“It seems unfortunate if we do this work and get our partners to do the work and the result is that Linux works great without having to do the work” said Bill Gates in 1999 (pdf). While we don’t know if he actually managed to do just that (creating problems to other OSes to work well with ACPI), but if he did, it is a good explanation why ACPI has been flaky on the majority of x86 computers with anything else other than Windows (the older, APM standard, seemed more compatible with alternative OSes).
Bill Gates on Making ACPI Not Work with Linux
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSAlert.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
hey, that is probably why my friend’s computer can not boot linux, but it sure runs windows xp great.
Could be, but usually -noacpi fixes acpi boot problems (it also disables all acpi features).
…of ACPI and Linux, that many ACPI bioses don’t adhere to the standard.
For example, the standard allows the Operating System to identify itself. There are many BIOSes out there, who just don’t talk to any other OS than “Windows”.
So AFAIK, Linux claims to be Windows, too. Much alike old Opera claiming to be the – much inferior – Internet Explorer.
While the last one was due to the stupidity of Web developers, I don’t know about the first one. Up till know, I assumed it was stupidity of the BIOS vendors.
Calling IE inferior is a pretty bold statement. If it’s your personal opinion, that’s fine but otherwise, then it really doesn’t hold any weight.
Have a look at the statistics and you’ll see that Opera goes as low as 0.2% usage. Month in, month out, Opera isn’t relevant.
http://www.w3schools.com/browsers/browsers_stats.asp
Firefox on the other hand is steadily growing (and deservedly so). Opera remains around 1.5% for at least the last year. Less even further back.
This site shows a similar snapshot of Opera popularity:
http://www.upsdell.com/BrowserNews/stat.htm
If it were superior, people would use it over IE – especially given the obvious security problems alone.
Edited 2007-04-13 11:48
If it were superior, people would use it over IE – especially given the obvious security problems alone.
Except for that little thing called lockin. Surely you’ve run across those sites that only work with IE? It’s not because IE is superior, it’s because MS created a non standards-compliant browser and shipped it with the OS most people use. That allowed web developers to be lazy, only certifying their site to work on IE, and in many cases using the IE only extensions. Lockin is what let the IE team cruise (to the point of not existing for a while), not any superiority of the browser.
Attempting to conflate popularity with quality is something that won’t get you anywhere. Opera is an amazing browser, and does what it can against a product that comes with folks’ OS and which has become (esp with 7) good enough.
“Surely you’ve run across those sites that only work with IE?”
You know, I haven’t actually ran across such a site in a *long* time. I used a while back but not in recent times.
Are web designers getting their shit together or is it Firefox/Konqueror that works around these IE quirks? I dunno, but I sure don’t run in to them as much as I used to, if at all.
Edited 2007-04-13 12:20
Both. There are more developers now building sites that are standards-driven and accessible than ever before.
Unfortunately, there are far more people creating web-sites than ever before, and the vast majority of new starts are still following circa-1996 tutorials or copying and pasting from the godawful code behind many major sites.
Firefox, et al, have had to put shims in place for compatibility, because not doing so would be a major barrier to adoption.
I don’t run into these sites, normally. But recently, I tried to find and book a holiday online. Well, those sites suck, at least the dutch ones. 50% of em doesn’t work at all in konqueror, and firefox doesn’t do much better. Dunno if they even work in IE but I bet they do…
Except for that little thing called lockin. It’s not because IE is superior, it’s because MS created a non standards-compliant browser and shipped it with the OS most people use.
Erm… those are strawman arguments. Lockin doesn’t exist when you can freely use other browsers (if you haven’t noticed, you are even asked if you want to install some of these other browsers during a normal installation of IE7), and many of the ‘standards’ that are being created today are anything but standard.
Your getting modded down pretty fast for this comment, but I want to address it anyway.
First let me say that I haven’t noticed the IE7 install asking if I want to install other browsers. Maybe it does, but I haven’t seen it.
