>Anyway, this all seems like it's a problem that >application developers simply don't use the tools >provided for them to make their programs >platform-independent. How would changing the >platforms themselves fix the problem---at least, >the one you were having?
The only solution I've seen for this problem that seems to work is FreeBSD's ports, or something like it...in the sense that more popular applications (Joerg Schilling's cdrtools, as a wonderful example) then have a "Makefile maintainer." It's the Makefile maintainer's job to find whatever patches are needed/available for the given application, and to write a Makefile which caters to the individual application's installation quirks. Thus, only one person has to endure the pain and suffering involved in figuring out how to get cdrtools to build. Everyone else thereafter can use the Makefile that was written by that original brave soul who figured it out. Also, even though individual Makefiles might themselves be very different, in order to cope with different non-standard install methods, the end user's installation experience is completely uniform...in most cases all they have to do is type "make install" to compile/install *anything.*
This is why I believe personally that if *any* package management/installation method was going to be included in the LSB, it should be something based on ports...simply because ports is the only solution I've yet come across which works in a complete sense. Virtually every other proposed system I've seen has problems. My own proposed solution is a combination of this, which is something I wrote myself a bit back, and rpm. The above linked project only has files in CVS right now though unfortunately, so you'll have to use that if you want them.
What I plan to do though is incorporate the use of bmake (NetBSD's make) and a partial clone of ports as a replacement for much of the body of rpm's.spec files, as well as doing away with rpm's dreaded subpackaging entirely. What this would mean is that I'd get rpm's packaging and dep tracking capabilities, which are in themselves quite good as long as you manually specify dependencies, but bmake/acbuild would also mean I'd finally have a sane specfile format. Of course, a fair amount of kludging would have to be done...the primary thing would be to convert rpm's custom vars into vars that bmake can make sense of. My system however would also at least partially do away with the need for programs like apt/urpmi as well, because of acbuild's fetch target. Yet another good thing about this though is that it wouldn't break existing compatibility with anything, in the sense that rpm will of course still be able to use specfiles which I won't have converted for this...I'd make my own specfiles available, but only people who wanted to use acbuild would need to know about them.
>I'm curious about what sorts of problems you ran >into. I was under the impression that configure >scripts were designed to at least make source >code portable, by dynamically finding the various >libraries and such that the application requires.
Sure...they are. However, you're making two assumptions here which may not (unfortunately) be the case...they are in many instances, but not in all.
a) That said science apps even use configure/autotools at all, and b) That said science apps have *well written* or *coherent* configure scripts, which look for everything they need, and do so in a sane manner. You wouldn't believe the sort of garbage I've seen in configure scripts before...made me wonder what on earth the developer in question had been smoking.
There are lots of apps out there which don't use a sane method of installation/compilation, and which should. DocBook is one example which comes to mind. True, in that scenario nothing needs to be compiled...but I find myself wondering if a basic makefile or shell script for installing the files into something close to the standard location would be too much to ask. Even assuming the default offered in said script was the wrong location for your system, it'd at least give you something to work with.
Then as another example you've got people like Joerg "I'm too cool for the rest of the population" Schilling, who find it necessary to completely reinvent the wheel, and who don't necessarily do so in a sane or comprehensible manner. Thus, if you wish to install any of their applications, you actually have to download their re-interpretation (regurgitation would probably be a better word) of Make or whatever else...it basically requires you to step into their own demented private universe.;)
So yeah...Unfortunately, as far as software compilation/installation is concerned, we can invent as many different solutions and standards as we want. Some people will still feel for whatever reason that whatever rules we create don't apply to them...and won't understand that said rules generally exist in order to make *their* lives easier, as well as the lives of everyone else.
I'm assuming that the fact that RPM is even part of the specification is due to Red Hat's influence, nevertheless it most certainly should not be. For one thing, RPM's.spec format is beyond horrible in all sorts of different ways. I advocate crucifiction for whoever came up with the idea of sub-packaging from within a spec file in particular...this "feature" alone tends to make specfiles for glibc in particular an incomprehensible nightmare...and I can also virtually guarantee that this particular feature almost exclusively is responsible for every mangled RPM-based system anyone has ever read about...it is also responsible for the inability to compile single non-RPM based applications on an otherwise RPM based system. In short, sub packaging was an almost indescribably bad (not to mention completely unnecessary) idea, and IMHO should be removed from RPM.
In addition, the standard macros are also completely opaque...you've got no way of knowing what on earth %setup -q T b does unless you've looked it up.
b) Economic theory
Economic theory says that in order to sell something, it is ideally desirable to have what is called a Unique Selling Point...i.e., to have something unique which nobody else has, which is the reason why someone is going to buy from you, and not the zillions of other companies in the world. Thus, being non-standard is actually the *only* means a Linux vendor company has of creating an incentive for a consumer to purchase from that particular company, rather than any other. Standards (at least beyond a certain degree) would end up meaning that every vendor sold a completely identical, generic product...and therefore, no consumer would have any incentive to buy from any one vendor in preference to any other. *Customers* might want that, and vendors pay lip service to it because they know customers want it...but the vendors themselves in reality do not want it, because it would be very bad for their business...so beyond a certain point, complete standardisation ain't going to happen.
c) Standards aren't *always* a good thing. Standards are primarily a good idea when talking about communication. The Web is a good example. Because the Web itself relies on standard protocols/specifications, (HTML, HTTP and so on) the idea is that individual computers, sometimes with radically different operating systems *internally*, can reliably communicate with each other *externally* on the Internet. But for each one of those computers to always be identical *internally* can be an extremely bad thing, security wise...Microsoft's woes in terms of virii, worms, and trojan horses provide all the evidence needed to prove this point.
d) An operating system/standards monoculture is only really desired by people who are seeking an excuse to be able to avoid using their brains. Such people want things to be so completely simplistic that they can perform any task on rote, and most especially, so that they can perform any task without the truly horrifying prospect of needing to actually *think.* This is not to suggest that I am completely opposed to anything being user-friendly...quite the opposite. But I *am* aware that there is a very large segment of the computer using population who are willing to put Eric Raymond's "luxury of ignorance" ahead of virtually every other aspect of software design in terms of importance, including security. "Give us one single operating system, and do not allow the complexity of any aspect of it to be greater than a retarded five year old's ability to deduce!" Such is the rallying cry among these souls, and I hear it on an almost daily basis, albeit normally somewhat veiled.
