Implementing a file format as binary data or even a simple SGML structure such as RTF means less overhead. Using XML you have to run an XML parser, and the file is more freeform. There are no set data structures, it is just a stream of text.
The term you are looking for here is 'self describing structure'.
If you have data in the form of LISP S-Expressions you know where structures start and stop. If you have just one document in that format you can pretty much work out the entire file format - or at least the features being used.
If you have a binary document you have to do a lot more digging and it can take you days to just work out the basic structure.
This will make Word much more useful, it will be much easier to create documents with other applications and emit them in Word format. So for example if I have a report writer component in my server I could spit out a Word Document rather than HTML which I would use today.
I can also write filters to automatically convert from Word format to other formats, so I can take HTML source and spit out Word, I can take an XML data structure and emit word.
So why would I prefer Word format over HTML when I was one of the people who helped write HTML? Well the answer is that virtually all HTML editors are optimized for editing Web pages. I write books and other reports that really don't fit that structure. There is nothing in HTML land that I have seen that provides the power of the Word outline mode and has built in spell checking as you go.
I don't know any Linux user that uses the C-shell. the command line vs GUI argument is mostly silly. Most people that become experts in any specific app tend towards a "command line interface" where applicable, chat with commands such as:/list (or something, I generally don't use chat programs) for example. Typing really is faster in some situations.
The point is that Linux is optimized for the expert user and that creating a system for the mass market requires a lot more than just a thin GUI veneer. As you point out the command line interface is more efficient, in fact if you want to get most things done it is essential.
I don't think there is anything wrong in having a system that is designed for experts, just don't fool yourself into thinking that everyone in the world wants to be just like you. I think you massively underestimate the amount of effort required to get Linux up and running. It is OK if you are using supported hardware with good drivers but if you don't know enough to know what has good driver then you are completely stuffed. I have a Zaurus, someone nicked the WiFi card, I replaced it with an almost identical model from the same manufacturer, it does not work and none of the five wizards who have looked at it could fix it.
TiVO is mass-market linux, so is visiting web sites powered by linux in the background, and so will the new PDAs from Palm be (grammar?).
Tivo is not an open application, it is as closed as can be imagined. You have to pay a subscription to keep using the thing. I can buy a Windows Media PC that is far more open in every sense that matters, I can plug as many hard drives in as I like, I can connect the thing up to the net, I can use it until they stop broadcasting in formats my video tuner card supports. The only real disadvantage for me is that I can't use it with my satelite dish, on the other hand my dishplayer works fine.
Sure Linux makes a great choice for a large number of embedded systems, but don't fool yourself into thinking that the fact my DVR runs linux means that it is likely to appear on my parents' PC any time soon.
You missed my point. IBM/Microsoft DOS was based on CPM, an open source/free OS. The BIOS for the IBM PC was also open. This open source software/firmware/hardware is what created Microsoft's advantage.
No, I got your point, you were completely and utterly wrong. CP/M was proprietary, so was the IBM PC BIOS.
India and China will take the PC in a new direction. How could they not? The numbers of sales they will generate will create economies of scale that cannot be created in our Windows controlled market. Microsoft cannot coerce their marketing departments, cannot influence their governments.
You think that the Chinese government cannot be bought as easily as the US Congress? Are you trying to be funny? Microsoft has major development centers in India and arguably has more political leverage there than it does in the US.
China and India have only got populations of a billion and a quarter a piece. Combined they are ten times the size of the US population. The market is not as developed however, litteracy is lower. It is a significant market but nowhere near large enough to change the dynamics of the business as much as you think.
India and China are unlikely to want to follow the example of Brazil which did a great job of cutting itself off from the 1980s IT revolution by insisting that all computers be developed domestically. Market forces are going to drive the development of the software markets in those countries in the same way they have in the US. People who are looking to get high paid JOBS writing software for PAY are unlikely to be very interested in an open source phenomena that is largely driven by factors that depend on having a lot of people with a lot of disposable income and/or time.
I hear the argument against the demise of Microsoft, but consider Microsoft's own history; a free OS on an 'open source' hardware platform. This time, instead of CPM/DOS on the S100/IBM PC, it will be Unix (Linux) on a PC.
??? Microsoft never produced a free OS, the IBM PC was not an open source hardware platform. The only area where open standards played a role was that the manufacturers of PC 'clones' refused to support the proprietary closed microchannel architecture and OS/2 that IBM was trying to introduce to monopolize the market.
Open standards are not open source.
Like most slashdot stories on this topic the article is not thought provoking in the slightest, it is simply a repetition of the same prejudices that have been repeated ad-nauseam without any thought at all.
Falling hardware prices have been an issue for years, Dell were selling a full spec PC with LCD monitor for $400 six months ago. Microsoft themselves sell their X-Box for around $200. It isn't very long since the cheapest usefull PC cost over $2000.
The masses go off and pay $50 for one computer game. There is no way that Tomb Raider or EverQuest have even one percent of the intellectual effort of Windows put into them. Open Source games are practically non existent, people are still working on a rip off of Civ 3.
Linux is nowhere near providing a mass market user experience and most people working on Linux have absolutely no interest in making it mass market. What some of them want to do is to make the mass market realise how superior the C shell is to a GUI interface but most of the serious developers understand that they are producing something for techies.
Most companies purposefully choose short retention policies, in an attempt to avoid these kinds of settlements... it isn't a sysadmin's fault.
Yes but Financial institutions are required to keep the material for three years by law.
