'Stable' just means it has a stable binary compatible API. It doesn't mean they won't add new features to it.
No Linux kernel has ever advertised a compatible ABI between revisions, and it's fairly unlikely any ever will. Instead stable generally means that the kernel source API will remain fairly constant through it's lifecycle (so that all the drivers don't end up broken until they are updated to a new API) and few sweeping kernel changes will occur. The 2.4 kernel series has not followed this pattern at least partially due to the fact it's had some rather severe performance, scalability and reliability problems since release. Linus has repeatedly refused to support any type of ABI for the Linux kernel, at least partially in fear of having to constantly support buggy interfaces that can never be removed.
[...] two changes I would make Microsoft release all the API specs in a public forum and make them freely available, instead of just on MSDN.
Microsoft already publishes all the API specs included in MSDN in a free public forum. They don't even make you logon to passport anymore to get at the stuff. The problem is that MSDN is no where near complete, and undocumented APIs, protocols and file formats abound in Windows.
My error. I remembered it was wrong shortly after I posted. The entire 256 colour thing is still rather deceptive, since I seem to recall some SNES demo writers getting at least 4,000 colours on screen at once. Then there were things like the fact the SNES could do an additive or subtractive blend of two backgrounds, which alone generated far more than the 256 on screen.
Can you give examples of games that best take advantage of H-DMA?
Most games used it in some regard. For example, the game F-Zero used the mode7 rotation and scaling and changed the scaling factor each scanline to create the illusion of a 3D track. Games that used a swirling background used H-DMA to change the scroll offset of each scanline to achieve the effect. Final Fantasy 3/6j used it to reprogram the palette for the gradient text dialog boxes. Other games used it to do a video mode switch in the middle of the screen (the only clean way to do it was to switch it vertically, because trying to do a horizontal split would create a seam on the screen). The thing about H-DMA was that these same effects could be done with the hardware of the Sega Genesis (with the exception of mode 7 stuff), except you had to drain a lot of CPU power just waiting for the appropriate events to happen, whereas with H-DMA, the CPU had to do nothing -- the CPU would get temporarily halted while the DMA took place at the speed of the RAM/GPU, and would resume execution the moment it was finished.
The more important factor was the SNES's custom graphics chip. : the SNES can display a maximum of 128 simultaneous sprites
The custom graphics chip and the sony designed sound chip were the hallmarks of the SNES, however, it's big selling feature wasn't the number of sprites on screen, or the number of colours it could display The most powerful feature of the SNES is one that's difficult to explain -- Horizontal DMA (H-DMA). What it was, is the ability to schedule DMA transfers on the GPU unit to occur at specific horizontal and vertical scanlines. The specs for the SNES said it could display 256 colours of a palette of 16768 colours, except when you bring in H-DMA, you could reprogram the palette during the H-blank and display many more than that. Or the sprite limitation could likewise be worked around by using this H-DMA. The SNES's sound CPU was likewise rather incredible. It was an entirely self-contained CPU, RAM and DAC system, and you fed your program into it via a set of I/O registers, and it ran parallel to the main CPU. Inside the chip itself was a system that could best be described as a compressed wavetable synth engine (compared to the Genesis's FM synth)
Well, to connect to the Exchange server, it's just a matter of reverse engineering two DCE RPC services -- the name service provider interface (nspi), and the exchange message store message database (emsmdb). The name service provider seems to be the easier of the two, since it's only for the address boook, and resolving names, and uses a fairly straightforward interface, the emsmdb on the other hand consists mainly of four RPC calls, one is for connecting, one for disconnecting, and two that are simply named DoRpc and GetMoreRpc. For those of you that don't understand, that means they used RPC to encapsulate RPC (probably MAPI). Simply trying to expose CDO would pose other problems, including security and an additional single point of failure. If simply contacting the Exchange server was the goal, it might be just as easy to extend Wine to be able to make RPC calls, and then use a lightweight version of wine with the native Outlook Exchange DLLs to call the server.
