This article jibes with my own personal experience.
This is a generalization, of course, but a lot of schools nowadays (including my own, it seems) seem to be intent on producing a bunch of trained monkeys who know the motions for a specific set of software -- basically Microsoft Excel + Microsoft Visual Basic + Microsoft Access. There are some places were C/C++ isn't even a requirement anymore, whereas Excel w/ VBA is!
Too bad very nearly everything us students are learning now is going to have to be re-learnt in some manner with the next revision of the software...
The problem is that they're not teaching generalized skills and abstract thinking as part of the curriculum. I think this one section of the article is very telling in this regard:
> He won't choose a high-tech subject, he said, > because he's interested in examining more > abstract ideas.
> "I'm here at college mainly to learn to explore, > because I love thinking about ideas," he said.
I know a lot of students (many of them ex-CS majors or soon-to-be-ex-CS majors) who feel the same way. The really brilliant people are either learning this stuff on their own, or are going into other fields altogether.
Something is fundamentally wrong with the current state of CS education in the U.S. -- shouldn't CS _by definition_ be about abstract ideas?
The point is that you need to teach the kids the _theory_ either first or in parallel with the specific skills. Once they have the theory down, they can pick up individual skills pretty easily.
If, however, the students train for a bunch of individual skills without first understanding how they relate, it makes it very difficult to learn new things. Sometimes (I have experienced this personally), it can even interfere with a student's ability to learn the theory later.
One thing that would help is if students were exposed to several vendors' software, instead of just one. You can't generalize very well when you only have one example to reason from.
Another thing would be if they actually started teaching _logic_ and _reason_ in schools again, instead of rote memorization. I see a lot of interest in rhetoric (vis a vis the popularity of debate teams), but next to none in logic or critical thinking. That's not a healthy balance.
If you don't get that preparation before college, you're pretty screwed if your college is going to have any kind of worthwhile curriculum itself.
The last thing that would help tremendously would be to return to teaching CS in a primarily Unix environment (note I said _primarily_, most certainly not exclusively).
It's not that Unix is the "magic OS of knowledge", it's that it does a better job of exposing the structure of things to the user. Yet, it still has a healthy amount of abstraction. There aren't really any other widely-used environments that are especially good at both.
That is, Unix is designed primarily to abstract the underlying system so that it can be manipulated and restructured effectively. It allows the student to play with and think a lot more about abstract ideas.
If you design a system so that the user is presented with a complete set of tools that require some intelligence to manipulate and use together, it becomes very effective for teaching a student generalized skills which can later be applied in specific situations.
If, on the other hand, you design something so that "a trained monkey could operate it", you'll find that it's hard to use it to teach people to do any more than what a trained monkey could do.
The problem is that students aren't being educated anymore. They're being trained.
You {under,over}estimate Katz
on
Why Kids Kill
·
· Score: 1
> I cant believe [Katz] didnt see the trap here..
> jezus, if playing violent games such as > "doom/quacke/etc" alleviates violent tendencies, > then by the same token wouldnt kiddie porn > alleviate abusive tendencies?
Uh... if I read his sexbots article correctly, he actually suggested something to that effect too; i.e. child sex robots would provide a healthy outlet for pedophiles. That was among the many things in that article that motivated me to give the/. "self-censorship" (as Katz would say) facilities a spin.
I've gotta stop clicking on Katz articles out of morbid curiousity when I'm not logged in and can see them. I don't I do him or myself any favors with my comments.
> Nasdaq is currently run on umax machines running > nt and microsoft nt server in a cluster > configuration.
That's actually not strictly true. Some of Nasdaq's non-critical systems are NT boxes in the configuration you describe, but that's about the extent of their NT deployment.
Unfortunately, their use in non-mission-critical stuff has somehow reduced to the meme "Nasdaq uses NT", which is correct, but misleading. They don't rely on it for mission-critical systems.
Ad Hocery on both sides of the aisle...
on
Red Hat 6.0
·
· Score: 1
> Gnome isn't an ad hoc design, like kde.
Although at first KDE kind of fell in around the Qt toolkit design-wise, from a design standpoint, it's getting pretty well structured and documented now. GNOME, on the other hand, seems to be stuck in an implement-first document-later cycle.
If you look carefully, KDE's (SOM/OpenDoc-influenced) object model is of a rather nicer design than Baboon (a cleaned-up COM), too.
At some point in the future, I strongly suspect that KDE and GNOME's object models will have to merge, in much the same way they finally ended up agreeing on a DnD protocol. Honestly I'd best like to see KDE's model adopted by both.
That being said, aside from the object models, it looks like the GNOME standards that _are_ specified are definitely turning out better than their KDE counterparts, both in terms of design and implementation.
I personally prefer GNOME, but they really need to get their design act together a little more.
(by the way, it's refreshing to see someone who can spell "ad hoc" correctly; it saddens me to see so many people who spell "fare" as "fair" and "their" as "there" and so on...)
What really gets me is that years back (think 70's) Xerox had optical mice that were capable of working on more or less any surface with a trivial amount of onboard logic and three LEDs or so. I have no idea why they didn't catch on.
Well, except the Xerox folks seem to be completely incompetent when it comes to marketing an idea...
Katz seems to be getting bitter...
on
ShutUp Software
·
· Score: 1
--- snip ---
And now I'm blocked by hundreds of Slashdotters as well. The people who've posted messages about this or e-mail me don't cite choice, convenience or time constraints - they simply claim to dislike my writing style or ideas, and want me to know that they're banning them and me, from their own individual experience.
This is everybody's absolute right; nobody should be forced to read me if he or she don't want to. But the ShutUp Software will almost certainly be used in unforeseen, sometimes unintended ways. What can be used against me can - and will -- be used against you.