The real problem here is that lawyers tend to worry about the incriminating email being found and they don't think about the email containing an alibi being found, or for that matter the email that helps the company win a major contract or do business more efficiently being found.
Some of the concerns can be reasonable, I have been in cases where it has cost hundreds of thousands of dollars to perform discovery. But that is really a consequence of the material being spread thin which is in turn a consequence of the fact that people need it and store the stuff they need.
It was called DivX (not to be confused with the encoding scheme) circuit city was tried launching it at the same time as normal DVDs. In case you forgot, you paid around $10 for a DVD and old watch it for 2 days, if you wanted to watch it again you could pay $5 and watch it for 2 more days, o pay another $10 to unlock the video forever.
People hated it, the only remnent is a character on penny-arcade.
They tried to hire me, I worked out the scheme they were proposing during the telephone call with the recruiter and started laughing.
The scheme was idiotic for four reasons. First there was simply no way that the rest of the industry was going to allow Circuit City a retail monopoly over the players or the content. Secondly the entire industry had tooled up to launch DVD which was set to become their biggest bonanza since the launch of CD, the Circuit City scheme meant launching a totally new standards war after the standard had launched. Third it was not even profitable for the studios, the DVDs had to sell at a discount to unrestricted DVDs but most people are unlikely to pay to watch the content a second time. Fourth, people want to collect DVDs and the copyright restriction meant that their collection would be worthless.
I don't see how this new scheme would be any more acceptable, it is idiotic to think that consumers would accept biometric identification to watch a DVD, they don't need to do that today there is no way that they would buy a player or content with that feature as long as unrestricted content was available. The legacy base of DVD players means that there is no chance that the studios could take unrestricted content off sale.
The industry is starting to get concerned about what happens in a few years time when the DVD patents expire and manufacturers start producing official players that are not zone locked and do not have navigation restrictions. Their current hope is that everyone will have upgraded to blue ray or whatever by then, I don't think so. Very few people have HDTV today and very few people care as much about high definition as techy types think.
Sure I would like a better picture on my TV, but not if I have to give a DNA blood sample and pay five times as much for the content.
I seem to recall that Neal Stephenson parodied this type of attitude in Snow Crash, the federal govt. workers who have to give blood samples to log into their computers etc.
Technologists can try treat people like machines but people don't have to let them.
My claim is that the RSA algorithm is patent-worthy. True, the patent should have been granted to Ellis and Cocks rather than Rivest, Shamir and Adelman, but that's a separate issue.
Actually not, the original idea of patents was to discourage people from maintaining trade secrets and encourage the free exchange of information. Ellis and Cocks never published.
The problem with the patent system is that it is no longer meeting the original aims. I have been asked to remove innovative concepts from several of my specs just in case they might be patented. At last count there are something like 100 US patents issued where other people make retrospective claims to work I was involved in.
The US system is especially broken. One of the real problems is that unlike every other country an applicant is allowed to backdate his claim to a year before the filing date. So a corrupt applicant can read something on a mailing list, apply for a patent and then sue and the defense has to come up with prior art that was published a year before the original post. This type of corruption is not unusual, it is routine.
Don't judge the whole patent system by the corrupt US system. Software patents are not inherently wrong, the problem is that there are simply too many trivial patents issued for obvious ideas. The idea of taking a long established business process and taking it to the Internet should not be considered patentable.
Also the legal process for deciding patent claims should be made much simpler and put a much higher burden of proof on the plaintif. It should not cost $5 million to get a patent case dismissed. Plaintifs should be required to state in their initial claim exactly how the defendant is alleged to infringe the patent and the specific patent claims being infringed. I am currently answering a claim involving a patent with about 60 claims, the complaint is purely pro-forma and gives no information as to even the products that are alleged to infringe.
The othe aspect of the USPTO racket is that it allows claims that are ridiculously broad. The rule should be that there is one standard for interpreting the claims. So if the claim is being interpreted broadly for the purposes of determining infringement it should be interpreted equally broadly for the purposes of prior art and any prior art should demolish the entire claim.
This point is exactly what I was thinking the moment I read that story and they seem to be in violation.
Could people stop the open source religious wanking for a few minutes and work out if the code that has been released allows something usefull to be done? Its not like anything in this thread has not been said a few million times already.
OK they have checksums, big deal, look for the file with the checksums in it or find where the public key is stored. I would guess they do this because they update the player over the air.
There are three important mods to a DishPlayer that would dramatically improve the utility. The first is to enable that external expansion port so that an external disk can be plugged in.
The second is to enable some form of serial port for command input so that the device can be set to record remotely.
The third is to have a program that allows files stored on the hard drive to be read and copied. An NFS daemon would be nice but something less would be acceptable.
I would guess that the reason that Disphlayer do not want people changing the code in the dishplayer is that there might be a way to disable the 'call home' feature on the modem. If that is their concern then it is easy enough to fix, in fact I will even tell them how to put the fix in place for free if they make the other modifications.
Erm, use a second hard drive, or tape, or something that was designed to make backups on? Seriously, why in hell would you ever wish to backup your workstation to DVD???
Because a '30GB' tape drive which is of course only a 15GB tape drive costs $2,000 which is more than the price of most workstations these days and ten times the price of a hard drive.
Tape systems are slow, expensive and flakey. The media they use costs several times the cost of the media they are backing up. They are bulky, prone to being temperamental and there is always a high probability that the tape will not read back in another machine.
For $7,500 you can buy an 'archival' HP opto-magnetic store which has a 250Gb capacity.
