Illustrator and rendering from AfterEffects. Other than that...

Re:Adversity

[realnowhereman]

I wasn't really commenting on whether it's possible to run it in that configuration - there must be all manner of set ups that people use. My point is that the minimum hardware requirements for Windows go up and up. On the other hand the minimum hardware requirements for Linux have remained as 386 for ever.

I can still remember running windows 95 on a 386. I've still got the CD so it would still be possible for me to do so. I could not do the same with windows XP. My supposition is that the reason older hardware is ignored (and thus options for pareing down software to meet everyone's requirements) is that developers are too keen on having the latest and greatest and thus forget about everyone not in the same lucky position.

I believe that Alan Cox, as an example, still uses a console with vi for editing the kernel. I have some confidence therefore that he will make sure that that configuration remains available to us all -- including those who might want to run a 3GHz P4 at blistering text console speeds.

Ok, I have a solution for them.

[Anonymous Coward]

Get together with the Gnome geeks, combine the two and spend 1 year making it smaller and faster.

to hell with feature bloat that they love so dearly, and to hell with chasing the unuseable do everything file manager... rip the damned html and other readers fro mthe file managers... Cripes you guys, Nautilus and Kfm both suck horribly now... and it's dragging the wm down with it. Just because microsoft does something doesnt mean we need to copy their mistakes too...

what's next xmms is going to be increased in size by a factor of 10?

Re:Ok, I have a solution for them.

[Seli]

> Get together with the Gnome geeks, combine the two and spend 1 year making it smaller and faster.

ONE year? Oh come on, you have absolutely no idea what you're talking about.

> and to hell with chasing the unuseable do everything file manager... rip the damned html and other readers fro mthe file managers... Cripes you guys, Nautilus and Kfm both suck horribly now... and it's dragging the wm down with it.

And here you don't know what you're talking about either. FIrst of all, KDE's filemanager is called Konqueror, since several years already. Second, the performance of a filemanager is hardly going to drag down the wm. And third, khtml and other readers are not integrated in Konqueror, so they can't be ripped off.

> what's next xmms is going to be increased in size by a factor of 10?

Why? You don't think it's already bloated enough for a thing that just plays music?

Re:Ok, I have a solution for them.

[fault0]

> to hell with feature bloat that they love so dearly, and to hell with chasing the unuseable do everything file manager... rip the damned html and other readers fro mthe file managers...

Uh, if you aren't using the html part from konqueror (and nautilus), it's not even loaded into memory. Like IE, Konq and Nautilus are mostly plugin shells. This contributes to their smallness greatly.

Next are you going to say to remove all modules capablities from Linux?

Re:this world has plenty of really helpless out th

[realnowhereman]

I think you've misunderstood -- IT'S MY MONEY AND I'LL DO WHAT I LIKE WITH IT.

If you want to give money to an african child, go ahead. If I want to burn my cash in front of an african child, I will.

I do not wish to minimise the plight of these poorer people. This is a separate issue. Where does anyone get the audacity to instruct others what they should do with their money? The mere fact that you are writing on a technical website demonstrates that you are most likely in front of a computer in a westernised country and have chosen not to sell all your excess and unneeded posessions, fly to africa and start an educational programme to pull third world countries into a better state. I have no problem with the fact that you haven't done this, you are not required to. Please do the rest of us the same courtesy.

Re:heh!!

[ccie_sec]

you just raised my adrenaline
WTG :)

I use a mixture

[gazbo]

Best tool for the job and all that. For example, our servers are all linux based; it's cheaper for our clients. Of course, none of this really helps the open source developers (except for the warm glow they will get that hopefully keeps their cardboard-box home warm on a winter's night).

Re:I use a mixture

[Anonymous Coward]

Uh, you should change your sig, then, I suppose. "Running only a mixture since 96" doesn't have the same catch, though.

Re:I use a mixture

[gazbo]

Dude, there's more to free/open software than just Linux, and there's more to proprietary software than just Windows.

Unless you actually thought I just used the kernel (and maybe the shell), then it's fairly obvious that I run more than Linux - for example the fucking applications that make the computer actually do something.

It should be fairly obvious that "Running only Linux..." means "...As opposed to Windows", rather than "I boot up a kernel then sit staring at a blank screen, smug with my 1337 purity".

Re:this world has plenty of really helpless out th

[IamTheRealMike]

Did it ever occur to you that maybe these people give to 3rd world charities as well?

Sponsoring a hacker and giving money to oxfam, concern or whatever are not mutually exclusive.