Then you're classified as an evil hacker and must be destroyed at any and all costs. Unless you're a licenced academic in good standing at a recognised educational institution, researching other people's encryption is fast becoming a dangerous hobby. It's been said that having a false sense of security is more dangerous than insecurity itself, but that's what these kinds of laws and lawsuits are effectively doing.
The RedHat suggestion was just a transparent PR ploy to try and cash in on the problems in the proposed settlement. IMHO, Microsoft should just be donating that money with no strings attached for the schools to do whatever they want with. I imagine the poorest schools probably need building maintainance and supplies rather than computers. Giving computers to poor schools where poor students attend isn't going to cure the so-called "electronic-divide", since that's far closer to the fact they're economically divided than anything else.
Windows is Microsoft's "Cash Cow", so to speak. It's where they get all the money they devote to research and development, and how they subsidise new projects until their competitors are out of business. However, as I understand it, the $10 off isn't against the price of a WinXP upgrade or OEM version(about $100), but instead against the full WinXP retail cost (roughly $200). Quite frankly, it's silly since most people could easily go to any local retailer, buy a $10 mouse and an OEM copy of Windows for less cost.
Lets be honest. This entire settlement has very little to do with Linux. In fact, I'd wager that this settlement has more to do with the fact that Microsoft would like to break into Apple's other big niche -- schools. Apple has a huge installed base in the school system, and those computers that are there aren't moving out anytime soon. Microsoft says they're willing to supply Macintosh software, but what do you want to bet the PC alternative (shipped with Windows, of course) will be cheaper or more systems for fewer "dollars". If an educator had to choose between 7 macintoshes or 10 PCs with Windows, what do you think they'd choose? Apple should be the one stinging from this more than anyone else -- not only has Microsoft potentially reduced their sales for the next years, they've also increased their installed base, and come out of the entire anti-trust lawsuit smelling like a rose.
What they have is a trademark on the compact disc logos, and license it out to companies that make CDs and hardware to read those CDs. If you made a wind chime and put the CD or CDDA logo on it, you'd be in trouble.
I think all this has more to do with the absolute lack of consistency in WIPO rulings. First the rule that adding sucks to a domain name is a recognised method for protesting a company, and deny the request to have the domain turned over. Later on, when Vivendi Universal gets hit with a sucks.com domain, WIPO rules that [companyname]sucks.com is "confusingly similar", and goes to great lengths do legitimize handing over the domain to them.
I often have to wonder if either ICANN or WIPO have heard of the term "integrity". Can anyone remember when ICANN's mandate was to be democratically elected? Seems like such a long time ago.
My observation of the Liberals in Canada is they tend to take up any position on the political scale that isn't dominated by anyone else. In places where the decidedly left-wing NDP is dominant, they adopt the right-wing. In places where the right-wing conservatives are dominant, they take up the left-wing. In the national arena, they seem to be mostly left, favouring social programs and spending. They seem to remind me of a chameleon for some strange reason.
When was the last time you saw a game that didn't have an ESRB rating on it? The only ones I've seen are those from small vendors that are just trying to start up and haven't the resources to get ESRB rated, or are games that are unlikely to fall foul of any ESRB K-A guidelines (card games, etc)...
Food is FORCED to list its indgediants (I know, a SHOCKING violation of the rights slashdoters hold dear). Hope to see BC reverse its decision on this as well.
Oh, I don't know, maybe perhaps because many people have violent allergic reactions to certain ingredients and can die if they consume them. It has far more to do with public safety than giving people more information.
Rating make it straightforward for folks to take a look at a game and have an idea of where it falls in the violance catagory.
Ratings is the one area that the entertainment software industry has already gone out of their way to provide. Self-regulation is generally preferred over government regulation, because it does tend to work somewhat more efficiently and quickly, and I believe the BC government is dropping the issue because they recognise the work they were trying to demand is already available and working.