As for standards. Here’s how I view what is standard and what is not. First of all the W3C and ECMA don’t actually create standards. They create recommendations for a standard way of doing things. Not all recommendations are followed by those who choose to use them. I don’t know of any browser who fully supports every W3C recommendation.
To me a standard is when more than one company (in the case of browsers) chooses to follow a given recommendation. It’s not based on market share. However, I will concede that IE, in it’s own way, has a coding standard. It’s just a very poor one. IE7 is attempting to fix that by more closely following the W3C recommendations.
It’s my opinion that browsers should not fix poorly coded web pages. We might have better web pages if browser developers elected to stop making browsers fix bad coding. However, that tends to acceptable behavior as not fixing bad code only makes the browser look inferior.
Almafeta, can you give me a screenshot of the setup program for Internet Explorer asking you if you wouldn’t rather be installing Firefox or Opera? I’d really love to see that because it seems like you’re confusing the selection of your search engines during the first run of IE with the idea of Microsoft actually letting people know that there are other browsers out there that have been doing what it thinks are innovative and revolutionary features of recent IE revisions for years.
Lockin with IE exists because IE is a flat-out requirement for many places, not the least of which are governmental and financial institutions. At the end of your comment you underline the main issue that’s been around since the Netscape/IE war began, ‘standards’ being created by a company and supported only by their browser. Not to mention standards simply not being supported at all. It’s obvious you’re not a web developer and haven’t had to deal with the lameness that is IE when so many other browsers just do things right.
Except for that little thing called lockin. Surely you’ve run across those sites that only work with IE? It’s not because IE is superior
But at the time it took over the world, IE was the superior browser.
But at the time it took over the world, IE was the superior browser.
You’re right. At the time IE took over it was the superior browser. Of course that hasn’t been the case for years now yet they still own the browser “market”. In fact even IE7 is inferior to most other browsers out there including Firefox, Konq, Opera, and Safari.
I take your points but how on earth my point of view fits into any of the -1 categories is beyond me. It just goes to show how stupid the comment rating feature of this site is.
“Saying that Opera is inferior to IE because more people use Windows is like saying that all restaurants are inferior to McDonalds.”
This is the best analogy I’ve seen in a long time. I’m gonna remember this.
So IE6’s complete ignorance to the published agreed standards didn’t make it inferior? The fact that IE7 still isn’t on the ball doesn’t make it inferior?
Inferior and popular are two very different things.
As a professional web developer, I’m long since past being surprised at people not knowing about the existence of anything besides IE—hell, most of them don’t even know that it’s IE, it’s just ‘the Internet’. On that basis, it should be no surprise that IE’s market share is so high: the only, quite frankly stunning, feat is that IE’s market share is as low as it is when compared to Windows’.
Some people prefer IE over other browsers. I can tell you I’ve had a few tech savvy people who just didn’t like Firefox or Opera.
I didn’t use to like Opera and chose Firefox instead, but as of a few days ago I now using Opera instead of Firefox. I still keep Firefox around though, as some things don’t display correctly in Opera.
As for IE7. My feelings are that Microsoft will make IE7+ as standard compliant as Firefox or other browsers. The reason is that many more sites are following the W3C’s recommendation and those sites aren’t working in IE. At some point even the dumbest user isn’t going to keep blaming the websites especially when their smarter friends show them how well the site works in other browsers.
If IE becomes as standard compliant someday as the other browsers then I don’t really feel that IE will lose much more ground to Firefox or the other browsers. The simple reason is that the majority of computer users don’t really care what it is they are using as long as it works reasonably well.
The main reasons more people don’t use Opera are they’ve never heard of it, or they have and are happy with their current browser (usually Firefox or Safari, but not so much Internet Explorer–IE users are usually computer illiterate and are the ones that have never heard of Opera). It has nothing to do with superiority.
If it were superior, people would use it over IE – especially given the obvious security problems alone.