I can always find an audience for my ranting (in addition to other sites and links, there are those 79,350 Slashdot readers who can somehow survive having my work appear on the site). But angry kids posting messages can be marginalized, even obliterated, in a snap.
--- snip ---
Read: "Well, fine. If you losers don't want to listen, I'll take my writing someplace else, and then you'll all be screwed!"
Really.
To be fair, he's probably just blowing off steam.
I normally have Katz filtered -- the SexBots article was kind of the last straw for me. (I say, yeah! Robot sex slaves! Heralding a new age of sociosexual dysfunctionality, totally disconnected from real human social experience! We'll never have to relate quasi-functionally with another human being again and we can Get It On any time we like with no responsibility or compassion!)
It's kind of sad. Despite his content (or lack thereof), he's a pretty good writer, and is at least capable of using the language properly. Sometimes he has actually written something cool. I'm just not willing to gamble my time away on that anymore.
People don't generally value freedom (for its own sake) that much anymore.
*knock knock knock*
Lemming: "Hello?"
Idealist: "Have you heard about products X, N or B?"
Lemming: "Yeah. Everyone's talking about those. But I already use product Y from vendor Z. You're not selling one of them, are you?"
Idealist: "No. I'm just concerned for the long-term welfare of humanity. Maybe you shouldn't use product Y."
Lemming: "Why not? I like it better. Products X, N and B aren't as good."
Idealist: "Have you used products X, N or B?"
Lemming: "No."
Idealist: "Then why do you say that?"
Lemming: "Because I like product Y and it's what I alrady use. Why switch?"
Idealist: "Well, among other things, using product Y serves to prevent you from using products not from vendor Z..."
Lemming: "So? All I ever use is vendor Z's products anyway."
Idealist: "There are less expensive alternatives to vendor Z's products that work better for most of the things I know you do."
Lemming: "But those don't work as good with vendor Z's products."
Idealist: "Look, using products from vendor Z is denying you the freedom to use better and less expensive products when they're availible? Doesn't that bother you?"
Lemming: "So?...not really. Product Y works good enough for me, and it's what everyone else is using."
Idealist: "Don't you care about having the freedom to try and use something better (and cheaper)?"
Lemming: "Well, I guess. But I don't really have a choice, because everyone else uses product Y too..."
Idealist: "But you DO have a choice! There are people using other, better, products that all work together. You're choosing to limit your own freedom by choosing product Y. The more people that use product Y, the less free everyone (including you) becomes."
Lemming: "Yeah, but most people are already using product Y with vendor Z's stuff. Could you leave me alone please?"
> the GNU/Linux business makes as much sense to > say X/Linux
Not really. Take away X, the system still boots. Take away the BSD-derived tools, the system still boots. Take away the GNU tools, the system won't get beyond/sbin/init. I think that says something about the relative importance of the GNU tools in the system.
This is the same reason that it doesn't make too much sense to say that the GNU tools on a Solaris box make GNU/Solaris. Unless you rewrite all of the system scripts and some of the other utilities, you'll still need to keep the Solaris toolset around -- the system will still depend on them, not the GNU toolset.
> The term "software piracy" is considered > inappropriate because illegal copying of > software is not comparable to robbery of a ship > at sea.
> Using propriety software, however, is comparable > to "slavery".
Note that the former metaphor is implicit, the latter is explicit (a simile, really).
What's so insidious about the former is that it is the normal term used in this case. Anyone using has to incorporate the implicit metaphor into their mental framework, and it DOES colour your thinking.
As for the latter, even RMS doesn't use "slavery" as a normal term for proprietary software. There, even he uses the (relatively) neutral term.
Explicit metaphors influence people's thinking too, of course, but they sink in at a much more conscious level, giving you a better opportunity to think about it and say, "hey, that really isn't right..." (as you did)
> That's like saying Ford should get sued when > there's a hit-and-run
With a hit-and-run, the design of the car is not a major factor.
No, it's like saying Ford should get sued if, say, a poorly designed car door falls off of the car at 60mph and someone goes flying. Negligent design; that sort of thing.
Can you really say that Microsoft's repeated and continuous failures to implement proper security measures (i.e. levels of priviledge combined with appropriate sandboxing to enforce same) for macro code are _acceptable_?
> No, there really needs to be something in > between GPL and BSD. Something that says that > the code can be used in proprietary projects > only if the authors of the code agree to such.
The authors (copyright holders) of code that is released under the GPL are perfectly free to allow proprietary use if they want; it basically amounts to relicenseing the code. It just requires the unanimous agreement of the copyright holders of the affected code -- exactly as in this mythical license you describe.
Scenario:
I write a GPLed app.
Company X approaches me, asking if they could use such-and-so portion of the code (or even the whole thing) in a piece of proprietary software.
I say "yes", and negotiate some special licensing agreement with them that allows them to use the code. That's perfectly legal, because I hold the copyright to the code. With multiple copyright holders, you do need to reach a consensus first, oc.
The copyright holder(s) has/have complete control over the way their software is licensed, no matter what license(s) is/are involved.
His intent is that GNU/Linux be used to refer to a primarily GNU toolset running on top of a Linux kernel. I.e. Linux = kernel, GNU/Linux = kernel + GNU utilities.
As long as it's understood that GNU/Linux only applies to the GNU tools + Linux kernel combination, I think the title is appropriate. It has to do with the relative importance of the GNU component of the system.
Take your average Linux distribution, and try each of these in turn:
Try removing X11. You don't get your GUI.
Try removing the BSD-derived tools. You don't get much of your network software anymore, as well as missing ps and df and some other genuinely useful things.
