Then you wouldn't be able to taste it, which apparently is the whole point. If all you wanted was nutrition by pumping into the stomach, you wouldn't use liquefied hamburgers.
Unless - you could chew on the hamburgers while having something else piped into your stomach. The bits of hamburger you swallow get diverted away from your throat and straight down a waste pipe, where they are compressed and used as building materials.
Splitting Microsoft into two, three or N separate companies, each with control over a particular area, is not the answer. It would just create a number of smaller monopolies in particular areas.
For example, the OS group would still be able to lock out competitors by fiddling with Kerberos. The applications group would still keep Word's overwhelming market share and its proprietary file formats, and keep changing them to keep out competitors. The 'Internet' group could still break standards for HTML, messaging and so on.
In many cases the individual products such as Office have just as much a monopoly as Windows itself. And even when the market share is smaller, the users are often locked in and unable to switch. The key issue here is *competition* - the three parts of Microsoft would not compete with each other, so nothing would improve. They might not be able to tie applications to the OS quite as much, but still the OS group will be able to control application development by deciding which standards not to support (eg, dropping or breaking Java). And so on.
I'm not sure what the best answer is, but how about splitting Microsoft into several companies, each of which has rights to _all_ Microsoft programs. Then we'd start to see some real competition among Windows vendors (rather than the single vendor as at present), and probably some big price drops for consumers too.
You might worry that this would let to fragmentation and incompatible solutions, but let the market take care of that. People won't buy and use an OS if it doesn't behave well towards competing products. It's only where you have monopolies that this mechanism cannot work.
An alternative would be to systematically auction off rights to one program after another, with the rule that a minimum of three independent bidders get the rights to each item. After that, you keep handing out rights to the highest bidder remaining until the highest bid is 'stop' (which would be the winners so far bidding money to try to stop others from climbing on board). This hopefully would get a reasonable balance between ensuring competition, and raking in lots of money. The money raised would compensate Microsoft shareholders.
So what's the difference between a web site being shut down because it carried information the government didn't like, and being shut down because it carried information some large company didn't like?
One problem is that people such as ISPs and governments may block ports used for ssh. What I'd like to see is a way to transparently tunnel all IP traffic across https.
In other words, when host A wants to send a packet to host B, it makes an https connection to B (if one isn't already open) and sends the packet along that. At the other end, B interprets the packet as coming from some special 'crypto' network interface, and handles it just as if it had come from the network card or modem.
The advantages of doing this would be that ISPs wouldn't want to block https, since it is used for ecommerce. Likewise governments. And because https is encrypted, there's no easy way to tell that you're engaging in subversive activities (eg encrypted telnet) rather than approved activities which involve buying lots of stuff on the net. (please bear in mind that this whole post has been run through a conspiracy-paranoia filter.)
Also, it could be totally transparent to the user; if such a feature got put as standard into the Linux kernel (for *example*), traffic between Linux boxes would form a sort of 'cryptobone' (!) while communications to other OSes would proceed as normal.
Do IBM not make x86 chips any more? They used to have their 'Blue Lightning' clock-tripled 486 clone, and later they manufactured Cyrix's stuff. But nowadays, they don't seem to be making 686es. That's a shame, since all the cool stuff they're doing for other architectures (copper, SOI) might benefit the PC world too.
In some cases, people have been known to answer the telephone while the radio is playing, thus ripping off the work of hard-working performers and stealing from the record companies. It's just like walking into a store and taking records of the shelf without paying.
This sort of theft cannot be allowed to continue; we'd better ban telephones at once, or at least replace them with an industry-approved version that cuts off after thirty seconds, and has region coding to stop music from being heard over the phone in countries where it hasn't yet been released.
Cool, now we can have lots of new laws lobbied for by Penthouse magazine. Perhaps they will press for censorship of Usenet to filter out all the filth and smut that corrupts our children and damages Penthouse's business.
Alternatively, every model used by Penthouse could have a barcode tatooed onto her buttocks, so that pirated pictures can be easily and automatically recognized.
When copyright was first introduced (in England and the USA at least, AFAIK), the term was fourteen years, with a possible renewal for another fourteen years during the author's lifetime.
Since then, the world has become faster and the pace of change has increased. News is now minute-by-minute rather than week-by-week, technologies come and go in years rather than decades, the popularity of music waxes and wanes far faster than it did then. Yet the copyright term has become longer and longer, for some reason. It now ranges from life plus fifty years, to life plus ninety years. That's an effective copyright term of a century and a half in some cases - totally unjustified by any substantially increased incentive to create new works.
Unfortunately, international treaties prevent countries from adopting more sensible policies, even if they wanted to. Or rather, it means that greater political will is required to get past the 'hiding behind our international "obligations"' stage. (IMHO intellectual property law, like taxation, is best decided individually on a country-by-country basis.)
I reckon that the copyright term will be put right during my lifetime; if it ever became an 'issue' it would certainly provoke a lot of support for a reduction to the original 14 years, or some other reasonable term. Unfortunately, the large media conglomerates such as Disney are those who keep pushing for another twenty years every decade or so, so it's unlikely that any of the mainstream media would start actively campaigning on this issue. But sooner or later, it is bound to get attention.
Regular renewal sounds a bit bureacratic and could be an excuse to favour those who can afford the lawyers and paperwork. Having to prove (ultimately before a court) that your work is still worth copyrighting could bog the whole world down in endless appeals and wrangling. Better to have something which is clear-cut and automatic; such as, if the author is alive you can extend the copyright up to a maximum of 28 years; if a book is out of print, however, copyright expires after 10 years.
Linux support for MCA bus machines has been around for _years_ and is stable and reliable. It's just that it only recently got into the main kernel tree. But now you can build your 2.2 kernel with MCA support and it Just Works. A lot of the MCA hardware such as ESDI, SCSI, XGA and network cards is supported too.
