When I started reading the article there were 0 posts, by the time I finished over a hundred.
As with all large companies Microsoft files lots of ridiculous patents. They do it for the same reason mine does, so that if they are sued by another company thay have something for swapsies.
It would probably not be a good thing for Microsoft if their customers could not attach Linux file systems easily. SAMBA is simply collateral damage in the high stakes game between EMC and Microsoft. EMC servers are very expensive and Microsoft would love to play bigger on that turf.
The bigger problem is that in the crackpot US PTO scheme you never know if a patent has been applied for on something until the government awards a 20 year monopoly in practicing it. The rules have been improved, i.e. made less open to corrupt abuse but they are still an extortionists charter.
I can't remember the last time Microsoft was the plaintif in a Patent lawsuit. They have been the victim of many Patent Trolls.
It would be an idiotic strategy for Microsoft to try to use patents to make.NET proprietary. But then again the tax cut for the ultra-rich and breaking the ABM treaty to build a 21st century Maginot line are crackpot ideas.
Last time Tim got suckered into lending his name to one of these enterprises the result was Akamai. Tim sold all his shares right after the lock out to avoid 'conflicts of interest'. Good move AKAM cratered afterwards.
The business model for Curl is laughable. It is based on an MIT conceit that the programming language is all important. When I looked at the Curl stuff a few years back it looked like yet another attempt to make C look like Lisp. Yeah whatever but Lisp does quite a good job of looking like Lisp.
If you have to download a plug in to use the language you might as well write the code in C# and be done with it. The guys who wrote C# are at least as smart as the guys in 545 Tech square.
Ward did not strike me as being the live wire of the lab, nor for that matter did his students. Dertousos is just the Director, like Negroponte he does not do, he presents, unlike Negroponte however I doubt that many grad students would go to talk to Detousos for inspiration.
Bottom line is I did not see people queueing up to learn Curl inside Tech square so I don't see why anyone outside the building is likely to use it.
I use a laptop for applications, not for hacking code. These days I can't do that on anything with less than two 20" displays, 1Gb RAM and the rest.
The most important applications are web surfing, Word and the ability to play DVDs on the plane with Civilization II next in line. Unfortunately I can't get Tomb Raider to work on my Sony Vaio Z505
What I would really like to get hold of is one of those tablet PCs with pen input. I used to have a couple of CrossPads but the software sucketh somewhat and is is something of a kludge. It is not bad at handwriting recognition but it is tedious having to tell it each time you turn the page. I would be happy to pay $3K for a tablet PC with a decent spec, but getting hold of one in the US appears impossible. Anyone know if they work?
The Sony is pretty cute, the revamp looks slightly better, I have the plug in DVD which is a pain for plane flights.
It's hard to prove a negative. The PTO is supposed to do some research to see if they can find prior art, but when it comes to software, they don't, even when to do so would apparently be quite trivial.
No it isn't. The PTO system relise upon the honest of the applicants. The PTO does not have the expertise to search for prior art, never has, never will.
The US PTO can't even perform searches of already issued patents competently, so what is the chance that in the 10 hours allowed for the average examination the clerk with a law degree is going to find prior art? An examiner does not need any actual experience of the field they are examining, just a relevant(ish) degree.
What is meant to be the bar to malicious/criminal patent applications is the difficulty of enforcing the pieces of utter crap that get issued. Then Lemelson came along with his perjured 'bar codes' continuation of 'machine vision' and extorted a billion dollars.
The system stinks and filling open source patents won't fix it. Unless someone wants to lay out $2,000,000 trying to enforce the patent it is worthless and pointless.
People have ruyn one O/S under another for years, IBM are currently selling their ability to run multiple Linux VMs under their MVS O/S - which had the idea so long ago that the patents have expired years ago.
I have only ever seen one patent that I could not work out a way to bust - Diffie-Helleman and that is long expired now. If the Open Source community start to use patents to try to force people to do everything their way I think they will have stopped being the solution and become the new problem. Its the type of mind control, silly-ass games that turn people off. I don't think the patents filled by the Open Source Community are likely to inconvenience me, just another patent to bust. The Open Source Community would probably have more problems if I started filling broad patents on my own work rather than putting everything into the public domain which I have done to date.
You can't take the low road to the moral high ground.
I have a Vodoo 3 card as well, only my problem is that I can't get it to work with my existing O/S, Tomb Raider Chronicles that is. But the problem is that 3DFX being a gonna and not having released their drivers open source the hardware is going to be useless sometime or other regadless.
My question is how much is microsoft actually comming up with themselves and how much are they hacking away from the opensource community? I heard that Active Directory is just bind with a microsoft twist to it. Is IIS just apache tweeked to hell and back?
Active Directory is an LDAP interface, BIND is a DNS interface. Active Directory also provides DNS support but the underlying data model is LDAP and the probability that any BIND code would be useful is zero.
At the time IIS first appeared Apache did not exist, it was still the NCSA Web server with a bunch of third party patches. Thau was still doing major surgery on the first release of Apache while I was running IIS in the office across the hall from him. IIS could conceivably contain some of the CERN Libwww code, but that was put in the public domain, it is not open source restricted. The Microsoft lawyers called up to ask what the status of the CERN code was before MSFT downloaded it.
But still it is easier to make completely unsubstantiated allegations, admitting that you have no evidence apart from your belief that Microsoft >= absolute evil => If it is evil Microsoft must be doing it.
Since you appear to be a Newbie Microsoft-basher I will help you with some hints:
The Register article itself states that the blocking of the old incompatible application versions is taking place with the knowledge and co-operation of the companies themselves who are not complaining. Therefore Microsoft must bave blackmailed the companies into not complaining
The mechanism is a blacklist that lists bad programs that cannot be run. Therefore Microsoft csn stop you running your own software by not including it on the blacklist.
Here you are confused. Kodak's bitch was that you couldn't make their software the default even if you wanted to. MS's photo software would re-register itself as the default spontaneously and for no reason.