The advantage of DVD over a disk is that it is a passive medium with no electronic parts to die over time. There is of course an issue with the possible degredation of the medium, but that is the case for tape as well.
At issue currently IS NOT ABOUT A COMITTEE PROCEDURE. That's the disconnect in logic in all the Democrat's arguments. You can't compare appointments killed in committee with refusal to vote on cloture in the full Senate. These two activities are COMPLETELY DIFFERENT.
No they are not, the GOP changed the rules of the senate to deny the Democrats the use of the procedures that they had used to block 60 Clinton nominees. It is only because they are brassed necked hypocrites that they are squawking now about the Democrats blocking 10 nominees.
It was Frist who originally called it the nuclear option. The only reason that the GOP is considering it is that they realize that they have already lost the entire session, there is not one part of the Bush program that is not already dead on arrival. The social security phase-out plan is do dead it is unlikely to even get a majority. The energy bill gives a multi-billion dollar handout to the most profitable companies in the US. The deficit makes it highly unlikely he will succeed with implementation of the existing tax cut plans, let alone making them 'permanent' - as if any Congress can bind its sucessors anyway.
Smuggling provisions like Real-ID into other measures is a foolish strategy for any program that is trying to build a long lasting infrastructure. The issue is now politicized and partisan, but only some of the Republicans support it. It is not likely to last long enough to go into force.
If you want to stop Real-ID the best way to do that is to make as many people aware of the corruptions of DeLay. He is openly accepting illegal bribes from lobbyists, $100,000 golfing holidays for him and his wife. In fact he is not just accepting, he is asking, demanding the bribes. That is corruption and the only reason the Republican party does not admit that it is corruption is that they have become utterly corrupted to the core. Even poor old John McCain has been bought off at this point.
The stench of Tom DeLay is going to cost the GOP the house in 2006. Once that happens the only challenge for the enquiries will be what rock to look under first, Enron, Harken, Halibutron, the corruption of this crew is astonishing.
Oh and they also got 1,600 US soldiers killed because they lied about Saddam's weapons of mass destruction.
Unless the parents have enough muscle power to enforce their offices to use openoffice, this wouldn't work. Also, to be a loss-leader, you must actually set the sales price at less than the cost of production. I don't think sicrosoft actually sells it at a loss here.
The marginal cost of production is the cost of stamping out a CD.
I took a look at the BECTA site and found no evidence of the report cited. There is an old rule in politics that when someone is talking about a report that has not yet issued it is because they are misrepresenting it.
The question is not whether FOSS software is cheaper, its whether it is equally functional for educational use. If you think that Linux is ready for use by a typical primary school teacher then you have really started believing your own propaganda.
One thing I found was that BECTA has a role in agreeing UK wide educational licenses with suppliers, it then acts as a reseller to the schools. Most of its suppliers are hardware vendors. Microsoft is conspicuously absent.
I would read this as an attempt by BECTA to strongarm Microsoft into giving them a better deal than the schools can get direct so that the schools are forced to order through them and BECTA can demonstrate how much money they are saving.
As a side, note, because the site owner claims to monitor the site -- specifically because she claims that all the content is "fact-based" -- the ceases to be protected by the laws that protect, say, Slashdot from being held responsible for the idiocy of Anonymous Coward. She has taken responsibility for the content of her site, and so she is now going to have to prove that all those angry posts are based entirely in fact.
Actually not.
IANAL. There are two standards for libel in the US depending on who the plaintif is. Public figures and most corporations are subject to a very strict standard of proof, the plaintif has to prove that the defendant knew the claims were false or was reckless or careless as to whether they were true.
In this case the company is performing a public service and ALL discussion of the company is protected political speech. It is a comment on the state provision of education services, not just the corporation.
It is very difficult to see how the defendant could now the claims made were false in this particular instance. The refusal to specify the claims found objectionable and the demand to take down the entire forum point to supression of debate as the clear objective of the corporation.
If the US actually have a press corps that questioned the activities of Republicans or their cronies this would be front page news. As it is it will be either unreported or burried on the inside pages along with reports of DeLay's numerous corruptions.
I'm well up on the background to the case, it's the legalese surrounding this particular event that needs translating.
The legalese is not that interesting, the real question is why SCO would be filing for an extension.
If SCO had found a smoking gun they would not be needing an extension. The fact they are asking for one means that they have not yet found one in that morass of discovery.
The refusal of summary judgement means nothing until the discovery process is over. The judge is not very likely to end the case while discovery is pending, SCO might find the proof they need. In this particular case it is unlikely that IBM would discover proof they were innocent and especially not until the specific copyright works have been identified.
The real story is in the responses that the judge has been giving to the motions. Her patience is wearing thin, and not with IBM.
Actually this is very significant, goodwill is a measure of the difference between the price paid for a company that is aqcuired and the hard assets of the company. Most IT companies have huge figures for goodwill because they will buy a company for maybe $100 million and its hard assets might be less then $10 million, the rest is the value of the ongoing business.
In normal circumstances goodwill depreciates at a steady rate but if a company suffers a major financial catastrophe the 'goodwill' has to be written off faster. SCO makes no money from the SCO business acquired by Caldrea, the assets have to be written down to zero as a result.
What this means is that SCO has no ongoing business from any of its acquisitions. It litterally has no goodwill, that is the relationships that are the real value of a modern company.
Or there was nothing there. I can't count how many start ups I've worked with that have shown me slideware of "patent pending" technology that was not implemented, not thought through or otherwise junk.