AFAIK FreeBSD also has working MCA support. It's good to see NetBSD getting it too - at about the same time as it's being dropped from Windows NT. (Win2k doesn't support MCA bus machines.)
If you have a PS/2 lying around in the cupboard, you have no excuse for not sticking Slackware on it right away. Also I'd like to insert a plug here for the newsgroup comp.sys.ibm.ps2.hardware, which is cool.
Here's some comments on the article, in true Usenet style made by interspersing quotations with responses to them.
3. THE ECONOMICS OF FREE SOFTWARE
... In practice the only possible cases are the following: - The software developer may have a personal fortune... - The software developer may have other sources of income, paid by an employer or client as compensation for services unrelated to the free software.... - Many public institutions such as universities will release for general use most of the software developed by their employees (although, as universities around the world are being pressed by the purse-string holders to enhance their economic value, and recognize the economic potential of the software they develop, this generous attitude is not as universal as it used to be).
Interesting mistake here - the economic potential of the software developed is usually _reduced_ if it is distributed in binary form, to only a small number of people. It would be more economically useful if everyone could use and improve it. There is the situation where a third party might have the money to improve the software and 'productize' it, but that does not require that the third party need be given a monopoly on the software. Economically, it would be more useful to release the software under a BSD-type licence and get some competition among the companies building on it.
I expect that rather than 'economic value of the software', Bertrand Mayer means 'monetary value to the university', which is not the same thing. Anyway, let's go on:
- Companies may find it beneficial to release some of their software products without asking for a fee....
The categories identified here -- donated, taxpayer-funded, privately funded, taxpayer-sponsored and privately-sponsored -- seem to exhaust the economic possibilities;
I think a category has been left out. This is the case where companies develop software and sell it for a fee, but decide to make it free software. There are plenty of people doing this - Red Hat is the most obvious example, they charge for their Linux distribution.
Then there are the cases where the company makes money from technical support, rather than selling copies. Cygnus was for many years the prime example of this, now they're owned by Red Hat. People like AbiSource or Helix Code intend to make money from custom enhancements to the software they write.
Many of the contributions of the free software community are admirable. Highly disturbing, however, is its widespread slander and hatred of the commercial software world.
One man, however high-profile, is not the same as 'widespread'. Most free software advocates do not agree with RMS, as far as I can tell. Indeed Eric Raymond, mentioned at length later on in the article, explicitly rejects Stallman's views on this. But anyway, RMS would reject your view that free software is anti-commercial per se, eg in this LinuxWorld interview:
When people said, "Don't pour poison in the river," they were called communists. But they didn't want to abolish business. They wanted to abolish pouring poison into the river. The free software movement is a lot like that. It's a lot like the environmental movement because the goal is not to abolish business, the goal is to end a certain kind of pollution. But in this case, it's not pollution of the air or the water, it's pollution of our social relationships.
Note again that when RMS says 'the free software movement', he doesn't refer to everyone who supports free software. Back to the article:
Nowhere in the hundreds of pages of GNU and FSF literature is there any serious explanation of why it is legitimate, for example, to make a living selling cauliflowers, or lectures (as a professor does), or videotapes of your lectures, but criminal to peddle software that you have produced by working long hours, sweating your heart out, thinking brilliantly, and risking your livelihood and that of your family.
There are plenty of explanations of the FSF's view on this. For example, see Selling Free Software. The idea is that when selling cauliflowers, you don't impose restrictions on what the purchaser can do with those cauliflowers. You don't make them give up any freedom (in the GNU sense). If you buy potatoes, you can plant them and grow more potatoes. (The closest analogy to the way you're not allowed to copy proprietary software is a genetically modified crop where the farmer is forbidden from saving some and planting it next year.)
To take your professor analogy, it would be okay (in FSF terms) to charge money for a lecture. But it would not be acceptable to stop your students passing on the knowledge they had learnt. RMS has said as much:
We see programmers as providing a service, much as doctors and lawyers now do--both medical and legal knowledge are freely redistributable entities for which the practitioners charge a distribution and service fee.
So the objection is not to 'selling' but to stopping the purchaser from changing and sharing what he has bought. The question then becomes, what if copyright is the only way to ensure a reasonable selling price for the author? RMS would prefer that the software never be written at all (work as a waiter instead), but personally I cannot agree with that. More from the article...
The only stated justification for the indictment of commercial software [apart from nostalgia] - is that software is different from other wares since it can be reproduced so easily. But this does not stand a minute's scrutiny. The difference is a matter of degree, not nature; software reproduction always costs something, even if it is as little as a dollar for a CD, ten cents of network connection time for an Internet download, or the marginal cost of using up more memory. With a good scanner or photocopier, you can reproduce a book, too, for very little money these days.
As the FSF site says over and over again, price is not the issue. Of course zero marginal cost is not a good enough reason for zero price - not in a capitalist society anyway. Have you seen the prices that the FSF charges for its software? It's bloody expensive. The question is - does the extra benefit to users from copyright on software outweigh the disadvantages? Fifty years ago, copyright on books was not a major restriction on people's actions, since copying a book would be difficult anyway. But for something that is naturally easy to copy and change such as source code, the restrictions placed by copyright law are more onerous. I think the loss of freedom is worth it in order to get more software produced, but it's not an open and shut case.
The article then goes on to criticize RMS for a 'skewed moral perspective', use of extreme analogies, and accusing the free software movement of hijacking the word 'free'. But this is a little unfair. Since the FSF concerns itself only with software, it's not surprising that the word 'freedom' on their pages is used only as it relates to computers. RMS is the first to admit that proprietary software is not the only problem the world has, or even the most serious. If every FSF page were prefixed with a disclaimer saying 'this is less important than other moral issues in the world', that would avoid accusations of a 'skewed moral perspective', but what would be the point? It's a mistake to assume that somebody concerned about one issue is a single issue proponent. It might just be that they don't have time to deal with everything, and have decided to focus on one specific area. (I do agree about the analogies getting a bit out of hand sometimes, but most of them work quite well.)