What actually happened (and as was discussed at length on./ already was that Kodak and a bunch of other manufacturers agreed on a standard for hooking up a camera. Microsoft installed a default driver for the standard in the O/S. Clearly an evil thing to do.
Kodak wanted to distribute a modified driver that when tested was found to register the Kodak image software as the default for all manufacturer's cameras using the format. Needless to say the driver did not ask the user before reconfiguring their machine. The assumtion being that anyone who bought a Kodak camera was somehow Kodak property.
There is a big different between getting a bunch of pesterware drivers from the vendor of the O/S and getting them from everywhere. First thing I do with a windows box is to throw the MSN icon and the rest of the junk in the recycle bin and pull the chain on it. Then I know that they are gone for good. The Kodak driver is like the original Netscape browser, it reinstalls itself over your defaults each time it runs.
Personally I think the Kodak scheme to drive consumers to their web site to buy printed copies is somewhat naive. The first thing I discovered after buying my Nikon Coolpix was that I stopped using film for snapshots. The second was that with electronic pics you load them onto a Web site and bore your relations remotely. The print out on dead tree step can be omitted entirely.
>>I guess next we will be hearing how Microsoft own the media and negative views of microsoft can never be heard.
uhhh... your joking right... MSNBC ring a bell? Oh sure if they really owned the media Microsoft would probably have to add to their name, AOL-TimeWarner.
No joke. One of the distinguishing features of an idealogue is that they cannot process contrary views. This is is called in the propaganda analysis field a 'stoccoma'. This then leads to the claim that all contrary views are due to 'the liberal media', a phrase that Murdoch's Fox news repeats several times an hour. Murdoch owns a larger media empire than Hurst, but he can't be heard because of the 'Liberal Media', yeah that would be Salon.com.
Notice in this thread people are already accusing anyone questioning the article as being a Microsoft PR flack. The point raised about MSNBC is important, yes Microsoft do own half of a calbe TV news station and do have an alliance with an infulential media owner. However their opponents Time Warner/AOL-Netscape happen to own more cable stations, channels magazines etc. than anyone else.
The same mindset could be seen on both sides during the 'red scares' orchestrated by Hoover and McCarthy. The communists refused to believe anything bad about Stalin's regime, denouncing the capitalist press. The scare-mongers believed that anyone who opposed them for any reason had to be a communist.
As I pointed out elsewhere, the original piece is an opportunity for Microsoft. If I were a Microsoft flack I would have written the original piece so others would knock it down and with it other anti-MSFT views that might be valid. The trial fell victim of the same problem, a perfectly reasonable tying claim was hijacked by Netscape et. al. and a turned into a vehicle for a bunch of their own complaints which is the reason the appeals court sent it back down.
Ummmm, IIRC, the injunction was against distributing something and calling it Java that wasn't really Java
The final settlement of the case was that Microsoft could continue to distribute the JVM with existing products for up to six years. However the question of whether Windows XP is a new version of an old product and thus covered or if it is a new product is arguable. Given Sun's previous behavior attempting to get an injuction against Microsoft releasing products despite a contract clause specifically denying that relief I don't see why people would expect Microsoft to do anything else.
The settlement was not about naming alone, it also covered the code that Microsoft had written. What the settlement came down to is that Sun now has sole control over the future of Java. Microsoft has no say in the future of the language, nor does anyone in the 'open source' community. Why should any company be forced to distribute a closed proprietary system in the name of 'open standards'?
The removal of Java from Windows XP is due soley to the lawsuit. The party that started the lawsuit is Sun, so blame Sun for the removal of the Java VM.
I'm astounded that Apple hasn't gone down that path. With the name recognition of "Apple" and the developer pool for *BSD, as well as commercial giants like Microsoft and Adobe, OS-X on the PC could definately compete.
And who would buy it? What programs would run on it?
To be astounded at Apple's failure to act must be an exhausting condition. Apple has been sleeping at the wheel at least fifteen years. Even Steve Jobs couldn't think of anything to do with the company that was more innovative than pretty boxes designed for niche marketing to architects and hairdressers.
The idea is utterly crackpot. Jobs has already failled to sell Nextstep on x86 once and almost ten years has gone by since. The sole value in the Apple O/S is that the software manufacturer controls the hardware. So the whole system is guaranteed to work together.
As for the name recognition of Apple, don't be too sure that it has a good reputation outside the US. Resentment over Apple's past discriminatory pricing is still remembered. US Apple weenies might think the company the embodiment of good but I think of it as the company that wanted to charge me $3000 more for a computer because I lived in Europe.
A while back Kitty Kelley wrote a biography of Nancy Reagan creating a new litterary genre, the nasty-ography. The people in the media who thought Reagan a complete fool loved the book and it pretty much reversed the growing Reagan myth for a while until the dope's Altzheimer's disease became public.
Then Kelley tried to repeat the trick and wrote a book about the UK Royal familly, oh dear. The problem was not that people did not want to hear bad things about the Royals, quite the contrary, after the soap opera divorces, familly feuds etc. the monarchy had become very unpopular. But Kelley's book made a whole series of unsubstantiated tabloid rumours that the Buck house PR team could explode with little difficulty. At the very time when the country was sickened by their reaction to Princess Diana's death the Kitty Kelley nasty-ography brought them undeserved sympathy.
I think that the Gillmore article and others like it are likely to cause the same reaction. It is very noticable that the Slashcrew have got seriously out of sync with the readership on this one. Most of the posts are saying 'why give us this ill informed made up crap?".
After all if we are going to start attacking MSFT on the basis of made up stuff it might as well be good made up stuff.
The article is ridiculous. What the guy who wrote it does not tell the readers is:
Sun got an injunction to stop Microsoft distributing Java
The 'key features' for which Passport is required are the instant messaging service. It is kind of hard to use an IM without some sort of identity registration and AOL is keeping AIM closed
Kodak want to install a driver that directs all pictures taken in an open format to Kodak's own web site. Microsoft has told them that it does not meet their requirements for drivers that ship with the O/S. Kodak is using FUD to try to get its own way.