I've kept a vague eye on this game for a while, on and off. It's main problem is that, for my money anyway, the game it's based on wasn't all that great...and FreeCiv for the most part also doesn't have most of the things that the original/s had which were capable of keeping people's attention. (Lots of in-game cinematics, AI using diplomacy until very recently, etc etc) Although FOSS is strong on code, multimedia seems to be a major problem for FOSS developers...Presumably because at least some of them are used to playing games like Nethack. (Graphics? What on earth do you need those for?)
The main game I really want to see a FOSS clone of is the Sims, in all honesty. The reason why is that with the Sims 2, EA basically killed the main reason why the original game was so popular...i.e., the degree to which it was editable with new wallpapers, floors, objects etc...and of course, the possibility of modifying something is one of FOSS's main strengths.
It would need to be given a completely different name though, and probably kept underground for a long while...EA in particular are extremely paranoid about the proactive amateur phenomenon...Their TOS for The Sims Online specifically outlaws the construction of server emulators by anyone who plays the game.
I realise that a Sims clone would also be very difficult from a coding point of view, so it perhaps isn't possible. I can dream, though.
>According to statements by both Stallman and >Perens, they consider programmers are "evil" unless >they give away their work.
This is one misconception that I am growing desperately tired of...however like some kind of many-tentacled octopus, no matter how many times I try and step on it, it just keeps rising from the grave. There also admittedly aren't that many times when I'll come to the defense of RMS...but this *is* one of them. Go and read this. In it, RMS states that he doesn't have a problem with people selling GPL bound software at all.
The free/beer and free/libre debates and the associated misconception were ESR's entire motivation for adopting the term Open Source, so far as I know. Basically what Stallman says at the above link is that the ONLY demand that the GPL makes is that source be provided with binaries. End of story. To quote from his article:-
"Since free software is not a matter of price, a low price isn't more free, or closer to free. So if you are redistributing copies of free software, you might as well charge a substantial fee and make some money. Redistributing free software is a good and legitimate activity; if you do it, you might as well make a profit from it." (Emphasis his)
I've started realising that it isn't actually so much RMS' philosophies themselves which bother me...it's the misconceptions that a lot of other ill-informed morons who claim to be his supporters develop and spread. RMS doesn't have anything against people making money from GPL-bound software AT ALL...He did it himself at one point with Emacs. He isn't Communist, and the GPL itself isn't, either. I repeat, all the GPL demands of people is that if you're going to distribute binaries in whatever manner, you also distribute accompanying source, so that said software has a chance to propogate itself. As long as both are sold together in the one package, you can charge as much money for said package as you like.
Just another pundit trying to secure his livelihood by promoting the status quo. I've commented on that particular syndrome before. The way that I know that that is what he is doing is because he says he can't refute Linux's technical superiority...but then goes on to claim that Windows is better overall, which they all do. By initially making the technical concession first, and then making their second claim, they try to have their cake and eat it...appear to have journalistic integrity while still maintaining the status quo.
I agree with one of the earlier comments that the editors need to start screening articles of this nature more carefully. They don't tell us anything we don't know, or that hasn't been said before.
The author of this article is right in one respect, though...Strictly speaking, Linux won't kill Windows...Microsoft are going to be responsible for their own demise, to a large degree.
Rather than "Starter Edition," here's some suggestions, if anyone from Redmond just happens to read this. (I know they won't do it - it's more a mental exercise while I eat)
1. Go download this, and make it natively multi-user if it isn't already. Give it a strong native security model, too...you can get some ideas here, and the best part is, they won't mind you doing that if you don't try and patent said ideas. Also, modularise your GUI, and don't prevent users from accessing the CLI when they want to.
2. Have the CLI composed of this and this for us CLI types.
3. Make the Add/Remove Programs panel essentially a net-aware frontend for either this or this.
4. Use this for hardware detection. Also re drivers, get rid of the suicidal policy of seeing third-party hardware vendors as the enemy, and actually support them...via tools, docs, etc. These people are your friends...they'll help you stay relevant.
5. Download this and use it as your default FS, and then get this and this, (although you already seem to know about this last one) and incorporate both of those into your stock UI. You've essentially got WinFS right there, without all the added complexity you'd no doubt throw into it if you tried to code it from scratch.
6. For the Agent angle, incorporate the last point, as well as putting help/docs in a non-binary format, making them searchable with this, converting said search results for use with this, and then use the AIML output as input for something like this. Also, instead of making the agent a tightly anthropomorphic personality, make it more generic, and more as though it's simply "the operating system" communicating with a user, rather than that dog or Clippit instead.
7. Give Outlook a major overhaul. This and this are examples of directions it IMHO should go in.
Just some random ideas, anywayz. Dreaming's fun.;) I'll probably get modded Offtopic, but it was worth it.
Attempting to strangle ADSL adoption, killing the national BBS community when the Internet first became mainstream in Australia in order to force adoption of Big Pond, and a host of other offenses meant that after an extended period of shopping around, I finally stopped using Telstra as a carrier completely last year, and they can now consider themselves permanently boycotted as far as I'm concerned. They are one of the most short-sighted, destructive, and generally amoral corporations I've heard of. They were also vocally criticised by Bill Gates during one of his visits here, for their strangulation of broadband adoption.
Apart from the above, to some degree there are now price incentives to use other carriers as well, particularly for voice. If you've got a credit card, you also might want to check out TPG for ADSL...they probably have the best deals I've seen.
There's Gaim. That will let them use AIM, but it will also mean that if their friends use MSN or ICQ instead, they can talk to all of them from within the same program.
>No Kazaa.
Heard of mldonkey? eDonkey, Kazaa, Gnutella, BitTorrent, Napster, and Direct Connect, all in the one app. Got more clients than you can shake a stick at, too.