I think the problem is that most parents choose to ignore the ratings that already exist, and I doubt that the situation would be improved by government mandated ratings.
There's a lot more to it than that. Netscape's webserver used to be quite popular on Windows NT 3.5x. Many people would simply purchase a copy of Windows NT 3.51 Workstation, and install Netscape's webserver on there, and have a perfectly functioning webserver for as many hits as they wanted to get. Microsoft caught onto this, and modified the license agreement in Windows NT 4.0, to restrict the maximum number of concurrent connections to 10, in order to force more people to buy Windows NT Server (that cost $800 more), which just incidently came with a free copy of IIS. Lets compare here, shall we? In the NT 3.51 days, people often bought a $100 copy of NT Workstation and a couple hundred on Netscape server, versus ~$900 for the MS server, but with NT4, they HAD to buy the $900 MS server along with the Netscape webserver product if they wanted to run Netscape, or they could just settle for IIS that came with that already expensive server software package.
Microsoft insisted that the NT 3.51 license didn't allow use of a webserver, but it was vague enough that it probably couldn't have been enforced, but they made certain that it could be in NT 4.0. Netscape wasn't the only one that was bitten by this license change -- O'Reilly's Website server also fell victim to it. Today, the only places Netscape/iPlanet's webservers still exist are on Novell (it's bundled with certain versions of Netware) and UNIX servers.
but if I find a vulnerability and report it to Microsoft, I think the correct thing to do is wait until Microsoft releases a patch before telling the world about it.
The problem is that Microsoft doesn't want the public to EVER know the problem existed. They want security professionals to withhold all the information except for the bits that Microsoft publishes in their security bulletins. Microsoft says they want complete control over the security bulletins for their products, because of some wishful thinking that by filtering the flow of that information, they will stop people from creating worms like CodeRed and such.
I see two options for this -- either they want this control so they can downgrade the severity of discovered holes so that groups like Gartner stop telling people to dump IIS, or they misunderstand the underground cracker community and it's ability to spread information about exploits and to discover holes on their own. I believe that by having nothing but filtered reports that are approved by Microsoft, we'll end up in much greater danger. First, we'll be in the position to be unable to verify Microsoft patches actually fix the problems they are released to fix. And second, legitimate administrators will be at a far worse disadvantage by it than malicious users will (collecting exploit information from less mainstream souces is not an activity you want to perform at work -- most are littered with porn popups and banners).
I can hear the retort, that most Administrators never use the sample exploits and vulnerability tests that are attached to advisories, and that's true. However, if you've been cracked into for the third time this week, and you've applied all the security patches and it's still happening, those sample exploits and vuln tests might suddenly come in useful. Especially on the Microsoft platform, several patches replace previous ones, and if you accidently apply a previous patch over a newer one, you can end up looking like you have the fixed version installed, but still have the vulnerable version installed.
Intuit did a good job of giving Microsoft the shaft when they locked MS out from online bill paying so that Money would still have trouble competing with Quicken.
Well, I don't know about anyone else, but my upgrade copy of Windows95 came with a coupon for a free copy of MS Money 95 for the price of shipping and handling. Or perhaps the fact they've been pushing OEMs to bundle Works Suite with all new computers, which also contains Microsoft Money. They might not have managed to stab Intuit in the back, but they haven't stopped trying.
if hard proof of "knifing in the back" can be provided, I would like to see it.
Well, there was always the test that beta versions of Windows 3.1 did for MS-DOS that did a completely synthetic "compatibility test" on the host OS and gave an error if it wasn't MS-DOS (or PC-DOS). The test didn't make it into the final version, but the beta version's error message was enough to disuade OEMs from selling it. Caldera eventually won a lawsuit against MS for exactly that (the lawsuit was initiated by either Digital Research or Novell, can't remember which).
Then there was the Stacker incident with MS-DOS 6, where Microsoft was found to be guily of patent infringement for the disk compression technology in MS-DOS 6.2 (which is why 6.21 contained no compression, and was reintroduced in 6.22).