Code signing has been used in Active-X and Java downloads for five years. Microsoft has never attempted to use the scheme to exclude software vendors and is not actually a CA for code signing certificates.
The idea of smart-tags was that anyone could set up an annotation service. Hooking up to Encarta as a default seemed a good idea.
The finding of the appeals court was that Microsoft was a monopoly, the tying claim in the windows case that Gilmor claims several times was affirmed was in fact reversed.
But who cares about facts when you are a silicon valley journalist and your readers will suck up anything thats anti-Microsoft even if it is utter lies.
The guy sounds like he wants to be the Rush Limbaugh of the tech sector. I guess next we will be hearing how Microsoft own the media and negative views of microsoft can never be heard.
If Ricochet can't make a go of wireless data I am at a loss to see how the 3G stuff is meant to catch on. 3G has much higher infrastructure costs and if Ricochet can only get 50K users @ $70 I don't see where the millions of 3G users are comming from paying $100 a month...
The big problem with Ricochet was the coverage area. I spend almost enough time in silicon valley to make it worthwhile. But I'm not too keen on spending $70 a month for a service I can't use arround my home in Boston. If I lived in Palo Alto I would not be too happy paying that amount for a system I can't make use of when I am on the road.
If every airport, starbucks and business class hotel in the US deployed 802.11B I suspect there would be practically no need for 3G, Ricochet or the rest.
I was always taught that because pi is infinite (i.e. never ends), it must repeat itself somewhere in itself... (Admittedly it can be argued against)
Then you were taught wrong. In fact I can prove the opposite. If PI reccurs within PI in any number base then PI is a rational number, therefore it cannot reccur.
The reason is that if we have 0.123123123... then we can divide it by 1000 and subtract it from itself to get 0.123, so if x - 0.001 x = 0.123 then x = 0.123 / 0.999 => x = 123/999
The proof can be trivially extended to cover numbers of the form 0.1111123123123123...
The flaw in the original logic is that a proof that every finite string occurs in PI does not consititute a proof that every infinite string occurs in PI.
I strongly suspect that Sony and Macrovision are both blowing smoke. Both companies are making unsubstantiated claims that are on their face pretty suspicious.
One thing to consider is that for the RIAA purposes of discouraging piracy they don't need to break people's equipment, they just need to make them worried enough that they might. Equally they don't need to introduce no-rip CDs, they merely need to convince people that ripping may not be possible in the future.
The various technologies are not particularly believable for several reasons. Not least the fact that they all depend upon implementation artifacts that are almost certain to be corrected in response to their actions. In the case of Macrovision the company appears to be doing no more than exploiting some already irritating bugs in the standard Microsoft Windows CD Rom driver. The 'halt on read error' was always a problem, try ripping an older CD and you will see the same problem.
In the case of the Sony scheme the system can only possibly work by exploiting some design error in certain players.
The schemes might be viable as a legislative prop, develop some ridiculous copy protection scheme then go to congress to insist that all future CD Rom players have stuff in them to detect 'copy-protected' disks.
I suspect that the RIAA has been wearing out its wellcome on the hill though. There are several formerly pro-RIAA Senators who are publicly saying they believe they were lied to. The RIAA would be taking a big risk if it went to buy fresh legislation, could end up seeing its earlier gains reversed.
Once you get issued a patent, you must put it up for auction. After the auction is over, flip a coin. Heads, and the government pays you 120% of the winning bid and puts the patent in the public domain. Tails, and the high bidder buys the patent.
The protocol is broken. Consider the following case, I file for 10 patents, I put them up for auction per the protocol. I then bid for the patents myself (possibly through one of my offshore companies).
Under such a scheme I can bid a billion dollars for my own patent, secure in the knowledge that if I am forced to pay I will be paying myself.
The CMU scheme sounds like the work of someone who has spent too much time reading Aynn Rand and too litte time noticing that most markets are rigged.
In a field like computer science why are we all recomending books almost twenty years old?
A lot of folk recomend Knuth, how many people actually use it for more than a reference? The biggest problem with TAOCP is that it is asurvey work that is now twenty odd years out of date. Lots of stuff has been developed since. The other is that the algorithms are presented in pseudo-machine code.
Don is a great guy, but he has been promising volume 4 since before I started grad school twelve years ago. If people are still using TAOCP as a guide to computer algorithms has it become a liability rather than a benefit? Would people read the litterature for newer and better techniques if they did not always go to TAOCP?
Equally, the dragon book is somewhat questionable. yacc is a great tool for writing LR(1) parsers, only problem is that Chomsky's model of human grammars are not the best plan for computer languages. Back in the 80s we thought this stuff was wonderful, now it seems more like a backwards step.
One book I can recommend without reservation is the Handbook of Applied Cryptography by Menzes, van Orschot et. al. This is the academic equivalent of Bruce Schneir's excellently readable but somewhat unreliable Applied Cryptography.
Secret's and lies would qualify for being in the library but for the ridiculous claim that Bruce only just realized that security was about risk control. He showed no surprize when I mentioned that to him three years ago, and he certainly does not credit me with the discovery (which of course I am not). But he is certainly right that it is a good spin to sell books 'I have found the alchemists stone'.
You can download SOAP, UDDI and WSDL implementations for free from IBM, and they're all in Java. Most big Java application servers already support Web Services. I see little value in.NET.
The implementation language is irrelevant. The VM strategy is relevant insofar as it affects performance. With C# Microsoft are not sacrificing any performance, Java implementations are still processor and memory hogs. Yes you can get a Java coder who can write code faster than a bad C programmer, the same is true for LISP, but the architecture still extracts a performance penalty.
If you think that.NET is simply a language war then you just don't get it, C# is not central to the.NET architecture. It is however central to getting developers to pay for another edition of Visual Studio.