And the use of Eric Raymond to try to criticize free software proponents as a whole is even sillier. He may be a bit of a nut, but remember that unlike RMS, ESR is _not_ in the business of making moral proclamations, at least not in the software area. He does his best to make a practical case for free software - or 'open source software' as he calls it to avoid frightening managers. Bertrand Meyer seems to be arguing:
Eric Raymond supports free software
Eric Raymond supports gun ownership
Gun ownership is a bad idea
Therefore, the whole free software movement is a bit wacky
The article asks:
Is it right, one might ask, to make a connection between Mr. Raymond, who is only one person, and the rest of the free software community? The answer is yes, for at least three reasons:
- His propaganda is prominent in his Web pages...
And? That still doesn't mean that it represents the whole free software movement. ESR doesn't even claim that his 'Open Source' writings represent the whole free software movement, let alone his barrel-of-a-gun writings. Some free software advocates support gun ownership, some don't. (I don't.) The fact that one person has written a web page and managed to get a large number of hits on it is neither here nor there.
- Eric Raymond has been one of the most visible proponents of the Open Source movement... his views, unless disavowed strongly and publicly, inevitably commit the rest of the movement.
That is a bizarre statement. This is not a political party where people are expected to follow the party line. It's not some corporation where any press release represents official company policy. The 'movement' is a lot looser and a lot harder to pin down that Mr Meyer seems to think. There is no requirement on somebody who supports free software to publicly disavow anything that ESR says. His gun ramblings should be treated as what they are, a quaint irrelevance.
Given the choice between
- a society where all software would be proprietary, and civilized measures would be in place preventing (for example) a disturbed white supremacist from buying a police gun... - a society where all software would be free and Mr. Raymond's views on gun "freedom" were fully realized, any ethically-conscious person would choose the former.
I don't see how this has any relevance. Especially since ESR does not want to remove copyright on software, and since the free software movement has nothing to do with gun advocacy - a couple of oddly-placed links on one guy's Web page notwithstanding.
we have had to cancel one major project, and reengineer a product completely, after wasting many person-months and disappointing customers, because of the deficiencies of two separate GNU products (the GCC compiler for Windows and the editor under GTK). In both cases the scenario was the same: fixes to well-known bugs being promised and promised again; everyone waiting for months and months, until it becomes clear that nothing will happen; in the end, having to write off all the affected developments. Since no one is in charge, and you didn't pay for the products, there is no one to blame.
But 'nobody in charge' and 'didn't pay' are not consequences of the software being free. There are companies more than willing to take your money in exchange for providing a guaranteed response to bug fixes. The difference is that with free software, you can shop around and get the best deal for such support, rather than being limited to the company which owns the copyright.
Even though the GNU products are often good, the licenses which accompany them are no better, in the warranties (or rather absence thereof) they offer to the user, than commercial software.
Of course not. Would you expect the FSF to become liable to warranty claims from someone who downloaded the software gratis from an FTP site? However, the GPL explicitly allows someone distributing the software to 'offer warranty protection in exchange for a fee'. It's up to the free market to provide this service for those who can pay for it.
Warranty provision has nothing to do with 'freedom' - unless you count as 'freedom' being able to drag a software author through the courts because you downloaded his software and it didn't work. Again, if you want to provide a warranty for GCC as a profitmaking business, go ahead and do so. But saying 'use this at your own risk' is very different to saying 'you may not change the software; you may not copy the software'. One restricts what the user can do; the other is just an arse-covering legal measure - unfortunately necessary in today's litigious world.
Product F is free software. It comes with the standard no-warranty warranty.
Product P is proprietary software. It costs $50 for the binary-only version. [and comes with a full and comprehensive warranty] ... I would consider the second solution more ethical
I don't see how the first is unethical, provided it is absolutely clear about there being no warranty. Making false claims would be unethical of course. I don't see how you can expect Product F to provide a warranty if you did not pay anything for it - the author could be bankrupted by lawsuits from users who downloaded a copy or got one from you. There is a third option of course, Product F for $50 _with_ a warranty provided by the author.
I think the point is that the GPL doesn't _automatically_ grant a warranty. But it does provide a base on which you can provide warranties to specific individuals, if they pay for it. That seems sensible to me.
The free software axioms hold, as we have seen, that although charging for software is wrong it is all right to charge for services associated with the software, such as maintenance and training. The risk here is that such an attitude may lead to products with known deficiencies, giving the provider a ready-made source of juicy service contracts.
I don't think that any company could get away with this - how would the product establish itself in the market if it were broken? Don't forget that this problem exists with proprietary software too, if it has a support contract. Do you think that Oracle insert bugs so that their customers will subscribe to the most expensive support contracts for guaranteed fixes? It's all rather far-fetched. If you have the source, you'd probably be able to spot if something fishy were going on.
The article continues by accusing free software advocates of character asassination towards proprietary software developers. Again it makes the mistake of equating RMS with 'the free software movement'. There is no collective view on such things, any more than on gun ownership.
Finally, I think that most free software projects do indeed acknowledge where they have used ideas from other programs, free or proprietary. The GIMP's credits section describes it as a 'Photoshop-like' program, and that's about as clear as you could be. Similarly it is obvious where GNUstep, bison, less and so on got their inspiration.
Well.. unfortunately it's not that simple any more. Although the European Patent Convention specifically excludes software from being treated as a patentable 'invention', the European Patent Office ruled that this didn't mean what it said, and computer programs can be patented after all. It just goes to show that patent offices will try as hard as they can to increase what is patentable, without concern for whether such patentability is desirable. The EPO seems no more trustworthy than the USPTO. Following the EPO's decision, the UK patent office has revised its practice on software patents.