I have idiots come and claim that they 'invented' ideas that are in several of my specifications. I then have to point out that the ideas were not new when I wrote the spec which was written long before said idiot sent me any material. It is getting to the point where I am going to have to get an admin to read all unsolicited mail so that unsolicited ideas can be returned unseen.
I don't work for Microsoft but I work with a lot of Microsoft people on standards. On several occasions they have objected to ideas I have wanted to put in because they are too interesting, too innovative and too likely to be patented by someone. The patent system is having the exact reverse effect of the one intended.
Having read the article and the comments I have yet to see anything that gives me any information on what the Chimney technology is. I doubt that it is going to be very important however since Longhorn is not yet out and this is the first time I have heard any mention of it.
As for Microsoft supporting the patent system in its current form, they have made frequent complaints about the abuse of the patent system and the issue of trivial patents. There is a middle ground between thinking that all software patents are evil and should be abolished and thinking that all are valid. Everyone who actually makes stuff is agreed that the vast majority of software patents are garbage. But even though 99% are garbage that should never be issued there are a few legitimate ones, the RSA patents for example, Diffie Hellman, Merkle Trees, the Lamport hash chain and the patent I have pending. Microsoft is spending several billion a year on Microsoft Research and it is not unfair for them to see some of that returned as patent royalties. Nor is it unfair for IBM or Texas Instruments to see a return on their investments.
But that does not mean that Eolas have a right to see a dime for their bogus patent or that Unisys should ever have received a cent for the Gif patent or that RAMBUS should profit from their schemes.
Win2k+ won't allow an application to overwrite a DLL. It will detect it and restore it. NT3.5x and 4 would, but the OS would usually crash right then and there.
DLL hell is not unique to either Linux or Windows. It has been a serious problem from the moment the first person updated a shared library on the first system to support them.
The logical symbols mechanism in VMS allowed many problems to be avoided, instead of loading a filename the program would load a logical symbol. It was easy to run a system with different versions of the same shared library in use simultaneously. UNIX and Windows never used symbols in quite the same way.
I think that folk need to look at what they get from a shared library. At one time it made great sense to save memory by having multiple processes share an in memor image of an executable. It makes particular sense if you are running something like Apache where child processes are being spawned off from the parent and share resources with it. I don't think it makes a lot of sense as a general approach when memory cost $50 per gigabyte.
We could probably do much better than shared memory if we went back to static libraries and instead used more intelligent linker technology. When I link to stdio I pull in maybe 500K of code and use at most 40% of the code paths. When I link to more recent libraries the library is much larger and the fraction I use much, much less. A shared library is an all or nothing affair, every part of the library has to be loaded in case another image might need it. Even if the code page is never touched the memory has to be allocated.
As for the question of user interfaces, I think that the way they are designed today is worse than sub-optimal. I would prefer to go back to an architecture similar to the one that the NextStation had. Instead of having the program implement the user interface as code it should send a description of the user interface to the windows manager and have it perform all the necessary animation.
This approach is similar to what we did in the early HTML days but the idea is to take the approach much further. I really dislike the fact that most programs are single threaded and the UI goes to sleep every time it is asked to do anything computationally intensive of requiring the network. The architecture I just described allows the window manager to keep the user interface alive even though the program logic is 'thinking' and the programmer does not have to do any work to achieve this.
The other advantage of this approach is a bit more controvertial, it limits the scope of the UI designer. This is a bad thing if you really, really love to foist a bizaro UI onto the user. On the other hand it means that every application can be skinned so implementing the bizarro UI is simply a matter of telling the program manager how to do it for every program on the machine.
A.177 caliber pellet fired upward shouldn't have enough terminal energy to damage the skylights, especially if you hit the baloon. Remember, skylights must be designed to absorb the energy of random stuff falling on them, like at least marble-sized hail, ice blown from adjacent rooftops, birds, and so on.
No they are not. The 'skylight' in question is a stained glass decorative overlay. Above that there is a difuser and above that there is the actual skylight. The skylight has been damaged during previous hacks so I see no reason at all to be so sure that it would not be damaged by a pellet fired at it.
The balloons would deflate of their own accord within a week. If there was a need to remove the balloons earlier the obvious approach would be to get another balloon, tie a string to it and attach a tack or series of tacks to the top of it.
Even better would be selling the right to shoot them down and using the money for some charity.
Who would pay for the repairs to the glass skylight in the ceiling? I.e. the one the ballons are directly underneath?
Seems to me as if Caltech is admitting that their own campus is so lame that they have to MIT to hack. Perhaps some of the pre-freshers will think up their own reply.
It is also simpler, much easier to use and maintain, and so much more reliable than BIND or Windows DNS. It also has never had a buffer overflow or other security problem.
This is absolutely correct, slashdot got this one wrong, DJBDNS has always been very secure, Windows NT DNS was insecure until SP4 went out and then they made the mistake of not making the security checks on by default.
A major security flaw was found in BIND as recently as 2002, if you are still using BIND 8.4.3 you are at risk and need to upgrade to BIND 9 now. Unfortunately a lot of people have not done that because there are some infrastructure issues.
BIND was cleaned up to a major degree by Paul Vixie, but the history absolutely does not justify folk using BIND to go throwing stones. What Microsoft has been doing is reminding people to turn on the security mechanism in a version of Windows that is generally considered obsolete in response to recent attacks.
Its like Red Hat issuing an urgent advisory for version 5.3
The main issue with DJBDNS is whether it will support the final DNSSEC spec. DJB is somewhat pissed with the DNSEXT working group (and with justification).