Java and C# are both at base merely a way of adding objects to C than the botched job they did in C++. C# is at least open in a way that Java is not, with Java Sun reserves to itself absolute control of the future of the language. Hello, this is the future of Open Source, I don't think soooo. Not only that but Sun will sick lawyers onto folk who don't do it their way. It isn't just Microsoft that Gosling and co don't want to hear from, there are plenty of people arround who are serious heavyweights in the language design arena that got the brush off. Microsoft were very interested to hear my ideas, Gosling was not. Sure they may 'steal' my ideas, but that is what I want them to do, use them. In fact I give them for free.
C# has several very nice ideas about how to write programs that use XML. It is not the only language that can be used to write XML programs, but it does make it easier and the advantage is enough to justify the cost of the compiler. I would be very happy if someone would do a C# front end for gcc. If that happened I would probably do all my coding in C#. I am not going to move to Java however because I don't like the baggage that comes with it, nor the 20Mb run time that the luser has to download to run the program, or for that matter the crappy GUI widgets that the loosers at Sun want me to use.
I don't give a honk about write once, run anywhere. There are only two platforms that matter today, Linux and Windows. Running on Apple, Solaris, VMS, Irix, HPUX, Genera, BeOs, etc. is not something I am going to sacrifice 80%+ of my CPU for or spend three times as long writing the program putting in optimizations that depend on the structure of the VM implementation or JIT.
The report quoted states that Minrad applied for the trademark. Another slashdotter checked the PTO site to find the Minrad application and the Lucas trademark grants.
But no, the sensationalist coverage in Yahoo has to be definitive, even though as you yourself quote it as saying:
when it asked permission from the government to call its new group of tools Light Sabers
This is the US, not the USSR, you don't ask permission from the government to name your product, you ask permission to stop other people using your name
Do you really think the court records of a district court are bogus? I don't believe you
You gibber on about checking facts and then on the basis of an idiot reporting in Yahoo claim that your equally dim witted opinion is backed by court records you have not read. You are not by any chance a relative of Jeffrey Archer? Somehow you don't seem to be practising what you preach.
The only thing I could make out in the article was that the guy does not know anything much about.NET but is very paranoid about what he does not know.
Passport and Hailstorm are great things for folk to get paranoid about. But they are only a means of breaking open the AOL Instant Messenger position. Microsoft wants to stop AOL from being able to leverage the IM login as a universal interface.
.NET is really about Microsoft's entry into the one area of the software industry it does not dominate - enterprise resource planning. Go to SAP or Oracle and they will charge you $10 million plus for a pile of barely implementable crap. Microsoft think they have a better idea and have been hiring and strategising accordingly. The most significant part of.NET is that Microsoft has co-opted IBM as an aly.
The way to make ERP software pay is for the company that wrote the software to run it as hosted software. This is for several reasons. First the costs of 365x24 support are amortized over hundreds of companies. More importantly however for the customer the company that wrote the software bears the cost of maintenance and the pain of all the unreliability, bugs etc.
The core of the.NET strategy however is somewhat subtler. Traditional ERP systems force you to rip out your existing installation and replace. SOAP allows you to take existing databases and applications, write a thin layer wrapper around them and have them integrate with other Web Services. Microsoft has a two tier strategy, everything runs SOAP, Windows 2000 however will be the SOAP platform with the most, the most APIs, the most tools, etc. etc..NET will fail however unless you can run the interface parts of it on other systems - including MVS, VMS, Solaris and of course Linux.
An open source version of parts of.NET is not a beachead against the open source community, it is denying Microsoft competitors revenue. Sun is already in freefall as UNIX types realise that a low cost Intel box running linux runs faster and more reliably than an overpriced Sun box. So deny Sun the revenues they might gain from selling Web Services boxes.
Sun, Oracle and Netscape ganged up to stop Bill with the anti-trust lawsuit. The story of how the suit was filled is a pretty disgusting case of special pleading by one group of corporations against another, especially if you don't like Microsoft. What could have been a successful anti-trust case became the explanation of why Netscape did not replace Microsoft.
.NET is simply Bill's way of getting revenge. However unlike McNealy, Ellison and Clarke Bill is not stupid enough to blab his mouth off in public about the companies he wants to destroy. The fact that he is not talking about Sun or Oracle is an insult to them, he is saying that they are not going to be players in the future of the software industry. Open source on the other hand is the only serious competitor left.
.NET may be just hype, but hey so is Java. But for Java Sun's lackluster processor performance would have consigned the company to history along with SGI and DEC. Instead Java put Sun right at the center of the Internet boom. Even if you dismiss.NET as meaningless hype, the point is that.NET has captured the airtime and there is none left for Sun or Oracle to launch JavaII or the like.
Lucas is not suing Minrad, they are objecting to Minrad's application to stop anyone else calling the device a light sabre.
So all the weenies who are gibbering on about how evil Lucas is for defending his trademark have to ask why Minrad should have exclusive trademark rights to the name 'Light Sabre'.
The trademark categories are not definitive, an application in one category does not foreclose a dilution claim in another category. In this case I think Lucas's lawyers have very good case. Minrad want to trade on the name recognition that Lucas has created. If they want to do that presumably the greedy bastard lawyers at Lucas will be happy to license the light sabre trademark to the greedy bastards running Minrad.
Lucas is almost certainly not directing this as a personal vendetta. His trademark lawyers are simply doing their job.
Actually, I think McCarthyism was principally instigated by Joseph McCarthy. He used Hoover to do some of the dirty work, sure.
Other way arround. McCarthy was pretty stupid which is why he over-reached. Hoover was very much the driving force. He fed the HUC with dirt on their targets, suggested many of them, hounded HUC critics and generally acted as one of the ring leaders, certainly he was a bigger driving force than McCarthy.
At the same time the FBI was refusing to act against organized crime and aiding and abetting the Klu Klux Klan.
I am not an anti-statist loonitarian. However the fact is that a lot of the suspicion of the state in the US is justified and the FBI is the principal cause of that suspicion.
perhaps if they beefed up the encryption in passwd files, shadow passwords would not be such an issue, i mean, that is what encryption is for is it not?
