You're confusing total system memory with total process memory. 32 bit CPU's have a 32 bit instruction pointer, which means 4GB is the largest address space a single process can address (whithout PAE tricks).
It's quite possible to have much more than 4GB in a 32 bit architecture, it just doesn't allow the programs to address more than 4GB, which 99.9% of the time is just fine. For example, the Xeon (and possibly the P4) can utilize 64GB of memory.
Cores don't come cheap. Think about the power requirements for 8 cores, then think about return you're getting in terms of actual utilization of those cores. Remember, we're talking desktop, not server. Sure, there will be people that want the bragging rights, but that's about it.
I suppose you could transcode a DVD in 5 minutes with that many cores... or a Blu-ray disk in an hour;)
I think he was talking about the foreseeable future.
1 core is really enough for most users. 2 cores is enough for most power users. 4 cores will be enough for all but the most demanding jobs. Workstations are different, however and are not usually considered part of the "desktop". For example, I could see 3D artists using 4 or 8 cores easily. In fact, there's simply no such thing as a computer that's "too fast" for certain purposes.
The issue, though, is one of moderation. Why would a desktop user want 8 cores, which are drawing insane amounts of power, when they're not even utilizing 4 to full advantage? Word processing, accounting, and surfing the web don't need any of this. Games? I can imagine in 10+ years we'll have some photo-realistic 3D games that run in real-time, but the vast majority of the work will likely be handled by GPU's and won't need 8 cores to deal with it.
I simply cannot fathom a purpose for 8 cores for any "desktop" application that isn't in the "workstation" class.
No, the Groklaw analysis is not spot-on. It seems to deliberately confuse the reading of the language. It talks about "conformant implementations", but that language does not exist in the covenant. It simply says, Microsoft cannot sue you for any intellectual property related to conforming to the specification.
For starters, that analysis is severely flawed for a number of reasons. The first, is that it applies only to the Office 2003 XML schemas. The second is that it is based on an invalid reading of the language. Here is the language:
Microsoft irrevocably covenants that it will not seek to enforce any of its patent claims necessary to conform to the technical specifications for the Microsoft Office 2003 XML Reference Schemas
The Groklaw analysis reads this as Microsoft will not sue you if it's a conforming implementation, but what this really says is that if, in the process of conforming to the specification (not a conforming implementation, as the groklaw article suggests) you are required to implement something that Microsoft owns IP on, then they won't sue you. That's a VERY different thing that what the Groklaw analysis says, and is in fact even farther reaching in it's indemnity than the Sun covenant, since sun only says they won't sue you for implementing any version they participate in, meaning that if Sun backs out of OASIS, then they can sue anyone for any version that comes after it.
We will have to see what the final covenant is when the OpenXML specification is published.
Caller ID wasn't a standard, and was rightfully rejected because of the patent issues. That's the standards process working as it's supposed to. I still stand by the point, if it's a standard, it can't be an "extension" by definition, and by definition, it can't be proprietary.
Nope.. I just can't seem to put the pieces together here. If Microsoft writes things into the standard, how could they be extensions? How could they be proprietary?
The terms Alpha, Beta, and Gamma used to mean this:
Alpha: Feature incomplete code Beta: Feature complete, but still buggy Gamma: We think we have all the bugs, but let's do some wider testing anyways to be sure
Gamma's most closely map to Release Candidates, which Beta 2 of Vista is not.
Beta's have gone through a lot of change since the invention of iterative development models. Now you can have parts of the code "feature complete" before other parts are. Betas are now issued during various feature complete milestones rather than when the entire product is feature complete.
Unless you, oh, write XHTML with a correct xml header
This was, in fact, deliberate. I can't seem to locate the reference right now, but they choose to have the XML prolog force IE into quirks mode as an easy way do this, while maintaining well formed code. I think this was because the W3C even recommends not including the XML doctype if it's not going to be served as application/xml+xhtml.
But that's really beside the point. A web developer should know what causes the browser to run in quirks mode or not. That's part of their job.
Your correct that there are bugs in IE6 still in certain cases that can affect the box model, though.
Look, the fact is, a ton of bugs have been fixed. Everything at positioniseverything.com, for instance.
I didn't ask you want Microsoft knows, I asked you what YOU know. FYI, the box model has been fixed since IE5.5. Most people don't know this, though, because they don't have any clue what strict mode is, and how the box model is rendered differently.
Yeah, they fixed like five CSS parsing bugs
Yeah, like the Peekaboo bug, Guillotine bug, Duplicate Character bug, Border Chaos, No Scroll bug, 3 Pixel Text Jog, Magic Creeping Text bug, Bottom Margin bug on Hover, Losing the ability to highlight text under the top border, IE/Win Line-height bug, Double Float Margin Bug, Quirky Percentages in IE, Duplicate indent, Moving viewport scrollbar outside HTML borders, 1 px border style, Disappearing List-background, Fix width:auto, min/max height, etc.. etc..