Then there's the active wars. Microsoft continues to try to leverage Microsoft clients to encourage people to drop their non-MS servers (such as Novell Netware servers). They've also introduced Windows Media Player, and are agressively pushing it on portable player manufacturers and including it in Windows XP to try and flush out Real. Then there's AOL and Microsoft trying to push people to MSN instead, again by using their OS to try and make it more difficult to use AOL than it would be to just break down and use MSN.
Most people who buy pre-installed Windows computers barely know how to operate them, let alone install an OS on them. People mess up where to plug in the phone cable and speakers all the time, and lets not even get into the idiot who managed to plug a USB device into the serial port.
If I was a MS stockholder, I'd prefer that behemoth was split, a la ATnT. In few years they'd be more valuable than the current single entity
I'd wager to say that if they were broken up, the keyword Microsoft has been harping about through this entire trial might become more of a reality. For a long time now, Microsoft has been trickling changes to consumers, afraid or unwilling to make any real drastic changes. The changes between Win95, Win98, and WinME are minor to say the least, and likewise between Win2K and WinXP (NT4->W2K was a little more drastic, then again, it was in development for a LONG time). If MS was split in two, the layers that squish new ideas before they're formed would likely diminish and both of the new MS companies would probably be much more likely to introduce new innovative products that may or may not end up being a flop.
The copy protection in the Lucasarts Monkey Island games consisted of code wheels and things like that. It's unlikely the scummvm interpreter would block them (in fact, I checked, and it doesn't).
Re:Response time of thermal diode is less than 1ms
on
AMD And THG update
·
· Score: 1
I wouldn't be surprised if THG ever gets in legal problems over issues like this. First they slander AMD, secondly they violate copyright??? Or DID they ask the author of the soundtrack for his written permission? Doubtful...
Mr. Voelkel needs to review journalism ethics. His review his lacking tact & neutrality. I'm pretty sure if he pulled the same stint with Intel, he probably would have a cease & desist letter in his mailbox by now...
Why is it everytime that Tom's Hardware Guide comes up with something praising a feature of Intel, and showing a weakness of AMD's line, he's the big devil, and in leagues with Intel trying to destroy AMD, yet time and time again, his benchmarks prove that AMDs are much faster than Intels. People bring up the advertising that Intel did on the site, and neglect to mention that he had AMD advertisements on there at one point in the past. People complain about the fact they talked to Siemens when the CPU failed instead of AMD. No one ever bothers to think that Siemens might actually have a support department that would be much easier to contact than AMD.
Please, take things with a grain of salt. They're trying to report the news, and you'll have to excuse them if they're not omnipotent, and able to see all possible omissions. When this new video came out, THG got turned into the devil spreading FUD for Intel, and no one ever really bothered to find out if they were using different boards (for the Athlons, they definately were), or the differences in testing. This much, however, is certain. The Intel chips have hard-wired thermal protection in them -- if they get too hot, they turn themselves off, whereas AMD does not, and has to rely on external hardware devices to do the job.
Re:It's actually WORSE if the fan fails.
on
AMD And THG update
·
· Score: 1
A dead fan will trap heat in the space between heatsink fans and effectively turn the heatsink into a heater. It will heat up your chip.
Not exactly. There should still be a power supply fan, and some air circulation in the case. The CPU fan dying can cause many nasty problems (not the least of which is a seized fan generates heat), but it's still not worse than the heatsink/fan being removed.
certs can help with #2. The other steps are left as an exercise for the reader.
Not exactly. The cert will only verify the binary has not changed since being signed. It doesn't verify that the code was writting by the company or individual, it doesn't verify that a virus isn't hitching a ride along with the code, that someone didn't break into the company network and implant a backdoor into the code, or anything else. It only verifies that there has been no corruption of the data since the initial signing of the binary.