That depends on how they've set up their hostclasses. If you are in the same host class with people who do use those programs, then they are correct to withhold deployment until everything is tested. They can then update everybody at once. If there are enough people like you, then maybe they should have you people in your own hostclass, I dunno.
Hostclasses? shmostclasses! Who would want to do that (well apart from me!)
The point I was making here is that what Microsoft is doing is removing the veto they gave to IT support, they are not forcing anyone to download SP2.
Microsoft said that the blocking was only a temporary measure when they started it. So anyone who has not planned for this event has only themselves to blame.
SP2 has a dramatic effect on the vulnerabilities that a machine is exposed to. If sysops are blocking the update they are having a major negative impact on their system security.
I'd like to see someone compromise a default Red Hat install from at least the past couple of years. Hint: no listening services == no way you can touch me in the fashion you describe. I'm told the same goes for OSX, but I can't speak from experience on this front.
The problem is making real sure that you have no listening services. If you have an O/S where programs have been written using RPC you will find that unless you work really, really hard to make sure it does not happen you have open RPC ports.
A while ago I spent some time stripping down a Digital Unix install to remove everything that was unnecessary. We thought that we could safely disable RPC, we didn't use NFS or any network type service. When we disabled the port the tape drive stopped working.
Once people produced good monitoring tools the systems moved from 'pretty good' lockdown to 'real lockdown'. But the empirical testing does not suggest that linux is as good enought to justify the level of complacency I hear.
Slashdot Mirror
The term you are looking for here is 'self describing structure'.
If you have data in the form of LISP S-Expressions you know where structures start and stop. If you have just one document in that format you can pretty much work out the entire file format - or at least the features being used.
If you have a binary document you have to do a lot more digging and it can take you days to just work out the basic structure.
This will make Word much more useful, it will be much easier to create documents with other applications and emit them in Word format. So for example if I have a report writer component in my server I could spit out a Word Document rather than HTML which I would use today.
I can also write filters to automatically convert from Word format to other formats, so I can take HTML source and spit out Word, I can take an XML data structure and emit word.
So why would I prefer Word format over HTML when I was one of the people who helped write HTML? Well the answer is that virtually all HTML editors are optimized for editing Web pages. I write books and other reports that really don't fit that structure. There is nothing in HTML land that I have seen that provides the power of the Word outline mode and has built in spell checking as you go.
The point is that Linux is optimized for the expert user and that creating a system for the mass market requires a lot more than just a thin GUI veneer. As you point out the command line interface is more efficient, in fact if you want to get most things done it is essential.
I don't think there is anything wrong in having a system that is designed for experts, just don't fool yourself into thinking that everyone in the world wants to be just like you. I think you massively underestimate the amount of effort required to get Linux up and running. It is OK if you are using supported hardware with good drivers but if you don't know enough to know what has good driver then you are completely stuffed. I have a Zaurus, someone nicked the WiFi card, I replaced it with an almost identical model from the same manufacturer, it does not work and none of the five wizards who have looked at it could fix it.
TiVO is mass-market linux, so is visiting web sites powered by linux in the background, and so will the new PDAs from Palm be (grammar?).
Tivo is not an open application, it is as closed as can be imagined. You have to pay a subscription to keep using the thing. I can buy a Windows Media PC that is far more open in every sense that matters, I can plug as many hard drives in as I like, I can connect the thing up to the net, I can use it until they stop broadcasting in formats my video tuner card supports. The only real disadvantage for me is that I can't use it with my satelite dish, on the other hand my dishplayer works fine.
Sure Linux makes a great choice for a large number of embedded systems, but don't fool yourself into thinking that the fact my DVR runs linux means that it is likely to appear on my parents' PC any time soon.
No, I got your point, you were completely and utterly wrong. CP/M was proprietary, so was the IBM PC BIOS.
India and China will take the PC in a new direction. How could they not? The numbers of sales they will generate will create economies of scale that cannot be created in our Windows controlled market. Microsoft cannot coerce their marketing departments, cannot influence their governments.
You think that the Chinese government cannot be bought as easily as the US Congress? Are you trying to be funny? Microsoft has major development centers in India and arguably has more political leverage there than it does in the US.
China and India have only got populations of a billion and a quarter a piece. Combined they are ten times the size of the US population. The market is not as developed however, litteracy is lower. It is a significant market but nowhere near large enough to change the dynamics of the business as much as you think.
India and China are unlikely to want to follow the example of Brazil which did a great job of cutting itself off from the 1980s IT revolution by insisting that all computers be developed domestically. Market forces are going to drive the development of the software markets in those countries in the same way they have in the US. People who are looking to get high paid JOBS writing software for PAY are unlikely to be very interested in an open source phenomena that is largely driven by factors that depend on having a lot of people with a lot of disposable income and/or time.
??? Microsoft never produced a free OS, the IBM PC was not an open source hardware platform. The only area where open standards played a role was that the manufacturers of PC 'clones' refused to support the proprietary closed microchannel architecture and OS/2 that IBM was trying to introduce to monopolize the market.
Open standards are not open source.
Like most slashdot stories on this topic the article is not thought provoking in the slightest, it is simply a repetition of the same prejudices that have been repeated ad-nauseam without any thought at all.
Falling hardware prices have been an issue for years, Dell were selling a full spec PC with LCD monitor for $400 six months ago. Microsoft themselves sell their X-Box for around $200. It isn't very long since the cheapest usefull PC cost over $2000.