At the same time, you can't really say people should only give money to charities that give food to starving children in Africa. People give to what seems important to them. I can understand those who'd give contributions to KDE that might directly benefit them in terms of a better desktop, as opposed to a charity that works in the 3rd world which doesn't.

Also remember that although these charities do good work and should be supported, they are effectively running at full speed to keep things where they are. There's a reason Africa is still such a hellhole, when South America and Asia are dragging themselves out of grinding poverty. Every time a part of Africa looks like it might be about to make serious progress, various tribal tensions are played off against each other and it degenerates into civil war. Of course that's a gross over-exagguration, South Africa for instance is doing quite well, but considering that Zimbabwe has basically gone downhill since they were given independance, largely because Mugabe leveraged tribal mistrust and favoratism, I think it's perfectly reasonable for people to want to give to a cause that they know stands a good chance of moving things forward immediately.

Don't get me wrong, since about a month ago I started giving some money by direct debit to Concern, who do a lot of such work in 3rd world countries. But I'm not kidding myself. My money will do some good, but it's unlikely to actually improve things, it'll only stem the misery.

Re:Don't do it until you read this

[IamTheRealMike]

Wow, these projects compete in every way. KParts is to Bonobo as GNOME Armageddon is to KDE Myths :)

How to help?

[Anonymous Coward]

> Been wondering what you can do to help out?
> Here's your chance!"

Actually, I'm a die-hard GNOME user (I tried KDE but I found it toyish, lame, and frankly, suckish). I'm wondering what I can do to help SABOTAGE the KDE project. Please give me advice on how I might engage in such activities.

Re:How to help?

[Anonymous Coward]

Check your head !
Why sabotage a inferior product/project ?

(Personally i like KDE)

Re:How to help?

[calethix]

That's easy enough, donate some large hard drives but before you send them, open them up and sneeze on them. Just be sure not to donate anything that can be used for easy backup. ;)

Re:How to help?

[Anonymous Coward]

I know! Since those KDE geeks don't come up with any good ideas of their own, they probably steal them from the Gnome project. If you want to sabotage KDE, hit them at the source and kill the Gnome project. Then you can kill yourself knowing full well all the best ideas in the world are dieing with you and KDE will never get them.

Hope this works out well for you.

Re:How to help?

[JoeBuck]

If you want to sabotage KDE, then I suggest doing lots of postings to public forums like the following:

"Actually, I'm a die-hard KDE user (I tried GNOME but I found it toyish, lame, and frankly, suckish). I'm wondering what I can do to help SABOTAGE the GNOME project. Please give me advice on how I might engage in such activities."

Posts of that sort will reflect poorly on the people you advocate for and damage their reputation. But perhaps that's what you have in mind.

Alternatively

[yatest5]

do *paid* work, then you either get provided with a computer or given 'money' to buy things with.

Maybe they should stick with the older hardware

[jthorpe]

KDE runs extremely slow on anything less than a 500MHz PC with 128mb of RAM+ and that's before you start running applications ontop of that. Maybe the KDE developers should develop a more efficient desktop that doesn't need such powerful systems for the task it performs.

Building KDE on my modern desktop (1.4 GHz Athlon, 512 MB RAM) takes 6-8 hours, what BS! Maybe they should rewrite the entire system if that's the case.

If you use KDE (I personally prefer WindowMaker), I recommend you DON'T donate higher-end hardware to them, it will just give them a chance to create an even more inefficient system and they'll turn around and say "well, it works on my Athlon XP 2800+ with 1Gb of RAM..., I guess your one year old computer is just toooooo slow to run our inefficient X beautifier.".

Re:Maybe they should stick with the older hardware

[Anonymous Coward]

Show me a single comparable desktop that can run well with 128 MB RAM... (the 500 Mhz CPU is not really a problem). MacOS X or WinXP definitely cant.
The goal of KDE is to create a competitive desktop, not a desktop for your C64...

Re:Maybe they should stick with the older hardware

[Anonymous Coward]

I'm running KDE 3.0.5a on an AMD K6-2 400 with 128MB ram, and it runs just fine. As far as speed is concerned, it is just as fast generally as win2k (and it actually boots faster than win2k on this machine).

I also have an old PII w/64MB ram, though KDE *is* as slow as a dog on that box.

While KDE runs fine on my AMD box, I would never dream of trying to compile KDE on it unless I was about to leave for a weekend vacation. :-)

Mark

Re:Maybe they should stick with the older hardware

[Anonymous Coward]

My bash prompt runs pretty nice on a 486. I tried KDE, what an awful lump of manure. The KDE developers are unemployed for a reason, they cannot code. Don't give them shit!