Anyway, it is still not certain whether such patents granted are actually enforceable. The European Commission are considering changing the law to catch up with the EPO's behaviour, so that software patents will become officially permitted (rather than de facto permitted as at present). Until then it isn't certain whether a lawsuit based on a European software patent would succeed (IANAL); big companies such as IBM are however stocking up on software patents in the hope of one day being able to use them. (Have a look at freepatents.org for more about stopping this from happening.)
Unisys claim to have a patent on LZW in several European countries, including the UK. I don't know whether they've actually tried to shake down British companies for money though. But you may not be as safe as you think.
Something to try would be to send another threatening letter to Oxford, pretending to be the MPAA. Claim that their discussion of why they pulled DeCSS infringes copyright / trade secrets or some other bullshit, and demand they remove it. It would be interesting to see just how ridiculous the demands have to be before they'd start to question them.
You could take this further; start writing to universities, ISPs and small companies up and down the land, with letters written in threatening lawyerese demanding that they pull particular Web pages. There could even be a 'form letter' program that puts something together from a set of stock phrases, adds a convincing-looking letterhead, and prints it out to be signed and posted. 'DIY Internet Censorship Kit' or something.
All this would be fairly pointless, unless it managed to generate a public outcry over freedom of speech and censorship. (I'm thinking of the UK here, but there's no reason it shouldn't apply to other countries.)
Some sort of bionic-chroot-on-steriods might be the answer for running untrusted binary-only software. FreeBSD has jail() which is like an improved version of chroot(), but what's needed here is something more sophisticated.
Ideally you'd make it so that _all_ access to the outside world can be filtered through some userland process. Preferably a Perl program, hehehe. This is roughly what I have in mind:
filter_syscall() { if (call is 'open file for reading') { if (filename one of those allowed) { return OKAY; } else { modify the 'open' call so the mode is set to read-only; return OKAY; } } else if (call is 'read') { return OKAY; } else if (call is 'send data over network') { check to see if the data is being sent back to RealNetworks - if it is then return FAIL_SILENTLY;
else return OKAY; } else { /* Not recognized */ warn("program is doing systemcall we haven't seen before"); return FAILURE; } }
It could be quite a lot of work to do this _fully_ for any large program, but a quick hack to allow all system calls except those sending private data, or to overwrite any private data being sent with 'X' characters, might be quite easy.
I think you can get S/390 cards for your PC. You won't get the mainframe reliability - which is the whole point of having a mainframe - but you will get a good chunk of the speed.
Even an S/390 emulator might run Linux at reasonable speeds, if you run the emulator on a modern PC or workstation.
A cautionary tale of Web design
on
Boo No More
·
· Score: 5
IIRC, the original boo.com site was overblown with too many images, JavaScript, probably Flash and lots of other junk. Nobody could use the site and within a few weeks they had to redo it as something simpler.
Now the company has gone bankrupt. Could this be related to customers being driven away by their early, over-flashy Website?
The tale of boo.com might be a useful weapon when trying to persuade your customers that JavaScript rollovers, MIDI files and Flash are not the last word in sophisticated Web design.
I'm beginning to think that the law needs to be made clearer, not by precedent-making lawsuits, but by legislation. Yes, I know that most legislation relating to the 'Information Age' (or 'Digital Millennium') has been stupid, misguided or downright corrupt - but for the sake of argument, imagine for a moment that legislatures in whatever country were capable of making intellectual property law to benefit consumers.
Explicity define what is fair dealing for works which are made electronically available, via the Web or otherwise. I would say that the following should be allowed:
- Saving copies of a file to local disk, to look at later. (Like timeshifting a TV programme.)
- Printing out copies for personal use. (If you can read it on screen, why not on paper?)
- Distributing copies to others, provided those other people would themselves be able to get a copy, if they could be bothered to. Thus if you save yahoo.com/index.html on disk and give it to a friend, that's explicitly allowed. And if both of you subscribe to the same paid service (eg stock quotes) it should be allowed for you to pass on information which your friend could access, if his modem weren't broken. He pays the fees too, after all.
- Caching and indexing; such labelling and modification of the document as is necessary to store it in a cache or index it efficiently. Returning the document or part of it, perhaps highlighted, as the result of a search.
- Hyperlinks are allowed. This is a no-brainer, but just as well to get it in there explicitly.
- Archives of historical documents and files which were publicly available. There might have to be exceptions in here for libellous material and other nasties, but in general I don't believe that copyright law should be a tool for suppressing leaks.
- Finally, translation and parody, as long as it is made clear that the content has been modified in this way, and of course the people who are given the modified copy could also have gotten the original copy for themselves.
Now, I realize that some of these are probably already allowed by copyright law in most countries. Perhaps they all are. But I'd rather have these things laid down with certainty; uncertainty about what the law says makes anyone vulnerable to an expensive lawsuit from a firm with deep pockets. They might just win, or at least manage to bankrupt you before they lose.
I don't think that any of the above would reduce the incentive for authors and publishers to create new works. And that is the purpose of copyright law, with the intent of benefiting the public. The Web could become a very unpleasant place with legally enforceable bits of JavaScript preventing you from printing pages, people being sued over hyperlinks, and other nastiness. Some of this is happening already.
There probably isn't a snowball's chance in hell of any legislative body working out fair dealing for the Web any time soon. But it's another thing to add to the wishlist.
Yes, I've written such a program myself. (Try to write it as the shortest possible shell script!) But some method based on meaning - or at least on something a bit more sophisticated than word frequency - might produce better results. Eg you could ignore not only stopwords such as 'and' and 'the' but also hackneyed phrases such as 'if you thought that X, think again'.
What I don't understand is why make the secondary cache smaller than the primary cache?