The masses go off and pay $50 for one computer game. There is no way that Tomb Raider or EverQuest have even one percent of the intellectual effort of Windows put into them. Open Source games are practically non existent, people are still working on a rip off of Civ 3.
Linux is nowhere near providing a mass market user experience and most people working on Linux have absolutely no interest in making it mass market. What some of them want to do is to make the mass market realise how superior the C shell is to a GUI interface but most of the serious developers understand that they are producing something for techies.
Yes but Financial institutions are required to keep the material for three years by law.
The real problem here is that lawyers tend to worry about the incriminating email being found and they don't think about the email containing an alibi being found, or for that matter the email that helps the company win a major contract or do business more efficiently being found.
Some of the concerns can be reasonable, I have been in cases where it has cost hundreds of thousands of dollars to perform discovery. But that is really a consequence of the material being spread thin which is in turn a consequence of the fact that people need it and store the stuff they need.
They tried to hire me, I worked out the scheme they were proposing during the telephone call with the recruiter and started laughing.
The scheme was idiotic for four reasons. First there was simply no way that the rest of the industry was going to allow Circuit City a retail monopoly over the players or the content. Secondly the entire industry had tooled up to launch DVD which was set to become their biggest bonanza since the launch of CD, the Circuit City scheme meant launching a totally new standards war after the standard had launched. Third it was not even profitable for the studios, the DVDs had to sell at a discount to unrestricted DVDs but most people are unlikely to pay to watch the content a second time. Fourth, people want to collect DVDs and the copyright restriction meant that their collection would be worthless.
I don't see how this new scheme would be any more acceptable, it is idiotic to think that consumers would accept biometric identification to watch a DVD, they don't need to do that today there is no way that they would buy a player or content with that feature as long as unrestricted content was available. The legacy base of DVD players means that there is no chance that the studios could take unrestricted content off sale.
The industry is starting to get concerned about what happens in a few years time when the DVD patents expire and manufacturers start producing official players that are not zone locked and do not have navigation restrictions. Their current hope is that everyone will have upgraded to blue ray or whatever by then, I don't think so. Very few people have HDTV today and very few people care as much about high definition as techy types think.
Sure I would like a better picture on my TV, but not if I have to give a DNA blood sample and pay five times as much for the content.
I seem to recall that Neal Stephenson parodied this type of attitude in Snow Crash, the federal govt. workers who have to give blood samples to log into their computers etc.
Technologists can try treat people like machines but people don't have to let them.
Actually not, the original idea of patents was to discourage people from maintaining trade secrets and encourage the free exchange of information. Ellis and Cocks never published.
The problem with the patent system is that it is no longer meeting the original aims. I have been asked to remove innovative concepts from several of my specs just in case they might be patented. At last count there are something like 100 US patents issued where other people make retrospective claims to work I was involved in.
The US system is especially broken. One of the real problems is that unlike every other country an applicant is allowed to backdate his claim to a year before the filing date. So a corrupt applicant can read something on a mailing list, apply for a patent and then sue and the defense has to come up with prior art that was published a year before the original post. This type of corruption is not unusual, it is routine.
Don't judge the whole patent system by the corrupt US system. Software patents are not inherently wrong, the problem is that there are simply too many trivial patents issued for obvious ideas. The idea of taking a long established business process and taking it to the Internet should not be considered patentable.
Also the legal process for deciding patent claims should be made much simpler and put a much higher burden of proof on the plaintif. It should not cost $5 million to get a patent case dismissed. Plaintifs should be required to state in their initial claim exactly how the defendant is alleged to infringe the patent and the specific patent claims being infringed. I am currently answering a claim involving a patent with about 60 claims, the complaint is purely pro-forma and gives no information as to even the products that are alleged to infringe.
The othe aspect of the USPTO racket is that it allows claims that are ridiculously broad. The rule should be that there is one standard for interpreting the claims. So if the claim is being interpreted broadly for the purposes of determining infringement it should be interpreted equally broadly for the purposes of prior art and any prior art should demolish the entire claim.
Could people stop the open source religious wanking for a few minutes and work out if the code that has been released allows something usefull to be done? Its not like anything in this thread has not been said a few million times already.
OK they have checksums, big deal, look for the file with the checksums in it or find where the public key is stored. I would guess they do this because they update the player over the air.
There are three important mods to a DishPlayer that would dramatically improve the utility. The first is to enable that external expansion port so that an external disk can be plugged in.
The second is to enable some form of serial port for command input so that the device can be set to record remotely.
The third is to have a program that allows files stored on the hard drive to be read and copied. An NFS daemon would be nice but something less would be acceptable.
I would guess that the reason that Disphlayer do not want people changing the code in the dishplayer is that there might be a way to disable the 'call home' feature on the modem. If that is their concern then it is easy enough to fix, in fact I will even tell them how to put the fix in place for free if they make the other modifications.
Because a '30GB' tape drive which is of course only a 15GB tape drive costs $2,000 which is more than the price of most workstations these days and ten times the price of a hard drive.
Tape systems are slow, expensive and flakey. The media they use costs several times the cost of the media they are backing up. They are bulky, prone to being temperamental and there is always a high probability that the tape will not read back in another machine.
For $7,500 you can buy an 'archival' HP opto-magnetic store which has a 250Gb capacity.
The advantage of DVD over a disk is that it is a passive medium with no electronic parts to die over time. There is of course an issue with the possible degredation of the medium, but that is the case for tape as well.
