Slashdot Mirror
Essay on the GNU Community
Rowan writes "I've
written a longish (but nice, I hope) publication about
various matters in the GNU Software Community. It ends with describing how important
conflicts are to learn and discover errors. It's meant to be the first in a row and I think
they'll all be worth the read. "
In real life, names can be overloaded. There,
as in programming languages, overloading is well
defined, but may lead to confusion for the uninitiated.
A more suitable model for the space of ideas can
be found in the OMG Common Object Services Spec Vol 1:
Naming Service. I believe that each name has a relatively precise
meaning for the speaker, but may be understood differently by a listener.
Namespace federation is an ongoing research problem and automatic mapping
is not going to come about easily. A rigid framework (like rdf, if it wasn't just for commercial purposes) would help,
but who in this world can muster the discipline (well, computers, maybe).
Well, hope you orcs had fun torching Rowan, but perhaps
Taco is developing some editorial style, finally A great article.
I look forward to more from Rowan..
The anger I can somewhat understand, but there is also a call
for dialouge. GNU means nothing if people can't use the software
for much. I have said it before, and I'll say it again, that many
regular posters here ARE using Windows to connect to Slashdot
because regardless of their pretensions, they *can't* get their
LInux boxes configured at home, sysadins or no. Look at Taco's
about link. Yet, most of you deny it.
Windows can be frustrating when it crashes and when one comes
upon its limitations, but so can Linux be frustrating to very
intelligent people. I think a large part of the frustration is that much
of unix, at least the documentation, is intentionally arcane to protect
the job security of sysadmins, who are not necessarily very intelligent
or creative people. They are just people who have been through
the hazing and have learned some of the ropes. Therefore, one
can argue that GNU software which is intenionally arcane and
poorly documented is *NOT FREE". It is only of use to a few people,
usuallly people working fo large institutions (the usual unix
clients) who parcel out rights and priviledges to "users". And, just
like MSCEs, reboot machines and install most of the time.
Yet, you think so much of yourselves. The fact that you
occasionally cut a and past a perl script does not make you
a developer. Sort of like the "tech men" managing the dying
atomic power plants of the old empire in Asimov's Foundation.
So, is, in effect, much of GNU only for a very narrow and selfish
guild or fraternity, which has probably done more to retard the advance
or computing than to promote it? That can be argued. These
are by no means creative geniuses - just careerists. Persons with
any kind of creativity or wisdom would not post the kinds of
comments I've read here. I say these negative posts are motivated
by envy at something real and refreshing.
Well, Rowan has done some work and research to "hack"
his own system, yet still admits that the whole system is very
frustrating to work with, sometimes. He also points out possibilities
in ways GNU can advance computing into the 21's century.
I agree with much of the specifics (though Rowan wasn't too
specific) except that I like KDE. "Themes" are the least important
aspect of a desktop to most users, and I like the way the
default KDE looks anyway.
Really, there is a lot that can be done to make computing more
creative for most users - even with text based applicatioins.
Especially in the area of "idea" databases using AI and fuzzy
searches, where users don't need to work with files at all.
(Unless they are repairing the system). Much of that can be
done with XML already - it just isn't.
We need more articles like this from people who have actually
tried to use GNU and Linux who are free from much of the
pretension (and dishonesty about one's own experiences with
Linux) I see here. A refreshing change from carefully edited,
pompous and preach articles by known personalities in the
Free Software movement. The rambling style and lack of
editing is ok. Content rules here.
skinbark
Linux GUI isn't ready for prime time and I believe most people know it. There is such fragmentation in the X world it makes the Unix (R) world look unified.
You have a few dozen toolkits, several window managers where approx 80% of the code is duplicated, an immense codebase consisting of code designed to be backwards compatible with features people haven't used in 10 years, and a company which until recently wanted to close X off to the open source community.
Hack ontop of hack ontop of hack to make what was originally a glorified drawing API, a graphical user interface.