No, the security of the password file has almost nothing to do with the security of the password algorithm. I am not aware of anyone breaking the DES encryption algorithm to break passwords.
The vulnerability comes from the ability to obtain passwords by brute force examination of the password space - a dictionary attack. Using AES would not make a difference because the issue is the number of possible passwords. And adding a random letter or number in the middle of the password 'pass3word' does not change the search difficulty as much as the naive sysops think.
In fact the stupid constraints (case sensitive, must have symbol) used to patch up weak password infrastructres should really count as security through obfustication at this point. The added security is measurable but woefully insufficient and really inconvenient for the user - why can't the password alg check for both positionws of CapsLock?
There are much better security measures than shaddow passwords, however at this stage relying on the encryption alone is nowhere near sufficient.
There are no statues to Beria left in the former USSR but the name of J. Edgar Hoover still disgraces the FBI building. Hoover was the principal instigator of what is known as McArthyism and the FBI has failed to throw off his tactics after his death.
What is the excuse for denying bail in this case? Suspect might write more software that would harm the interests of corporations?
Back about thirty years ago most computer security depended on physical security alone, once you were in the door of the building you could pretty much do anything you wanted with just a little determination. Password systems to the extent they existed were pretty rudimentary and security was not considered as part of the O/S design, or to the extent that it was the 'security' mainly consisted of feelgood measures for marketing purposes.
Security Through Obscurity was one of the slogans thrown back at the complacent computer vendors and sysadmins. As with any slogan though the medium does not allow for a sophisticated or subtle message.
About ten years ago the UNIX comunity started to wake up to security issues. Previously the UNIX approach had been pretty rudimentary, most UNIX machines were departmental or personal workstations and security was a nice to have but relatively few people depended on it actually working.
Then the Internet grew beyond academia and suddenly security started to matter an awful lot.
At the same time traditional UNIX security nostrums started to be re-examined, like the lack of shaddow passwords. At the time the idea of protecting a password file with a password was considered by most UNIX advocates to represent the evil of security through obscurity. The UNIX one way encrypted password file was held up as the exemplar, to want to protect it was to fall foul of the sin of 'security through obscurity' and obviously anyone like myself who advocate3d the practice had to be a fool. Today anyone who doesn't use shaddow passwords is considered foolish.
What it boils down to is that slogans have a tendency to turn into dogma. Then someone comes along and points out that the real world is not as simple as the slogan.
Slashdot Mirror
As with all large companies Microsoft files lots of ridiculous patents. They do it for the same reason mine does, so that if they are sued by another company thay have something for swapsies.
It would probably not be a good thing for Microsoft if their customers could not attach Linux file systems easily. SAMBA is simply collateral damage in the high stakes game between EMC and Microsoft. EMC servers are very expensive and Microsoft would love to play bigger on that turf.
The bigger problem is that in the crackpot US PTO scheme you never know if a patent has been applied for on something until the government awards a 20 year monopoly in practicing it. The rules have been improved, i.e. made less open to corrupt abuse but they are still an extortionists charter.
I can't remember the last time Microsoft was the plaintif in a Patent lawsuit. They have been the victim of many Patent Trolls.
It would be an idiotic strategy for Microsoft to try to use patents to make .NET proprietary. But then again the tax cut for the ultra-rich and breaking the ABM treaty to build a 21st century Maginot line are crackpot ideas.
The business model for Curl is laughable. It is based on an MIT conceit that the programming language is all important. When I looked at the Curl stuff a few years back it looked like yet another attempt to make C look like Lisp. Yeah whatever but Lisp does quite a good job of looking like Lisp.
If you have to download a plug in to use the language you might as well write the code in C# and be done with it. The guys who wrote C# are at least as smart as the guys in 545 Tech square.
Ward did not strike me as being the live wire of the lab, nor for that matter did his students. Dertousos is just the Director, like Negroponte he does not do, he presents, unlike Negroponte however I doubt that many grad students would go to talk to Detousos for inspiration.
Bottom line is I did not see people queueing up to learn Curl inside Tech square so I don't see why anyone outside the building is likely to use it.
The most important applications are web surfing, Word and the ability to play DVDs on the plane with Civilization II next in line. Unfortunately I can't get Tomb Raider to work on my Sony Vaio Z505
What I would really like to get hold of is one of those tablet PCs with pen input. I used to have a couple of CrossPads but the software sucketh somewhat and is is something of a kludge. It is not bad at handwriting recognition but it is tedious having to tell it each time you turn the page. I would be happy to pay $3K for a tablet PC with a decent spec, but getting hold of one in the US appears impossible. Anyone know if they work?
The Sony is pretty cute, the revamp looks slightly better, I have the plug in DVD which is a pain for plane flights.
No it isn't. The PTO system relise upon the honest of the applicants. The PTO does not have the expertise to search for prior art, never has, never will.
The US PTO can't even perform searches of already issued patents competently, so what is the chance that in the 10 hours allowed for the average examination the clerk with a law degree is going to find prior art? An examiner does not need any actual experience of the field they are examining, just a relevant(ish) degree.
What is meant to be the bar to malicious/criminal patent applications is the difficulty of enforcing the pieces of utter crap that get issued. Then Lemelson came along with his perjured 'bar codes' continuation of 'machine vision' and extorted a billion dollars.
The system stinks and filling open source patents won't fix it. Unless someone wants to lay out $2,000,000 trying to enforce the patent it is worthless and pointless.
People have ruyn one O/S under another for years, IBM are currently selling their ability to run multiple Linux VMs under their MVS O/S - which had the idea so long ago that the patents have expired years ago.
I have only ever seen one patent that I could not work out a way to bust - Diffie-Helleman and that is long expired now. If the Open Source community start to use patents to try to force people to do everything their way I think they will have stopped being the solution and become the new problem. Its the type of mind control, silly-ass games that turn people off. I don't think the patents filled by the Open Source Community are likely to inconvenience me, just another patent to bust. The Open Source Community would probably have more problems if I started filling broad patents on my own work rather than putting everything into the public domain which I have done to date.