And what are the 'merits' of accepting a $100,000 golfing holiday from a lobbyist?
The reason you can only reply with insults is that you know that it was a corrupt bribe demanded by a corrupt politician who leads a corrupt party.
No they are not, the GOP changed the rules of the senate to deny the Democrats the use of the procedures that they had used to block 60 Clinton nominees. It is only because they are brassed necked hypocrites that they are squawking now about the Democrats blocking 10 nominees.
It was Frist who originally called it the nuclear option. The only reason that the GOP is considering it is that they realize that they have already lost the entire session, there is not one part of the Bush program that is not already dead on arrival. The social security phase-out plan is do dead it is unlikely to even get a majority. The energy bill gives a multi-billion dollar handout to the most profitable companies in the US. The deficit makes it highly unlikely he will succeed with implementation of the existing tax cut plans, let alone making them 'permanent' - as if any Congress can bind its sucessors anyway.
Smuggling provisions like Real-ID into other measures is a foolish strategy for any program that is trying to build a long lasting infrastructure. The issue is now politicized and partisan, but only some of the Republicans support it. It is not likely to last long enough to go into force.
If you want to stop Real-ID the best way to do that is to make as many people aware of the corruptions of DeLay. He is openly accepting illegal bribes from lobbyists, $100,000 golfing holidays for him and his wife. In fact he is not just accepting, he is asking, demanding the bribes. That is corruption and the only reason the Republican party does not admit that it is corruption is that they have become utterly corrupted to the core. Even poor old John McCain has been bought off at this point.
The stench of Tom DeLay is going to cost the GOP the house in 2006. Once that happens the only challenge for the enquiries will be what rock to look under first, Enron, Harken, Halibutron, the corruption of this crew is astonishing.
Oh and they also got 1,600 US soldiers killed because they lied about Saddam's weapons of mass destruction.
The marginal cost of production is the cost of stamping out a CD.
I took a look at the BECTA site and found no evidence of the report cited. There is an old rule in politics that when someone is talking about a report that has not yet issued it is because they are misrepresenting it.
The question is not whether FOSS software is cheaper, its whether it is equally functional for educational use. If you think that Linux is ready for use by a typical primary school teacher then you have really started believing your own propaganda.
One thing I found was that BECTA has a role in agreeing UK wide educational licenses with suppliers, it then acts as a reseller to the schools. Most of its suppliers are hardware vendors. Microsoft is conspicuously absent.
I would read this as an attempt by BECTA to strongarm Microsoft into giving them a better deal than the schools can get direct so that the schools are forced to order through them and BECTA can demonstrate how much money they are saving.
Actually not.
IANAL. There are two standards for libel in the US depending on who the plaintif is. Public figures and most corporations are subject to a very strict standard of proof, the plaintif has to prove that the defendant knew the claims were false or was reckless or careless as to whether they were true.
In this case the company is performing a public service and ALL discussion of the company is protected political speech. It is a comment on the state provision of education services, not just the corporation.
It is very difficult to see how the defendant could now the claims made were false in this particular instance. The refusal to specify the claims found objectionable and the demand to take down the entire forum point to supression of debate as the clear objective of the corporation.
If the US actually have a press corps that questioned the activities of Republicans or their cronies this would be front page news. As it is it will be either unreported or burried on the inside pages along with reports of DeLay's numerous corruptions.
"Well Duhh!", she replies
"No Duhh Duhh Dit Dit Duhh Duhhh"
<thump>
If you think about it there is no risk whatsoever in throwing stones at glass houses if you live in a rubber house.
There is however a serious downside to throwing stones at a rubber house someone else is living in.
The legalese is not that interesting, the real question is why SCO would be filing for an extension.
If SCO had found a smoking gun they would not be needing an extension. The fact they are asking for one means that they have not yet found one in that morass of discovery.
The refusal of summary judgement means nothing until the discovery process is over. The judge is not very likely to end the case while discovery is pending, SCO might find the proof they need. In this particular case it is unlikely that IBM would discover proof they were innocent and especially not until the specific copyright works have been identified.
The real story is in the responses that the judge has been giving to the motions. Her patience is wearing thin, and not with IBM.
Actually this is very significant, goodwill is a measure of the difference between the price paid for a company that is aqcuired and the hard assets of the company. Most IT companies have huge figures for goodwill because they will buy a company for maybe $100 million and its hard assets might be less then $10 million, the rest is the value of the ongoing business.
In normal circumstances goodwill depreciates at a steady rate but if a company suffers a major financial catastrophe the 'goodwill' has to be written off faster. SCO makes no money from the SCO business acquired by Caldrea, the assets have to be written down to zero as a result.
What this means is that SCO has no ongoing business from any of its acquisitions. It litterally has no goodwill, that is the relationships that are the real value of a modern company.
I have idiots come and claim that they 'invented' ideas that are in several of my specifications. I then have to point out that the ideas were not new when I wrote the spec which was written long before said idiot sent me any material. It is getting to the point where I am going to have to get an admin to read all unsolicited mail so that unsolicited ideas can be returned unseen.
I don't work for Microsoft but I work with a lot of Microsoft people on standards. On several occasions they have objected to ideas I have wanted to put in because they are too interesting, too innovative and too likely to be patented by someone. The patent system is having the exact reverse effect of the one intended.
Having read the article and the comments I have yet to see anything that gives me any information on what the Chimney technology is. I doubt that it is going to be very important however since Longhorn is not yet out and this is the first time I have heard any mention of it.