Linux is ready for corporate computing, server environments where you don't need a 600 meg GUI to run a 2 meg web server or mail server.
Linux will be ready for the desktop arena soon, just not right now.
--
The world is neither black nor white nor good nor evil, only many shades of CowboyNeal.
The article is posted from a ".nl" domain, so it is possible, indeed likely that the writer is not a native English speaker. Please keep that in mind when assessing the essay. Also keep it in mind when (as someone already has) questioning the notion of this person writing a book.
English is the lingua franca of the Internet. Since it's the only language I speak and for other reasons I won't go into to avoid a flamewar I think that's a good thing. However, it can put people who aren't fluent in it at a disadvantage. Even people who can speak/read/write it fairly well might not be able to write fully grammatically correct idiomatic English. Unfortunately, this can put people at a disadvantage in discussions as it can make them seem confused when in fact their ideas might be clear. Of course, their ideas might possibly be murky as well, which makes comprehension doubly a problem!
No disrespect to Rowan, but I think slashdot articles need to more concise, better written, better edited and better presented than this. If we want an unmoderated source of amateurish rambles then we still have Usenet.
Rowan: it has a certain charm and I accept that you may go on to say something interesting, but that's an awful lot of grey text to wade through before reaching it.
"I am not getting into the KDE customizability argument", followed by:
;-) and my opinion holds no weight, but who am I to say that?
"it sucks", "it looks worse than a windows theme", "it will get better using Qt 2.0".
Gee, wonder what would happen if you *got* into the argument.
Ok, let me explain you a few things, please take it kindly.
1) The goal of KDE is not to provide themes to you. There, I said it. Themes are 99% of the time pointless, distracting uglyness, which only detract from your productivity.
2) Configurability != themes.
Configurability is providing the means to adjust the behaviour of your environment to different ways of working.
KDE's mac-like menu bar is more configurability than all the gtk pixmap themes. Why? Because it's useful. Because it provides you a different feel, not just a different look.
It's an option, you change it, you adjust KDE to your preferred way of using it. It's configurability.
3) Why leave windows and go to KDE?
Put like that it makes no sense, of course.
Then again, that should come as no surp[rise, since it actually makes no sense.
People don't like windows for KDE (or GNOME). They leave windows for Linux, or FreeBSD or whatever, +X +KDE.
So, what do you get by making that switch: you get multiuserness, you get remotability, you get stability, you get some extra software, you get some software cheaper (say, WP), and you get a GUI that is not terrible, from the point of view of a person who is used to windows.
You see, if you switched instead to Linux+X+twm, several of the advantages you would get are not accessible to you, or are accessible through a higher learning curve, which often means the switching user won't care.
It's a golden rule of UI: if the user doesn't know it's there, or if the user can't *use* it, it could as well not exist.
Hope this helps you, it looks like you have something you want to say, but I can't see it behind the things I see as wrong or disagree too deeply with. If you want to reach a reader, you need to make sense to him.
Then again, I may not be your intended reader (I'd say it's likely I am not
Disclaimer: I am a KDE developer.
someone with enough guts to point out reality. Linux is far too complicated to head for the mass desktop market anytime soon. And the documentation is really poor. I don't know about other distributions, but Red Hat's documentation is chaotic, incomplete, and often incorrect. There is virtually no comprehensive documentation to explain how modules work together. Maybe I've overlooked something.
If I knew more, I'd write it myself. But if I had the documentation, learning it would be a lot easier. It's a vicious circle.
* And remember, it's spelled N-e-t-s-c-a-p-e, but it's pronounced "Mozilla."
However, aside from his advice about approaching conflict in a positive way, he had little to say about what to do about the problems he raised.
So in that sense, he's preaching to the choir here at Slashdot, to very little point. It's relatively easy to identify problems, but it's often quite difficult to solve problems.
On the other hand, there's a place for this... I've been complaining about Microsoft for ages, but I didn't have a concrete suggestion about what to do about them, I've just been frustrated.