Try removing all of your GNU tools. _NOTHING_ will work anymore.
That's also why something like cygwin32-compiled GNU utilities don't make GNU/Win32 -- you take the GNU stuff away in that case, the system still works fine.
Ironically, if the Daemon Linux folks have their way, we'll _need_ to use this appellation to differentiate "normal" Linux environments from a pure BSD toolset running on top of the Linux kernel. The latter would -- suprise -- need to be termed something like BSD/Linux.
(as an aside, I think that Daemon Linux is dangerous; do we really want to reintroduce the BSD/AT&T split that was the first wedge in the original fragmentation of Unix? Before you answer, see what something basic like $ does in an AT&T-style make(1) [i.e. GNU make], then try it in a BSD-style make(1).)
In a similar manner to the GNU/Linux being used to differentiate toolsets, this distinction also serves to differentiate a Linux-based GNU system from a GNU toolset running on top of the HURD.
For instance, Debian has both GNU/Linux and GNU/HURD distributions now.
Anyway, personally, "GNU/Linux" as a term isn't that important to me at this point; all of the current Linux distros use the GNU toolset, so differentiation isn't important (yet) -- I generally just say "Linux".
The GNU/Linux label is perfectly fair and valid when applied to a compound GNU/Linux system, so don't criticize anyone for using it correctly. I personally don't have any qualms about using it when I need to.
...and other sundry Unixen. Some of the developers are avid FreeBSD users, which is one of the reasons for the switch to an X/BSD-esque licence from the LGPL.
phew. I think you win this one. Sorry if I've taken a while to respond; I'm in lynx now; it seems that there's a serious memory leak in the text area widget supplied by the version of Motif statically linked with Netscape, aggravated by long replies. (my last two were done from work, where I apparently have a less buggy Motif implementation)... I wish I could relink the thing. I guess I needed the time to think about this anyway. I guess I can at least agree unreservedly for small markets -- the small markets thing was nagging in the back of my mind for some months, actually, due to unrelated conversations.
> This solution does have a problem in that after > one has recouped ones expenses one can continue > raking in pure profit.
Pure profit is (in and of itself) fine, but inflating the value of your product by creating scarcity (which is what proprietary software licensing amounts to) is to my mind immoral, and an unprovoked application of force. I think this is especially true of critical products, like food and (soon, if not already in some cases) software.
> I disagree with the last sentance. I am selling > you software, I value it by the work that I put > in to create the software, not by the cost of > the media it's on.
You have no way to directly recoup that cost; you're not selling the well, you're selling the water. Bleah. I still can't find a good physical analogy for this. Software just can't be validly compared to a physical good.
Anyway, the thing is that charging for media is one thing -- you can include the value of your development work in the price you charge. That goes for any other physical goods or services associated with the software.
Forcing people to pay you for additional copies of the software, however, is wrong. If they made a copy of the software once they had it, it would expend _none_ of your resources. They're not taking anything away from you that you already posessed. Why, then, should they be obligated to pay you?
> Why is this unethical?
Because you're coercing people into paying you money for a service that they do not need you to provide. In your terms, it's an unprovoked use of force.
> Are you under the false assumption that I use > Microsoft's products (or any other proprietary > products)?
No, I didn't know enough to assume that. I suspected you were an author of proprietary software, however.
I'll stop here because Netscape has started flaking out on me.
> What product is naturally unlimited? The years > of man hours invested by the program's authors?
Nope. that's real-world capital, unlike the software. You're confusing the capital used for production with the good produced.
> False. Proprietary software is like a well you > dug up closer to the town than the river. You > spent the time, energy, and resourcefullness to > create the well,
> not the towns folk
This applies only to "from scratch" software developed under a so-called "cathedral model", then.
> hence you have the right to charge whatever you > want, and impose arbitrary limitations on it's > use.
Yes, you did find a valid problem with my own analogy here -- I didn't include anything that adequately paralleled the initial capital investment required to make the software in the first place.
Hrm. Okay, I'll bite. A well _would_ be a better analogy, except that well water is a finite, exhaustible, resource -- river water (at least as far as the needs of a single small town are concerned) is not, provided nobody upstream interferes.
Basically, your well analogy would be valid if:
- the groundwater was inexhaustible (note that digging another well would be analgous to creating a new software product) - you (or the townspeople) could take water from the well at zero cost (like duplicating software) - an essentially infinite number of people could have access to the well at once without causing problems - the physical nature of the well didn't imply concerns regarding land use rights
> The part about sharing the water is a false > analogy.
I don't think so. Very nearly all proprietary software licences that I've read restrict the use of a particular copy of the software to a specific set of people (generally one).
I'm not sure either of our analogies are necessarily that valid, given that water is itself a physical, finite, resource (although in my river analogy, it is, considering the meagre needs of a small town, unlimited).
Software itself (I'm not talking about production captial) can be duplicated and shared at zero cost; it's not a finite resource at all, and creating artificial scarcity (so you can get more money) in such a case is just unethical and wrong.
> I have heard the opinion expressed in a previous > debate that people shouldn't be babied with > computers, that is, they should have to > understand how they work and be able to fix > them.
To a limited extent, yes. Like people who drive cars know at least the basics of how the car works, and can open the hood if they need to and fix basic stuff. Beyond that, I agree with you in that users can't be expected to know everything about the internals of the system (even programmers can't), but I disagree with you in that they should have the chance to learn about the internals if they want to.
> This "support" argument dumbfounds me.
Yeah, support isn't the best way to get funds if you're a software producer, but yet even proprietary software companies are expected by their customers to provide support. Hrmm....