Slashdot Mirror
Yes, Windows 2000 is based on New Technology Technology.
That's for Windows CE, not Windows XP Embedded. They are different products. CE is much smaller than XP, and lower featured.
The custom graphics chip and the sony designed sound chip were the hallmarks of the SNES, however, it's big selling feature wasn't the number of sprites on screen, or the number of colours it could display The most powerful feature of the SNES is one that's difficult to explain -- Horizontal DMA (H-DMA). What it was, is the ability to schedule DMA transfers on the GPU unit to occur at specific horizontal and vertical scanlines. The specs for the SNES said it could display 256 colours of a palette of 16768 colours, except when you bring in H-DMA, you could reprogram the palette during the H-blank and display many more than that. Or the sprite limitation could likewise be worked around by using this H-DMA. The SNES's sound CPU was likewise rather incredible. It was an entirely self-contained CPU, RAM and DAC system, and you fed your program into it via a set of I/O registers, and it ran parallel to the main CPU. Inside the chip itself was a system that could best be described as a compressed wavetable synth engine (compared to the Genesis's FM synth)
Well, to connect to the Exchange server, it's just a matter of reverse engineering two DCE RPC services -- the name service provider interface (nspi), and the exchange message store message database (emsmdb). The name service provider seems to be the easier of the two, since it's only for the address boook, and resolving names, and uses a fairly straightforward interface, the emsmdb on the other hand consists mainly of four RPC calls, one is for connecting, one for disconnecting, and two that are simply named DoRpc and GetMoreRpc. For those of you that don't understand, that means they used RPC to encapsulate RPC (probably MAPI). Simply trying to expose CDO would pose other problems, including security and an additional single point of failure. If simply contacting the Exchange server was the goal, it might be just as easy to extend Wine to be able to make RPC calls, and then use a lightweight version of wine with the native Outlook Exchange DLLs to call the server.
Then you're classified as an evil hacker and must be destroyed at any and all costs. Unless you're a licenced academic in good standing at a recognised educational institution, researching other people's encryption is fast becoming a dangerous hobby. It's been said that having a false sense of security is more dangerous than insecurity itself, but that's what these kinds of laws and lawsuits are effectively doing.
The RedHat suggestion was just a transparent PR ploy to try and cash in on the problems in the proposed settlement. IMHO, Microsoft should just be donating that money with no strings attached for the schools to do whatever they want with. I imagine the poorest schools probably need building maintainance and supplies rather than computers. Giving computers to poor schools where poor students attend isn't going to cure the so-called "electronic-divide", since that's far closer to the fact they're economically divided than anything else.
Windows is Microsoft's "Cash Cow", so to speak. It's where they get all the money they devote to research and development, and how they subsidise new projects until their competitors are out of business. However, as I understand it, the $10 off isn't against the price of a WinXP upgrade or OEM version(about $100), but instead against the full WinXP retail cost (roughly $200). Quite frankly, it's silly since most people could easily go to any local retailer, buy a $10 mouse and an OEM copy of Windows for less cost.
Lets be honest. This entire settlement has very little to do with Linux. In fact, I'd wager that this settlement has more to do with the fact that Microsoft would like to break into Apple's other big niche -- schools. Apple has a huge installed base in the school system, and those computers that are there aren't moving out anytime soon. Microsoft says they're willing to supply Macintosh software, but what do you want to bet the PC alternative (shipped with Windows, of course) will be cheaper or more systems for fewer "dollars". If an educator had to choose between 7 macintoshes or 10 PCs with Windows, what do you think they'd choose? Apple should be the one stinging from this more than anyone else -- not only has Microsoft potentially reduced their sales for the next years, they've also increased their installed base, and come out of the entire anti-trust lawsuit smelling like a rose.
What they have is a trademark on the compact disc logos, and license it out to companies that make CDs and hardware to read those CDs. If you made a wind chime and put the CD or CDDA logo on it, you'd be in trouble.