Opera is objectively superior to Internet Explorer in nearly every aspect, from rendering engine (Opera 9.x passes ACID2; IE 7 doesn’t even come close) to security (Opera has fewer issues on Securnia than Firefox and IE) to customizability (The toolbars are more customizable, and Opera has skins). The only thing where Internet Explorer takes the cake without a doubt is site compatibility, and that is not Opera’s fault per se; Opera better adheres to web standards, but IE is so ubiquitous that some web designers make sites for IE and its quirks and shortcomings.
“So AFAIK, Linux claims to be Windows, too. Much alike old Opera claiming to be the – much inferior – Internet Explorer.”
flanque, understand that Opera fake the browser identification by ID itself as IE 5. Much like Linux fake ACPI identification by ID itself to the BIOS as Windows…
So it is NOT about IE being inferior. It is about inverior IE version…
Kochise
It’s definitely Microsoft’s fault. There are two compilers available to generate AML (ACPI Machine Language) — Intel’s reference implementation, and Microsoft’s. Guess what most manufacturers use to provide ACPI support in their BIOS?
Naturally, Microsoft-generated AML works fine with Microsoft’s own ACPI driver.
This is a very severe problem, because on modern hardware ACPI is in charge of things like fan control and CPU throttling, which if not handled properly, could have disastrous consequences for your laptop.. as happened to mine:
http://hircus.wordpress.com/2007/04/14/hp-hp-lama-sabachthani/
One begins to see why HP has been known to refuse warranty claims if the customer admits to running Linux. Though it’s entirely the fault of Microsoft and the hardware manufacturers. Unless the ACPI interpreter in Windows does not handle Intel-compiled AML properly, leaving manufacturers with no choice?
I’m never buying a computer with Windows pre-installed again, if only because of this. Give me bare-bones, Linux preinstalled, or a Mac. MS is not getting a single more dollar, even if I have to pay a bit more for that privilege (for a Dell, it’s a reasonable $19)
ACPI support has been pretty damn flaky on Windows, never mind other operating systems, without some form of third party drivers – which don’t work well either. Ask any OEM about hibernation or standby problems and you’ll be told they don’t support it. Now that Windows Vista basically relies on having a working ACPI implementation for the things it does, the situation doesn’t look good.
I’d also love to know what Bill meant when he said “if we do this work”, and I’d be interested to know what that work actually is.
He also goes on to write:
“Or maybe we could patent something related to this.”
Which is why the Linux and open source worlds need to take a firm stand on it – which they’re not doing. If anyone had bothered to read the whole Cleartype thing and the e-mail by David Turner:
http://permalink.gmane.org/gmane.comp.fonts.freetype.user/1912
“Of course, all of this is my personal opinion, and I would *love* to be proven wrong !!”
So, the whole issue of Microsoft’s patent coverage on this was based on his own personal opinion – which Novell and others just believe! In other words, he swallowed all the FUD whole.
>I’d be interested to know what that work actually is
“[product] isn’t done until [competitor] won’t run”
I’d also love to know what Bill meant when he said “if we do this work”, and I’d be interested to know what that work actually is.
Me too.
ACPI establishes industry-standard interfaces enabling OS-directed configuration, power management, and thermal management of mobile, desktop, and server platforms. (emphasis added, from http://www.acpi.info )
OS Directed. How could Linux possibly benefit from something like that “without having to do the work”? Linux devs have to do the same work MS did: create a way for the OS to direct power, thermal, etc management, using a standard interface. It’s hard to see that Gates has a point here, other than to throw up roadblocks for Linux. Unfortunately in the process he seems to have thrown up some roadblocks for Windows too. Where’s the sense in that?
Maybe thinking and writing all the ACPI specifications ? It was co-developed by Microsoft.
So Linux can use the specification, but no Linux developers have participated to the writing of the specs.
Perhaps I am wrong, but it is how the statement can be read.
And? Public specifications are public. AFAIK, Microsoft didn’t participated in the basic but major TCP/IP specifications, but benefit from it like everyone.