Yeah, that looks like just 5 parsing bugs.
The actual bugs are things like positioning and sizing boxes and cutting off content inside divs and all sorts of problems. Microsoft hasn't fixed any of these.
Bull, all of the above, plus more have been fixed in IE7. So again, what specific flaws still remain? You clearly don't know because you think things that were fixed as far back as beta 1 still aren't fixed. You haven't even LOOKED at it, have you?
Further, you know so little about it that you think conditional comments are in CSS. They're not. They're in the HTML, which you can use to include a specific style sheet for IE, or specific versions of IE if you like. Further, this isn't new, it's been in IE since 5.0. Again, you haven't even *LOOKED* at it, have you?
Why don't you stop saying they haven't fixed any bugs, when you really don't know what you're talking about?
Actually, yes, you can run two versions on the same machine via several mechanisms. There's the unsupported hack method detailed here http://www.quirksmode.org/browsers/multipleie.html, and you can also download a free VM from VMWare or Microsoft and run a development version of your OS with the other version.
ACID2 does not test consistnecy across browsers. It's merely a snaphot of one set of features (and some invalid ones). Passing Acid2 doesn't mean you're fully standards compliant. It just means you pass those specific tests. It would be like giving someone a dozen questions from the SAT, and if they pass them claiming they got a perfect score.
Now, if the Web standards project wanted to come up with a *COMPLETE* compliance test, that would be a different story. In fact, one could argue that the lack of a comprehensive validation suite is the real problem with the web, and that faul should fall squarely on the shoulders of the W3C.
If there were a comprehensive validation suite, there would be a hell of a lot less "interpreting" of the standard, and developers would be able to have something to shoot for.
Which reasons might those be? Do you have specific cases where there are still CSS bugs? That doesn't include unimplemented features, since those will be ignored anyways. If you do, you should report them. Also, if you want to target stuff just at IE, rather than use the star hack, why not just use conditional comments? It's supported.
Yeah, yeah, they've had umpteen years to work on this, but they didn't. They've only been working on this for less than 2. IE7 is a stopgap release, focusing on security and fixing bugs in the rendering with a few new UI enhancements that people have been screaming for. A new parser would have made it impossible to get out the door in a reasonable time frame.
The development leads have indicated that they are going to be doing more agressive updates in the future, and i wouldn't be surprised to see an xhtml+xml support in the next version. It's just beyond the scope of this one. Live with it, and don't be a petulant child because you can't have everything instantly. It's in the works.
Apart from the fact that you're grossly wrong about fixing bugs (They've fixed a ton of them, including the Guilotene bug which you claim isn't fixed, which was fixed back in beta 1 http://www.mezzoblue.com/archives/2005/07/28/ie7_c ss_upda/ ), the issue with fixing the css parsing bugs is that yes, they are bugs and should be fixed. Need to target IE specifically? Use conditional comments.
This is really only going to be a problem with people that were using strict doctypes and hacks and may have one of the few remaining issues bug issues, which aren't very many. The fact that IE7 has worked flawlessly on every CSS site with hacks i've used it on says a lot.
Uhh.. dude. The box model has been fixed for 5 years. IE6's box model, running in strict mode, is standard compliant.
The quicks mode box model will, by definition, always be non-standard, because that's there for compatibility. Want a compliant box model, make sure the browser is running in strict mode.
Actually, no. A lot of people don't really understand how the political system works. We elect people to represent us for office, but we do so by voting for the people we agree with in the hopes that they will make decisions similar to what we ourselves would make. This doesn't, by itself, guarantee that an elected official will vote as his constituents would. He is under no obligation to do so. If he diverges far enough from what his constituents want, he will likely not get re-elected, but that's an entirely different matter.
In other words, we try to elect people that think like we do, not elect people that will be mindless slaves voting only for what they think people in their districts want. As such, donating to a candidate that thinks the way you do is the best way to show support for them, not buying their vote on key issues.
Now of course it doesn't always work that way, but my point is that donating money doesn't automatically mean you're buying a vote, it could simply mean "I like how you think and I want to help you get (re)elected so you will continue to support these interests which we both agree on".
Slashdot Mirror
You're confusing total system memory with total process memory. 32 bit CPU's have a 32 bit instruction pointer, which means 4GB is the largest address space a single process can address (whithout PAE tricks).
It's quite possible to have much more than 4GB in a 32 bit architecture, it just doesn't allow the programs to address more than 4GB, which 99.9% of the time is just fine. For example, the Xeon (and possibly the P4) can utilize 64GB of memory.