I often have to wonder if either ICANN or WIPO have heard of the term "integrity". Can anyone remember when ICANN's mandate was to be democratically elected? Seems like such a long time ago.
My observation of the Liberals in Canada is they tend to take up any position on the political scale that isn't dominated by anyone else. In places where the decidedly left-wing NDP is dominant, they adopt the right-wing. In places where the right-wing conservatives are dominant, they take up the left-wing. In the national arena, they seem to be mostly left, favouring social programs and spending. They seem to remind me of a chameleon for some strange reason.
I think the problem is that most parents choose to ignore the ratings that already exist, and I doubt that the situation would be improved by government mandated ratings.
Microsoft insisted that the NT 3.51 license didn't allow use of a webserver, but it was vague enough that it probably couldn't have been enforced, but they made certain that it could be in NT 4.0. Netscape wasn't the only one that was bitten by this license change -- O'Reilly's Website server also fell victim to it. Today, the only places Netscape/iPlanet's webservers still exist are on Novell (it's bundled with certain versions of Netware) and UNIX servers.
I see two options for this -- either they want this control so they can downgrade the severity of discovered holes so that groups like Gartner stop telling people to dump IIS, or they misunderstand the underground cracker community and it's ability to spread information about exploits and to discover holes on their own. I believe that by having nothing but filtered reports that are approved by Microsoft, we'll end up in much greater danger. First, we'll be in the position to be unable to verify Microsoft patches actually fix the problems they are released to fix. And second, legitimate administrators will be at a far worse disadvantage by it than malicious users will (collecting exploit information from less mainstream souces is not an activity you want to perform at work -- most are littered with porn popups and banners).
I can hear the retort, that most Administrators never use the sample exploits and vulnerability tests that are attached to advisories, and that's true. However, if you've been cracked into for the third time this week, and you've applied all the security patches and it's still happening, those sample exploits and vuln tests might suddenly come in useful. Especially on the Microsoft platform, several patches replace previous ones, and if you accidently apply a previous patch over a newer one, you can end up looking like you have the fixed version installed, but still have the vulnerable version installed.
Then there was the Stacker incident with MS-DOS 6, where Microsoft was found to be guily of patent infringement for the disk compression technology in MS-DOS 6.2 (which is why 6.21 contained no compression, and was reintroduced in 6.22).
Then there's the active wars. Microsoft continues to try to leverage Microsoft clients to encourage people to drop their non-MS servers (such as Novell Netware servers). They've also introduced Windows Media Player, and are agressively pushing it on portable player manufacturers and including it in Windows XP to try and flush out Real. Then there's AOL and Microsoft trying to push people to MSN instead, again by using their OS to try and make it more difficult to use AOL than it would be to just break down and use MSN.
Most people who buy pre-installed Windows computers barely know how to operate them, let alone install an OS on them. People mess up where to plug in the phone cable and speakers all the time, and lets not even get into the idiot who managed to plug a USB device into the serial port.
Not exactly. While they are releasing their CLX libraries under GPL, you still need the proprietary Kylix compiler to compile it all up.
The copy protection in the Lucasarts Monkey Island games consisted of code wheels and things like that. It's unlikely the scummvm interpreter would block them (in fact, I checked, and it doesn't).
Please, take things with a grain of salt. They're trying to report the news, and you'll have to excuse them if they're not omnipotent, and able to see all possible omissions. When this new video came out, THG got turned into the devil spreading FUD for Intel, and no one ever really bothered to find out if they were using different boards (for the Athlons, they definately were), or the differences in testing. This much, however, is certain. The Intel chips have hard-wired thermal protection in them -- if they get too hot, they turn themselves off, whereas AMD does not, and has to rely on external hardware devices to do the job.
Not exactly. There should still be a power supply fan, and some air circulation in the case. The CPU fan dying can cause many nasty problems (not the least of which is a seized fan generates heat), but it's still not worse than the heatsink/fan being removed.