Re:Maybe they should stick with the older hardware

[thryllkill]

MacOS X runs fine on my ibook, and it only has 128 megs of ram. Not to mention they aren't complaining about KDE not running well on their systems, they are complaining about how long it takes to compile. RTFA!

Re:Maybe they should stick with the older hardware

[fault0]

Funny, KDE 3.x runs fine on my pII 450 with 128 mb ram.

I remember KDE 2.1.x running fine on my powerpc 603e 200 mhz (603e=slow chip.. slower than PPC 604 and G3) with 48 mb of ram.

Re:Maybe they should stick with the older hardware

[Anonymous Coward]

Yay, I'm glad for you. How many timees a week were you able to build KDE as a whole from source?

Re:Maybe they should stick with the older hardware

[fault0]

As a whole? never? I use ccache?

Compiling KDE on _that_ machine? barely ever. although I do use it for distcc once in a while. On that machine, I barely ever go into X anyways, and I skipped KDE 3.0 as a whole (KDE 2.2.2, I beleive to KDE 3.1 rc3)

Re:Nice to see

[smallfries]

Why should I say fine wines? How will it help you in your pitiful bourgeois existence, I could sell you fine wines but that would certainly cost you a tidy sum.

Fucking proles Ahh, now here we can at least agree on something. Especially when their rent is due and they become desperate. Exquisite

A troll posing as a FUD-fighter

[Johan Veenstra]

> GNOME version number is bumped (currently it is at 1.4).

Next time at least post something up to date

Adopting a geek ?

[Fred_A]

Jeez I don't know, is it house trained ? Will I have to have it neutered ?

Remember, a geek is for life, it's a big decision...

Re:Adopting a geek ?

[Scorchio]

I don't think neutering will be an issue.

Re:Adopting a geek ?

[Tar-Palantir]

It depends on if you have any pretty teenage girls in the house... then you'll have to have it neutered.

Re:Adopting a geek ?

[Anonymous Coward]

There will still be the stink and stains of geek cum. There just won't be any sperm in it.

Re:Adopting a geek ?

[Anonymous Coward]

Only if the teenage girls have green hair, improportionally large heads and long legs, and the area of their eyes make up 40% of their faces.

Oh, they've got to wear school/sailor uniforms with miniskirts too.

That's the only "girls" a geek knows of.

Why send hardware?

[Anonymous Coward]

I'd be more than willing to really ADOPT a geek. In the sense that the geek moves to my place and I pay for his food and living. I've got lots of hardware here that I just might need one day and dont want to give away, but I wouldnt mind if some geek used it. Besides, I could always use a little help while coding and I sometimes feel lonely, so I'll have someone to cuddle. Geeks are just adorable - kinda like having a cat to play with. Only geeks play with a ball of ethernet cable, not a ball of yarn.

Plus I heard that fat geeks are really warm - could save some on the heating bill.

RTFA jack-ass

[Fizzl]

And my obsolote motherboard or CPU will help them live another day?

I have no intention to give my money to anyone for free. I can, however, give away my obsolete motherboard or CPU which I couldn't sell for a price that would justify the hassle of auctioning/whatever it.

This sounds like a vast conspiracy...

[sean.m.bober]

to build the ultimaate Duke Nukem Forever gaming machine. Oh wait...Nevermind. Sean

supporting free software in this way is good...

[bob@dB.org]

but for the ones who don't like kde that much and would rather support gnome, there is always The GNOME Foundation

Re:supporting free software in this way is good...

[21mhz]

Come now, everybody knows that GNOME developers are a bunch of filthy rich corporate servants, starting from age 16. Let the non-profit status of their Foundation not fool you.

Obviously a ploy...

[sn0wcrash]

to create a bigger beowolf cluster!

Why Dont the Rich KDE Geeks

[Zapdos]

share?

You asked for it...

[barnaclebarnes]

You might want to check out the Woman of KDE Website. Not quite what you were looking for but I guess you are ugly and beggars can't be choosers right?

Re:You asked for it...

[IIRCAFAIKIANAL]

DEAR FUCKING GOD, [kde.org] that just turned me GAY.

Good, less competition for the rest of us. The female on the right is pretty hot.

(Well, except I'm practically married to an intelligent hottie already:)

Re:You asked for it...

[carlos_benj]

So, you really liked that guy on the left, eh?

