1) The GPL is neither MORE nor LESS 'free' than the BSDL or AL.
Proof by assertion now? I think not. The GPL imposes restrictions on the licensees, restrictions of which the other mentioned licences are perfectly free of. Consequently, they are more free.
2.) The GPL is not the 'ONE TRUE LICENSE' any more than the BSDL, or AL are (or any others for that matter).
That's nice of you to say. If you've never met the folks who believe all software should be GPL'd, then I envy you your innocence.
3.) THE GPL IS NOT 'VIRAL'!
Well, fine. It doesn't give you influenza, and it doesn't rewrite your BIOS. But it certainly spreads itself from one place to other, irrespective of any sense of proportion. Would you be happier if we just called it homeopathic instead? Viral has many meanings, and this one is certainly close enough to count. But if you prefer, I'll call it homepathic tomfoolery instead.
You can count angels on the head of a pin if you'd like, but I don't care to.
What you write is yours. What I write is mine. And I can do whatsoever I please with what I write. I choose to make it a free gift. I don't care what others do with it. That's not up to me. If it were, it wouldn't have been a free gift.
Well my flavor of ethics is not the same as yours. Please tell me: Would you sue me if I put your AL-only code and someone else's GPL-only code in the same program and distributed it (under GPL as the GPL would require)? Would the AL-only author be likely to win such a case?
I would not sue you. I give things away, and ask nothing back.
Why don't slashdotters like BSD? BSD licenses allow code that you write to come under the control of other people, and you can't stop them.
You are profoundly mistaken. The BSD licence on your code merely lets them make their own choice about their own code. Your code, however, is still free. Forever.
That's exactly what the original poster said: if it wasn't for the GPL (say, if Linux was released unde the BSD license), then Microsoft could embrace and extend with proprietary extensions.
I really hope that you don't think that's stopping them. They can make a complete O/S using closed source, even the drivers, leaving the kernel alone GPL'd. Other factors apply here.
The problem with the BSD and the Artistic for me is that I'm not interested in facilitating someone else's proprietary software without getting something back from them - I am only interested in sharing with people who give me the same rights that I give them.
In other words, by making sure that others treat you as charitably as you treat them. I hope you understand why so many of us prefer to treat others the way we want them to treat us. We produce free gifts, without requiring anything back. It's nice if it happens, but only if that return is itself freely given.
The GPL is about freedom. The whole point of it is that, once a piece of software is GPLd, that source will forever be available.
This is one of the big confusions out there. Perhaps some people are lying, but I hope that most are merely misunderstanding the matter.
The simple fact is that a free licence like the BSDL or the AL makes exactly the same guarantee, without the sting.
Of course, the licences do differ, but not with respect to what you wrote right there. The place they differ is in the non-free part of the GPL, the part where it tries to force other people to follow its own moral notions due not to free choice but because of a stiff requirement.
If MS did release "MS Linux - All Singing, All Dancing and Whiter Whites" then I think they would come under a great deal of scrutiny which would (hopefully) prevent them from subverting the GPL.
Microsoft has never considered the operating system to be the kernel alone. They could certainly take the GPL'd Linux kernel and assemble their own fully-fledged operating system, complete with closed-source device drivers, utilities, and user-targeted application programms, and call that "MS-Linux".
In fact, I predict that someday they shall, at which point we call get to buy Linus a beer. But I think he'll have one as soon as MS starts producing programs for Linux, not their own MS-Linux operating system. At least, that's what Linus once remarked.
I can't understand why you folks can't see that the GPL absolutely produces the effect described, and that the AL does not. I guess I can see why you don't like the term "poison pill", descriptive though it may be. But the effect is inarguable. It seeks to apply itself to your own work. It makes demands and restrictions on how you use the original, and what you do with your own work product. It forbids you from using GPL'd software in conjunction with commercial software like Oracle libraries. It effectively stops your from selling your own work in a meaningful way that generates revenues from each copy.