I've got a Netcomm 56k internal based on the Lucent Mars chipset. Contrary to the amount of flak I've heard levelled at the hardware, the drivers from here work solidly in my experience, with both 2.4 and 2.6 series kernels.
However, not all winmodems are created equal. I'm not entirely sure about this, but I believe it depends on whether or not the winmodem in question actually DOES have a controller chip and just needs proprietary drivers, as opposed to a true controllerless modem. From what I've read, the Lucent AMR modems are genuinely controllerless and thus not supported under Linux.
In a nutshell, because the term "winmodem" is a catchment word and actually describes quite a large number of different devices, you need to make sure you have reasonably intimate knowledge of which specific breed of winmodem you've got. Some will work, some won't...but again, in my experience anyway with my own chipset, both the 2.4 and 2.6 Lucent modules work well.
Of course, a standard hardware modem is always more desirable if you can get one...particularly seeing as the Lucent modules taint the kernel, which may be a problem for some people. (it doesn't particularly bother me) The advantages of winmodems however are price, greater level of availability these days from what I've seen, and marginally better throughput than their standard cousins in some instances. I'm hoping to eventually save up for a standard/external one of these days, and as I said they are more desirable if you can find/afford it...but I'm at least surviving on my Lucent right now.
Do you really think it does Linux's image any good to have idiots burying her with flames...and *especially* phoning her house in the wee hours and harassing her? (Assuming that is true, of course) That just adds grist to her mill about what a pack of wild eyed, unwashed fanatical hippies we supposedly are.
People here keep talking about how she supposedly has no ethics and her opinion is completely for sale, etc. What some of the dimmer bulbs among you are perhaps missing is that that could actually be useful to us...because assuming she really *is* that cynical, if we tried applying some diplomacy, the opinion she eventually ends up selling could be ours.
Microsoft have traditionally used people like Laura to communicate with computer laypeople, and in the past at least, it was very effective for them. If Laura's values really are for sale, we ought to try and recruit her ourselves...because every individual like her that we can convert, is one less such individual who's still relaying Microsoft's party line to the masses, and can actually be relaying a pro-Linux message to them instead.
Yeah, I know I sound like a disgusting, slimey American spindoctor here most likely...but the problem is, this shit *works*. That is the reason why Microsoft got where they did, and why despite the popularity of the GPL/GNU's *software*, RMS is currently battling impending irrelevance.
As far as influencing people's thought is concerned, Stallman's methods don't work. It's as simple as that. This has been proven, and it is being proven...Over and over and over again. RMS has been tried, and found sorely wanting. Given that, if we want Linux to continue to gain traction among the greater portion of the population, we need to abandon Stallman's ideas and attitude once and for all, and focus on methods which *do* work instead.
And please, any of his advocates are welcome to come forward and refute me on this, because it also doesn't mean that I'm in opposition to the entire paradigm of FOSS in general. Those of you who give Stallman sole credit for said concept's creation are chronically misguided.
Even apart from that though, I have one simple comeback for any of Stallman's advocates which I'm guessing they won't have an answer for...Namely, that they are an endangered species. RMS's amen corner are a tiny minority, and getting smaller by the day, from everything I've seen. The pat answer that Stallman himself gives to this, that the majority aren't interested in being right, also doesn't wash. Any idea can be effectively communicated if said communicator knows what they're doing.
The reason why I keep bringing this up is because I really believe that Stallman is an anvil around Linux's neck. Stallman can be held indirectly responsible for virtually all of the zealotry and pedantry that Linux users are at times accused of...because he himself is a zealot and a pedant, and his followers adopt his example. He represents a major diplomatic problem that Linux truly does not need, and the fact that other people emulate his fringe-dwelling behaviour means that the problem only gets amplified.
We get rid of Stallman, and trolls like Laura DiDio lose all their ammunition, as far as calling us extremists is concerned, more or less instantly...Because Stallman *is* both an extremist and an advocate of extremism. It's that simple.
as though I've just been exonerated. I've been saying for a while why I have issues with RMS, and if this license goes ahead as depicted here, there will finally be irrefutable evidence of what I've been saying...Namely, that the man is a megalomaniac whose intentions are not anywhere near as glorious as he has made out.
Stallman needs to be careful. He exists in a community whose tendency is, once it has identified problems, to simply route around them. If the mask truly comes off and he exposes his real agenda of wanting control over people, rather than being the advocate of libertarianism that he has claimed, he will run the risk of being rendered completely irrelevant. Although I'm actually hoping he continues to be imprudent, because I want people to see the truth about this man. Once enough people have, then we can all move on to other licenses which more genuinely serve our purpose.
These laws are almost entirely unenforceable. As an ISP, if any government wanted to force me to monitor individual users' bandwidth, I'd ask them if they were going to provide me with extra staff to help me do it, since there would be no possible way that I could do such a thing myself. Let's also see if they can persuade their usually understaffed, underpaid police forces to do the job, as well...My guess is that that is unlikely.
Any government that wants to can pass as many laws like this as it wants, and then sit back and watch as the general public completely ignores them. Governments, the RIAA, and WIPO need to get it into their empty heads once and for all:- We *want* to pirate music, we're *going* to pirate music, and apart from a few token lawsuits here and there against the odd big fish, for the most part there is exactly jack shit you can do about it. Get used to it, because it (and we) are not going away.
>You know what would be a good idea. A bunch of >geeks getting together with a bunch of researchers >in their respective fields. Creating honest, >non-biased "this is the way it is" anaylsis and >reports on TCO/Software/Hardware/etc. Sort of like >Consumer Reports(tm) but with more detail and >analysis of specific topics.
The problem is, there have been analyses done that have at least try to make it sound like that. The worst ones of all are the whining, seemingly apologetic ones written by trolls like Eugenia from OSNews. They start with statements like, "I really love Linux, and want to see it succeed in the marketplace, *but*..." and then proceed to launch into a litany of ignorant misconceptions and FUD.
The main problem is that these journalists who keep doing Microsoft's PR work for them, are I think people who unfortunately believe that their livelihood is tied to the maintenance of Microsoft's dominance. They think that if Microsoft were to collapse, the "ecosystem" around Microsoft (as MS themselves call it) would collapse along with it.