Re:You asked for it...

[1u3hr]

Actually, see the women of KDE here. About half of them are worth a look...

Re:You asked for it...

[Trinn]

I'm also, as you put it, practically married to an intelligent hottie, but I'll put it this way. If I were female, this might just "turn me lesbian" (though personally I don't believe someone can be "turned", just commenting on what was already said.)

Sad

[dpete4552]

I'm sure this gives everyone at the Microsoft camp a good laugh.

jackass my ass --- Re:RTFA jack-ass

[Anonymous Coward]

there are also poor schools and much more who dont even have shit like old 386 or 486... why not donating there idiot?

the rich only want to get richer and the wealthy complain all the time that their lives is so fucking bad....

jeez think about the really important projects and donate your hardware there where isnt anything yet.

kde are doing a good job already, so where is the problem???

Re:jackass my ass --- Re:RTFA jack-ass

[Anonymous Coward]

Because my fucking tax dollars already go to the schools.

I'm only guessing, but the poor probably only want to get richer also. And cryin' about how their lives are so fucking bad, but not doing anything to better themselves.

Since I use it every day. KDE is an important project.

RTFA jack-ass

A better solution

[Anonymous Coward]

Instead of donating them money and hardware, how about giving them their local want ads? I am sure if they were employed in decent to good jobs they could afford all the spiffy new electronics they want. There is no reason that a programmer on a project as successful as KDE would be out of work once they learned optimization. That is probably the little problem hindering them from gooey hardware indulgence. Come on, a part time job couldn't be that hard to cope with anyways. Most geeks still live in their parent's basement and if you put them in a fast food enviroment, they would fit right in with all the other overweight greasy employees there.

Let a geek log in on your system

[NKJensen]

I guess a lot of us would let a geek log in on our system(s) and compile there as a niced process. The boxes are running anyway...

The geek can then get the compiled code for testing.

Would that help you guys?

huh, maybe someone should adopt me

[calethix]

I was curios so I went to see what kind of stuff they're looking for:
Examples of useful items:
* Memory in quantities > 64 MB
* Desktop processors > 800 MHz
* Motherboards that support such processors
* Hard drives > 10 GB
* Laptops > 300 MHz
* Monitors, graphics cards, other similar goodies

Not to sound like a troll or anything but I'm running Linux on a K6-233 with 64mb ram right now. Granted, I just built a new main PC so that will soon be upgraded to an 850 with 256mb but even that is barely above the requirements.

Re:huh, maybe someone should adopt me

[Anonymous Coward]

And i'm running a p4 2.53 ghz with 120 gb of HD space, an ATI radeon 9700 pro, and 1 gb of ram.

Maybe I should donate something? nah.. wait, I'm a poor college student too.

Re:huh, maybe someone should adopt me

[Anonymous Coward]

very funny but do you compile as much code as they do?

stock options?

[peter303]

Like when the Linux companies gave some to open source programmers.

Great!

[Tuxinatorium]

But unfortunately, this adopt-a-geek program could cause my investment in parents' basement futures to become worthless.

Apples and Oranges...

[crimson30]

Dude... is it me or aren't geeks a little more proven? I mean... if this were to take off, it would have some sort of positive effect (on KDE). Meanwhile, feeding the children of some 3rd world country accomplishes what? Feeds them so they can grow up and have more starving children, right?

At least helping geeks isn't likely to contribute to overpopulation (:

Re:Apples and Oranges...

[Anonymous Coward]

oh so now its you who decides who is worth to be livin and whos not...

some government or dictator will select you some day to be unworthy, just the same way you just decided about whos worthy and not, would like to see your face then...

are you so wise to be their judge? can you give it to them what they deserve?

remember your wise words cos one day u will be judged, and it will not just be some dictators or government....

Re:Apples and Oranges...

[crimson30]

You're right. No one should be judged any more worthy than another. On a small scale, my comment reads like pure flamebait... but on a larger scale, I would like to see *my* money accomplish something visibly positive. Contributing to the welfare of people who are likely to generate more welfare-requiring progeny isn't my idea of something positive.

Maybe they should put out something useful first..

[Anonymous Coward]

Instead of a WM that crashes and hangs all the time. Take the bloat down and they won't have to whine about not having fast enough hardware to write it on!

Faster hardware ??

[dreez]

This is all wrong, the KDE-developpers should have slower hardware, so they are motivated to make KDE as fast as possible and get the source small enough to compile within an hour or so. So here's the deal, i will do a system swap with a KDE developper , My 266 Pentium for his 2+ Ghz P4 !! Grtz Dr--z

Re:Faster hardware ??