If a memory location is used frequently enough to be stored in the secondary cache, it will surely also be in the bigger primary cache, which is faster and looked at first. Thus the secondary cache is effectively useless, except maybe as a write-back buffer. Maybe this is a mistake in the article, or I've not understood?
(BTW wouldn't it be cool to write software that fits entirely in the primary cache - 128Kbyte used to be considered a lot! If there were some way to initialize the cache at boot time, you wouldn't need RAM at all:-)
I realize that neither of these is particularly original, but then neither have they been perfected by anybody; there is still lots of room for improvement, and plenty of scope for adding as much natural-language AI as you like:
A precis program which will condense long websites or discussions. Jon Katz's articles on Slashdot might make a useful set of tests, also try to precis the comments posted to them:-) Seriously though, it would be useful for busy people who haven't got time to read things from top to bottom. It could have some way of determining which links are relevant, following them, and adding their information to the precis generated (with attributions if necessary). Or you could use it to cut out types of information you don't want - to filter out marketingspeak, if you're a techie, or to filter out techspeak, if you're not.
The other suggestion is a remembrance agent which looks at the website you're reading and suggests links (from your browser history, from search engines or from some big collaborative database) which might be relevant. This might finally be a use for those sidebars that recent browsers seem to have spouted. Again, this has been done before, but it's not something which has been done perfectly. You might also be able to use it as a fact-checker for postings you make to Usenet - although that would be rather difficult to implement, I imagine.
Slashdot Mirror
Then you wouldn't be able to taste it, which apparently is the whole point. If all you wanted was nutrition by pumping into the stomach, you wouldn't use liquefied hamburgers.
Unless - you could chew on the hamburgers while having something else piped into your stomach. The bits of hamburger you swallow get diverted away from your throat and straight down a waste pipe, where they are compressed and used as building materials.
I thought they chopped the heads off. Crushing skulls would splatter brains everywhere, which is probably a Bad Thing in these BSE-paranoid days.
Splitting Microsoft into two, three or N separate companies, each with control over a particular area, is not the answer. It would just create a number of smaller monopolies in particular areas.
For example, the OS group would still be able to lock out competitors by fiddling with Kerberos. The applications group would still keep Word's overwhelming market share and its proprietary file formats, and keep changing them to keep out competitors. The 'Internet' group could still break standards for HTML, messaging and so on.
In many cases the individual products such as Office have just as much a monopoly as Windows itself. And even when the market share is smaller, the users are often locked in and unable to switch. The key issue here is *competition* - the three parts of Microsoft would not compete with each other, so nothing would improve. They might not be able to tie applications to the OS quite as much, but still the OS group will be able to control application development by deciding which standards not to support (eg, dropping or breaking Java). And so on.
I'm not sure what the best answer is, but how about splitting Microsoft into several companies, each of which has rights to _all_ Microsoft programs. Then we'd start to see some real competition among Windows vendors (rather than the single vendor as at present), and probably some big price drops for consumers too.
You might worry that this would let to fragmentation and incompatible solutions, but let the market take care of that. People won't buy and use an OS if it doesn't behave well towards competing products. It's only where you have monopolies that this mechanism cannot work.
An alternative would be to systematically auction off rights to one program after another, with the rule that a minimum of three independent bidders get the rights to each item. After that, you keep handing out rights to the highest bidder remaining until the highest bid is 'stop' (which would be the winners so far bidding money to try to stop others from climbing on board). This hopefully would get a reasonable balance between ensuring competition, and raking in lots of money. The money raised would compensate Microsoft shareholders.
So what's the difference between a web site being shut down because it carried information the government didn't like, and being shut down because it carried information some large company didn't like?
This robot is only one part of the picture. What's needed is:
Robotic cow-rearing units with automated grass planting
Fully automated cow-killer
Robot which chops up the cows into beef
Robot to grind the beef and press it into burgers
Robot to cook the burgers (ie, this one)
Serving machine in drive-through fast-food outlet
Self-driving cars which take customers automatically to McDonald's
Bionic jaw muscle attachments, so you don't have to spend effort chewing
Computer-controlled automatic toilet on wheels that follows you around to be ready when you need a dump
Automated sewage disposal (I think we have this).
One problem is that people such as ISPs and governments may block ports used for ssh. What I'd like to see is a way to transparently tunnel all IP traffic across https.
In other words, when host A wants to send a packet to host B, it makes an https connection to B (if one isn't already open) and sends the packet along that. At the other end, B interprets the packet as coming from some special 'crypto' network interface, and handles it just as if it had come from the network card or modem.
The advantages of doing this would be that ISPs wouldn't want to block https, since it is used for ecommerce. Likewise governments. And because https is encrypted, there's no easy way to tell that you're engaging in subversive activities (eg encrypted telnet) rather than approved activities which involve buying lots of stuff on the net. (please bear in mind that this whole post has been run through a conspiracy-paranoia filter.)
Also, it could be totally transparent to the user; if such a feature got put as standard into the Linux kernel (for *example*), traffic between Linux boxes would form a sort of 'cryptobone' (!) while communications to other OSes would proceed as normal.
I don't want to Ask The Man, I want to Stick It To The Man!
Do IBM not make x86 chips any more? They used to have their 'Blue Lightning' clock-tripled 486 clone, and later they manufactured Cyrix's stuff. But nowadays, they don't seem to be making 686es. That's a shame, since all the cool stuff they're doing for other architectures (copper, SOI) might benefit the PC world too.
In some cases, people have been known to answer the telephone while the radio is playing, thus ripping off the work of hard-working performers and stealing from the record companies. It's just like walking into a store and taking records of the shelf without paying.
This sort of theft cannot be allowed to continue; we'd better ban telephones at once, or at least replace them with an industry-approved version that cuts off after thirty seconds, and has region coding to stop music from being heard over the phone in countries where it hasn't yet been released.