I'm not sure where you get that idea, but this chart says the x2 4000 uses between 120 and 150 watts
i ndex.x?pg=15
http://techreport.com/reviews/2005q2/athlon64-x2/
Cores don't come cheap. Think about the power requirements for 8 cores, then think about return you're getting in terms of actual utilization of those cores. Remember, we're talking desktop, not server. Sure, there will be people that want the bragging rights, but that's about it.
;)
I suppose you could transcode a DVD in 5 minutes with that many cores... or a Blu-ray disk in an hour
I think he was talking about the foreseeable future.
1 core is really enough for most users. 2 cores is enough for most power users. 4 cores will be enough for all but the most demanding jobs. Workstations are different, however and are not usually considered part of the "desktop". For example, I could see 3D artists using 4 or 8 cores easily. In fact, there's simply no such thing as a computer that's "too fast" for certain purposes.
The issue, though, is one of moderation. Why would a desktop user want 8 cores, which are drawing insane amounts of power, when they're not even utilizing 4 to full advantage? Word processing, accounting, and surfing the web don't need any of this. Games? I can imagine in 10+ years we'll have some photo-realistic 3D games that run in real-time, but the vast majority of the work will likely be handled by GPU's and won't need 8 cores to deal with it.
I simply cannot fathom a purpose for 8 cores for any "desktop" application that isn't in the "workstation" class.
No, the Groklaw analysis is not spot-on. It seems to deliberately confuse the reading of the language. It talks about "conformant implementations", but that language does not exist in the covenant. It simply says, Microsoft cannot sue you for any intellectual property related to conforming to the specification.
The Groklaw analysis reads this as Microsoft will not sue you if it's a conforming implementation, but what this really says is that if, in the process of conforming to the specification (not a conforming implementation, as the groklaw article suggests) you are required to implement something that Microsoft owns IP on, then they won't sue you. That's a VERY different thing that what the Groklaw analysis says, and is in fact even farther reaching in it's indemnity than the Sun covenant, since sun only says they won't sue you for implementing any version they participate in, meaning that if Sun backs out of OASIS, then they can sue anyone for any version that comes after it.
We will have to see what the final covenant is when the OpenXML specification is published.
You're missing the point. If it's in the standard, it can't be an extension, now can it?
Caller ID wasn't a standard, and was rightfully rejected because of the patent issues. That's the standards process working as it's supposed to. I still stand by the point, if it's a standard, it can't be an "extension" by definition, and by definition, it can't be proprietary.
Standards organizations usually require RAND policies on any patents.
How is that, exactly? They've published a covenant not to sue, exactly like Sun did with ODF. You don't need to get a license to implement OpenXML.
Uhhh... think... hmmm.. thing some more...
Nope.. I just can't seem to put the pieces together here. If Microsoft writes things into the standard, how could they be extensions? How could they be proprietary?
What were you trying to say?
Ummm.. no. This is a beta.
The terms Alpha, Beta, and Gamma used to mean this:
Alpha: Feature incomplete code
Beta: Feature complete, but still buggy
Gamma: We think we have all the bugs, but let's do some wider testing anyways to be sure
Gamma's most closely map to Release Candidates, which Beta 2 of Vista is not.
Beta's have gone through a lot of change since the invention of iterative development models. Now you can have parts of the code "feature complete" before other parts are. Betas are now issued during various feature complete milestones rather than when the entire product is feature complete.
No. UIPI fixes this supposedly "unfixable" bug.
according to this http://blogs.zdnet.com/Ou/?p=196 Sun themselves say that ODF is significantly slower than Microsoft's XML format.
I would hazard a bet that there are FAR more applications that read and write Office formats than there are that read ODF.
Unless you, oh, write XHTML with a correct xml header
This was, in fact, deliberate. I can't seem to locate the reference right now, but they choose to have the XML prolog force IE into quirks mode as an easy way do this, while maintaining well formed code. I think this was because the W3C even recommends not including the XML doctype if it's not going to be served as application/xml+xhtml.
But that's really beside the point. A web developer should know what causes the browser to run in quirks mode or not. That's part of their job.
Your correct that there are bugs in IE6 still in certain cases that can affect the box model, though.
Look, the fact is, a ton of bugs have been fixed. Everything at positioniseverything.com, for instance.
Tell you what, even if that were true, it wouldn't matter. Filters are not standards compliant.
So, you're complaining that when doing non-standards compliant tricks, the result is non-standard. How imaginitive of you.
I didn't ask you want Microsoft knows, I asked you what YOU know. FYI, the box model has been fixed since IE5.5. Most people don't know this, though, because they don't have any clue what strict mode is, and how the box model is rendered differently.