[fault0]

> and get the source small enough to compile within an hour or so.

great.. go rewrite g++ for them so that compiling large C++ apps (mozilla, kde, OO) takes less time.

Adopt a Geek

[MissMoneypenny]

mmmmmm I already did that; he's 26-years old and also known as 'boyfriend'. I must say, they don't cost much these geeks and they do come with extra features - though I don't know if that also goes for a KDE geek ;-)
-
MissMp

This is What Happens When You Code For Free

[Anonymous Coward]

Free as in poor.

Re:this world has plenty of really helpless out th

[rseuhs]

adopt some poor asian or african kid, rather than adding more stuff and efforts in areas where people arent really poor compared to the huge problems in the rest of the world out there...

The sad (and cynical) truth is that adopting a 3rd world child will only make the problem worse because instead of the kid dying now, her/his 6 or more kids die later. (unless you triple your donations every generation)

Ethiopia went from 8 Million to 80 Million inhabitants from the 1920's to the 1990's.

Imagine the US going from about 150 Million to 1.5 Billion or western Europe going from 300 Million to 3 Billion in the same period.

Food donations won't solve Africa's problems, only postpone them.

It might be better if

[narfbot]

It might be better if they were support all poor open source geeks. You see, I yesterday was compiling stuff on a 1.2 ghz athlon yesterday under knoppix (this system didn't have linux on it, but I was testing it). And it was surely a whole lot faster than my 333 Mhz comp I use all the time. If I made that jump (which would be at such an insignificant cost to me, but I am poor), I would be so happy to do much more programming. But see, I don't use KDE at all, and probably never will, but I want to work some more on projects like Wine, or maybe some kernel modules/drivers to get some of my hardware to work better. Even though I wouldn't be working on KDE at all, I could be assisting the KDE project in an indirect way, like the kernel, by improving hardware/software compatibility, which is definately necessary for Linux today to continue to grow.

+1 Insightful

[schlach]

Thank you sir for proposing a simple solution to what should have been an easy problem. You should post this on the original page, too.

Oh come on now

[bogie]

If you *really* want to help out...why not join the peacecorps and due a tour of Africa?

People could go on and on about more "worthy" causes. Let the KDE people request some help without trying to guilt others into donating to a different cause.

Oh I forgot

[oliverthered]

I'm all for supporting hard up developers, I was once one my self.
If anyone lives in the Newbury/Reading/Basingstoke area (UK) and could really do with some extra kit, I've probably got some spare bits floatings around (256MB ram a couple of HDD's, boxed Mandrake 8.0) drop me a mail and I'll see what I can do.

Feed them?

[mu51c10rd]

Do I have feed this geek? Do their laundry? Hope not...I debate if some of the geeks who spend all the hours of the day posting on /. wash their clothes ( and themselves ). On a serious note, I think it is a great idea. The open source world could certainly use some more programmers, this appears to be a good way of obtaining them. Nothing like having a young padowan learner...

I think it's not very fun...

[Peterus7]

I mean, you don't get to keep him? You don't get to let him live in the basement and program computer things for you in exchange for food, bandwidth, and sleeping on the couch? Where's the fun in that?

I'd like to see a rent a geek service. Now that would be cool.

Re:I think it's not very fun...

[pizza_milkshake]

hey i'll do that, as long as the bandwidth is good and the food is spicy and chinese

Can I Keep Him?

[witten]

Woah! This topic is eerily familiar. Check out this comic on adopting KDE users.

Performance

[Anonymous Coward]

Maybe if it takes two days to compile your program on a 300Mhz system, and then takes 2 hours to draw the windows in the resulting program, maybe that is a clue that you need to take a look at the performance to eyecandy ratio and see if you can make the system a little faster.

Neither KDE nor Gnome run very well on my 233 laptop anymore, and it is not because my laptop is too slow to provide the services that Gnome and KDE provide or that the architecture of either needs that much (though both could improve in this area as well). They run poorly because of things, like Nautilus for example, that take up all the system's resources drawing simple icons in as fancy a way as possible. Before Nautilus, in the gmc days, Gnome ran just fine on that 233.

There are numerous ways that KDE and Gnome could improve performance. Unfortunately both projects are headed in the exact opposite direction.