Yes, my post was meant as a joke.
Cool, now we can have lots of new laws lobbied for by Penthouse magazine. Perhaps they will press for censorship of Usenet to filter out all the filth and smut that corrupts our children and damages Penthouse's business.
Alternatively, every model used by Penthouse could have a barcode tatooed onto her buttocks, so that pirated pictures can be easily and automatically recognized.
What's this whole 'octane' thing anyway? It can't have much to do with the proportion of octane molecules.
When copyright was first introduced (in England and the USA at least, AFAIK), the term was fourteen years, with a possible renewal for another fourteen years during the author's lifetime.
Since then, the world has become faster and the pace of change has increased. News is now minute-by-minute rather than week-by-week, technologies come and go in years rather than decades, the popularity of music waxes and wanes far faster than it did then. Yet the copyright term has become longer and longer, for some reason. It now ranges from life plus fifty years, to life plus ninety years. That's an effective copyright term of a century and a half in some cases - totally unjustified by any substantially increased incentive to create new works.
Unfortunately, international treaties prevent countries from adopting more sensible policies, even if they wanted to. Or rather, it means that greater political will is required to get past the 'hiding behind our international "obligations"' stage. (IMHO intellectual property law, like taxation, is best decided individually on a country-by-country basis.)
I reckon that the copyright term will be put right during my lifetime; if it ever became an 'issue' it would certainly provoke a lot of support for a reduction to the original 14 years, or some other reasonable term. Unfortunately, the large media conglomerates such as Disney are those who keep pushing for another twenty years every decade or so, so it's unlikely that any of the mainstream media would start actively campaigning on this issue. But sooner or later, it is bound to get attention.
Regular renewal sounds a bit bureacratic and could be an excuse to favour those who can afford the lawyers and paperwork. Having to prove (ultimately before a court) that your work is still worth copyrighting could bog the whole world down in endless appeals and wrangling. Better to have something which is clear-cut and automatic; such as, if the author is alive you can extend the copyright up to a maximum of 28 years; if a book is out of print, however, copyright expires after 10 years.
Linux support for MCA bus machines has been around for _years_ and is stable and reliable. It's just that it only recently got into the main kernel tree. But now you can build your 2.2 kernel with MCA support and it Just Works. A lot of the MCA hardware such as ESDI, SCSI, XGA and network cards is supported too.
AFAIK FreeBSD also has working MCA support. It's good to see NetBSD getting it too - at about the same time as it's being dropped from Windows NT. (Win2k doesn't support MCA bus machines.)
If you have a PS/2 lying around in the cupboard, you have no excuse for not sticking Slackware on it right away. Also I'd like to insert a plug here for the newsgroup comp.sys.ibm.ps2.hardware, which is cool.
Interesting mistake here - the economic potential of the software developed is usually _reduced_ if it is distributed in binary form, to only a small number of people. It would be more economically useful if everyone could use and improve it. There is the situation where a third party might have the money to improve the software and 'productize' it, but that does not require that the third party need be given a monopoly on the software. Economically, it would be more useful to release the software under a BSD-type licence and get some competition among the companies building on it.
I expect that rather than 'economic value of the software', Bertrand Mayer means 'monetary value to the university', which is not the same thing. Anyway, let's go on:
I think a category has been left out. This is the case where companies develop software and sell it for a fee, but decide to make it free software. There are plenty of people doing this - Red Hat is the most obvious example, they charge for their Linux distribution.
Then there are the cases where the company makes money from technical support, rather than selling copies. Cygnus was for many years the prime example of this, now they're owned by Red Hat. People like AbiSource or Helix Code intend to make money from custom enhancements to the software they write.
One man, however high-profile, is not the same as 'widespread'. Most free software advocates do not agree with RMS, as far as I can tell. Indeed Eric Raymond, mentioned at length later on in the article, explicitly rejects Stallman's views on this. But anyway, RMS would reject your view that free software is anti-commercial per se, eg in this LinuxWorld interview:
Note again that when RMS says 'the free software movement', he doesn't refer to everyone who supports free software. Back to the article:
There are plenty of explanations of the FSF's view on this. For example, see Selling Free Software. The idea is that when selling cauliflowers, you don't impose restrictions on what the purchaser can do with those cauliflowers. You don't make them give up any freedom (in the GNU sense). If you buy potatoes, you can plant them and grow more potatoes. (The closest analogy to the way you're not allowed to copy proprietary software is a genetically modified crop where the farmer is forbidden from saving some and planting it next year.)
To take your professor analogy, it would be okay (in FSF terms) to charge money for a lecture. But it would not be acceptable to stop your students passing on the knowledge they had learnt. RMS has said as much:
So the objection is not to 'selling' but to stopping the purchaser from changing and sharing what he has bought. The question then becomes, what if copyright is the only way to ensure a reasonable selling price for the author? RMS would prefer that the software never be written at all (work as a waiter instead), but personally I cannot agree with that. More from the article...
As the FSF site says over and over again, price is not the issue. Of course zero marginal cost is not a good enough reason for zero price - not in a capitalist society anyway. Have you seen the prices that the FSF charges for its software? It's bloody expensive. The question is - does the extra benefit to users from copyright on software outweigh the disadvantages? Fifty years ago, copyright on books was not a major restriction on people's actions, since copying a book would be difficult anyway. But for something that is naturally easy to copy and change such as source code, the restrictions placed by copyright law are more onerous. I think the loss of freedom is worth it in order to get more software produced, but it's not an open and shut case.