You can't take the low road to the moral high ground.
I have a Vodoo 3 card as well, only my problem is that I can't get it to work with my existing O/S, Tomb Raider Chronicles that is. But the problem is that 3DFX being a gonna and not having released their drivers open source the hardware is going to be useless sometime or other regadless.
Active Directory is an LDAP interface, BIND is a DNS interface. Active Directory also provides DNS support but the underlying data model is LDAP and the probability that any BIND code would be useful is zero.
At the time IIS first appeared Apache did not exist, it was still the NCSA Web server with a bunch of third party patches. Thau was still doing major surgery on the first release of Apache while I was running IIS in the office across the hall from him. IIS could conceivably contain some of the CERN Libwww code, but that was put in the public domain, it is not open source restricted. The Microsoft lawyers called up to ask what the status of the CERN code was before MSFT downloaded it.
But still it is easier to make completely unsubstantiated allegations, admitting that you have no evidence apart from your belief that Microsoft >= absolute evil => If it is evil Microsoft must be doing it.
Since you appear to be a Newbie Microsoft-basher I will help you with some hints:
The Register article itself states that the blocking of the old incompatible application versions is taking place with the knowledge and co-operation of the companies themselves who are not complaining. Therefore Microsoft must bave blackmailed the companies into not complaining
The mechanism is a blacklist that lists bad programs that cannot be run. Therefore Microsoft csn stop you running your own software by not including it on the blacklist.
What actually happened (and as was discussed at length on ./ already was that Kodak and a bunch of other manufacturers agreed on a standard for hooking up a camera. Microsoft installed a default driver for the standard in the O/S. Clearly an evil thing to do.
Kodak wanted to distribute a modified driver that when tested was found to register the Kodak image software as the default for all manufacturer's cameras using the format. Needless to say the driver did not ask the user before reconfiguring their machine. The assumtion being that anyone who bought a Kodak camera was somehow Kodak property.
There is a big different between getting a bunch of pesterware drivers from the vendor of the O/S and getting them from everywhere. First thing I do with a windows box is to throw the MSN icon and the rest of the junk in the recycle bin and pull the chain on it. Then I know that they are gone for good. The Kodak driver is like the original Netscape browser, it reinstalls itself over your defaults each time it runs.
Personally I think the Kodak scheme to drive consumers to their web site to buy printed copies is somewhat naive. The first thing I discovered after buying my Nikon Coolpix was that I stopped using film for snapshots. The second was that with electronic pics you load them onto a Web site and bore your relations remotely. The print out on dead tree step can be omitted entirely.
uhhh... your joking right... MSNBC ring a bell? Oh sure if they really owned the media Microsoft would probably have to add to their name, AOL-TimeWarner.
No joke. One of the distinguishing features of an idealogue is that they cannot process contrary views. This is is called in the propaganda analysis field a 'stoccoma'. This then leads to the claim that all contrary views are due to 'the liberal media', a phrase that Murdoch's Fox news repeats several times an hour. Murdoch owns a larger media empire than Hurst, but he can't be heard because of the 'Liberal Media', yeah that would be Salon.com.
Notice in this thread people are already accusing anyone questioning the article as being a Microsoft PR flack. The point raised about MSNBC is important, yes Microsoft do own half of a calbe TV news station and do have an alliance with an infulential media owner. However their opponents Time Warner/AOL-Netscape happen to own more cable stations, channels magazines etc. than anyone else.
The same mindset could be seen on both sides during the 'red scares' orchestrated by Hoover and McCarthy. The communists refused to believe anything bad about Stalin's regime, denouncing the capitalist press. The scare-mongers believed that anyone who opposed them for any reason had to be a communist.
As I pointed out elsewhere, the original piece is an opportunity for Microsoft. If I were a Microsoft flack I would have written the original piece so others would knock it down and with it other anti-MSFT views that might be valid. The trial fell victim of the same problem, a perfectly reasonable tying claim was hijacked by Netscape et. al. and a turned into a vehicle for a bunch of their own complaints which is the reason the appeals court sent it back down.
The final settlement of the case was that Microsoft could continue to distribute the JVM with existing products for up to six years. However the question of whether Windows XP is a new version of an old product and thus covered or if it is a new product is arguable. Given Sun's previous behavior attempting to get an injuction against Microsoft releasing products despite a contract clause specifically denying that relief I don't see why people would expect Microsoft to do anything else.
The settlement was not about naming alone, it also covered the code that Microsoft had written. What the settlement came down to is that Sun now has sole control over the future of Java. Microsoft has no say in the future of the language, nor does anyone in the 'open source' community. Why should any company be forced to distribute a closed proprietary system in the name of 'open standards'?
The removal of Java from Windows XP is due soley to the lawsuit. The party that started the lawsuit is Sun, so blame Sun for the removal of the Java VM.
And who would buy it? What programs would run on it?
To be astounded at Apple's failure to act must be an exhausting condition. Apple has been sleeping at the wheel at least fifteen years. Even Steve Jobs couldn't think of anything to do with the company that was more innovative than pretty boxes designed for niche marketing to architects and hairdressers.
The idea is utterly crackpot. Jobs has already failled to sell Nextstep on x86 once and almost ten years has gone by since. The sole value in the Apple O/S is that the software manufacturer controls the hardware. So the whole system is guaranteed to work together.
As for the name recognition of Apple, don't be too sure that it has a good reputation outside the US. Resentment over Apple's past discriminatory pricing is still remembered. US Apple weenies might think the company the embodiment of good but I think of it as the company that wanted to charge me $3000 more for a computer because I lived in Europe.