We need to somehow reassure ZDNet and their spiritual kin that if Linux was to gain more market share than Windows, it wouldn't automatically mean that they would be out of a job. Linux has heaps of events, activity, and issues of various kinds that media people can report on...ZDNet's peeps only need to look at Newsforge or LWN to find that out.
I guess my point here is that the trade press do not need to feel that they have to keep trashing Linux in order to protect themselves and their employment. ZDNet could quite easily go from being Microsoft's unofficial Ministry for Propaganda to being wholehearted Linux advocates if they wanted to...and aside from still making a very satisfactory income, they'd sleep better at night knowing that they were engaging in a far greater level of journalistic integrity than they have in the past, as well. It *is* very possible, guys. Just ask O'Reilly.
IANAL, so...now that we have all this wonderfully detailed information about spammers, how do we use it to have them locked up for 20 years or so with their very own girlfriend named Spike?;-)
I don't believe Debian's dying, but I'd almost be glad if it was...At least then I wouldn't have to put up with the egocentric attitude of its' users. I will admit that Debian users have actually become probably my primary pet hate as far as this site is concerned. I think the only group I find more smug, obnoxious, and generally annoying are OSX fanboys.;)
I'm also assuming that if it was dying, most of the Slashbots who use it would also make the assumption that Linux in general was, since we all know that Debian is the only distribution in existence.;)
Nothing random about it. I've only made two out of 10+ submissions that have been accepted. I submitted info about the Microsoft "Get the Facts" campaign in particular before ESR had updated the Halloween Docs about it, and it got rejected...and then of course when someone submitted ESR's article on it, that went straight through.
The thing that I primarily find ironic about this is the fact that the alleged degeneracy of various different elements of popular culture (video games are only one item on a very long list) are almost exclusively protested by those individuals who together, comprise the single most criminally insane, morally degenerate, and maniacally dangerous group of people currently in existence on this planet; namely, the American Right.
Rightwing politicians and their advocates endorse the slaughter in Iraq, the errosion of education, the erradication of science, apocalyptic foreign policy, the worst forms of cronyism and corruption, both corporate and political, and fear, hatred and xenophobia of every kind, and more dangerous than anything else, the proliferation and use of nuclear weapons. The neoconservative movement ranks in my mind as one of the most dire threats that humanity as a whole have ever faced. For any of their number to express concern about the morality of virtually anything in comparison with their own crimes goes beyond hypocrisy...There is no word in existence to describe it.
I think it's vital that we somehow find a way to prevent this individual and anybody else who thinks like him from getting what they want. Contrary to what he might say, the system we have right now *does* work, and has worked well. The only reason why he is trying to make people believe that it hasn't worked is so that he can institute his own system.
The UN wants to have jurisdiction over *everything.* Nothing makes Kofi Annan start foaming at the mouth more than when he hears the phrase "national sovereignty." In practice however, the only thing the UN really amounts to is the Third World's vehicle for trying to fulfill Pinky and the Brain's primary objective. Everyone wants to take over the world...no huge conspiracy theory there. The US govt, the Israeli govt, the EU govts; everyone. The thing we have to somehow try and do though is make sure that none of them manage it...and the UN succeeding in doing it wouldn't be any more desirable than anyone else managing to.
The UN, contrary to what some highly emotive left-wing types will have you believe, (and before you make the assumption, no, I'm not a fascist...I've been called left myself) is most definitely NOT "our last, best hope for peace." This is not Star Trek, and the UN are NOT the Federation. This is the real world, and the only thing the UN are our "last, best hope" for is a centralised global government...something which would ultimately cause a greater level of tyranny and suffering possibly than we've ever seen before. (And yes, I know we saw a lot in WW2. That's part of my point.)
If I currently had modpoints, I'd devote all five of them to modding this post Overrated, and thus blast this juvenile, unfunny attempt at humour back to AC-level oblivion where it belongs.
What sort of mindless sheep are you? Heaven forbid the idea that you either a) Make a.deb file yourself, or b) Compile/install it outside of dpkg (horrors!)
Forgive me, though. I realise that following either of those suggestions would require using your own brain and your own initiative, which going by your sig as a George W Bush supporter, probably *is* a little too much to ask.
>The main work and value of distributions like >Debian is in the extensive testing that goes into >making sure that all the packages work together.
One thing...not everyone uses the same distribution. And before you say it, no, we shouldn't. The lack of a monoculture is one of Linux's primary defenses against viruses...To newbs, lack of a single universal interface seems like a weakness...but apart from strong multi-user, one of the other things that impedes virus writing is the lack of a consistent environment...because it means a virus writer cannot make a single set of assumptions about how a virus will need to behave. If a virus was written to attack something written for KDE, for example, the fact that there are other desktops means that a Gnome user may not be affected by it. People keep calling for "unity" all the time, when what they don't realise is that the universal interface is actually one of Microsoft's biggest problems...it's not a strength.
Autopackage is something that would be distro independent, and as such could be extremely beneficial...people could still ship distro-specific patches with an autopackage quite easily if they wanted to. The added benefit though would also be that someone who either wasn't using Debian/RH or who was actually using something distro-neutral like LFS could still install it.
Linux has always been about choice. If someone wants to use Debian, great...but there are a lot of other distributions out there which do different things. Autopackage could give all of them the means of installing packages easily, while still retaining their uniqueness. That's a good and important thing.
Although some may question the advisability of advocating a controlled substance, I believe that a short course of marijuana is something you may find extremely therapeutic. I myself used to be as pedantic as this, but after smoking weed intermittently for around 18 months, (late 2001-2003) was no longer so, and have not been to this day.
Although it may be argued that my becoming less pedantic was actually due to the marijuana causing neurological damage, if it was damage it was in such a way that I was still profoundly grateful for. It made me less likely to care about being correct in instances where being correct truly did not matter, and where the only purpose being correct served was to alienate and annoy my fellow man.