The article then goes on to criticize RMS for a 'skewed moral perspective', use of extreme analogies, and accusing the free software movement of hijacking the word 'free'. But this is a little unfair. Since the FSF concerns itself only with software, it's not surprising that the word 'freedom' on their pages is used only as it relates to computers. RMS is the first to admit that proprietary software is not the only problem the world has, or even the most serious. If every FSF page were prefixed with a disclaimer saying 'this is less important than other moral issues in the world', that would avoid accusations of a 'skewed moral perspective', but what would be the point? It's a mistake to assume that somebody concerned about one issue is a single issue proponent. It might just be that they don't have time to deal with everything, and have decided to focus on one specific area. (I do agree about the analogies getting a bit out of hand sometimes, but most of them work quite well.)
And the use of Eric Raymond to try to criticize free software proponents as a whole is even sillier. He may be a bit of a nut, but remember that unlike RMS, ESR is _not_ in the business of making moral proclamations, at least not in the software area. He does his best to make a practical case for free software - or 'open source software' as he calls it to avoid frightening managers. Bertrand Meyer seems to be arguing:
The article asks:
And? That still doesn't mean that it represents the whole free software movement. ESR doesn't even claim that his 'Open Source' writings represent the whole free software movement, let alone his barrel-of-a-gun writings. Some free software advocates support gun ownership, some don't. (I don't.) The fact that one person has written a web page and managed to get a large number of hits on it is neither here nor there.
That is a bizarre statement. This is not a political party where people are expected to follow the party line. It's not some corporation where any press release represents official company policy. The 'movement' is a lot looser and a lot harder to pin down that Mr Meyer seems to think. There is no requirement on somebody who supports free software to publicly disavow anything that ESR says. His gun ramblings should be treated as what they are, a quaint irrelevance.
I don't see how this has any relevance. Especially since ESR does not want to remove copyright on software, and since the free software movement has nothing to do with gun advocacy - a couple of oddly-placed links on one guy's Web page notwithstanding.
But 'nobody in charge' and 'didn't pay' are not consequences of the software being free. There are companies more than willing to take your money in exchange for providing a guaranteed response to bug fixes. The difference is that with free software, you can shop around and get the best deal for such support, rather than being limited to the company which owns the copyright.
Of course not. Would you expect the FSF to become liable to warranty claims from someone who downloaded the software gratis from an FTP site? However, the GPL explicitly allows someone distributing the software to 'offer warranty protection in exchange for a fee'. It's up to the free market to provide this service for those who can pay for it.
Warranty provision has nothing to do with 'freedom' - unless you count as 'freedom' being able to drag a software author through the courts because you downloaded his software and it didn't work. Again, if you want to provide a warranty for GCC as a profitmaking business, go ahead and do so. But saying 'use this at your own risk' is very different to saying 'you may not change the software; you may not copy the software'. One restricts what the user can do; the other is just an arse-covering legal measure - unfortunately necessary in today's litigious world.
I don't see how the first is unethical, provided it is absolutely clear about there being no warranty. Making false claims would be unethical of course. I don't see how you can expect Product F to provide a warranty if you did not pay anything for it - the author could be bankrupted by lawsuits from users who downloaded a copy or got one from you. There is a third option of course, Product F for $50 _with_ a warranty provided by the author.
I think the point is that the GPL doesn't _automatically_ grant a warranty. But it does provide a base on which you can provide warranties to specific individuals, if they pay for it. That seems sensible to me.
I don't think that any company could get away with this - how would the product establish itself in the market if it were broken? Don't forget that this problem exists with proprietary software too, if it has a support contract. Do you think that Oracle insert bugs so that their customers will subscribe to the most expensive support contracts for guaranteed fixes? It's all rather far-fetched. If you have the source, you'd probably be able to spot if something fishy were going on.
The article continues by accusing free software advocates of character asassination towards proprietary software developers. Again it makes the mistake of equating RMS with 'the free software movement'. There is no collective view on such things, any more than on gun ownership.
Finally, I think that most free software projects do indeed acknowledge where they have used ideas from other programs, free or proprietary. The GIMP's credits section describes it as a 'Photoshop-like' program, and that's about as clear as you could be. Similarly it is obvious where GNUstep, bison, less and so on got their inspiration.
Well.. unfortunately it's not that simple any more. Although the European Patent Convention specifically excludes software from being treated as a patentable 'invention', the European Patent Office ruled that this didn't mean what it said, and computer programs can be patented after all. It just goes to show that patent offices will try as hard as they can to increase what is patentable, without concern for whether such patentability is desirable. The EPO seems no more trustworthy than the USPTO. Following the EPO's decision, the UK patent office has revised its practice on software patents.
Anyway, it is still not certain whether such patents granted are actually enforceable. The European Commission are considering changing the law to catch up with the EPO's behaviour, so that software patents will become officially permitted (rather than de facto permitted as at present). Until then it isn't certain whether a lawsuit based on a European software patent would succeed (IANAL); big companies such as IBM are however stocking up on software patents in the hope of one day being able to use them. (Have a look at freepatents.org for more about stopping this from happening.)
Unisys claim to have a patent on LZW in several European countries, including the UK. I don't know whether they've actually tried to shake down British companies for money though. But you may not be as safe as you think.
Something to try would be to send another threatening letter to Oxford, pretending to be the MPAA. Claim that their discussion of why they pulled DeCSS infringes copyright / trade secrets or some other bullshit, and demand they remove it. It would be interesting to see just how ridiculous the demands have to be before they'd start to question them.
You could take this further; start writing to universities, ISPs and small companies up and down the land, with letters written in threatening lawyerese demanding that they pull particular Web pages. There could even be a 'form letter' program that puts something together from a set of stock phrases, adds a convincing-looking letterhead, and prints it out to be signed and posted. 'DIY Internet Censorship Kit' or something.
All this would be fairly pointless, unless it managed to generate a public outcry over freedom of speech and censorship. (I'm thinking of the UK here, but there's no reason it shouldn't apply to other countries.)
Is it the mail clients that are intelligent, or the users?