Then Kelley tried to repeat the trick and wrote a book about the UK Royal familly, oh dear. The problem was not that people did not want to hear bad things about the Royals, quite the contrary, after the soap opera divorces, familly feuds etc. the monarchy had become very unpopular. But Kelley's book made a whole series of unsubstantiated tabloid rumours that the Buck house PR team could explode with little difficulty. At the very time when the country was sickened by their reaction to Princess Diana's death the Kitty Kelley nasty-ography brought them undeserved sympathy.
I think that the Gillmore article and others like it are likely to cause the same reaction. It is very noticable that the Slashcrew have got seriously out of sync with the readership on this one. Most of the posts are saying 'why give us this ill informed made up crap?".
After all if we are going to start attacking MSFT on the basis of made up stuff it might as well be good made up stuff.
But who cares about facts when you are a silicon valley journalist and your readers will suck up anything thats anti-Microsoft even if it is utter lies.
The guy sounds like he wants to be the Rush Limbaugh of the tech sector. I guess next we will be hearing how Microsoft own the media and negative views of microsoft can never be heard.
The big problem with Ricochet was the coverage area. I spend almost enough time in silicon valley to make it worthwhile. But I'm not too keen on spending $70 a month for a service I can't use arround my home in Boston. If I lived in Palo Alto I would not be too happy paying that amount for a system I can't make use of when I am on the road.
If every airport, starbucks and business class hotel in the US deployed 802.11B I suspect there would be practically no need for 3G, Ricochet or the rest.
Then you were taught wrong. In fact I can prove the opposite. If PI reccurs within PI in any number base then PI is a rational number, therefore it cannot reccur.
The reason is that if we have 0.123123123... then we can divide it by 1000 and subtract it from itself to get 0.123, so if x - 0.001 x = 0.123 then x = 0.123 / 0.999 => x = 123/999
The proof can be trivially extended to cover numbers of the form 0.1111123123123123 ...
The flaw in the original logic is that a proof that every finite string occurs in PI does not consititute a proof that every infinite string occurs in PI.
One thing to consider is that for the RIAA purposes of discouraging piracy they don't need to break people's equipment, they just need to make them worried enough that they might. Equally they don't need to introduce no-rip CDs, they merely need to convince people that ripping may not be possible in the future.
The various technologies are not particularly believable for several reasons. Not least the fact that they all depend upon implementation artifacts that are almost certain to be corrected in response to their actions. In the case of Macrovision the company appears to be doing no more than exploiting some already irritating bugs in the standard Microsoft Windows CD Rom driver. The 'halt on read error' was always a problem, try ripping an older CD and you will see the same problem. In the case of the Sony scheme the system can only possibly work by exploiting some design error in certain players.
The schemes might be viable as a legislative prop, develop some ridiculous copy protection scheme then go to congress to insist that all future CD Rom players have stuff in them to detect 'copy-protected' disks.
I suspect that the RIAA has been wearing out its wellcome on the hill though. There are several formerly pro-RIAA Senators who are publicly saying they believe they were lied to. The RIAA would be taking a big risk if it went to buy fresh legislation, could end up seeing its earlier gains reversed.
The protocol is broken. Consider the following case, I file for 10 patents, I put them up for auction per the protocol. I then bid for the patents myself (possibly through one of my offshore companies).
Under such a scheme I can bid a billion dollars for my own patent, secure in the knowledge that if I am forced to pay I will be paying myself.
The CMU scheme sounds like the work of someone who has spent too much time reading Aynn Rand and too litte time noticing that most markets are rigged.
A lot of folk recomend Knuth, how many people actually use it for more than a reference? The biggest problem with TAOCP is that it is asurvey work that is now twenty odd years out of date. Lots of stuff has been developed since. The other is that the algorithms are presented in pseudo-machine code.
Don is a great guy, but he has been promising volume 4 since before I started grad school twelve years ago. If people are still using TAOCP as a guide to computer algorithms has it become a liability rather than a benefit? Would people read the litterature for newer and better techniques if they did not always go to TAOCP?
Equally, the dragon book is somewhat questionable. yacc is a great tool for writing LR(1) parsers, only problem is that Chomsky's model of human grammars are not the best plan for computer languages. Back in the 80s we thought this stuff was wonderful, now it seems more like a backwards step.
One book I can recommend without reservation is the Handbook of Applied Cryptography by Menzes, van Orschot et. al. This is the academic equivalent of Bruce Schneir's excellently readable but somewhat unreliable Applied Cryptography.
Secret's and lies would qualify for being in the library but for the ridiculous claim that Bruce only just realized that security was about risk control. He showed no surprize when I mentioned that to him three years ago, and he certainly does not credit me with the discovery (which of course I am not). But he is certainly right that it is a good spin to sell books 'I have found the alchemists stone'.
The implementation language is irrelevant. The VM strategy is relevant insofar as it affects performance. With C# Microsoft are not sacrificing any performance, Java implementations are still processor and memory hogs. Yes you can get a Java coder who can write code faster than a bad C programmer, the same is true for LISP, but the architecture still extracts a performance penalty.
If you think that .NET is simply a language war then you just don't get it, C# is not central to the .NET architecture. It is however central to getting developers to pay for another edition of Visual Studio.
Java and C# are both at base merely a way of adding objects to C than the botched job they did in C++. C# is at least open in a way that Java is not, with Java Sun reserves to itself absolute control of the future of the language. Hello, this is the future of Open Source, I don't think soooo. Not only that but Sun will sick lawyers onto folk who don't do it their way. It isn't just Microsoft that Gosling and co don't want to hear from, there are plenty of people arround who are serious heavyweights in the language design arena that got the brush off. Microsoft were very interested to hear my ideas, Gosling was not. Sure they may 'steal' my ideas, but that is what I want them to do, use them. In fact I give them for free.
C# has several very nice ideas about how to write programs that use XML. It is not the only language that can be used to write XML programs, but it does make it easier and the advantage is enough to justify the cost of the compiler. I would be very happy if someone would do a C# front end for gcc. If that happened I would probably do all my coding in C#. I am not going to move to Java however because I don't like the baggage that comes with it, nor the 20Mb run time that the luser has to download to run the program, or for that matter the crappy GUI widgets that the loosers at Sun want me to use.