Maybe you like this effect. Maybe you don't like the term "poison pill". But by any other name, it still has these effects. Of that there is no question, for it's in plain black and white for all to read. All that's left is whether you deem this effect good or not, and what you call it.
And it certainly shouldn't come as a great surprise to you that some people prefer to instead use a free licence, charitably giving their work away to the improve the lot of programmers everywhere without passing moral judgment on those programmers' lifestyle choices. Some of us have never been happy in that kind of church, full of pushy self-righteousness and telling others what to do.
Yes, the GPL does forbid me from making a profit on the fruits of my own labors using the traditional royalty scheme. You're lying or deceived if you think otherwise.
I find it amazing that after all the anger and hostility you've shown towards GNU and the Free Software Foundation, you have summed up the GPL and blessed it with your stamp of approval as a 'true free software license'.
You misread. That's not what's being said. The licence I cited is not infective. It makes no claims about others' code. Only this original code, not the fruits of your own subsequent labors.
If you want to make your own changes, your changes are not subject to its terms. This licence guarantees that the original code remains forever free, which is usually all anybody wants. It leaves completely up to you the decision about what to do with your own code that you created, modifing the original. The original is guaranteed to remain forever free. That's what this licence says. Nothing more, nothing less.
This licence merely leaves up to you the free choice of what to do with your own work. If it did not, if it sought to impose its morality on you, it would render you incapable of making a moral decision--which is itself a fundamentally immoral act. It's distressing how many people take perverse pleasure in committing this kind of immoral act.
The referenced licence is a charitable gift. There is no poison pill. It's a free gift.
"Do unto others as you would have them do onto you" is completely different than "Coerce others into doing what you want them to do". I know which rule I prefer. It's really about morality, you see.
Oh, no--the GPL requires reinvention. The LGPL does not. Consider what happens when you want to sell a program that uses both GNU DBM and an Oracle library. You're screwed, and must reinvent.
Of course, there are ways around that, effectively converting the GPL into the LGPL as applied to libraries. Better not to use the viral version in the first place, though.
I can't think of any restriction on "freedom" in the AL that wouldn't also constitute criminal fraud. For example, making an incompatible version that no longer compiles standard Perl programs, yet continuing to pretend that this is merely the standard Perl. This is what Larry was most trying to avoid, because this very thing did in fact really happen to him once before (albeit with a different program). He caught a lot of flak because of this unpleasantry. The burnt child fears fire, and all that.
The AL certainly doesn't purport to stop you from adding your own extensions. And if you do that, it certainly doesn't tell you under what conditions you can or cannot distribute or charge for this work, nor does it say anything about whether you must provide source for your own work. (Actually, it says that it doesn't say that.:-) That would be wicked because it would mean trying to exert control over some other software besides the original; that is, stuff that whoever issued the licence didn't themself write. I don't even know whether it's legal, but it's certainly not programmer-friendly.
As I dimly understand these matters, Larry just doesn't want you to write something and then pretend that it was Larry who really wrote it. I don't blame him, and I'd be surprised if anybody did. I doubt you'd want somebody other than a legitimate owner of that name putting "written by [your name here]" all over their own software.
fraud, not about restricting anybody's freedom. I hardly see these two matters as alternate faces of the same issue, but perhaps some people do.
If you intend to make your software as useful as it can be to as many people as you can, then you should make it free software. Which is a terrible word, because of word games from the FSF. I mean free as in "gift". As in "free of restrictions" or as in "no strings attached". There are plenty of licences out there that do this. Short licences are better than long ones. The best license is "do as thou wilt".
Here's one that's been floating around:
You may use, modify, distribute, and sell this program in any way you wish, provided you do not restrict others from doing the same.
And here's another:
You are permitted and encouraged to use this code in your own programs for fun or for profit as you see fit. A simple comment in the code giving credit would be courteous but is not required.