Actually, with free software, I tend to see a proliferation of companies, some of which specialize in support, some of which specialize in software production and/or packaging. All around, the same amount of software generates more wealth (in general) than the equivalent proprietary software would.
You'd be suprised how many people would rather buy a CD than download something.
> This opinion fits quite nicely with the > idea of selling support -- you tell the customer > that you aren't going to make things easier for > them in the software, instead they'll have to > pay for training
Yeah, and you've never seen people shell out big bucks to be trained on, say, basic usage of Microsoft Excel (and that is IMO, one of the best designed products of it's class, user-interface-wise). Convince me that people don't need to be trained to use proprietary software.
Hrm. I kind of avoided your argument there. I think to remedy that I'll just point out the various projects that are working on increasing the usability of free software, many funded by companies that are selling it.
(By the way... I do agree with that making a poor product just to do more business with support is just wrong. I just don't see companies producing free software doing that, though.)
> and bug fixes.
Could you please explain to me how a company producing could (on a practical level) charge money for bug fixes if they didn't place restrictions on the redistribution of the software? Charging for bugfixes is only seen in the world of proprietary software.
> Go ahead, take this route. It won't be the > software produced by this reasoning that will > drive software progress to the next level.
Oh... I guess Mozilla w/ NGLayout comes to mind first here. Definitely shoddy, uninnovative, inefficient, non-standards-compliant stuff.
> Why should anyone have any rights to my > brainchild except those that I grant them? In > the well analogy, I can sell the town the well, > but what right do they have to force me to tell > them how I built it, or to help them fix it?
This is about the water in the well (the software product) not the well itself (the intial investment needed to produce the software). You're still confusing the capital used to produce the good with the good itself... in the case of software, a limited capital investment allows you to produce an infinite amount of the good.
Presumably you're not selling your programmers to the end users, are you? (or is there a black market in programmers that I don't know about?)
Presumably you're not being compelled to give them support (helping them fix the well) for _free_, are you? In most situations, the licensing fee you pay for proprietary software also pays for support.
> -> Caveat: On the other hand, I don't have any > problems with people reverse engineering my > well, that would be imposing on their right to > reason about things.
Okay. That's certainly fair.
> Don't work for them if they don't pay you what > you think you're worth. If you do, then it is > you who is ripping yourself off.
Can you think of any examples of proprietary software companies that do pay their programmers per copy licenced? ANY examples?
> Why should a company do you any favours?
Oh, I dunno. Moral obligation to pay their employees fairly?
> By the same token, why should I do a company any > favours?
Indeed. I could just make unauthorized copies for all my friends. But I don't, because that would be immoral (breach of contract, at minimum).
> If everyone just did what was right for them > (without using force upon another) and quit > worrying about the other guy I think we'd all be > better off.
By "right", do you mean moral, or beneficial? The context indicates the latter...
The problem is that people who are only looking out for their own (immediate; people are short-sighted) interests will inevitably resort to force to resolve conflicts between their interests and other people's interests.
Maybe you're right... let's see... does a consumer who makes unauthorized copies of software do so because:
a. the software company doesn't use force to prevent them from doing that? b. they are looking out for their own interests, and not those of the software companies?
It cuts both ways, man.
Despite this, I will say that I think that looking out for the interests of others (occasionally, at least) is the only thing that's kept us (as a species) alive for this long.
> Keeping source closed is not fueled by an > intentional desire to rob others of their > freedom. It is simply the most convenient way to > create a barrier to entry in a market.
I think I'll refrain from comment.
> As RMS says, these companies will never make as > much money as Microsoft.
So, there are other proprietary software companies that make as much money as Microsoft?
> Don't buy a companies software if you don't > think it's worth it. If it's the only thing on > the market then tough luck, do without, create > your own solution, or buy the software.
Unfortunately, as software is becoming more and more important to our society, doing without is becoming less and less of an option daily. I would venture to say that in the space of a few decades, access to software will be as important as, say, access to water or food (moreover, our access to those resources may be at least partially dependent on software).
Water and food are finite resources; software (itself, separate from production methods) is not. More's the pity that people try and play tollkeeper to software resources it as if they were.
Now, as for making my own solution, at home, I do just that. I have not payed money to license any of the software on my machine, and yet I'm 100% legal. (yet note that I have still exchanged money for goods with software companies like RedHat... how's that work, I wonder?)
The problem is that, unlike in the physical world, there is no inherent scarcity with software. We're not talking about refusing to give away a physical product here, we're talking about preventing people from using a resource that is naturally unlimited.
Proprietary software is like (and about as moral and ethical as) damming up a river, the only source of water for a town, and then selling it back to the townspeople in bottles under the stipulation that you can't even share your bottle of water with anyone else.
"Now," you may say, "what about the limited real-world capital that goes into the initial production of the software?" The thing is, with any software, there are also associated real-world products and services (media, support, etc) that you can charge for and make a decent profit and pay your programmers too.
There are companies do this for free software. Exclusively. They make money. Plenty of money. Cash. Moolah.
Profit.
I mean, come on, we're all capitalists here, right?
In other words, it has been demonstrated many times in the real world that you won't starve (and in fact can still make a healthy profit) if you sell "free" (i.e speech) software.
Another thing -- has the inherent inequity of most programmers being payed per project, and the companies they work for charging per copy of proprietary software, ever occurred to you?
Companies that make proprietary software rip off their programmers as well as the consumers.
Slashdot Mirror
This article jibes with my own personal experience.
This is a generalization, of course, but a lot of schools nowadays (including my own, it seems) seem to be intent on producing a bunch of trained monkeys who know the motions for a specific set of software -- basically Microsoft Excel + Microsoft Visual Basic + Microsoft Access. There are some places were C/C++ isn't even a requirement anymore, whereas Excel w/ VBA is!