As for Microsoft supporting the patent system in its current form, they have made frequent complaints about the abuse of the patent system and the issue of trivial patents. There is a middle ground between thinking that all software patents are evil and should be abolished and thinking that all are valid. Everyone who actually makes stuff is agreed that the vast majority of software patents are garbage. But even though 99% are garbage that should never be issued there are a few legitimate ones, the RSA patents for example, Diffie Hellman, Merkle Trees, the Lamport hash chain and the patent I have pending. Microsoft is spending several billion a year on Microsoft Research and it is not unfair for them to see some of that returned as patent royalties. Nor is it unfair for IBM or Texas Instruments to see a return on their investments.
But that does not mean that Eolas have a right to see a dime for their bogus patent or that Unisys should ever have received a cent for the Gif patent or that RAMBUS should profit from their schemes.
DLL hell is not unique to either Linux or Windows. It has been a serious problem from the moment the first person updated a shared library on the first system to support them.
The logical symbols mechanism in VMS allowed many problems to be avoided, instead of loading a filename the program would load a logical symbol. It was easy to run a system with different versions of the same shared library in use simultaneously. UNIX and Windows never used symbols in quite the same way.
I think that folk need to look at what they get from a shared library. At one time it made great sense to save memory by having multiple processes share an in memor image of an executable. It makes particular sense if you are running something like Apache where child processes are being spawned off from the parent and share resources with it. I don't think it makes a lot of sense as a general approach when memory cost $50 per gigabyte.
We could probably do much better than shared memory if we went back to static libraries and instead used more intelligent linker technology. When I link to stdio I pull in maybe 500K of code and use at most 40% of the code paths. When I link to more recent libraries the library is much larger and the fraction I use much, much less. A shared library is an all or nothing affair, every part of the library has to be loaded in case another image might need it. Even if the code page is never touched the memory has to be allocated.
As for the question of user interfaces, I think that the way they are designed today is worse than sub-optimal. I would prefer to go back to an architecture similar to the one that the NextStation had. Instead of having the program implement the user interface as code it should send a description of the user interface to the windows manager and have it perform all the necessary animation.
This approach is similar to what we did in the early HTML days but the idea is to take the approach much further. I really dislike the fact that most programs are single threaded and the UI goes to sleep every time it is asked to do anything computationally intensive of requiring the network. The architecture I just described allows the window manager to keep the user interface alive even though the program logic is 'thinking' and the programmer does not have to do any work to achieve this.
The other advantage of this approach is a bit more controvertial, it limits the scope of the UI designer. This is a bad thing if you really, really love to foist a bizaro UI onto the user. On the other hand it means that every application can be skinned so implementing the bizarro UI is simply a matter of telling the program manager how to do it for every program on the machine.
No they are not. The 'skylight' in question is a stained glass decorative overlay. Above that there is a difuser and above that there is the actual skylight. The skylight has been damaged during previous hacks so I see no reason at all to be so sure that it would not be damaged by a pellet fired at it.
The balloons would deflate of their own accord within a week. If there was a need to remove the balloons earlier the obvious approach would be to get another balloon, tie a string to it and attach a tack or series of tacks to the top of it.
Who would pay for the repairs to the glass skylight in the ceiling? I.e. the one the ballons are directly underneath?
Seems to me as if Caltech is admitting that their own campus is so lame that they have to MIT to hack. Perhaps some of the pre-freshers will think up their own reply.
class fruit_derived_topping:
# etc.
No, in Python thats:
Sausage egg and jam, bacon egg and jam,
(Viking chorus) Jam! Jam! Jam! Jam! Jam! Jam! Jam! Jam! Lovely Jam! Wonderful Jam!
This is absolutely correct, slashdot got this one wrong, DJBDNS has always been very secure, Windows NT DNS was insecure until SP4 went out and then they made the mistake of not making the security checks on by default.
A major security flaw was found in BIND as recently as 2002, if you are still using BIND 8.4.3 you are at risk and need to upgrade to BIND 9 now. Unfortunately a lot of people have not done that because there are some infrastructure issues.
BIND was cleaned up to a major degree by Paul Vixie, but the history absolutely does not justify folk using BIND to go throwing stones. What Microsoft has been doing is reminding people to turn on the security mechanism in a version of Windows that is generally considered obsolete in response to recent attacks.
Its like Red Hat issuing an urgent advisory for version 5.3
The main issue with DJBDNS is whether it will support the final DNSSEC spec. DJB is somewhat pissed with the DNSEXT working group (and with justification).
Hostclasses? shmostclasses! Who would want to do that (well apart from me!)
The point I was making here is that what Microsoft is doing is removing the veto they gave to IT support, they are not forcing anyone to download SP2.
Microsoft said that the blocking was only a temporary measure when they started it. So anyone who has not planned for this event has only themselves to blame.
SP2 has a dramatic effect on the vulnerabilities that a machine is exposed to. If sysops are blocking the update they are having a major negative impact on their system security.
The problem is making real sure that you have no listening services. If you have an O/S where programs have been written using RPC you will find that unless you work really, really hard to make sure it does not happen you have open RPC ports.
A while ago I spent some time stripping down a Digital Unix install to remove everything that was unnecessary. We thought that we could safely disable RPC, we didn't use NFS or any network type service. When we disabled the port the tape drive stopped working.
Once people produced good monitoring tools the systems moved from 'pretty good' lockdown to 'real lockdown'. But the empirical testing does not suggest that linux is as good enought to justify the level of complacency I hear.