If one can trade patches freely within the community, then it is obvious that good patches would propagate. It would also make sense for Sun to implement them if it so desired and if they make sense. If you own a copy of a painting you can draw whatever you want on top and trade your modifications with anyone else; the owner of the original is the only one who can modify the original.
The fundamental problems are several. First, resources are limited. Even given unlimited energy and the ability to reclaim wastage and to recycle anything (which I do not believe will happen), there is still time to consider. All else being equal (this is important), if it takes me two days to make X and you three minutes to make Y, X should be worth rather more than Y. There are only a finite number of objects which can be produced by a given person within a given time; some method for rationing them must be created. This is an economic fact.
Energy will probably never be limitless. Even fusion requires raw materials which must be ripped out of the earth--the miners must devote their hard labour to this task. I have a nasty feeling that were solar power to produce all of the world's electricity the climatological effects would be intense. We may be someday able to release and control huge amounts of energy, but I am rather confident that, much like hard drive space, land, laws and everything else in life, the uses of energy will always remain one step ahead. Remember back when 1MB was huge? Now 12GB is tight for me...
Let us reconsider those miners above. Why would anyone be a miner when he can be an artist? Because he wants to? But how many people want to spend their days in a black pit? Or consider the case of sewer workers. These poor folks literally spend their days knee-deep in sh*t. How do you convince someone to do that? Not all jobs are fun; not all are pleasant, in air-conditioned offices in front of computers. Some mechanism for rationing the limited supply of fun jobs must be determine.
Another problem is that different things are useful in differing degrees to different people. Size 8 shoes (8 in. long; about 20 cm) are useless to me, but extremely useful to my father. I have no use for eggplant--as far as I'm concerned it is compost fodder--but to a Frenchman that aubergine might be just the thing for his ratatouille. Some mechanism must be developed to account for these differences in taste--eggplants should be rationed to the Frog while I get hamburgers.
I have shown, very roughly and in a form I am sure which would send an econ major into apoplexy and a logician into fits, that rationaing of resources and labour is necessary. I think we can agree that labour is a resource, so I will treat it as such from here on out.
There are many schemes for rationing. We could have a first-come, first-served scheme: all the grain (or electricity, or jobs or workers) is in a big heap and whoever gets there first can take all he wants. Well, if he's smart he'll take it all, or as much as he can carry. He would then take that surplus and go down to the fellow who has cornered whatever it is that our grain-hoarder really wants (perhaps pictures of Miss Portman). He would then trade the useless (to him) grain for the useful (to him) pictures. He might collaborate with several such hoarders and produce a common unit to ease the chose of bartering. Thus, with the invention of money, we are back to the current situation, except that some people have everything and everybody else starves.
The above is a utopian model: producers produce and just give stuff away to the first comer, which as we see cannot work. Well, what about a socialist system, under which a benevolent commission rations everything? Unfortunately this does not work either. The commission cannot accurately know the myriads of ways in which resources may be valued. They can set the price of beef lower than it ought to be. This is the situation in the US: the government has subsidised the production of beef and thus more beef is produced (I do not believe it would be possible to satiate the American demand for beef; I know I sure love it). The natural limiting factors (expense of grazing land, cost benefit of farming vs. ranching) thus do not apply and the end result is that we have too many cattle--we are thus inefficient. Conversely, we tax alcohol at a rate designed to cut consumption. Thus less alcoh and we inefficiently produless alcohol thatn would be desirable. No commission can possibly comprehend the full interplay of factors: people's varying desires for thing; seasonally-affected variations; regional variations &c.
And, in a socialist set-up, the commissions tend to become Big Brothers. No-one wants to be a sewer worker--the commission must force somebody to do it. Few people desire to give 90+% of their earnings away to others, but the commission makes them do it. So this non-omniscient commission forces its ill-informed opinions upon the masses? Whither right? Whither justice?
The free market, while imperfect, is a way to fix all this. No-one is forced into anything, except through economic means. The stick is eschewed in favour of the carrot for the most part. Everyone is given the hope of improving his lot and the freedom to realise that hope. How does this work?
A free market assumes a multitude of producers and consumers. Everyone is both at one point or another; I produce labour, while I also consume all sorts of goods. Any producer may sell his product to any consumer at a price the two arrange between themselves. If my work is worth at least $10/hr. to me, and my work is worth at most $30/hr. to my employer, then we can settle on a price somewhere in between. If my work is worth $30/hr. to me and $10/hr. to my employer, then I can try to find another consumer to buy my labour. If there is no other customer to buy at that price, I had better consider lowering it.