Too bad very nearly everything us students are learning now is going to have to be re-learnt in some manner with the next revision of the software...
The problem is that they're not teaching generalized skills and abstract thinking as part of the curriculum. I think this one section of the article is very telling in this regard:
> He won't choose a high-tech subject, he said,
> because he's interested in examining more
> abstract ideas.
> "I'm here at college mainly to learn to explore,
> because I love thinking about ideas," he said.
I know a lot of students (many of them ex-CS majors or soon-to-be-ex-CS majors) who feel the same way. The really brilliant people are either learning this stuff on their own, or are going into other fields altogether.
Something is fundamentally wrong with the current state of CS education in the U.S. -- shouldn't CS _by definition_ be about abstract ideas?
The point is that you need to teach the kids the _theory_ either first or in parallel with the specific skills. Once they have the theory down, they can pick up individual skills pretty easily.
If, however, the students train for a bunch of individual skills without first understanding how they relate, it makes it very difficult to learn new things. Sometimes (I have experienced this personally), it can even interfere with a student's ability to learn the theory later.
One thing that would help is if students were exposed to several vendors' software, instead of just one. You can't generalize very well when you only have one example to reason from.
Another thing would be if they actually started teaching _logic_ and _reason_ in schools again, instead of rote memorization. I see a lot of interest in rhetoric (vis a vis the popularity of debate teams), but next to none in logic or critical thinking. That's not a healthy balance.
If you don't get that preparation before college, you're pretty screwed if your college is going to have any kind of worthwhile curriculum itself.
The last thing that would help tremendously would be to return to teaching CS in a primarily Unix environment (note I said _primarily_, most certainly not exclusively).
It's not that Unix is the "magic OS of knowledge", it's that it does a better job of exposing the structure of things to the user. Yet, it still has a healthy amount of abstraction. There aren't really any other widely-used environments that are especially good at both.
That is, Unix is designed primarily to abstract the underlying system so that it can be manipulated and restructured effectively. It allows the student to play with and think a lot more about abstract ideas.
If you design a system so that the user is presented with a complete set of tools that require some intelligence to manipulate and use together, it becomes very effective for teaching a student generalized skills which can later be applied in specific situations.
If, on the other hand, you design something so that "a trained monkey could operate it", you'll find that it's hard to use it to teach people to do any more than what a trained monkey could do.
The problem is that students aren't being educated anymore. They're being trained.
> I cant believe [Katz] didnt see the trap here..
/. "self-censorship" (as Katz would say) facilities a spin.
> jezus, if playing violent games such as
> "doom/quacke/etc" alleviates violent tendencies,
> then by the same token wouldnt kiddie porn
> alleviate abusive tendencies?
Uh... if I read his sexbots article correctly, he actually suggested something to that effect too; i.e. child sex robots would provide a healthy outlet for pedophiles. That was among the many things in that article that motivated me to give the
I've gotta stop clicking on Katz articles out of morbid curiousity when I'm not logged in and can see them. I don't I do him or myself any favors with my comments.
> Nasdaq is currently run on umax machines running
> nt and microsoft nt server in a cluster
> configuration.
That's actually not strictly true. Some of Nasdaq's non-critical systems are NT boxes in the configuration you describe, but that's about the extent of their NT deployment.
Unfortunately, their use in non-mission-critical stuff has somehow reduced to the meme "Nasdaq uses NT", which is correct, but misleading. They don't rely on it for mission-critical systems.
> Gnome isn't an ad hoc design, like kde.
Although at first KDE kind of fell in around the Qt toolkit design-wise, from a design standpoint, it's getting pretty well structured and documented now. GNOME, on the other hand, seems to be stuck in an implement-first document-later cycle.
If you look carefully, KDE's (SOM/OpenDoc-influenced) object model is of a rather nicer design than Baboon (a cleaned-up COM), too.
At some point in the future, I strongly suspect that KDE and GNOME's object models will have to merge, in much the same way they finally ended up agreeing on a DnD protocol. Honestly I'd best like to see KDE's model adopted by both.
That being said, aside from the object models, it looks like the GNOME standards that _are_ specified are definitely turning out better than their KDE counterparts, both in terms of design and implementation.
I personally prefer GNOME, but they really need to get their design act together a little more.
(by the way, it's refreshing to see someone who can spell "ad hoc" correctly; it saddens me to see so many people who spell "fare" as "fair" and "their" as "there" and so on...)
What really gets me is that years back (think 70's) Xerox had optical mice that were capable of working on more or less any surface with a trivial amount of onboard logic and three LEDs or so. I have no idea why they didn't catch on.
Well, except the Xerox folks seem to be completely incompetent when it comes to marketing an idea...
--- snip ---
And now I'm blocked by hundreds of Slashdotters as well. The people who've
posted messages about this or e-mail me don't cite choice, convenience or time
constraints - they simply claim to dislike my writing style or ideas, and want me
to know that they're banning them and me, from their own individual
experience.
This is everybody's absolute right; nobody should be forced to read me if he or
she don't want to. But the ShutUp Software will almost certainly be used in
unforeseen, sometimes unintended ways. What can be used against me can -
and will -- be used against you.
I can always find an audience for my ranting (in addition to other sites and links,
there are those 79,350 Slashdot readers who can somehow survive having my
work appear on the site). But angry kids posting messages can be marginalized,
even obliterated, in a snap.
--- snip ---
Read: "Well, fine. If you losers don't want to listen, I'll take my writing someplace else, and then you'll all be screwed!"