As you see, a free licence is simple, to the point, and generous. It is not an insidious imposition of your person moral choices upon others. If you decide their choices for them a priori, they can make no moral decision. There is no goodness in being automata. You must let people choose for themselves.
Some people prefer to install poison-pills in their licences. Usually, this poison pill is about using the software to make money with. Sleepycat Software has that, the GPL has that, and so do lots of others. I suppose some selfish people have good reasons for this, but let's not be pretending that software with a poison-pill in it is somehow "free", or that it does the most good. It doesn't. A selfish poison pill tries to make sure that the original authors' socio-economic-political dogma gets spread through the world at the cost of helping fewer people. "Use" licences like this hamper code reuse and hurt programmers. A gift, on the other hand, comes without a price tag on it.
Every author has to make up their own mind here. I personally prefer software freely given away--without restrictions, without legislated morality, without poison pills, without any agenda beyond trying to help to make the world a better place. The AL seems to do a good job at that.
Try, please, to remember what the greatest gift of all is. If you know what it is and why, then you'll understand. If you do not, then I'm not sure I can convince you. But the answer is charity.
Have I ever told you how much I despise that these fricking webreaders don't give you your own fricking $EDITOR to use? Oh, right I have. Well, it sucks. And yes, I try with lynx. It dumps core.
Here's the corrected passave:
My private tragedy, which cannot, and indeed should not, be anybody's concern, is that I had to abandon my natural idiom, my untrammelled, rich, and infinitely docile Russian tongue for a second-rate brand of English, devoid of any of those apparatuses--the baffling mirror, the black velvet backdrop, the implied associations and traditions--which the native illusionist, frac-tails flying, can magically use to transcend the heritage in his own way.
Well, that was moderately amusing, although I'm not quite sure what you're trying to say. Do you propose that the same people will someday begin with Corel or Caldera or MS-Linux, then progress into Redhat or SuSE or VA-Linux, and eventually reach Debian or Slackware or Mammoth--with perhaps the truly adept finally topping out at creating their own lovingly hand-crafted version of Linux?
If that's not what you meant, kindly elucidate. If that is what you meant, then I've two serious doubts regarding your suggestion. The first doubt is your apparent assumption that the very same group of folks will necessarily advance through all of these stages--or even that they ought to do so. Why would they change from what they know? Seems to me that the only someone who wasn't happy with what they had would try something else.
The second doubt is one of focus. I remain convinced that this is the only axis about which the different flavors of Linux will cluster. But even if it were, I certainly wasn't intending to pronounce some kind of critical judgment upon this scenario. I was merely attepting to elicit opinions regarding where this inevitable clustering might eventually occur. My doubt on your content stems from being unsure whether this idiot-or-expert-friendliness really the most appropriate and likely criterion for how this will all settle out.
As for your condemnation of "big words", I find your standards of what counts as "big" to be surprisingly low. Here are all the words I used that had double-digit lengths, arranged by descending length. (Perl script available upon request:-)
15: diversification
14: considerations
13: idiot-hostile, microcomputer
12: biologically, experiencing, expert-proof, incompetence, installation, minicomputer, professional
I think you'll agree that those are, all in all, pretty simple words, words that any Junior High School student should find completely accessible. I was unaware that Slashdot had become a forum in which one's writing should be restricted to fifth-grade reading levels, or below. Should this be the case, please do me the kindless of informing me of this sad state of affairs, so that I might solicit the consider opinion and acute reasoning of the nearest fifth grader when I next wish to contribute.
Let me take issue with two likely sources of your apparent malaise.
Cambrian Explosion -- Anyone who's ever taken a class in biology, or who's been into non-purple dinosaurs is surely familiar with what happened to life in the pre-Cambrian/Cambrian transition. Perhaps residents of Kansas shall someday be deprived of such an understanding of the world, but one must not restrict oneself in a forum read round the world to writing for those poor unfortunates.