This carries over into everything. Land costs a certain amount per acre, which is an aggregation of the value it has for all uses. Land in the middle of West Texas may be near-worthless for business as it is far from big cities, but it may be worth quite a bit as ranch land. Land in Kansas may produce $10,000/yr. in cattle or $15,000/yr. in wheat. Everyone is free to produce as he will; the intelligent man produces the most efficient goods given the resources at his disposal, which means that he makes the most possible money which means that he can buy the largest posible amount of what he desires.
To try to bring this back on topic, how does this relate to software? Well, software takes time to write. The three weeks I spend writing Bob's Cool Program are three weeks that I do not spend farming, or sweeping streets, or making cars, or otherwise making money. I convert three weeks of time into an artifact, a piece of source code. Compiling that code takes little time; thinking by time alone, it might make sense to give away the binary and sell the source. But most home users do not have access to compilers. So perhaps it's a better idea to sell the binaries and give away my source. But if I give away the source then someone else may take my three weeks' work, add one days', compile the thing and sell it, gaining me naught. This is related to the problem of the BSD License that the GPL seeks to address.
Of course, if I GPL my work then the situation is the same, except that I get that other fellow's one day of work. If I can get twenty people each to give me a days' work thusly, then I have put three weeks' work into something and gotten somebody else to put four weeks' into it. I now have a work worth seven weeks of labour for the low price of three weeks. Thus the GPL is economically sound, as long as I actually value that software in the manner I described.
What if I could have made $10,000 from my software but instead end up get $5,000 worth from the GPLed programme I released? Then I have made an economically unsound decision and lost money. I have made a decision which I may find morally more comfortable, but I am worse off for it. And as far as morality goes, I have wasted $5,000; the world is that much poorer because I didn't make money--I lack $5,000 I could have spent. Remember that value is actually created; when I trade two goats for your one cow, I am better off and you are better off. When I get paid $25/hr, I'm better off and my employer is better off. When I sell a product worth $500 to you for $100, we're both better off. If I give away a product worth $100 to 100 people and get naught in return, they are better off but I am worse off. I have hurt myself and they are now able to spend their money in an inefficient manner. Thus I have hurt the economy.
So one should try to examine the cost/benefit of GPLing software. Which is what software producers are doing. RMS, OTOH, believes that it is a moral imperative so to do. I think that I have shown above how it might actually be harmful.
What, exactly, was wrong with the older Sun license? Yeah, it was not free software, but it gave full use of the source code to those who had purchased the product. This allowed bug fizing and feature enehancement by the community for the community (unless I greatly misremember the details). Quite honestly, this is the way I think software should work by defaut. If I buy a copy of Word I should get full source. I don't necessarily believe that I should then be able to give that source away. But I do not want to be locked into a buggy piece of software which I cannot fix or modify.
Perhaps the old license was too restrictive in other ways, in order to ensure that only paying members got source? Or did it demand that Sun be assigend copyright on mods? Or was there some other valid complaint? Or was it simply griping that Sun dared release non-free software?
Free software is cool stuff. I write it, and GPL it. But I do not demand that everyone else GPL their stuff. Although I do think that software copyrights should be like patents: short term (say, two to five years); can be renewed once for an additional term; the source is on file; the source becomes available at the expiration of the copyright. This way people can make money for a few years on their work, but we still get the source in the not-so-long-run.
I'll tell you why a utopia won't work: because I'd take advantage of it. And so would anyone else with half a mind. If it's free, I might as well take all of it. The reason that software works even when free is that a) it is a limitless resource and b) enough people enjoy making it on the side. They still need to produce limited resources using limited resources to put bread (another limited resource) on the table. If bread's ever free, I'm going to the store to get all of it. And I'll give it to others, too. In return for little slips of paper. Or perhaps for that funky yellow metal they use in necklaces...
Of course, that can be fixed by rationing food. An approach which worked so well in WWII. That's the problem with socialism. It is not very realistic. Instead of people buying as they can, it relies on a Big Brother to parcel out the goods. Of course, Big Brother (or Uncle Joe...) likes to have as much as he can for himself.
People work on incentive; it's the only way. Free Software and Open Source (which are one and the same, no matter what His Highness thinks) only work because people have an incentive to write excellent software; they wish to scratch their itch. Take that away, and we have no more free software.
Very few people have an internal desire to clean toilets or plow fields. That sort of work scratches precious few itches (although it does cause a few). Yet, these things need doing. Thus the free market. It arranges for a fully-functioning society, unlike any other scheme yet proposed.
This brings up an issue about which I ahve been thinking for quite some time now, ever since I read the circumstances which gave birth to the FSF. Stallman wanted the source to a printer driver because it was buggy. He went from that starting point to argue that all source should be free. I have often thought that his ends could have been meat by stating that source and modification are free within the community of customers. That is, if I purchase a copy of Word I have a right to the source code, and a right to share my modifications to that source with others who have bought it. But I cannot give it away to someone else who has not paid for it.