Really.
To be fair, he's probably just blowing off steam.
I normally have Katz filtered -- the SexBots article was kind of the last straw for me. (I say, yeah! Robot sex slaves! Heralding a new age of sociosexual dysfunctionality, totally disconnected from real human social experience! We'll never have to relate quasi-functionally with another human being again and we can Get It On any time we like with no responsibility or compassion!)
It's kind of sad. Despite his content (or lack thereof), he's a pretty good writer, and is at least capable of using the language properly. Sometimes he has actually written something cool. I'm just not willing to gamble my time away on that anymore.
> Small price to pay for a lot more freedom.
...not really. Product Y works good enough for me, and it's what everyone else is using."
People don't generally value freedom (for its own sake) that much anymore.
*knock knock knock*
Lemming: "Hello?"
Idealist: "Have you heard about products X, N or B?"
Lemming: "Yeah. Everyone's talking about those. But I already use product Y from vendor Z. You're not selling one of them, are you?"
Idealist: "No. I'm just concerned for the long-term welfare of humanity. Maybe you shouldn't use product Y."
Lemming: "Why not? I like it better. Products X, N and B aren't as good."
Idealist: "Have you used products X, N or B?"
Lemming: "No."
Idealist: "Then why do you say that?"
Lemming: "Because I like product Y and it's what I alrady use. Why switch?"
Idealist: "Well, among other things, using product Y serves to prevent you from using products not from vendor Z..."
Lemming: "So? All I ever use is vendor Z's products anyway."
Idealist: "There are less expensive alternatives to vendor Z's products that work better for most of the things I know you do."
Lemming: "But those don't work as good with vendor Z's products."
Idealist: "Look, using products from vendor Z is denying you the freedom to use better and less expensive products when they're availible? Doesn't that bother you?"
Lemming: "So?
Idealist: "Don't you care about having the freedom to try and use something better (and cheaper)?"
Lemming: "Well, I guess. But I don't really have a choice, because everyone else uses product Y too..."
Idealist: "But you DO have a choice! There are people using other, better, products that all work together. You're choosing to limit your own freedom by choosing product Y. The more people that use product Y, the less free everyone (including you) becomes."
Lemming: "Yeah, but most people are already using product Y with vendor Z's stuff. Could you leave me alone please?"
Idealist: "AUGH&*#@^$*(#@&^$&*#@^"
Lemming: "I'm calling the police..."
...but don't call it 1.0.
That's my philosophy.
ssia.
> the GNU/Linux business makes as much sense to
/sbin/init. I think that says something about the relative importance of the GNU tools in the system.
> say X/Linux
Not really. Take away X, the system still boots. Take away the BSD-derived tools, the system still boots. Take away the GNU tools, the system won't get beyond
This is the same reason that it doesn't make too much sense to say that the GNU tools on a Solaris box make GNU/Solaris. Unless you rewrite all of the system scripts and some of the other utilities, you'll still need to keep the Solaris toolset around -- the system will still depend on them, not the GNU toolset.
> The term "software piracy" is considered
> inappropriate because illegal copying of
> software is not comparable to robbery of a ship
> at sea.
> Using propriety software, however, is comparable
> to "slavery".
Note that the former metaphor is implicit, the latter is explicit (a simile, really).
What's so insidious about the former is that it is the normal term used in this case. Anyone using has to incorporate the implicit metaphor into their mental framework, and it DOES colour your thinking.
As for the latter, even RMS doesn't use "slavery" as a normal term for proprietary software. There, even he uses the (relatively) neutral term.
Explicit metaphors influence people's thinking too, of course, but they sink in at a much more conscious level, giving you a better opportunity to think about it and say, "hey, that really isn't right..." (as you did)
Just because you give up a freedom voluntarily (in a legal contract, in this case) does not mean that you aren't less free as a result.
To be honest, it rather disturbs me that a lot of people don't seem to understand that.
> That's like saying Ford should get sued when
> there's a hit-and-run
With a hit-and-run, the design of the car is not a major factor.
No, it's like saying Ford should get sued if, say, a poorly designed car door falls off of the car at 60mph and someone goes flying. Negligent design; that sort of thing.
Can you really say that Microsoft's repeated and continuous failures to implement proper security measures (i.e. levels of priviledge combined with appropriate sandboxing to enforce same) for macro code are _acceptable_?
> No, there really needs to be something in
> between GPL and BSD. Something that says that
> the code can be used in proprietary projects
> only if the authors of the code agree to such.
The authors (copyright holders) of code that is released under the GPL are perfectly free to allow proprietary use if they want; it basically amounts to relicenseing the code. It just requires the unanimous agreement of the copyright holders of the affected code -- exactly as in this mythical license you describe.
Scenario:
I write a GPLed app.
Company X approaches me, asking if they could use such-and-so portion of the code (or even the whole thing) in a piece of proprietary software.
I say "yes", and negotiate some special licensing agreement with them that allows them to use the code. That's perfectly legal, because I hold the copyright to the code. With multiple copyright holders, you do need to reach a consensus first, oc.
The copyright holder(s) has/have complete control over the way their software is licensed, no matter what license(s) is/are involved.
:%s/\$/\$/g
His intent is that GNU/Linux be used to refer to a primarily GNU toolset running on top of a Linux kernel. I.e. Linux = kernel, GNU/Linux = kernel + GNU utilities.
As long as it's understood that GNU/Linux only applies to the GNU tools + Linux kernel combination, I think the title is appropriate. It has to do with the relative importance of the GNU component of the system.
Take your average Linux distribution, and try each of these in turn:
Try removing X11. You don't get your GUI.