Professional Wild-Eyed Visionary
Rowan, you've made some valid points. Quite a few of them, in fact. I've got to admit, though, that they weren't easy to find.
I confess that I'm probably a bit more critical of others' writing than most (life's hell when you were a double Comp Sci/Englsh Lit major), but I hope you're open to a little constructive criticism.
Focus, my man! I think you tried to make too many points too quickly, and I often had a difficult time figuring out where one point ended and another began. I know it can be difficult to have a lot of things to say, and a limited space to say them. You're sure to lose readers, though, if you try to cram too many of them together at once.
As an example, CmdrTaco's title for your article was "Essay on the GNU Community," but that was only a small section at the end. I'd be very hard-pressed to come up with a title that fit the entire article... and that should be a sign you're trying to say too much. Select a topic, and write about that topic. You can always write more later.
You've got several distinct ideas there (and more than a few major complaints) which could easily be articles of their own. Think them out, come to conclusions where you can, and take some time to decide if they might be worthy of their own article. If they're not, see if they help make the main point if the article you're writing. If they don't, throw them away -- they're only cluttering things up.
This has absolutely nothing to do with the moderators - the article was posted by Rob and Rob alone - the moderators don't influence the content of the topics, just the rating of the responses.
Hi there,
:)
I'm Dutch too (assuming Rowan is Dutch). I don't think there's much of an excuse for the amount of misuse of the English language displayed by the article.
Granted, non-native speakers make mistakes. But Rowan chose to write this in English and submitted it to Slashdot for 'publication'. I think more care could easily have been taken.
And even if the language excuse were valid, there is no excuse for the lack of coherence of this rant. I've read rants by Rowan before (on the ggi mailing list and I think he's also been active on the fbcon mailing list). This is one rant too many. Now I'm going to give some potentially unflattering suggestions:
* I think you mentioned you wouldn't learn C on that list, but would start with C++. If you want to have a say in GGI development or Linux kernel development, I strongly advise you to learn C. If you then want to influence the direction of a project, *code* (or document, or test, but do something constructive). If you do that, you may find people may actually listen more carefully to what you have to say.
* If you haven't experience with any programming language yet (though I've seen a reference to Pliant), then with Eric Raymond I suggest you learn Python, which I consider to be a great language. If you come to comp.lang.python (or tutor@python.org) with questions I'll personally help you. But really, any language will do. Even Perl.
* Please read a book or web page about writing coherent essays.
Regards,
Martijn
Okay...everything negative that can be said about this article and about Rob's editorial policy has already been said.
I would like to point out that:
a) nobody's perfect, not even El Taco himself
b) everyone has a bad day
c) everyone makes the occassional blunder
Now, that said, I think that we can all agree that the general quality of slashdot is pretty damned high, and getting higher all the time (thanks be to the gods for comment moderation...*hallelujah!*).
Just 'cuz CT fumbled the ball this once doesn't mean that the overall quality of slashdot is suddenly dropping. Just 'cuz ol'TacoHead had a bad day doesn't mean that he's suddenly turned into a bumbling moron.
Rob has done a *phenomenal* job with this site and everything related to it. My god people, have some respect.
Now, since everything negative has already been said, let's give it a rest and move on, okay?
Thanks.
- deb
If it is elistist to demand proper communication, then I am by all means elitist.
Most of what Rowan said was _obvious_ to the informed individual. To the uninformed individual there is no argument supporting his positions. He certainly does not have the kind of clout for anyone to just take his word for it. The point is that this paper really has no point. Other than something for 15 year old slashdotters to clammor around.
For example, while he does point out that there is a lack of unity in window managers he doesn't really offer a good reason why there should be. Or the fact that "Open development" might actually have a few inherant flaws. While some believe that "Open Source is the ultimate standard", I would beg to differ. Empirical evidence backs me up on this, eg: window managers, GUI libraries, etc. Just because you release your source, does not mean people will rally around yours. It takes a great deal more than this. This is one strength that the corporate world has enjoyed.