This preserves the profit motive and makes it possible to write code whose source is open and whose bugs get fixed, while at the same time enabling authors to charge for what is, after all, their work and their expenditure of resources. It is true that eventually someone's modifications could be so great that he may want to charge for it; again, he would be able to do so. Let's call the original source A and he mods B. The fellow could sell B only to those who already own A, but could sell A+B to anyone, as long as he paid the original author his due for A.
I will admit that this could get out-of-hand if everyone wanted to charge even for slight mods. But again, that could be controlled--specifying that mods of less than 100 lines or something similar must be public domain, perhaps.
I do not believe that I have ever seen anything on the FSF site explaining why something along these lines (obv. not exactly this sort of thing) is not a good idea.
Re. traffic lights, wouldn't lt32767.lights.mot.london.uk make infinitely more sense? Are these people (not you; the London bunch) ignorant of DNS hierarchies, or are they simply stupid?
The world doesn't need new TLDs; it needs new third level domains...
This may actually be true, but somehow I doubt it. After all, what is to stop someone writing a programme which does certain things based on certain inputs? Is that not, after all, what _all_ programs do? So it is possible, by constructing certain inputs, to cause certain things to occur. From this, it is (in time and given the existence of bugs) possible to write a bootstrapper to then run a virus. Voila!
You do realise, do you not, that AbiWord has little profitability anyway? It's GPLed. This can only help it, as there may be some features of StarOffice which it can pick up. IMHO StarOffice looks much nicer, and it ahs incredible import capabilities, but it runs like a dog. AbiWord runs nicely, but it has that gtk-I-wanna-look-like-Win-3.1 feel:-(
Were I not completely at a loss for free time, I might put some work into that...
That's what the Qt folks are saying and a certain radical fringe is trying to deny that truth. IMHO you are right, but not in the NSHO of Stallman and the FSF (both of which I do have respect for). Their claim is that libraries are just as much modules as are modules. I firmly disagree, but c'est la vie.
The problem with the DUL is that a well-configured host should act as its own mail server and send mail directly from that host to whatever machines are the appropriate recipients. Instead, we are forced to configure our machines to use an external mail server, which slows things down, takes up bandwidth and in many other ways is a right royal pain in various body parts.
The DUL is highly annoying. It bounced mail from my Linux box because my sendmail was configured to deliver mail properly, i.e. to the appropriate machine. I had to reconfigure it to be stupid and send all mail to my ISP, which slows thigns down and adds another point of failure to things. Thanks loads.
Mail from dial-up accts has a higher probability of being spam, granted. But with the marked increase in intelligent hosts (i.e. hosts which are properly configured and act as their own mail servers) installed in user's homes increases, this probability will decrease. I call on all right-thinking mail admins to Do the Right Thing and turn off the DUL.
I also call on the US Congress, acting under its authority to regulate inter-state commerce, to ban spam sent across state lines, and on the legislatures of the states to do the same for intra-state spam, and on foreign legislative bodies to do the same. A federal law declaring the sending of spam across our national borders an actionable act would be nice too:-)
If the majority of people do something incorrectly the wrong way does not become right; they majority have become wrong. The majority is not necessarily correct. Democracy does not apply to the real world; the zebras cannot vote the lions into herbivorousness.
The reason for standards bodies is that intelligent people realise that they cannot foresee all possibilities and consider all viewpoints. A standards body is formed to attempt to do this as well as possible (it obv. cannot do a perfect job of it...). Engineers & computer scientists realise the utility of standards bodies. If the vast moajority of the population does not, standards bodies will remain useful; the opinion of the ignorant majority means nothing to the standards body. All it means to anyone is that the unwashed masses may not follow the standard because they are too proud to know when to listen to someone else.
The W3C may not be listened to, but that makes its standards no less useful, and vertainly does not make Microsoft's any more so. All it means is that most people will follow the one and ignore the other. The ignored one may very well be better, and very prob. is in this case.
No offense, but I think that you misunderstood some of what he meant.
a file with one name - possible
a file with multiple names - possible with links
many files sharing one name - possible if they are in separate directories (see below)
many files sharing one directory - possible
a file can be in one or more directories - again possible with links
a file with no name - I'm not sure about this. From my viewpoint, a thing (can of soda or a file) isn't distinct without some sort of name. For a can of soda, the name may not be just "Pepsi", it may be "the fourth Pepsi from the left in the fridge". Likewise with files - they don't need good names, but they do need to be distinguishable from their fellows, even if just by inode number. If you can't tell your 10000 head of cattle apart in some basic way, then how do you know you really have 10000?
a file in no directory - see below
I believe that by `many files sharing one name' he meant within on directory (or functional group), i.e. one pathname, in much the way, say, that no single beer in the fridge has a name; they're all just `beer' and it doesn't matter which is which. I don't know if this would transfer to a computer well: we'd just make one big beer file.