Try removing the BSD-derived tools. You don't get much of your network software anymore, as well as missing ps and df and some other genuinely useful things.
Try removing all of your GNU tools. _NOTHING_ will work anymore.
That's also why something like cygwin32-compiled GNU utilities don't make GNU/Win32 -- you take the GNU stuff away in that case, the system still works fine.
Ironically, if the Daemon Linux folks have their way, we'll _need_ to use this appellation to differentiate "normal" Linux environments from a pure BSD toolset running on top of the Linux kernel. The latter would -- suprise -- need to be termed something like BSD/Linux.
(as an aside, I think that Daemon Linux is dangerous; do we really want to reintroduce the BSD/AT&T split that was the first wedge in the original fragmentation of Unix? Before you answer, see what something basic like $ does in an AT&T-style make(1) [i.e. GNU make], then try it in a BSD-style make(1).)
In a similar manner to the GNU/Linux being used to differentiate toolsets, this distinction also serves to differentiate a Linux-based GNU system from a GNU toolset running on top of the HURD.
For instance, Debian has both GNU/Linux and GNU/HURD distributions now.
Anyway, personally, "GNU/Linux" as a term isn't that important to me at this point; all of the current Linux distros use the GNU toolset, so differentiation isn't important (yet) -- I generally just say "Linux".
The GNU/Linux label is perfectly fair and valid when applied to a compound GNU/Linux system, so don't criticize anyone for using it correctly. I personally don't have any qualms about using it when I need to.
...and other sundry Unixen. Some of the developers are avid FreeBSD users, which is one of the reasons for the switch to an X/BSD-esque licence from the LGPL.
phew. I think you win this one. Sorry if I've taken a while to respond; ... I wish I could relink the thing.
I'm in lynx now; it seems that there's a serious memory leak in the text area widget supplied by the version of Motif statically linked with Netscape, aggravated by long replies. (my last two were done from work, where I apparently have a less buggy Motif implementation)
I guess I needed the time to think about this anyway.
I guess I can at least agree unreservedly for small markets --
the small markets thing was nagging in the back of my mind for some months, actually, due to unrelated conversations.
> This solution does have a problem in that after
> one has recouped ones expenses one can continue
> raking in pure profit.
Pure profit is (in and of itself) fine, but inflating the value of your product by creating scarcity (which is what proprietary software licensing amounts to) is to my mind immoral, and an unprovoked application of force. I think this is especially true of critical products, like food and (soon, if not already in some cases) software.
> I disagree with the last sentance. I am selling
> you software, I value it by the work that I put
> in to create the software, not by the cost of
> the media it's on.
You have no way to directly recoup that cost; you're not selling the well, you're selling the water. Bleah. I still can't find a good physical analogy for this. Software just can't be validly compared to a physical good.
Anyway, the thing is that charging for media is one thing -- you can include the value of your development work in the price you charge. That goes for any other physical goods or services associated with the software.
Forcing people to pay you for additional copies of the software, however, is wrong. If they made a copy of the software once they had it, it would expend _none_ of your resources. They're not taking anything away from you that you already posessed. Why, then, should they be obligated to pay you?
> Why is this unethical?
Because you're coercing people into paying you money for a service that they do not need you to provide. In your terms, it's an unprovoked use of force.
> Are you under the false assumption that I use
> Microsoft's products (or any other proprietary
> products)?
No, I didn't know enough to assume that. I suspected you were an author of proprietary software, however.
I'll stop here because Netscape has started flaking out on me.
> What product is naturally unlimited? The years
... I guess Mozilla w/ NGLayout comes to mind first here. Definitely shoddy, uninnovative, inefficient, non-standards-compliant stuff.
... let's see ... does a consumer who makes unauthorized copies of software do so because:
... how's that work, I wonder?)
> of man hours invested by the program's authors?
Nope. that's real-world capital, unlike the software. You're confusing the capital used for production with the good produced.
> False. Proprietary software is like a well you
> dug up closer to the town than the river. You
> spent the time, energy, and resourcefullness to
> create the well,
> not the towns folk
This applies only to "from scratch" software developed under a so-called "cathedral model", then.
> hence you have the right to charge whatever you
> want, and impose arbitrary limitations on it's
> use.
Yes, you did find a valid problem with my own analogy here -- I didn't include anything that adequately paralleled the initial capital investment required to make the software in the first place.
Hrm. Okay, I'll bite. A well _would_ be a better analogy, except that well water is a finite, exhaustible, resource -- river water (at least as far as the needs of a single small town are concerned) is not, provided nobody upstream interferes.
Basically, your well analogy would be valid if:
- the groundwater was inexhaustible (note that digging another well would be analgous to creating a new software product)
- you (or the townspeople) could take water from the well at zero cost (like duplicating software)
- an essentially infinite number of people could have access to the well at once without causing problems
- the physical nature of the well didn't imply concerns regarding land use rights
> The part about sharing the water is a false
> analogy.
I don't think so. Very nearly all proprietary software licences that I've read restrict the use of a particular copy of the software to a specific set of people (generally one).
I'm not sure either of our analogies are necessarily that valid, given that water is itself a physical, finite, resource (although in my river analogy, it is, considering the meagre needs of a small town, unlimited).
Software itself (I'm not talking about production captial) can be duplicated and shared at zero cost; it's not a finite resource at all, and creating artificial scarcity (so you can get more money) in such a case is just unethical and wrong.
> I have heard the opinion expressed in a previous
> debate that people shouldn't be babied with
> computers, that is, they should have to
> understand how they work and be able to fix
> them.