Slashdot Mirror
>Anyway, this all seems like it's a problem that
.spec files, as well as doing away with rpm's dreaded subpackaging entirely. What this would mean is that I'd get rpm's packaging and dep tracking capabilities, which are in themselves quite good as long as you manually specify dependencies, but bmake/acbuild would also mean I'd finally have a sane specfile format. Of course, a fair amount of kludging would have to be done...the primary thing would be to convert rpm's custom vars into vars that bmake can make sense of. My system however would also at least partially do away with the need for programs like apt/urpmi as well, because of acbuild's fetch target. Yet another good thing about this though is that it wouldn't break existing compatibility with anything, in the sense that rpm will of course still be able to use specfiles which I won't have converted for this...I'd make my own specfiles available, but only people who wanted to use acbuild would need to know about them.
>application developers simply don't use the tools
>provided for them to make their programs
>platform-independent. How would changing the
>platforms themselves fix the problem---at least,
>the one you were having?
The only solution I've seen for this problem that seems to work is FreeBSD's ports, or something like it...in the sense that more popular applications (Joerg Schilling's cdrtools, as a wonderful example) then have a "Makefile maintainer." It's the Makefile maintainer's job to find whatever patches are needed/available for the given application, and to write a Makefile which caters to the individual application's installation quirks. Thus, only one person has to endure the pain and suffering involved in figuring out how to get cdrtools to build. Everyone else thereafter can use the Makefile that was written by that original brave soul who figured it out. Also, even though individual Makefiles might themselves be very different, in order to cope with different non-standard install methods, the end user's installation experience is completely uniform...in most cases all they have to do is type "make install" to compile/install *anything.*
This is why I believe personally that if *any* package management/installation method was going to be included in the LSB, it should be something based on ports...simply because ports is the only solution I've yet come across which works in a complete sense. Virtually every other proposed system I've seen has problems. My own proposed solution is a combination of this, which is something I wrote myself a bit back, and rpm. The above linked project only has files in CVS right now though unfortunately, so you'll have to use that if you want them.
What I plan to do though is incorporate the use of bmake (NetBSD's make) and a partial clone of ports as a replacement for much of the body of rpm's
>I'm curious about what sorts of problems you ran
;)
>into. I was under the impression that configure
>scripts were designed to at least make source
>code portable, by dynamically finding the various
>libraries and such that the application requires.
Sure...they are. However, you're making two assumptions here which may not (unfortunately) be the case...they are in many instances, but not in all.
a) That said science apps even use configure/autotools at all, and
b) That said science apps have *well written* or *coherent* configure scripts, which look for everything they need, and do so in a sane manner. You wouldn't believe the sort of garbage I've seen in configure scripts before...made me wonder what on earth the developer in question had been smoking.
There are lots of apps out there which don't use a sane method of installation/compilation, and which should. DocBook is one example which comes to mind. True, in that scenario nothing needs to be compiled...but I find myself wondering if a basic makefile or shell script for installing the files into something close to the standard location would be too much to ask. Even assuming the default offered in said script was the wrong location for your system, it'd at least give you something to work with.
Then as another example you've got people like Joerg "I'm too cool for the rest of the population" Schilling, who find it necessary to completely reinvent the wheel, and who don't necessarily do so in a sane or comprehensible manner. Thus, if you wish to install any of their applications, you actually have to download their re-interpretation (regurgitation would probably be a better word) of Make or whatever else...it basically requires you to step into their own demented private universe.
So yeah...Unfortunately, as far as software compilation/installation is concerned, we can invent as many different solutions and standards as we want. Some people will still feel for whatever reason that whatever rules we create don't apply to them...and won't understand that said rules generally exist in order to make *their* lives easier, as well as the lives of everyone else.
a) RPM
.spec format is beyond horrible in all sorts of different ways. I advocate crucifiction for whoever came up with the idea of sub-packaging from within a spec file in particular...this "feature" alone tends to make specfiles for glibc in particular an incomprehensible nightmare...and I can also virtually guarantee that this particular feature almost exclusively is responsible for every mangled RPM-based system anyone has ever read about...it is also responsible for the inability to compile single non-RPM based applications on an otherwise RPM based system. In short, sub packaging was an almost indescribably bad (not to mention completely unnecessary) idea, and IMHO should be removed from RPM.
I'm assuming that the fact that RPM is even part of the specification is due to Red Hat's influence, nevertheless it most certainly should not be. For one thing, RPM's
In addition, the standard macros are also completely opaque...you've got no way of knowing what on earth %setup -q T b does unless you've looked it up.
b) Economic theory
Economic theory says that in order to sell something, it is ideally desirable to have what is called a Unique Selling Point...i.e., to have something unique which nobody else has, which is the reason why someone is going to buy from you, and not the zillions of other companies in the world. Thus, being non-standard is actually the *only* means a Linux vendor company has of creating an incentive for a consumer to purchase from that particular company, rather than any other. Standards (at least beyond a certain degree) would end up meaning that every vendor sold a completely identical, generic product...and therefore, no consumer would have any incentive to buy from any one vendor in preference to any other. *Customers* might want that, and vendors pay lip service to it because they know customers want it...but the vendors themselves in reality do not want it, because it would be very bad for their business...so beyond a certain point, complete standardisation ain't going to happen.
c) Standards aren't *always* a good thing. Standards are primarily a good idea when talking about communication. The Web is a good example. Because the Web itself relies on standard protocols/specifications, (HTML, HTTP and so on) the idea is that individual computers, sometimes with radically different operating systems *internally*, can reliably communicate with each other *externally* on the Internet. But for each one of those computers to always be identical *internally* can be an extremely bad thing, security wise...Microsoft's woes in terms of virii, worms, and trojan horses provide all the evidence needed to prove this point.
d) An operating system/standards monoculture is only really desired by people who are seeking an excuse to be able to avoid using their brains. Such people want things to be so completely simplistic that they can perform any task on rote, and most especially, so that they can perform any task without the truly horrifying prospect of needing to actually *think.* This is not to suggest that I am completely opposed to anything being user-friendly...quite the opposite. But I *am* aware that there is a very large segment of the computer using population who are willing to put Eric Raymond's "luxury of ignorance" ahead of virtually every other aspect of software design in terms of importance, including security. "Give us one single operating system, and do not allow the complexity of any aspect of it to be greater than a retarded five year old's ability to deduce!" Such is the rallying cry among these souls, and I hear it on an almost daily basis, albeit normally somewhat veiled.