While a file can have many names, using hard or soft links, actions on one name do not necessarily affect the others: delete one hard link and the others are still there (a useful function, true); delete a symbolically linked file and all the symlinks go insane (hardly useful).
A file in many directories is possibly different from a file with multiple names. Can't see it myself, but then I'm no Ph.D.
Non-named files would be those that are linked to in some other way. Photographs or such might be especially good in this sense. I know that many people have photo collections which give them many views on their libraries; the photos themselves are named in sequential order. Why name them at all? Why not jsut have a `photo chooser' which displays each image and its comments? Why make the named server as comments on a file? Because that's the way we've been doing it for 30 years?
I do not know how workable his ideas are, but OTOH we can see that the current approach has its shortcomings. Some work needs to be done, anyway.
He never said that Linux is obsolete; he said that it is old. There is obviously a great deal of benefit to be derived from this (we know where the bugs are, we have perfected it &c.), but the state of the art does need to advance.
He's not afraid of anything. He is creating. He's trying to see beyond the current, trying to forecast the future. He may very well be wrong. But many of his points sound as things used to be on the net before it was commercialised; I like to think that some of these will come true.
Ummm, a lifestream was what he termed his replacement for files--nothing to do with software companies.
FWIW, I thought that much of what he described sounded like the Net of yesteryear, when clients for a multitude of servers proliferated and platforms did not matter nearly so much. And several ideas were interesting, although I do know how workable they might be (e.g. lifestreams). He is completely correct in one thing: the standard file structure we use is woefully inadequate to the millions-of-documents world in which we find ourselves. Obviously any system in the near future will be built on top of files, but we need to get beyond them for day-to-day work.
And yes, I'd rather say `I need the report I was working on yesterday' to my computer and get it.
Actually, most Mac users I have known have quite udnerstood my desire to use Linux. The two are complimentary, after all. What we Mac users (you see, I wear two hates) do take objection to is people who use Windows willingly as well as those who choose Windows for others. Windows is an uglified, crippled Mac OS. There is little competition between the Mac and Linux; their domains do not overlap by much.
OTOH, there is a great deal of competition between Linux and Windows on the one hand, and Mac OS and windows on the other. Windows cannot be a good server platform, yet it cannot be a good desktop platform. It is a bastard, attempting all and perfecting none. The Macintosh has perfected the desktop; I seriously doubt that we in the Linux world will ever have anywhere near as coherent, elegant, refined, intuitive and all-round cool as the Mac interface. Unix in general, and Linux in particular, has mastered the server realm. I seriously doubt that we Mac users will ever have a platform anywhere near as stable, expandable, programmable, scalable and all-round cool as the Linux platform.
Windows addicts will carry on in their folly, and we in both camps will laught at their misfortune.
That said, both platforms can get much better. WTF's up with Aqua? Linux is still not IMHO serious competition for Solaris, HP-UX or AIX. All things come to those who wait...
How exactly is what I posted flamebait? Sure, I insulted an AC, but certainly with reason: his post was mere assertion apparently uninformed by reading the article. I pointed out that the article itself addressed his points.
I'm quite willing to take moderation based on objective facts, but being modded down solely due to having expressed an unpopular opinion is a bitter pill to swallow. I know that when I moderate I take the job seriously. Generally I only mod down people whom I agree with, but whose method of posting is indelicate, for that is the only way that I can be objective.
Insulting an AC's mental skills is hardly flamebait, esp. when his post is obviously ignorant and he has not read the article referred to.
It was a highly amusing incident in which a Greenpeace fleet was going to `protest' (read: interfere with) nuclear tests. The French sank it. Legally, an act of self-defence against an attacker. Very funny IMHO; that sort of group (left- or right-wing) is just too used to a lack of resistance. People ought to exercise their right to fight back more often.
Actually, that sounds quite fair. We would have the patches; we could write a free library to implement that API. It's not as though it is some undocumented API; we would have the code, and thus have all the docs we need.
So what is the difference, exactly, between a command and a function in a shared library? Both implement an API; the one in a shell, the other in some other language.
And the bit about ls is exactly my point. It's an implementation of an API. Can non-GPLed work use the GNU extensions to ls?
Allow dynamic linking; disallow static linking. Static linking defeats the ability to rebuild th binary, but dynamic linking allows any library in the background. If we wanted we could write a GPLed Qt-alike.
Re:This is neither a huge surprise, nor a bad idea
on
Tripwire Going GPL
·
· Score: 2
I know it might sound silly trying to obtain a support contract for Tripwire, but at the last company I worked for, such a thing would not only be desired, but not too terribly hard to get upper management to sign off on.