To a limited extent, yes. Like people who drive cars know at least the basics of how the car works, and can open the hood if they need to and fix basic stuff. Beyond that, I agree with you in that users can't be expected to know everything about the internals of the system (even programmers can't), but I disagree with you in that they should have the chance to learn about the internals if they want to.
> This "support" argument dumbfounds me.
Yeah, support isn't the best way to get funds if you're a software producer, but yet even proprietary software companies are expected by their customers to provide support. Hrmm....
Actually, with free software, I tend to see a proliferation of companies, some of which specialize in support, some of which specialize in software production and/or packaging. All around, the same amount of software generates more wealth (in general) than the equivalent proprietary software would.
You'd be suprised how many people would rather buy a CD than download something.
> This opinion fits quite nicely with the
> idea of selling support -- you tell the customer
> that you aren't going to make things easier for
> them in the software, instead they'll have to
> pay for training
Yeah, and you've never seen people shell out big bucks to be trained on, say, basic usage of Microsoft Excel (and that is IMO, one of the best designed products of it's class, user-interface-wise). Convince me that people don't need to be trained to use proprietary software.
Hrm. I kind of avoided your argument there. I think to remedy that I'll just point out the various projects that are working on increasing the usability of free software, many funded by companies that are selling it.
(By the way... I do agree with that making a poor product just to do more business with support is just wrong. I just don't see companies producing free software doing that, though.)
> and bug fixes.
Could you please explain to me how a company producing could (on a practical level) charge money for bug fixes if they didn't place restrictions on the redistribution of the software? Charging for bugfixes is only seen in the world of proprietary software.
> Go ahead, take this route. It won't be the
> software produced by this reasoning that will
> drive software progress to the next level.
Oh
> Why should anyone have any rights to my
> brainchild except those that I grant them? In
> the well analogy, I can sell the town the well,
> but what right do they have to force me to tell
> them how I built it, or to help them fix it?
This is about the water in the well (the software product) not the well itself (the intial investment needed to produce the software). You're still confusing the capital used to produce the good with the good itself... in the case of software, a limited capital investment allows you to produce an infinite amount of the good.
Presumably you're not selling your programmers to the end users, are you? (or is there a black market in programmers that I don't know about?)
Presumably you're not being compelled to give them support (helping them fix the well) for _free_, are you? In most situations, the licensing fee you pay for proprietary software also pays for support.
> -> Caveat: On the other hand, I don't have any
> problems with people reverse engineering my
> well, that would be imposing on their right to
> reason about things.
Okay. That's certainly fair.
> Don't work for them if they don't pay you what
> you think you're worth. If you do, then it is
> you who is ripping yourself off.
Can you think of any examples of proprietary software companies that do pay their programmers per copy licenced? ANY examples?
> Why should a company do you any favours?
Oh, I dunno. Moral obligation to pay their employees fairly?
> By the same token, why should I do a company any
> favours?
Indeed. I could just make unauthorized copies for all my friends. But I don't, because that would be immoral (breach of contract, at minimum).
> If everyone just did what was right for them
> (without using force upon another) and quit
> worrying about the other guy I think we'd all be
> better off.
By "right", do you mean moral, or beneficial? The context indicates the latter...
The problem is that people who are only looking out for their own (immediate; people are short-sighted) interests will inevitably resort to force to resolve conflicts between their interests and other people's interests.
Maybe you're right
a. the software company doesn't use force to prevent them from doing that?
b. they are looking out for their own interests, and not those of the software companies?
It cuts both ways, man.
Despite this, I will say that I think that looking out for the interests of others (occasionally, at least) is the only thing that's kept us (as a species) alive for this long.
> Keeping source closed is not fueled by an
> intentional desire to rob others of their
> freedom. It is simply the most convenient way to
> create a barrier to entry in a market.
I think I'll refrain from comment.
> As RMS says, these companies will never make as
> much money as Microsoft.
So, there are other proprietary software companies that make as much money as Microsoft?
> Don't buy a companies software if you don't
> think it's worth it. If it's the only thing on
> the market then tough luck, do without, create
> your own solution, or buy the software.
Unfortunately, as software is becoming more and more important to our society, doing without is becoming less and less of an option daily. I would venture to say that in the space of a few decades, access to software will be as important as, say, access to water or food (moreover, our access to those resources may be at least partially dependent on software).
Water and food are finite resources; software (itself, separate from production methods) is not. More's the pity that people try and play tollkeeper to software resources it as if they were.
Now, as for making my own solution, at home, I do just that. I have not payed money to license any of the software on my machine, and yet I'm 100% legal. (yet note that I have still exchanged money for goods with software companies like RedHat
...or I'm especially humour-impaired tonight; not sure which.
The problem is that, unlike in the physical world, there is no inherent scarcity with software.
We're not talking about refusing to give away a physical product here, we're talking about preventing people from using a resource that is naturally unlimited.
Proprietary software is like (and about as moral and ethical as) damming up a river, the only source of water for a town, and then selling it back to the townspeople in bottles under the stipulation that you can't even share your bottle of water with anyone else.
"Now," you may say, "what about the limited real-world capital that goes into the initial production of the software?" The thing is, with any software, there are also associated real-world products and services (media, support, etc) that you can charge for and make a decent profit and pay your programmers too.
There are companies do this for free software. Exclusively. They make money. Plenty of money. Cash. Moolah.
Profit.
I mean, come on, we're all capitalists here, right?
In other words, it has been demonstrated many times in the real world that you won't starve (and in fact can still make a healthy profit) if you sell "free" (i.e speech) software.
Another thing -- has the inherent inequity of most programmers being payed per project, and the companies they work for charging per copy of proprietary software, ever occurred to you?
Companies that make proprietary software rip off their programmers as well as the consumers.