I've kept a vague eye on this game for a while, on and off. It's main problem is that, for my money anyway, the game it's based on wasn't all that great...and FreeCiv for the most part also doesn't have most of the things that the original/s had which were capable of keeping people's attention. (Lots of in-game cinematics, AI using diplomacy until very recently, etc etc) Although FOSS is strong on code, multimedia seems to be a major problem for FOSS developers...Presumably because at least some of them are used to playing games like Nethack. (Graphics? What on earth do you need those for?)
The main game I really want to see a FOSS clone of is the Sims, in all honesty. The reason why is that with the Sims 2, EA basically killed the main reason why the original game was so popular...i.e., the degree to which it was editable with new wallpapers, floors, objects etc...and of course, the possibility of modifying something is one of FOSS's main strengths.
It would need to be given a completely different name though, and probably kept underground for a long while...EA in particular are extremely paranoid about the proactive amateur phenomenon...Their TOS for The Sims Online specifically outlaws the construction of server emulators by anyone who plays the game.
I realise that a Sims clone would also be very difficult from a coding point of view, so it perhaps isn't possible. I can dream, though.
>According to statements by both Stallman and >Perens, they consider programmers are "evil" unless >they give away their work.
This is one misconception that I am growing desperately tired of...however like some kind of many-tentacled octopus, no matter how many times I try and step on it, it just keeps rising from the grave. There also admittedly aren't that many times when I'll come to the defense of RMS...but this *is* one of them. Go and read this. In it, RMS states that he doesn't have a problem with people selling GPL bound software at all.
The free/beer and free/libre debates and the associated misconception were ESR's entire motivation for adopting the term Open Source, so far as I know. Basically what Stallman says at the above link is that the ONLY demand that the GPL makes is that source be provided with binaries. End of story. To quote from his article:-
"Since free software is not a matter of price, a low price isn't more free, or closer to free. So if you are redistributing copies of free software, you might as well charge a substantial fee and make some money. Redistributing free software is a good and legitimate activity; if you do it, you might as well make a profit from it."
(Emphasis his)
I've started realising that it isn't actually so much RMS' philosophies themselves which bother me...it's the misconceptions that a lot of other ill-informed morons who claim to be his supporters develop and spread. RMS doesn't have anything against people making money from GPL-bound software AT ALL...He did it himself at one point with Emacs. He isn't Communist, and the GPL itself isn't, either. I repeat, all the GPL demands of people is that if you're going to distribute binaries in whatever manner, you also distribute accompanying source, so that said software has a chance to propogate itself. As long as both are sold together in the one package, you can charge as much money for said package as you like.
There's still OpenNap.
Just another pundit trying to secure his livelihood by promoting the status quo. I've commented on that particular syndrome before. The way that I know that that is what he is doing is because he says he can't refute Linux's technical superiority...but then goes on to claim that Windows is better overall, which they all do. By initially making the technical concession first, and then making their second claim, they try to have their cake and eat it...appear to have journalistic integrity while still maintaining the status quo.
I agree with one of the earlier comments that the editors need to start screening articles of this nature more carefully. They don't tell us anything we don't know, or that hasn't been said before.
The author of this article is right in one respect, though...Strictly speaking, Linux won't kill Windows...Microsoft are going to be responsible for their own demise, to a large degree.
Rather than "Starter Edition," here's some suggestions, if anyone from Redmond just happens to read this. (I know they won't do it - it's more a mental exercise while I eat)
;) I'll probably get modded Offtopic, but it was worth it.
1. Go download this, and make it natively multi-user if it isn't already. Give it a strong native security model, too...you can get some ideas here, and the best part is, they won't mind you doing that if you don't try and patent said ideas. Also, modularise your GUI, and don't prevent users from accessing the CLI when they want to.
2. Have the CLI composed of this and this for us CLI types.
3. Make the Add/Remove Programs panel essentially a net-aware frontend for either this or this.
4. Use this for hardware detection. Also re drivers, get rid of the suicidal policy of seeing third-party hardware vendors as the enemy, and actually support them...via tools, docs, etc. These people are your friends...they'll help you stay relevant.
5. Download this and use it as your default FS, and then get this and this, (although you already seem to know about this last one) and incorporate both of those into your stock UI. You've essentially got WinFS right there, without all the added complexity you'd no doubt throw into it if you tried to code it from scratch.
6. For the Agent angle, incorporate the last point, as well as putting help/docs in a non-binary format, making them searchable with this, converting said search results for use with this, and then use the AIML output as input for something like this. Also, instead of making the agent a tightly anthropomorphic personality, make it more generic, and more as though it's simply "the operating system" communicating with a user, rather than that dog or Clippit instead.
7. Give Outlook a major overhaul. This and this are examples of directions it IMHO should go in.
Just some random ideas, anywayz. Dreaming's fun.
Attempting to strangle ADSL adoption, killing the national BBS community when the Internet first became mainstream in Australia in order to force adoption of Big Pond, and a host of other offenses meant that after an extended period of shopping around, I finally stopped using Telstra as a carrier completely last year, and they can now consider themselves permanently boycotted as far as I'm concerned. They are one of the most short-sighted, destructive, and generally amoral corporations I've heard of. They were also vocally criticised by Bill Gates during one of his visits here, for their strangulation of broadband adoption.
Apart from the above, to some degree there are now price incentives to use other carriers as well, particularly for voice. If you've got a credit card, you also might want to check out TPG for ADSL...they probably have the best deals I've seen.
>No AIM
There's Gaim. That will let them use AIM, but it will also mean that if their friends use MSN or ICQ instead, they can talk to all of them from within the same program.
>No Kazaa.
Heard of mldonkey? eDonkey, Kazaa, Gnutella, BitTorrent, Napster, and Direct Connect, all in the one app. Got more clients than you can shake a stick at, too.
I've got a Netcomm 56k internal based on the Lucent Mars chipset. Contrary to the amount of flak I've heard levelled at the hardware, the drivers from here work solidly in my experience, with both 2.4 and 2.6 series kernels.