Many times the IT department or its outsourced equivalent has Service Level Agreements; I know we do. They're 24/7, so we buy 24/7 support for everything we have. As networking and systems continue to rise in importance, I expect that we will really see a boom in the support market. OS/FS are poised to really take advantage of this.
Re:Tripwire, ColdFusion, and Mission: Impossible
on
Tripwire Going GPL
·
· Score: 3
Tripwire is used to defend from anything which can change files. chmod o-w does not work, because one can do a chmod o+w. Ipchains and tcpdump do no good because one can still sit in front of the machine. No system is secure; tripwire is used to detect when security has been breached.
Do you know what it does? It calculates checksums for all files. These sums can then be stored on read-only media, such as a CD. Then a simple check is all that is nec. to detect modifications to system files.
I've never understood what it is about Francophones. Everywhere they go they seem to think that they are the bee's knees--c.f. Quebec, France, Louisiana and a thousand former French protectorates. Must be something about the language:-)
Minitel had some interesting features, but it was never as good as the Internet, which predated Minitel by over a decade. And I hate to say it, but London and New York will remain the financial centres of the world, with Silicon Valley and Germany doing just fine, thank you, in the technology end of things. France has a nice position in the world, but it will never be the `technological and financial centre.' It has its time: the entire Mediæval, Rennaisance and Early Modern periods. The mantle has passed.
What you're missing is that there are many people who would prefer the GPL to be viral, not in the usual sense that it's accused, but in the sense that it infects anything which touches it. Is it illegal for a proprietary app to call a GPLed ls? Of course not; that's utterly absurd. So why do some think that dynamically linking to GPLed software should be illegal? Because they are fanatics.
I appreciate the GPL; I dislike proprietary software. But I am willing to co-exist. As long as my code remains free and open, I am happy. If my code is dynamically linked against but the source is included, I am happy. Free software will win in the end, for the most part. I'm sure that there will always be some proprietary software. Big deal. I needn't exterminate my enemy, but simply defeat him.
Slashdot Mirror
It's not making as much noise because the PowerPC chip is much more efficient than the Intel family and its clones. Motorola & IBM made a better chip.
If one can trade patches freely within the community, then it is obvious that good patches would propagate. It would also make sense for Sun to implement them if it so desired and if they make sense. If you own a copy of a painting you can draw whatever you want on top and trade your modifications with anyone else; the owner of the original is the only one who can modify the original.
Energy will probably never be limitless. Even fusion requires raw materials which must be ripped out of the earth--the miners must devote their hard labour to this task. I have a nasty feeling that were solar power to produce all of the world's electricity the climatological effects would be intense. We may be someday able to release and control huge amounts of energy, but I am rather confident that, much like hard drive space, land, laws and everything else in life, the uses of energy will always remain one step ahead. Remember back when 1MB was huge? Now 12GB is tight for me...
Let us reconsider those miners above. Why would anyone be a miner when he can be an artist? Because he wants to? But how many people want to spend their days in a black pit? Or consider the case of sewer workers. These poor folks literally spend their days knee-deep in sh*t. How do you convince someone to do that? Not all jobs are fun; not all are pleasant, in air-conditioned offices in front of computers. Some mechanism for rationing the limited supply of fun jobs must be determine.
Another problem is that different things are useful in differing degrees to different people. Size 8 shoes (8 in. long; about 20 cm) are useless to me, but extremely useful to my father. I have no use for eggplant--as far as I'm concerned it is compost fodder--but to a Frenchman that aubergine might be just the thing for his ratatouille. Some mechanism must be developed to account for these differences in taste--eggplants should be rationed to the Frog while I get hamburgers.
I have shown, very roughly and in a form I am sure which would send an econ major into apoplexy and a logician into fits, that rationaing of resources and labour is necessary. I think we can agree that labour is a resource, so I will treat it as such from here on out.
There are many schemes for rationing. We could have a first-come, first-served scheme: all the grain (or electricity, or jobs or workers) is in a big heap and whoever gets there first can take all he wants. Well, if he's smart he'll take it all, or as much as he can carry. He would then take that surplus and go down to the fellow who has cornered whatever it is that our grain-hoarder really wants (perhaps pictures of Miss Portman). He would then trade the useless (to him) grain for the useful (to him) pictures. He might collaborate with several such hoarders and produce a common unit to ease the chose of bartering. Thus, with the invention of money, we are back to the current situation, except that some people have everything and everybody else starves.