Ubi sunt? -- You hardly need to be able to extemporaneously recite Shelly's Ozymandias or Bryon's Don Juan to be familiar with the age-old Ubi sunt? cry. You don't even need to know the rest of the phrase, Ubi sunt qui ante nos fuerunt? (loosely: "Where are they now, those who have gone [actually, perfect tense of `to be'] before us?"). Based as it is on the starkest of realities, it's an ancient and universal theme found throughout any culture's literature. The haunting words themselves have become something of an iconic representation of this familiar trope.
Much as a writer, who, referring to a "sea of troubles", calls up for the reader Hamlet's famous quandary, I in both those particular cases above specifically elected turns of phrase that evoked connections back to a larger world, a richer and deeper and older world, that, while existing outside of our current discussion, remains nevertheless intimately associated with it through the interconnections of the Human Experience.
I should like, please, to cite for you two passages that seem of particular relevance here:
But why should a syndicated columnist use the word? I can hear Mr. Williams re-asking. Well, not really, just to show off--one doesn't Williams re-asking. Well, not really, just to show off--one doesn't "show off" one's workaday equipment. You see, that word, and a hundred or so others, are a part of my
working vocabulary, even as a C augmented eleventh chord with a raised ninth can be said to be an operative resource of the performing jazz pianist.
Are we now closing in on the question, by using the exlusivist word "performing"?
Yes, in a way we are, I suppose. Because just as the discriminating ear greets gladly the C augmented eleventh, when just the right harmonic moment has come for it, so the fastidious eye encounters happily the word that says exactly what the writer wished not only said but conveyed, here defined as a performing writer sensitive to cadence, variety, marksmanship, accent, nuance, and drama.
What of the reader who misses the refinement? Well, what of the listener deaf to the special reach of the C augmented eleventh? The reader has the usual choices: he can ignore the word; attempt, from the context, to divine its meaning precisely or roughly (not hard, in the narrative above, on Professor Weiss's liberal politics); or he can look it up. Are these alternatives an imposition? Yes, if the newspaper's columnist that day is giving instructions on how to treat a rattlesnake byte. You would not instruct the reader to fight the poison à outrance.
But newspapers, in particular in one-paper cities, tend to acknowledge an obligation beyond merely reporting the news. The very idea of a "feature", whether designed to advice (Ann Landers), amuse (Art Buchwald), satirize (G. R. Trudeau), or opine (the syndicated (Art Buchwald), satirize (G. R. Trudeau), or opine (the syndicated columnist), presupposes that the performer should use the full range of his relevant skills, even if the percentage of readers who turn to that feature is reduced. Surely there is a corner, in spacy papers that carry five pages on sports, for Addison and Steele? It required a Pulitzer Prize to alert some editors to the very existence of Murray Kempton, the most entertaining analytical belletrist in town, and now we read him, hungrily, in the Stamford Advocate. Readers have diverse interests, resources, skills, and appetites. The Latin Mass Committee in London petitioned for the resumption of a single mass to be said in Latin after the postconciliar ban of 1965, and was turned down--on the grounds that Latin was only "for the educated few." Evelyn Waugh said in a letter to the Times: "Surely," he wrote, "in all her charity, Mother Church can make a little room, even for the educated few"?
(William F. Buckley, on page 36 of The Right Word, 1996.)
Or, if Buckley's politics blind you to his words, then here is shorter quotation:
My private tragedy, which cannot, and indeed should not, be anybody's concern, is that I had to abandon my natural idiom, my untrammelled, rich, and infinitely docile Russian tongue for a second-rate brand of English, devoid of any of those apparatuses--the baffling mirror, the black velvet backdrop, the implied associations and traditions--which the native illusionist, frac-tails flying, can magically use to the native illusionist, frac-tails flying, can magically use to transcend the heritage in his own way.
So... let's see. If I'm running an Intel chip, I can choose between Windows NT (slow) and RedHat Linux (why? When there's so much better to be had?) with KDE installed. If I've got a SPARC-based computer, I can only use it on Solaris.