I don't give a honk about write once, run anywhere. There are only two platforms that matter today, Linux and Windows. Running on Apple, Solaris, VMS, Irix, HPUX, Genera, BeOs, etc. is not something I am going to sacrifice 80%+ of my CPU for or spend three times as long writing the program putting in optimizations that depend on the structure of the VM implementation or JIT.
The report quoted states that Minrad applied for the trademark. Another slashdotter checked the PTO site to find the Minrad application and the Lucas trademark grants.
But no, the sensationalist coverage in Yahoo has to be definitive, even though as you yourself quote it as saying:
when it asked permission from the government to call its new group of tools Light Sabers
This is the US, not the USSR, you don't ask permission from the government to name your product, you ask permission to stop other people using your name
Do you really think the court records of a district court are bogus? I don't believe you
You gibber on about checking facts and then on the basis of an idiot reporting in Yahoo claim that your equally dim witted opinion is backed by court records you have not read. You are not by any chance a relative of Jeffrey Archer? Somehow you don't seem to be practising what you preach.
Passport and Hailstorm are great things for folk to get paranoid about. But they are only a means of breaking open the AOL Instant Messenger position. Microsoft wants to stop AOL from being able to leverage the IM login as a universal interface.
The way to make ERP software pay is for the company that wrote the software to run it as hosted software. This is for several reasons. First the costs of 365x24 support are amortized over hundreds of companies. More importantly however for the customer the company that wrote the software bears the cost of maintenance and the pain of all the unreliability, bugs etc.
The core of the .NET strategy however is somewhat subtler. Traditional ERP systems force you to rip out your existing installation and replace. SOAP allows you to take existing databases and applications, write a thin layer wrapper around them and have them integrate with other Web Services. Microsoft has a two tier strategy, everything runs SOAP, Windows 2000 however will be the SOAP platform with the most, the most APIs, the most tools, etc. etc. .NET will fail however unless you can run the interface parts of it on other systems - including MVS, VMS, Solaris and of course Linux.
An open source version of parts of .NET is not a beachead against the open source community, it is denying Microsoft competitors revenue. Sun is already in freefall as UNIX types realise that a low cost Intel box running linux runs faster and more reliably than an overpriced Sun box. So deny Sun the revenues they might gain from selling Web Services boxes.
Sun, Oracle and Netscape ganged up to stop Bill with the anti-trust lawsuit. The story of how the suit was filled is a pretty disgusting case of special pleading by one group of corporations against another, especially if you don't like Microsoft. What could have been a successful anti-trust case became the explanation of why Netscape did not replace Microsoft.
.NET is simply Bill's way of getting revenge. However unlike McNealy, Ellison and Clarke Bill is not stupid enough to blab his mouth off in public about the companies he wants to destroy. The fact that he is not talking about Sun or Oracle is an insult to them, he is saying that they are not going to be players in the future of the software industry. Open source on the other hand is the only serious competitor left.
So all the weenies who are gibbering on about how evil Lucas is for defending his trademark have to ask why Minrad should have exclusive trademark rights to the name 'Light Sabre'.
The trademark categories are not definitive, an application in one category does not foreclose a dilution claim in another category. In this case I think Lucas's lawyers have very good case. Minrad want to trade on the name recognition that Lucas has created. If they want to do that presumably the greedy bastard lawyers at Lucas will be happy to license the light sabre trademark to the greedy bastards running Minrad.
Lucas is almost certainly not directing this as a personal vendetta. His trademark lawyers are simply doing their job.
Other way arround. McCarthy was pretty stupid which is why he over-reached. Hoover was very much the driving force. He fed the HUC with dirt on their targets, suggested many of them, hounded HUC critics and generally acted as one of the ring leaders, certainly he was a bigger driving force than McCarthy.
At the same time the FBI was refusing to act against organized crime and aiding and abetting the Klu Klux Klan.
I am not an anti-statist loonitarian. However the fact is that a lot of the suspicion of the state in the US is justified and the FBI is the principal cause of that suspicion.
No, the security of the password file has almost nothing to do with the security of the password algorithm. I am not aware of anyone breaking the DES encryption algorithm to break passwords.
The vulnerability comes from the ability to obtain passwords by brute force examination of the password space - a dictionary attack. Using AES would not make a difference because the issue is the number of possible passwords. And adding a random letter or number in the middle of the password 'pass3word' does not change the search difficulty as much as the naive sysops think.
In fact the stupid constraints (case sensitive, must have symbol) used to patch up weak password infrastructres should really count as security through obfustication at this point. The added security is measurable but woefully insufficient and really inconvenient for the user - why can't the password alg check for both positionws of CapsLock?
There are much better security measures than shaddow passwords, however at this stage relying on the encryption alone is nowhere near sufficient.
What is the excuse for denying bail in this case? Suspect might write more software that would harm the interests of corporations?
Guns don't kill people, bullets kill people
Security Through Obscurity was one of the slogans thrown back at the complacent computer vendors and sysadmins. As with any slogan though the medium does not allow for a sophisticated or subtle message.
About ten years ago the UNIX comunity started to wake up to security issues. Previously the UNIX approach had been pretty rudimentary, most UNIX machines were departmental or personal workstations and security was a nice to have but relatively few people depended on it actually working.
Then the Internet grew beyond academia and suddenly security started to matter an awful lot.
At the same time traditional UNIX security nostrums started to be re-examined, like the lack of shaddow passwords. At the time the idea of protecting a password file with a password was considered by most UNIX advocates to represent the evil of security through obscurity. The UNIX one way encrypted password file was held up as the exemplar, to want to protect it was to fall foul of the sin of 'security through obscurity' and obviously anyone like myself who advocate3d the practice had to be a fool. Today anyone who doesn't use shaddow passwords is considered foolish.
What it boils down to is that slogans have a tendency to turn into dogma. Then someone comes along and points out that the real world is not as simple as the slogan.