The above is a utopian model: producers produce and just give stuff away to the first comer, which as we see cannot work. Well, what about a socialist system, under which a benevolent commission rations everything? Unfortunately this does not work either. The commission cannot accurately know the myriads of ways in which resources may be valued. They can set the price of beef lower than it ought to be. This is the situation in the US: the government has subsidised the production of beef and thus more beef is produced (I do not believe it would be possible to satiate the American demand for beef; I know I sure love it). The natural limiting factors (expense of grazing land, cost benefit of farming vs. ranching) thus do not apply and the end result is that we have too many cattle--we are thus inefficient. Conversely, we tax alcohol at a rate designed to cut consumption. Thus less alcoh and we inefficiently produless alcohol thatn would be desirable. No commission can possibly comprehend the full interplay of factors: people's varying desires for thing; seasonally-affected variations; regional variations &c.
And, in a socialist set-up, the commissions tend to become Big Brothers. No-one wants to be a sewer worker--the commission must force somebody to do it. Few people desire to give 90+% of their earnings away to others, but the commission makes them do it. So this non-omniscient commission forces its ill-informed opinions upon the masses? Whither right? Whither justice?
The free market, while imperfect, is a way to fix all this. No-one is forced into anything, except through economic means. The stick is eschewed in favour of the carrot for the most part. Everyone is given the hope of improving his lot and the freedom to realise that hope. How does this work?
A free market assumes a multitude of producers and consumers. Everyone is both at one point or another; I produce labour, while I also consume all sorts of goods. Any producer may sell his product to any consumer at a price the two arrange between themselves. If my work is worth at least $10/hr. to me, and my work is worth at most $30/hr. to my employer, then we can settle on a price somewhere in between. If my work is worth $30/hr. to me and $10/hr. to my employer, then I can try to find another consumer to buy my labour. If there is no other customer to buy at that price, I had better consider lowering it.
This carries over into everything. Land costs a certain amount per acre, which is an aggregation of the value it has for all uses. Land in the middle of West Texas may be near-worthless for business as it is far from big cities, but it may be worth quite a bit as ranch land. Land in Kansas may produce $10,000/yr. in cattle or $15,000/yr. in wheat. Everyone is free to produce as he will; the intelligent man produces the most efficient goods given the resources at his disposal, which means that he makes the most possible money which means that he can buy the largest posible amount of what he desires.
To try to bring this back on topic, how does this relate to software? Well, software takes time to write. The three weeks I spend writing Bob's Cool Program are three weeks that I do not spend farming, or sweeping streets, or making cars, or otherwise making money. I convert three weeks of time into an artifact, a piece of source code. Compiling that code takes little time; thinking by time alone, it might make sense to give away the binary and sell the source. But most home users do not have access to compilers. So perhaps it's a better idea to sell the binaries and give away my source. But if I give away the source then someone else may take my three weeks' work, add one days', compile the thing and sell it, gaining me naught. This is related to the problem of the BSD License that the GPL seeks to address.
Of course, if I GPL my work then the situation is the same, except that I get that other fellow's one day of work. If I can get twenty people each to give me a days' work thusly, then I have put three weeks' work into something and gotten somebody else to put four weeks' into it. I now have a work worth seven weeks of labour for the low price of three weeks. Thus the GPL is economically sound, as long as I actually value that software in the manner I described.
What if I could have made $10,000 from my software but instead end up get $5,000 worth from the GPLed programme I released? Then I have made an economically unsound decision and lost money. I have made a decision which I may find morally more comfortable, but I am worse off for it. And as far as morality goes, I have wasted $5,000; the world is that much poorer because I didn't make money--I lack $5,000 I could have spent. Remember that value is actually created; when I trade two goats for your one cow, I am better off and you are better off. When I get paid $25/hr, I'm better off and my employer is better off. When I sell a product worth $500 to you for $100, we're both better off. If I give away a product worth $100 to 100 people and get naught in return, they are better off but I am worse off. I have hurt myself and they are now able to spend their money in an inefficient manner. Thus I have hurt the economy.
So one should try to examine the cost/benefit of GPLing software. Which is what software producers are doing. RMS, OTOH, believes that it is a moral imperative so to do. I think that I have shown above how it might actually be harmful.
Perhaps the old license was too restrictive in other ways, in order to ensure that only paying members got source? Or did it demand that Sun be assigend copyright on mods? Or was there some other valid complaint? Or was it simply griping that Sun dared release non-free software?
Free software is cool stuff. I write it, and GPL it. But I do not demand that everyone else GPL their stuff. Although I do think that software copyrights should be like patents: short term (say, two to five years); can be renewed once for an additional term; the source is on file; the source becomes available at the expiration of the copyright. This way people can make money for a few years on their work, but we still get the source in the not-so-long-run.
Of course, that can be fixed by rationing food. An approach which worked so well in WWII. That's the problem with socialism. It is not very realistic. Instead of people buying as they can, it relies on a Big Brother to parcel out the goods. Of course, Big Brother (or Uncle Joe...) likes to have as much as he can for himself.