Yay! Let's hear it for "write once, run anywhere!".
As you seem to have noticed, "anywhere" in Sun-speak doesn't mean "anywhere" in regular English. It means, abstractly, "anywhere we (Sun) let you run it". Concretely, it means on a very, very short list of platforms. Java advocates I know basically say it's for Sun or Microsoft only.
So much for the much-vaunted portability story. It's just that--a story. But they've hyped it enough to trick the public into believing it. Please don't just blindly take my word for this. Believing without seeing is what got us into this. Stop listening to stories and try it yourself. Take a program written in Java, and try to get it to run on a dozen diverse platforms.
I fetched the source for w3m and couldn't even compiled. I even tried on two completely different operating systems: Redhat and OpenBSD. Different build failures, but still no go.
Well, I'm still looking forward to trying it. Maybe once that build process gets fixed up I'll try again.
Yes, it's nice to know which distributions are the most friendly to the new user, which have the most idiot-proof documentation, and so on. But it seems that a lot of the reviews focus on three things exclusively: support, idiot-proof documentation, and how easy it is to install.
Are those necessarily the orthogonal axes to guarantee viability of new species? Will any species that excel is areas outside of those survive?
Isn't ease of installation a red herring? Here's my thinking: you get a professional to install a professional system. No big deal. Don't ask you grandma to do it; it's not fair to either of you. Long-term stability and integration seems far more important, because its the quotidian use not the one-off installation that will take its toll.
Right now, we're seeing a phenomenal genetic diversification of Linux operating systems. We see lean Linuxes and porky ones. We see hybrids and half-breeds. Speciation is at the Cambrian level.
Cool.
But how long does this last? Please think back to all the different kinds of microcomputer operating systems that used to exist. Or, think of all the different kinds of minicomputer operating systems. Heck, at one time, we even had fair handful of supers, or at least, minisupers, running different operating systems.
Problem: ubi sunt?
In each category, through attrition, acquisition, or complete incompetence, we've been reduced to just a couple or so instances of each. There is every reason to believe that the virtually innumerable Linuces in the world today will, one way or the other, eventually become countable again.
So, which will these be? I'm not looking for names. I'm looking for which concepts will prove critical for clustering. Each cluster will, I believe, turn into one or at most two individual versions of Linux, the others having gone the way of all things.
You named a few criteria. Those are certainly important considerations for one cluster. What other loci will form clusters? What will drive them? I easily can see a slackware or debian style OS that caters to programmers forming one cluster, a Corel or Caldera another (there will eventually be an MS-Linux there, tool; mark my words) for the current turnkey consumer crowd, and possibly a Redhat or SuSE another for somewhere in between.
Will the idiot-proof MS-style Linuces prove expert-proof as well? Will the SlackDeb be idiot-hostile and expert friendly?
What are the affinities? How will the clusters clump?
Right now, we're experiencing a speciation explosion in Linux OSes that will someday become known as our "golden age", much as we've seen occur biologically as well as vis-a-vis other operating systems groups. Where are we going? When will the die back happen? Which OSes will we be left with?
ObJoke: The only the thing you can expect enterprise users to have is a communicator and a phaser.:-)
I'm getting increasingly confused by this whole enterprise thing. I really thought I started to know what it meant. Now I'm further behind than I was. My working definition was "enterprise=big".
I'm starting to think it has more to do with computing systems managed by business majors or something. Or data processing stuff, punch cards for the new millennium. It's a big word. I thought it meant big, or dumb, or turn-key, non-technical, or expensive, or something. Now I've no idea.
Dang it. What did people call this enterprise thingie ten years ago? What's "for the enterprise"? What's "enterprise class"? What's "enterprising businesses"?
I'm not sure what to say. You're half 100% right, and you're half 100% wrong.