Yeah, they fixed like five CSS parsing bugs
Yeah, like the Peekaboo bug, Guillotine bug, Duplicate Character bug, Border Chaos, No Scroll bug, 3 Pixel Text Jog, Magic Creeping Text bug, Bottom Margin bug on Hover, Losing the ability to highlight text under the top border, IE/Win Line-height bug, Double Float Margin Bug, Quirky Percentages in IE, Duplicate indent, Moving viewport scrollbar outside HTML borders,
1 px border style, Disappearing List-background, Fix width:auto, min/max height, etc.. etc..
Yeah, that looks like just 5 parsing bugs.
The actual bugs are things like positioning and sizing boxes and cutting off content inside divs and all sorts of problems. Microsoft hasn't fixed any of these.
Bull, all of the above, plus more have been fixed in IE7. So again, what specific flaws still remain? You clearly don't know because you think things that were fixed as far back as beta 1 still aren't fixed. You haven't even LOOKED at it, have you?
Further, you know so little about it that you think conditional comments are in CSS. They're not. They're in the HTML, which you can use to include a specific style sheet for IE, or specific versions of IE if you like. Further, this isn't new, it's been in IE since 5.0. Again, you haven't even *LOOKED* at it, have you?
Why don't you stop saying they haven't fixed any bugs, when you really don't know what you're talking about?
Actually, yes, you can run two versions on the same machine via several mechanisms. There's the unsupported hack method detailed here http://www.quirksmode.org/browsers/multipleie.html , and you can also download a free VM from VMWare or Microsoft and run a development version of your OS with the other version.
ACID2 does not test consistnecy across browsers. It's merely a snaphot of one set of features (and some invalid ones). Passing Acid2 doesn't mean you're fully standards compliant. It just means you pass those specific tests. It would be like giving someone a dozen questions from the SAT, and if they pass them claiming they got a perfect score.
Now, if the Web standards project wanted to come up with a *COMPLETE* compliance test, that would be a different story. In fact, one could argue that the lack of a comprehensive validation suite is the real problem with the web, and that faul should fall squarely on the shoulders of the W3C.
If there were a comprehensive validation suite, there would be a hell of a lot less "interpreting" of the standard, and developers would be able to have something to shoot for.
Which reasons might those be? Do you have specific cases where there are still CSS bugs? That doesn't include unimplemented features, since those will be ignored anyways. If you do, you should report them. Also, if you want to target stuff just at IE, rather than use the star hack, why not just use conditional comments? It's supported.
Yeah, yeah, they've had umpteen years to work on this, but they didn't. They've only been working on this for less than 2. IE7 is a stopgap release, focusing on security and fixing bugs in the rendering with a few new UI enhancements that people have been screaming for. A new parser would have made it impossible to get out the door in a reasonable time frame.
The development leads have indicated that they are going to be doing more agressive updates in the future, and i wouldn't be surprised to see an xhtml+xml support in the next version. It's just beyond the scope of this one. Live with it, and don't be a petulant child because you can't have everything instantly. It's in the works.
Apart from the fact that you're grossly wrong about fixing bugs (They've fixed a ton of them, including the Guilotene bug which you claim isn't fixed, which was fixed back in beta 1 http://www.mezzoblue.com/archives/2005/07/28/ie7_c ss_upda/ ), the issue with fixing the css parsing bugs is that yes, they are bugs and should be fixed. Need to target IE specifically? Use conditional comments.
This is really only going to be a problem with people that were using strict doctypes and hacks and may have one of the few remaining issues bug issues, which aren't very many. The fact that IE7 has worked flawlessly on every CSS site with hacks i've used it on says a lot.
Uhh.. dude. The box model has been fixed for 5 years. IE6's box model, running in strict mode, is standard compliant.
The quicks mode box model will, by definition, always be non-standard, because that's there for compatibility. Want a compliant box model, make sure the browser is running in strict mode.
Actually, no. A lot of people don't really understand how the political system works. We elect people to represent us for office, but we do so by voting for the people we agree with in the hopes that they will make decisions similar to what we ourselves would make. This doesn't, by itself, guarantee that an elected official will vote as his constituents would. He is under no obligation to do so. If he diverges far enough from what his constituents want, he will likely not get re-elected, but that's an entirely different matter.
In other words, we try to elect people that think like we do, not elect people that will be mindless slaves voting only for what they think people in their districts want. As such, donating to a candidate that thinks the way you do is the best way to show support for them, not buying their vote on key issues.
Now of course it doesn't always work that way, but my point is that donating money doesn't automatically mean you're buying a vote, it could simply mean "I like how you think and I want to help you get (re)elected so you will continue to support these interests which we both agree on".