When IBM entered the PC market, there were already many other machines out there. All of them incompatible. IBM became THE standard virtually overnight. The writing was on the wall, virtually everyone knew it. IBM had sufficient clout to make it happen. While some may argue that the IBM PC was not the optimal platform, I'll still argue that this was the best thing for the industry and the end user. Having ONE platform, meant that individuals and companies could code for ONE platform and be able to rely on it still being there when they're done. Companies could develop sound cards, and video cards, etc. As the market grew, IBM compatibles started rolling in... Overall this was a good thing.
This is one quality that the Open Source movement lacks simply lacks. This is in some ways a desirable feature. Perhaps some will say there is a certain amount of personal clout that will achieve the same effect. That RMS, Torvalds, Raadt, and co. can bring similar forces to bear. While this may be true to some extent, I do not believe it is anything as close. There will still be significant fragmentation.
Let us imagine if the Linux community wanted to create ONE 3d graphics API, sort of like OpenGL, only absolutely free. There is a certain neccessity for unity if we desire hardware accelerated video cards. I suspect it would never happen, atleast not on a large enough basis to really compete with the alternatives such as OpenGL. You would still have Joe Schmoe, saying. Hey wait, I can do it just a little bit better. And a few bleeding edge types who say 'you know, he's right'. This happens all the time. Unlike the commercial market Joe Schmoe is coding for his own personal edification, not for profits. The fact that he doesn't have a snowball's chance in hell in getting it off the ground won't deter him. The end result is fragmentation, and maybe excessive fragmentation.
My point is not that Linux, or OSS sucks. Rather that OSS has certain qualities. Not all of them are optimal. I wish people would speak frankly about these issues, so maybe they can be addressed. I'm sick of hearing 'arguments', that all needs will be meet(and better than the commercial markets) because 'if there is an itch, someone will scratch it'. This is not an argument, it is conjecture. Let us think as individuals, don't just take RMS' words at face value.
First: Just because you're a hacker and spend time configuring the guts of your system doesn't cut you any slack when you're trying to communicate. A good bit of your text was rambling, pointless, poorly constructed and ungrammatical. It is no excuse that you're a hacker. If you want to say something to the larger community, learn to say it clearly and well. Judging by other comments, you nearly lost most of your readership before they got to your point.
Speaking of points, wasn't yours, "We need to be willing to gracefully concede a point to others, to compromise for the good of the movement"? Or did I miss that?
In case that was his point, tho, speaking to the rest of the community, I'll say I can agree. Being prepared to work for your ideals is critical, particularly in the face of opposition. But above all, being willing to work even when the credit doesn't come to you, even if it means "trying on" someone else's vision for a day. That will be where advances can be made. On the other hand, there comes a time when compromise is not possible. When Vision demands that you move this way. There are going to be collisions. Live with them! Conflict is not evil, but what is done in the name of "resolution" can be. When conflict rears its head, we would do well to look beyond the shouting, to the root cause, and strike there.
So back to you Rowan. You've thrown the gauntlet; I'll flip it back. Pick one conflict in the Open Source/Free Software movement, and recommend a solution. Demonstrate the skill we need more than temper-control: discretionary judgement.
Believe nothing, not even if I say it, if it violates your sense of reason -- Buddha
Thingy would not have been the term I would have used if I wrote an piece like this. It seems to me that a lot of people are jumping on the 'write an editorial and get it posted on slashdot' band wagon. This can be a Good Thing, if only the same thing wasn't being said over and over.
This is my favorite line: :)
"My vision allows me to see a glimpse of what computers will be like in the long term and that glimpse makes me even more sick of the old." I wish I had that super power, too
Yikes!
/.??
This reads like some wordy, rambling and largely non-sensical post that would have been quickly moderated below my threshold.
What is this stuff doing getting posted on
This has got to be an all-time low.
Do you even know anything about perl? -- AC Replying to Tom Christiansen post.