What you're right about is that it's highly Good to have programs that let you get a lot of mileage out of only a simple introduction. Games like perl, vi, and nethack are follow this principle of yours. Yes, trial and error is the only way to learn. You have to play the game to develop any skill to it.
However, that's not all there is to them, and you'll never get to get the most you can out of these games using the million-monkey approach. You need some orientation, some documentation, some background and theory of operation. And because of this, they're far more powerful.
What you're wrong about is the expectation that no introduction is necessary. Think about the "just guess" generation. They see a function call like socketpair(Child_Side, Parent_Side, AF_UNIX, SOCK_STREAM, PF_UNSPEC) and they start typing random things trying to guess what happens when you diddle those arguments. It boggles the mind. Likewise when they're setting up networks using route and ifconfig. They just punch random things. It's damned scary. I think it's the post literate society where there's no book learning left.
The "just guess" generation is so brain-damaged by Microsoft's permanently-dumb-user mindset that they are severely handicapped when it comes to power tools. When a child wants to play with a power tool like an F-15, you need to take him by the hand a bit until he's ready for it. Otherwise, he's a hazard to himself and others.
Slashdot Mirror
What you write is yours. What I write is mine. And I can do whatsoever I please with what I write. I choose to make it a free gift. I don't care what others do with it. That's not up to me. If it were, it wouldn't have been a free gift.
You can't steal something that's free.
Funny how many people get this wrong.
Good luck getting permission from Oracle. Do you really not understand the problem?
The simple fact is that a free licence like the BSDL or the AL makes exactly the same guarantee, without the sting.
Of course, the licences do differ, but not with respect to what you wrote right there. The place they differ is in the non-free part of the GPL, the part where it tries to force other people to follow its own moral notions due not to free choice but because of a stiff requirement.
In fact, I predict that someday they shall, at which point we call get to buy Linus a beer. But I think he'll have one as soon as MS starts producing programs for Linux, not their own MS-Linux operating system. At least, that's what Linus once remarked.
Are we there yet?
Maybe you like this effect. Maybe you don't like the term "poison pill". But by any other name, it still has these effects. Of that there is no question, for it's in plain black and white for all to read. All that's left is whether you deem this effect good or not, and what you call it.
And it certainly shouldn't come as a great surprise to you that some people prefer to instead use a free licence, charitably giving their work away to the improve the lot of programmers everywhere without passing moral judgment on those programmers' lifestyle choices. Some of us have never been happy in that kind of church, full of pushy self-righteousness and telling others what to do.
Yes, the GPL does forbid me from making a profit on the fruits of my own labors using the traditional royalty scheme. You're lying or deceived if you think otherwise.
If you want to make your own changes, your changes are not subject to its terms. This licence guarantees that the original code remains forever free, which is usually all anybody wants. It leaves completely up to you the decision about what to do with your own code that you created, modifing the original. The original is guaranteed to remain forever free. That's what this licence says. Nothing more, nothing less.
This licence merely leaves up to you the free choice of what to do with your own work. If it did not, if it sought to impose its morality on you, it would render you incapable of making a moral decision--which is itself a fundamentally immoral act. It's distressing how many people take perverse pleasure in committing this kind of immoral act.
The referenced licence is a charitable gift. There is no poison pill. It's a free gift.
"Do unto others as you would have them do onto you" is completely different than "Coerce others into doing what you want them to do". I know which rule I prefer. It's really about morality, you see.
Of course, there are ways around that, effectively converting the GPL into the LGPL as applied to libraries. Better not to use the viral version in the first place, though.
The AL certainly doesn't purport to stop you from adding your own extensions. And if you do that, it certainly doesn't tell you under what conditions you can or cannot distribute or charge for this work, nor does it say anything about whether you must provide source for your own work. (Actually, it says that it doesn't say that. :-) That would be wicked because it would mean trying to exert control over some other software besides the original; that is, stuff that whoever issued the licence didn't themself write. I don't even know whether it's legal, but it's certainly not programmer-friendly.