Is there a technical solution to this?
/* Not recognized */
Some sort of bionic-chroot-on-steriods might be the answer for running untrusted binary-only software. FreeBSD has jail() which is like an improved version of chroot(), but what's needed here is something more sophisticated.
Ideally you'd make it so that _all_ access to the outside world can be filtered through some userland process. Preferably a Perl program, hehehe. This is roughly what I have in mind:
filter_syscall() {
if (call is 'open file for reading') {
if (filename one of those allowed) {
return OKAY;
} else {
modify the 'open' call so the mode is
set to read-only; return OKAY;
}
} else if (call is 'read') {
return OKAY;
} else if (call is 'send data over network') {
check to see if the data is being sent
back to RealNetworks - if it is then
return FAIL_SILENTLY;
else return OKAY;
} else {
warn("program is doing systemcall we haven't
seen before");
return FAILURE;
}
}
It could be quite a lot of work to do this _fully_ for any large program, but a quick hack to allow all system calls except those sending private data, or to overwrite any private data being sent with 'X' characters, might be quite easy.
I think you can get S/390 cards for your PC. You won't get the mainframe reliability - which is the whole point of having a mainframe - but you will get a good chunk of the speed.
Even an S/390 emulator might run Linux at reasonable speeds, if you run the emulator on a modern PC or workstation.
IIRC, the original boo.com site was overblown with too many images, JavaScript, probably Flash and lots of other junk. Nobody could use the site and within a few weeks they had to redo it as something simpler.
Now the company has gone bankrupt. Could this be related to customers being driven away by their early, over-flashy Website?
The tale of boo.com might be a useful weapon when trying to persuade your customers that JavaScript rollovers, MIDI files and Flash are not the last word in sophisticated Web design.
I'm beginning to think that the law needs to be made clearer, not by precedent-making lawsuits, but by legislation. Yes, I know that most legislation relating to the 'Information Age' (or 'Digital Millennium') has been stupid, misguided or downright corrupt - but for the sake of argument, imagine for a moment that legislatures in whatever country were capable of making intellectual property law to benefit consumers.
Explicity define what is fair dealing for works which are made electronically available, via the Web or otherwise. I would say that the following should be allowed:
- Saving copies of a file to local disk, to look at later. (Like timeshifting a TV programme.)
- Printing out copies for personal use. (If you can read it on screen, why not on paper?)
- Distributing copies to others, provided those other people would themselves be able to get a copy, if they could be bothered to. Thus if you save yahoo.com/index.html on disk and give it to a friend, that's explicitly allowed. And if both of you subscribe to the same paid service (eg stock quotes) it should be allowed for you to pass on information which your friend could access, if his modem weren't broken. He pays the fees too, after all.
- Caching and indexing; such labelling and modification of the document as is necessary to store it in a cache or index it efficiently. Returning the document or part of it, perhaps highlighted, as the result of a search.
- Hyperlinks are allowed. This is a no-brainer, but just as well to get it in there explicitly.
- Archives of historical documents and files which were publicly available. There might have to be exceptions in here for libellous material and other nasties, but in general I don't believe that copyright law should be a tool for suppressing leaks.
- Finally, translation and parody, as long as it is made clear that the content has been modified in this way, and of course the people who are given the modified copy could also have gotten the original copy for themselves.
Now, I realize that some of these are probably already allowed by copyright law in most countries. Perhaps they all are. But I'd rather have these things laid down with certainty; uncertainty about what the law says makes anyone vulnerable to an expensive lawsuit from a firm with deep pockets. They might just win, or at least manage to bankrupt you before they lose.
I don't think that any of the above would reduce the incentive for authors and publishers to create new works. And that is the purpose of copyright law, with the intent of benefiting the public. The Web could become a very unpleasant place with legally enforceable bits of JavaScript preventing you from printing pages, people being sued over hyperlinks, and other nastiness. Some of this is happening already.
There probably isn't a snowball's chance in hell of any legislative body working out fair dealing for the Web any time soon. But it's another thing to add to the wishlist.
Yes, I've written such a program myself. (Try to write it as the shortest possible shell script!) But some method based on meaning - or at least on something a bit more sophisticated than word frequency - might produce better results. Eg you could ignore not only stopwords such as 'and' and 'the' but also hackneyed phrases such as 'if you thought that X, think again'.
What I don't understand is why make the secondary cache smaller than the primary cache?
:-)
If a memory location is used frequently enough to be stored in the secondary cache, it will surely also be in the bigger primary cache, which is faster and looked at first. Thus the secondary cache is effectively useless, except maybe as a write-back buffer. Maybe this is a mistake in the article, or I've not understood?
(BTW wouldn't it be cool to write software that fits entirely in the primary cache - 128Kbyte used to be considered a lot! If there were some way to initialize the cache at boot time, you wouldn't need RAM at all
I realize that neither of these is particularly original, but then neither have they been perfected by anybody; there is still lots of room for improvement, and plenty of scope for adding as much natural-language AI as you like:
:-) Seriously though, it would be useful for busy people who haven't got time to read things from top to bottom. It could have some way of determining which links are relevant, following them, and adding their information to the precis generated (with attributions if necessary). Or you could use it to cut out types of information you don't want - to filter out marketingspeak, if you're a techie, or to filter out techspeak, if you're not.
A precis program which will condense long websites or discussions. Jon Katz's articles on Slashdot might make a useful set of tests, also try to precis the comments posted to them
The other suggestion is a remembrance agent which looks at the website you're reading and suggests links (from your browser history, from search engines or from some big collaborative database) which might be relevant. This might finally be a use for those sidebars that recent browsers seem to have spouted. Again, this has been done before, but it's not something which has been done perfectly. You might also be able to use it as a fact-checker for postings you make to Usenet - although that would be rather difficult to implement, I imagine.