However, not all winmodems are created equal. I'm not entirely sure about this, but I believe it depends on whether or not the winmodem in question actually DOES have a controller chip and just needs proprietary drivers, as opposed to a true controllerless modem. From what I've read, the Lucent AMR modems are genuinely controllerless and thus not supported under Linux.
In a nutshell, because the term "winmodem" is a catchment word and actually describes quite a large number of different devices, you need to make sure you have reasonably intimate knowledge of which specific breed of winmodem you've got. Some will work, some won't...but again, in my experience anyway with my own chipset, both the 2.4 and 2.6 Lucent modules work well.
Of course, a standard hardware modem is always more desirable if you can get one...particularly seeing as the Lucent modules taint the kernel, which may be a problem for some people. (it doesn't particularly bother me) The advantages of winmodems however are price, greater level of availability these days from what I've seen, and marginally better throughput than their standard cousins in some instances. I'm hoping to eventually save up for a standard/external one of these days, and as I said they are more desirable if you can find/afford it...but I'm at least surviving on my Lucent right now.
Do you really think it does Linux's image any good to have idiots burying her with flames...and *especially* phoning her house in the wee hours and harassing her? (Assuming that is true, of course)
That just adds grist to her mill about what a pack of wild eyed, unwashed fanatical hippies we supposedly are.
People here keep talking about how she supposedly has no ethics and her opinion is completely for sale, etc. What some of the dimmer bulbs among you are perhaps missing is that that could actually be useful to us...because assuming she really *is* that cynical, if we tried applying some diplomacy, the opinion she eventually ends up selling could be ours.
Microsoft have traditionally used people like Laura to communicate with computer laypeople, and in the past at least, it was very effective for them. If Laura's values really are for sale, we ought to try and recruit her ourselves...because every individual like her that we can convert, is one less such individual who's still relaying Microsoft's party line to the masses, and can actually be relaying a pro-Linux message to them instead.
Yeah, I know I sound like a disgusting, slimey American spindoctor here most likely...but the problem is, this shit *works*. That is the reason why Microsoft got where they did, and why despite the popularity of the GPL/GNU's *software*, RMS is currently battling impending irrelevance.
As far as influencing people's thought is concerned, Stallman's methods don't work. It's as simple as that. This has been proven, and it is being proven...Over and over and over again. RMS has been tried, and found sorely wanting. Given that, if we want Linux to continue to gain traction among the greater portion of the population, we need to abandon Stallman's ideas and attitude once and for all, and focus on methods which *do* work instead.
And please, any of his advocates are welcome to come forward and refute me on this, because it also doesn't mean that I'm in opposition to the entire paradigm of FOSS in general. Those of you who give Stallman sole credit for said concept's creation are chronically misguided.
Even apart from that though, I have one simple comeback for any of Stallman's advocates which I'm guessing they won't have an answer for...Namely, that they are an endangered species. RMS's amen corner are a tiny minority, and getting smaller by the day, from everything I've seen. The pat answer that Stallman himself gives to this, that the majority aren't interested in being right, also doesn't wash. Any idea can be effectively communicated if said communicator knows what they're doing.
The reason why I keep bringing this up is because I really believe that Stallman is an anvil around Linux's neck. Stallman can be held indirectly responsible for virtually all of the zealotry and pedantry that Linux users are at times accused of...because he himself is a zealot and a pedant, and his followers adopt his example. He represents a major diplomatic problem that Linux truly does not need, and the fact that other people emulate his fringe-dwelling behaviour means that the problem only gets amplified.
We get rid of Stallman, and trolls like Laura DiDio lose all their ammunition, as far as calling us extremists is concerned, more or less instantly...Because Stallman *is* both an extremist and an advocate of extremism. It's that simple.
as though I've just been exonerated. I've been saying for a while why I have issues with RMS, and if this license goes ahead as depicted here, there will finally be irrefutable evidence of what I've been saying...Namely, that the man is a megalomaniac whose intentions are not anywhere near as glorious as he has made out.
Stallman needs to be careful. He exists in a community whose tendency is, once it has identified problems, to simply route around them. If the mask truly comes off and he exposes his real agenda of wanting control over people, rather than being the advocate of libertarianism that he has claimed, he will run the risk of being rendered completely irrelevant. Although I'm actually hoping he continues to be imprudent, because I want people to see the truth about this man. Once enough people have, then we can all move on to other licenses which more genuinely serve our purpose.
...And I'll say it again.
These laws are almost entirely unenforceable. As an ISP, if any government wanted to force me to monitor individual users' bandwidth, I'd ask them if they were going to provide me with extra staff to help me do it, since there would be no possible way that I could do such a thing myself. Let's also see if they can persuade their usually understaffed, underpaid police forces to do the job, as well...My guess is that that is unlikely.
Any government that wants to can pass as many laws like this as it wants, and then sit back and watch as the general public completely ignores them. Governments, the RIAA, and WIPO need to get it into their empty heads once and for all:- We *want* to pirate music, we're *going* to pirate music, and apart from a few token lawsuits here and there against the odd big fish, for the most part there is exactly jack shit you can do about it. Get used to it, because it (and we) are not going away.
>You know what would be a good idea. A bunch of
>geeks getting together with a bunch of researchers
>in their respective fields. Creating honest,
>non-biased "this is the way it is" anaylsis and
>reports on TCO/Software/Hardware/etc. Sort of like
>Consumer Reports(tm) but with more detail and
>analysis of specific topics.
The problem is, there have been analyses done that have at least try to make it sound like that. The worst ones of all are the whining, seemingly apologetic ones written by trolls like Eugenia from OSNews. They start with statements like, "I really love Linux, and want to see it succeed in the marketplace, *but*..." and then proceed to launch into a litany of ignorant misconceptions and FUD.
The main problem is that these journalists who keep doing Microsoft's PR work for them, are I think people who unfortunately believe that their livelihood is tied to the maintenance of Microsoft's dominance. They think that if Microsoft were to collapse, the "ecosystem" around Microsoft (as MS themselves call it) would collapse along with it.