People work on incentive; it's the only way. Free Software and Open Source (which are one and the same, no matter what His Highness thinks) only work because people have an incentive to write excellent software; they wish to scratch their itch. Take that away, and we have no more free software.
Very few people have an internal desire to clean toilets or plow fields. That sort of work scratches precious few itches (although it does cause a few). Yet, these things need doing. Thus the free market. It arranges for a fully-functioning society, unlike any other scheme yet proposed.
This brings up an issue about which I ahve been thinking for quite some time now, ever since I read the circumstances which gave birth to the FSF. Stallman wanted the source to a printer driver because it was buggy. He went from that starting point to argue that all source should be free. I have often thought that his ends could have been meat by stating that source and modification are free within the community of customers. That is, if I purchase a copy of Word I have a right to the source code, and a right to share my modifications to that source with others who have bought it. But I cannot give it away to someone else who has not paid for it.
This preserves the profit motive and makes it possible to write code whose source is open and whose bugs get fixed, while at the same time enabling authors to charge for what is, after all, their work and their expenditure of resources. It is true that eventually someone's modifications could be so great that he may want to charge for it; again, he would be able to do so. Let's call the original source A and he mods B. The fellow could sell B only to those who already own A, but could sell A+B to anyone, as long as he paid the original author his due for A.
I will admit that this could get out-of-hand if everyone wanted to charge even for slight mods. But again, that could be controlled--specifying that mods of less than 100 lines or something similar must be public domain, perhaps.
I do not believe that I have ever seen anything on the FSF site explaining why something along these lines (obv. not exactly this sort of thing) is not a good idea.
The world doesn't need new TLDs; it needs new third level domains...
This may actually be true, but somehow I doubt it. After all, what is to stop someone writing a programme which does certain things based on certain inputs? Is that not, after all, what _all_ programs do? So it is possible, by constructing certain inputs, to cause certain things to occur. From this, it is (in time and given the existence of bugs) possible to write a bootstrapper to then run a virus. Voila!
Were I not completely at a loss for free time, I might put some work into that...
That's what the Qt folks are saying and a certain radical fringe is trying to deny that truth. IMHO you are right, but not in the NSHO of Stallman and the FSF (both of which I do have respect for). Their claim is that libraries are just as much modules as are modules. I firmly disagree, but c'est la vie.
I quite agree with the RBL.
Mail from dial-up accts has a higher probability of being spam, granted. But with the marked increase in intelligent hosts (i.e. hosts which are properly configured and act as their own mail servers) installed in user's homes increases, this probability will decrease. I call on all right-thinking mail admins to Do the Right Thing and turn off the DUL.
I also call on the US Congress, acting under its authority to regulate inter-state commerce, to ban spam sent across state lines, and on the legislatures of the states to do the same for intra-state spam, and on foreign legislative bodies to do the same. A federal law declaring the sending of spam across our national borders an actionable act would be nice too:-)
The RBL is a good idea, one which I agree with.
The reason for standards bodies is that intelligent people realise that they cannot foresee all possibilities and consider all viewpoints. A standards body is formed to attempt to do this as well as possible (it obv. cannot do a perfect job of it...). Engineers & computer scientists realise the utility of standards bodies. If the vast moajority of the population does not, standards bodies will remain useful; the opinion of the ignorant majority means nothing to the standards body. All it means to anyone is that the unwashed masses may not follow the standard because they are too proud to know when to listen to someone else.
The W3C may not be listened to, but that makes its standards no less useful, and vertainly does not make Microsoft's any more so. All it means is that most people will follow the one and ignore the other. The ignored one may very well be better, and very prob. is in this case.
No offense, but I think that you misunderstood some of what he meant.
I believe that by `many files sharing one name' he meant within on directory (or functional group), i.e. one pathname, in much the way, say, that no single beer in the fridge has a name; they're all just `beer' and it doesn't matter which is which. I don't know if this would transfer to a computer well: we'd just make one big beer file.
While a file can have many names, using hard or soft links, actions on one name do not necessarily affect the others: delete one hard link and the others are still there (a useful function, true); delete a symbolically linked file and all the symlinks go insane (hardly useful).
A file in many directories is possibly different from a file with multiple names. Can't see it myself, but then I'm no Ph.D.
Non-named files would be those that are linked to in some other way. Photographs or such might be especially good in this sense. I know that many people have photo collections which give them many views on their libraries; the photos themselves are named in sequential order. Why name them at all? Why not jsut have a `photo chooser' which displays each image and its comments? Why make the named server as comments on a file? Because that's the way we've been doing it for 30 years?
I do not know how workable his ideas are, but OTOH we can see that the current approach has its shortcomings. Some work needs to be done, anyway.
He never said that Linux is obsolete; he said that it is old. There is obviously a great deal of benefit to be derived from this (we know where the bugs are, we have perfected it &c.), but the state of the art does need to advance.