As I dimly understand these matters, Larry just doesn't want you to write something and then pretend that it was Larry who really wrote it. I don't blame him, and I'd be surprised if anybody did. I doubt you'd want somebody other than a legitimate owner of that name putting "written by [your name here]" all over their own software.
fraud, not about restricting anybody's freedom. I hardly see these two matters as alternate faces of the same issue, but perhaps some people do.
If you intend to make your software as useful as it can be to as many people as you can, then you should make it free software. Which is a terrible word, because of word games from the FSF. I mean free as in "gift". As in "free of restrictions" or as in "no strings attached". There are plenty of licences out there that do this. Short licences are better than long ones. The best license is "do as thou wilt".
Here's one that's been floating around:
And here's another:As you see, a free licence is simple, to the point, and generous. It is not an insidious imposition of your person moral choices upon others. If you decide their choices for them a priori, they can make no moral decision. There is no goodness in being automata. You must let people choose for themselves.
Some people prefer to install poison-pills in their licences. Usually, this poison pill is about using the software to make money with. Sleepycat Software has that, the GPL has that, and so do lots of others. I suppose some selfish people have good reasons for this, but let's not be pretending that software with a poison-pill in it is somehow "free", or that it does the most good. It doesn't. A selfish poison pill tries to make sure that the original authors' socio-economic-political dogma gets spread through the world at the cost of helping fewer people. "Use" licences like this hamper code reuse and hurt programmers. A gift, on the other hand, comes without a price tag on it.
Every author has to make up their own mind here. I personally prefer software freely given away--without restrictions, without legislated morality, without poison pills, without any agenda beyond trying to help to make the world a better place. The AL seems to do a good job at that.
Try, please, to remember what the greatest gift of all is. If you know what it is and why, then you'll understand. If you do not, then I'm not sure I can convince you. But the answer is charity.
Here's the corrected passave:
If that's not what you meant, kindly elucidate. If that is what you meant, then I've two serious doubts regarding your suggestion. The first doubt is your apparent assumption that the very same group of folks will necessarily advance through all of these stages--or even that they ought to do so. Why would they change from what they know? Seems to me that the only someone who wasn't happy with what they had would try something else.
The second doubt is one of focus. I remain convinced that this is the only axis about which the different flavors of Linux will cluster. But even if it were, I certainly wasn't intending to pronounce some kind of critical judgment upon this scenario. I was merely attepting to elicit opinions regarding where this inevitable clustering might eventually occur. My doubt on your content stems from being unsure whether this idiot-or-expert-friendliness really the most appropriate and likely criterion for how this will all settle out.
As for your condemnation of "big words", I find your standards of what counts as "big" to be surprisingly low. Here are all the words I used that had double-digit lengths, arranged by descending length. (Perl script available upon request :-)
I think you'll agree that those are, all in all, pretty simple words, words that any Junior High School student should find completely accessible. I was unaware that Slashdot had become a forum in which one's writing should be restricted to fifth-grade reading levels, or below. Should this be the case, please do me the kindless of informing me of this sad state of affairs, so that I might solicit the consider opinion and acute reasoning of the nearest fifth grader when I next wish to contribute.
Let me take issue with two likely sources of your apparent malaise.
Much as a writer, who, referring to a "sea of troubles", calls up for the reader Hamlet's famous quandary, I in both those particular cases above specifically elected turns of phrase that evoked connections back to a larger world, a richer and deeper and older world, that, while existing outside of our current discussion, remains nevertheless intimately associated with it through the interconnections of the Human Experience.
I should like, please, to cite for you two passages that seem of particular relevance here:
Or, if Buckley's politics blind you to his words, then here is shorter quotation:So much for the much-vaunted portability story. It's just that--a story. But they've hyped it enough to trick the public into believing it. Please don't just blindly take my word for this. Believing without seeing is what got us into this. Stop listening to stories and try it yourself. Take a program written in Java, and try to get it to run on a dozen diverse platforms.