So forgive me if I don't shed any tears, if I don't jump right out and give you a better system than *I* have (>800Mhz is more processing than my home PC). I don't WANT you to have that much processing space, if you did it would be used poorly and then I won't be able to run Gnome or KDE anymore even on my home PC. My system would be so busy drawing stupid little anti-aliased animations that it wouldn't be able to operate as a PC anymore.

Forgive the fire, but this is a major pet-peeve of mine. Computers should be useable first, pretty last and as far as I can tell there hasn't been enough gain in useability and useful features to warrant the resources that these desktops continue to require.

NR

Yum...

[Peterus7]

You know I think I'd enjoy doing that as well. What say we all start some sort of online/slashdot based company, rent a geek: Nomadic geeks, exchanging computer repair/programming/webbuilding services in exchange for a cool basement with a computer and bandwidth and spicy chinese food... Yum...

OBLIGATORY....

[Anonymous Coward]

Here we go .. the ol' obligatory

"In Soviet Russia, KDE Geeks adopt YOU!"

Depends a lot on change control

[billstewart]

How much of the system you have to build from scratch depends a *lot* on the quality of change control processes you're using and the amount of the code that's being actively worked on and modified at any given time. It's really critical for developers and testers to be able to work on something that looks very much like the environment the real user will have, and sometimes that means a lot of recompile-from-scratch-on-a-clean-box.

Disk space is a situation that's improved radically since the days when I was coding, and price/size has been on a deep faster-than-Moore's-law dive for a few years now. A 120-GB disk costs about US$120-200 these days, and the trip from 2GB->6GB->20GB->120GB only took about 3-4 years, but it rapidly crossed the boundaries of "how big is the biggest system I'm working on now with everything and the kitchen sink (except my MP3s) thrown in".

Also, one reason disk drive was always a critical resource was that corporate IT departments often forgot the difference between computers and systems - developers and testers often need large numbers of systems, but that doesn't have to cost a lot of money because one computer with a removable-disk-enclosure and a stack of 20 disks in $10 plastic drawers really costs a _lot_ less than a stack of 20 computers, and the IT department and/or the developer can keep a set of clean images available to duplicate more checkpointed-from-user's-perspective systems for testing on.

Developer vs. compiler horsepower

[billstewart]

It's really nice to know that many of the developers are still using slow, wimpy computers instead of the latest supersonic barnburners, because that says they'll resist bloatware and their products might work on my old desktop PC (not the really old one running TWM, just the 233MHz one). As a longtime user of corporate-IT-department laptops, I appreciate this for work use as well as for home, since the laptops are typically far more limited, and they're expensive enough they're usually a couple years old even beyond the laptop limitations. It would be nice if we could give the folks at Microsoft a bunch of old computers for all their developers and make sure they use them too :-)

Having said that, though, that doesn't mean that the machine you run your compiler on needs to be the same machine you install and test the end product on (and in fact it's really nice if it's not, because that forces you to make cross-compiling for production systems work well.) Running the compiler on a newer, somewhat faster CPU and adding a big disk drive helps a lot, and motherboards at Fry's seem to be running about $99 for ~1.7MHz Athlons these days. Of course, it's awfully tempting if you've got a new motherboard to install it on your desktop (even if you're not a gamer...)

Distributed compiling for students

[billstewart]

It's true that my disk drive bus is a lot faster than my LAN, which is a lot faster than my DSL or especially my modem. But in a university student environment, either for students in dorms or campus computer centers, there's often a fast LAN and a lot of spare horsepower much of the day that might be harnessed effectively. There's a big security assumption there (:-), and also it's much more useful if the students are all running Unix of some sort rather than Windows, but it can be a good deal.

Also, there's the question of where the data lives - do the master copies live on your PC, or on a server, and how do you check it in or out? It may actually be just about as fast in a parallel environment, where people are getting the data from fast LANs.

Re:Distributed compiling for students

[JohnFluxx]

Check out distcc, which does distributed c compiling.

At my uni you have an NFS drive. So I can ssh into any machine in the uni and my files appear to be on it.
One quick and dirty way to do this would be simply to compile each directory on a seperate machine. Most large programs have lots of sub componants so you can compile each part seperately.
Use distcc for more complex cases - plus it is probably easier to use.
From distcc.samba.org:

"distcc is nearly linearly scalable for small numbers of machines: Building Linux 2.4.19 on a single 1700MHz Pentium IV machine with distcc 0.15 takes 6 minutes, 45 seconds. Using distcc across three such machines on a 100Mbps switch takes only 2 minutes, 30 seconds: 2.6x faster. "

I took a quick look at the instructions. It really is dead easy to setup. I think I'll do it :)