He's not afraid of anything. He is creating. He's trying to see beyond the current, trying to forecast the future. He may very well be wrong. But many of his points sound as things used to be on the net before it was commercialised; I like to think that some of these will come true.
Try to read the articel next time.
Ummm, a lifestream was what he termed his replacement for files--nothing to do with software companies.
FWIW, I thought that much of what he described sounded like the Net of yesteryear, when clients for a multitude of servers proliferated and platforms did not matter nearly so much. And several ideas were interesting, although I do know how workable they might be (e.g. lifestreams). He is completely correct in one thing: the standard file structure we use is woefully inadequate to the millions-of-documents world in which we find ourselves. Obviously any system in the near future will be built on top of files, but we need to get beyond them for day-to-day work.
And yes, I'd rather say `I need the report I was working on yesterday' to my computer and get it.
OTOH, there is a great deal of competition between Linux and Windows on the one hand, and Mac OS and windows on the other. Windows cannot be a good server platform, yet it cannot be a good desktop platform. It is a bastard, attempting all and perfecting none. The Macintosh has perfected the desktop; I seriously doubt that we in the Linux world will ever have anywhere near as coherent, elegant, refined, intuitive and all-round cool as the Mac interface. Unix in general, and Linux in particular, has mastered the server realm. I seriously doubt that we Mac users will ever have a platform anywhere near as stable, expandable, programmable, scalable and all-round cool as the Linux platform.
Windows addicts will carry on in their folly, and we in both camps will laught at their misfortune.
That said, both platforms can get much better. WTF's up with Aqua? Linux is still not IMHO serious competition for Solaris, HP-UX or AIX. All things come to those who wait...
How exactly is what I posted flamebait? Sure, I insulted an AC, but certainly with reason: his post was mere assertion apparently uninformed by reading the article. I pointed out that the article itself addressed his points.
I'm quite willing to take moderation based on objective facts, but being modded down solely due to having expressed an unpopular opinion is a bitter pill to swallow. I know that when I moderate I take the job seriously. Generally I only mod down people whom I agree with, but whose method of posting is indelicate, for that is the only way that I can be objective.
Insulting an AC's mental skills is hardly flamebait, esp. when his post is obviously ignorant and he has not read the article referred to.
It was a highly amusing incident in which a Greenpeace fleet was going to `protest' (read: interfere with) nuclear tests. The French sank it. Legally, an act of self-defence against an attacker. Very funny IMHO; that sort of group (left- or right-wing) is just too used to a lack of resistance. People ought to exercise their right to fight back more often.
Actually, that sounds quite fair. We would have the patches; we could write a free library to implement that API. It's not as though it is some undocumented API; we would have the code, and thus have all the docs we need.
So what is the difference, exactly, between a command and a function in a shared library? Both implement an API; the one in a shell, the other in some other language.
And the bit about ls is exactly my point. It's an implementation of an API. Can non-GPLed work use the GNU extensions to ls?
Allow dynamic linking; disallow static linking. Static linking defeats the ability to rebuild th binary, but dynamic linking allows any library in the background. If we wanted we could write a GPLed Qt-alike.
I know it might sound silly trying to obtain a support contract for Tripwire, but at the last company I worked for, such a thing would not only be desired, but not too terribly hard to get upper management to sign off on.
Many times the IT department or its outsourced equivalent has Service Level Agreements; I know we do. They're 24/7, so we buy 24/7 support for everything we have. As networking and systems continue to rise in importance, I expect that we will really see a boom in the support market. OS/FS are poised to really take advantage of this.
Do you know what it does? It calculates checksums for all files. These sums can then be stored on read-only media, such as a CD. Then a simple check is all that is nec. to detect modifications to system files.
Minitel had some interesting features, but it was never as good as the Internet, which predated Minitel by over a decade. And I hate to say it, but London and New York will remain the financial centres of the world, with Silicon Valley and Germany doing just fine, thank you, in the technology end of things. France has a nice position in the world, but it will never be the `technological and financial centre.' It has its time: the entire Mediæval, Rennaisance and Early Modern periods. The mantle has passed.
What you're missing is that there are many people who would prefer the GPL to be viral, not in the usual sense that it's accused, but in the sense that it infects anything which touches it. Is it illegal for a proprietary app to call a GPLed ls? Of course not; that's utterly absurd. So why do some think that dynamically linking to GPLed software should be illegal? Because they are fanatics.
I appreciate the GPL; I dislike proprietary software. But I am willing to co-exist. As long as my code remains free and open, I am happy. If my code is dynamically linked against but the source is included, I am happy. Free software will win in the end, for the most part. I'm sure that there will always be some proprietary software. Big deal. I needn't exterminate my enemy, but simply defeat him.