So much for Java being "portable", eh?
I'm not quite sure, but I think I'm impressed. :-)
Well, I'm still looking forward to trying it. Maybe once that build process gets fixed up I'll try again.
And I must look into this w3m thing.
thanks,
Isn't ease of installation a red herring? Here's my thinking: you get a professional to install a professional system. No big deal. Don't ask you grandma to do it; it's not fair to either of you. Long-term stability and integration seems far more important, because its the quotidian use not the one-off installation that will take its toll.
Right now, we're seeing a phenomenal genetic diversification of Linux operating systems. We see lean Linuxes and porky ones. We see hybrids and half-breeds. Speciation is at the Cambrian level.
Cool.
But how long does this last? Please think back to all the different kinds of microcomputer operating systems that used to exist. Or, think of all the different kinds of minicomputer operating systems. Heck, at one time, we even had fair handful of supers, or at least, minisupers, running different operating systems.
Problem: ubi sunt?
In each category, through attrition, acquisition, or complete incompetence, we've been reduced to just a couple or so instances of each. There is every reason to believe that the virtually innumerable Linuces in the world today will, one way or the other, eventually become countable again.
So, which will these be? I'm not looking for names. I'm looking for which concepts will prove critical for clustering. Each cluster will, I believe, turn into one or at most two individual versions of Linux, the others having gone the way of all things.
You named a few criteria. Those are certainly important considerations for one cluster. What other loci will form clusters? What will drive them? I easily can see a slackware or debian style OS that caters to programmers forming one cluster, a Corel or Caldera another (there will eventually be an MS-Linux there, tool; mark my words) for the current turnkey consumer crowd, and possibly a Redhat or SuSE another for somewhere in between.
Will the idiot-proof MS-style Linuces prove expert-proof as well? Will the SlackDeb be idiot-hostile and expert friendly?
What are the affinities? How will the clusters clump?
Right now, we're experiencing a speciation explosion in Linux OSes that will someday become known as our "golden age", much as we've seen occur biologically as well as vis-a-vis other operating systems groups. Where are we going? When will the die back happen? Which OSes will we be left with?
Something to think about.
I'm getting increasingly confused by this whole enterprise thing. I really thought I started to know what it meant. Now I'm further behind than I was. My working definition was "enterprise=big".
I'm starting to think it has more to do with computing systems managed by business majors or something. Or data processing stuff, punch cards for the new millennium. It's a big word. I thought it meant big, or dumb, or turn-key, non-technical, or expensive, or something. Now I've no idea.
Dang it. What did people call this enterprise thingie ten years ago? What's "for the enterprise"? What's "enterprise class"? What's "enterprising businesses"?
What you're right about is that it's highly Good to have programs that let you get a lot of mileage out of only a simple introduction. Games like perl, vi, and nethack are follow this principle of yours. Yes, trial and error is the only way to learn. You have to play the game to develop any skill to it.
However, that's not all there is to them, and you'll never get to get the most you can out of these games using the million-monkey approach. You need some orientation, some documentation, some background and theory of operation. And because of this, they're far more powerful.
What you're wrong about is the expectation that no introduction is necessary. Think about the "just guess" generation. They see a function call like socketpair(Child_Side, Parent_Side, AF_UNIX, SOCK_STREAM, PF_UNSPEC) and they start typing random things trying to guess what happens when you diddle those arguments. It boggles the mind. Likewise when they're setting up networks using route and ifconfig. They just punch random things. It's damned scary. I think it's the post literate society where there's no book learning left.
The "just guess" generation is so brain-damaged by Microsoft's permanently-dumb-user mindset that they are severely handicapped when it comes to power tools. When a child wants to play with a power tool like an F-15, you need to take him by the hand a bit until he's ready for it. Otherwise, he's a hazard to himself and others.