We need to somehow reassure ZDNet and their spiritual kin that if Linux was to gain more market share than Windows, it wouldn't automatically mean that they would be out of a job. Linux has heaps of events, activity, and issues of various kinds that media people can report on...ZDNet's peeps only need to look at Newsforge or LWN to find that out.
I guess my point here is that the trade press do not need to feel that they have to keep trashing Linux in order to protect themselves and their employment. ZDNet could quite easily go from being Microsoft's unofficial Ministry for Propaganda to being wholehearted Linux advocates if they wanted to...and aside from still making a very satisfactory income, they'd sleep better at night knowing that they were engaging in a far greater level of journalistic integrity than they have in the past, as well. It *is* very possible, guys. Just ask O'Reilly.
IANAL, so...now that we have all this wonderfully detailed information about spammers, how do we use it to have them locked up for 20 years or so with their very own girlfriend named Spike? ;-)
>And all of a sudden it's dying?
;)
;)
I don't believe Debian's dying, but I'd almost be glad if it was...At least then I wouldn't have to put up with the egocentric attitude of its' users. I will admit that Debian users have actually become probably my primary pet hate as far as this site is concerned. I think the only group I find more smug, obnoxious, and generally annoying are OSX fanboys.
I'm also assuming that if it was dying, most of the Slashbots who use it would also make the assumption that Linux in general was, since we all know that Debian is the only distribution in existence.
Nothing random about it. I've only made two out of 10+ submissions that have been accepted. I submitted info about the Microsoft "Get the Facts" campaign in particular before ESR had updated the Halloween Docs about it, and it got rejected...and then of course when someone submitted ESR's article on it, that went straight through.
The thing that I primarily find ironic about this is the fact that the alleged degeneracy of various different elements of popular culture (video games are only one item on a very long list) are almost exclusively protested by those individuals who together, comprise the single most criminally insane, morally degenerate, and maniacally dangerous group of people currently in existence on this planet; namely, the American Right.
Rightwing politicians and their advocates endorse the slaughter in Iraq, the errosion of education, the erradication of science, apocalyptic foreign policy, the worst forms of cronyism and corruption, both corporate and political, and fear, hatred and xenophobia of every kind, and more dangerous than anything else, the proliferation and use of nuclear weapons. The neoconservative movement ranks in my mind as one of the most dire threats that humanity as a whole have ever faced. For any of their number to express concern about the morality of virtually anything in comparison with their own crimes goes beyond hypocrisy...There is no word in existence to describe it.
>which unlike your elitist pompous cry-baby whining,
>actually makes people smile.
Elitist? Pompous, cry-baby whining I can accept...I *have* engaged in that at times...But how am I elitist?
I think it's vital that we somehow find a way to prevent this individual and anybody else who thinks like him from getting what they want. Contrary to what he might say, the system we have right now *does* work, and has worked well. The only reason why he is trying to make people believe that it hasn't worked is so that he can institute his own system.
The UN wants to have jurisdiction over *everything.* Nothing makes Kofi Annan start foaming at the mouth more than when he hears the phrase "national sovereignty." In practice however, the only thing the UN really amounts to is the Third World's vehicle for trying to fulfill Pinky and the Brain's primary objective. Everyone wants to take over the world...no huge conspiracy theory there. The US govt, the Israeli govt, the EU govts; everyone. The thing we have to somehow try and do though is make sure that none of them manage it...and the UN succeeding in doing it wouldn't be any more desirable than anyone else managing to.
The UN, contrary to what some highly emotive left-wing types will have you believe, (and before you make the assumption, no, I'm not a fascist...I've been called left myself) is most definitely NOT "our last, best hope for peace." This is not Star Trek, and the UN are NOT the Federation. This is the real world, and the only thing the UN are our "last, best hope" for is a centralised global government...something which would ultimately cause a greater level of tyranny and suffering possibly than we've ever seen before. (And yes, I know we saw a lot in WW2. That's part of my point.)
Repeat after me, kids:- The UN is NOT my friend.
If I currently had modpoints, I'd devote all five of them to modding this post Overrated, and thus blast this juvenile, unfunny attempt at humour back to AC-level oblivion where it belongs.
>Damn, no Debian package for it. Oh well.
.deb file yourself, or
What sort of mindless sheep are you? Heaven forbid the idea that you either
a) Make a
b) Compile/install it outside of dpkg (horrors!)
Forgive me, though. I realise that following either of those suggestions would require using your own brain and your own initiative, which going by your sig as a George W Bush supporter, probably *is* a little too much to ask.
>The main work and value of distributions like >Debian is in the extensive testing that goes into >making sure that all the packages work together.
One thing...not everyone uses the same distribution. And before you say it, no, we shouldn't. The lack of a monoculture is one of Linux's primary defenses against viruses...To newbs, lack of a single universal interface seems like a weakness...but apart from strong multi-user, one of the other things that impedes virus writing is the lack of a consistent environment...because it means a virus writer cannot make a single set of assumptions about how a virus will need to behave. If a virus was written to attack something written for KDE, for example, the fact that there are other desktops means that a Gnome user may not be affected by it. People keep calling for "unity" all the time, when what they don't realise is that the universal interface is actually one of Microsoft's biggest problems...it's not a strength.
Autopackage is something that would be distro independent, and as such could be extremely beneficial...people could still ship distro-specific patches with an autopackage quite easily if they wanted to. The added benefit though would also be that someone who either wasn't using Debian/RH or who was actually using something distro-neutral like LFS could still install it.
Linux has always been about choice. If someone wants to use Debian, great...but there are a lot of other distributions out there which do different things. Autopackage could give all of them the means of installing packages easily, while still retaining their uniqueness. That's a good and important thing.
Although some may question the advisability of advocating a controlled substance, I believe that a short course of marijuana is something you may find extremely therapeutic. I myself used to be as pedantic as this, but after smoking weed intermittently for around 18 months, (late 2001-2003) was no longer so, and have not been to this day.
Although it may be argued that my becoming less pedantic was actually due to the marijuana causing neurological damage, if it was damage it was in such a way that I was still profoundly grateful for. It made me less likely to care about being correct in instances where being correct truly did not matter, and where the only purpose being correct served was to alienate and annoy my fellow man.