If they didn’t want the ACPI specs to be public, they should have convince firmware and hardware partners to keep it closed. I dunno if they tried, but if they tried, they failed.
Public specifications are public specifications are public specifications. Don’t blame the public for reading them, follow them, using them. Make your specifications private for that.
The guys who code ACPI for Linux are Intel guys (Intel actually has quite a few people working only for Linux-related things, it’s very nice)
And Intel have participated into the ACPI development, probably even more than MS. So it’s not that Linux hasn’t done “anything”.
Bingo!
OS directed means that “Windows” will do this part… but not actually Microsoft code, but drivers written by the hardware vendor and certified by Microsoft. Hence the double wammy. It’s dependent on both Microsoft’s documentation and the hardware vendor’s physical chips + drivers. You can assume the task is long and costly for the VENDOR to do all the work so they will cut corners wherever possible to “just meet” the MS specs..i.e. the features may not exist at all!! but the company’s driver might just lie to the OS to pass the cert and do something else.
Enter Linux people expecting that HARDWARE works as advertised on the spec sheet missing that little “*” that says some features enabled by drivers only.
“I’d also love to know what Bill meant when he said “if we do this work”, and I’d be interested to know what that work actually is.”
Most likely he meant working on defining/creating the standard together with HP, Toshiba, Intel and Phoenix.
It’s nice to be modded down for seriously answering an on-topic question. Hurrah.
Of course APM worked with other OSes just fine… it was putting all the burden on the BIOS. ACPI relies OTH on the OS itself to execute stuff… or not execute it if it’s non compliant to the standard dictated by MS.
… that we at (insert appropriate non-MS company) and our partners have worked on (TCPIP/Ethernet/SMTP/IMAP/Kerberos/other standard) and the result is that Windows works great without having to do the work.
Edited 2007-04-13 11:53
anyone has any experience with linuxbios concerning ACPI?
“It seems unfortunate if we do this work and get our partners to do the work and the result is that Linux works great without having to do the work.”
Funny he would say this given that Windows’ network stack is directly derived from BSD.
Actually, it was rewritten for Vista.
Yup, but still doesn’t change the fact that they used it.
Personally, it’s nice to see how they gave Linux so much ‘love’ even back in ’99
With this kind of love, who needs enemies
))
And more to the point, Windows is implementing all kinds of open standards that MS never did any work on except extending them with their own proprietary shite.
Edited 2007-04-13 12:46
And even then… they break the open specifications making them useless
.
It used to be, but is no longer, they rewrote it awhile ago. some of the utils are still derived from BSD, like FTP
I wonder why, given evidence as this, the world keeps supporting Windows?
Probably the same reasons lousey politicians are constantly rehired.
Or why companies with terrible track records socially/environmentally/quality wise are still succesful.
So many people just don’t care…
Probably the same reasons lousey politicians are constantly rehired.
Or why companies with terrible track records socially/environmentally/quality wise are still succesful.
So many people just don’t care…
That’s it! We have our 2009 Windows brand!
“Windows Apathy – When did we ever care where you want to go today?”
It should have a co-release of Windows Emo – “How could you want to take me there?!”
*LOL*
Aawww… My stomach… damn you and those cramps
Sorry! 8D
ACPI can be really flaky on windows too. And I’m not talking about discount whitebox desktops, I had a lot of problems on my compaq nx7000 (would not come back from hibernation) and I have a lot of issues with standby on my current self-assembled desktop (with lots of expensive hardware). To be earnest, suspend on ubuntu works just fine… so maybe if there’s a conspiracy, it didn’t work out too well after all.
What I don’t get is this rabid anti-Linux attitude from Gates and CO. It’s like they’re reaction out of pure fear or something.
Reminds me of the US’s response to Cuba for the last 40 years, rabid antipathy towards something they really should not be all that bothered about. It’s not like Cuba have any means of attacking the US and it’s not like Linux is going to take over the Desktop anytime soon.
I can understand being weary of a competitor but all this out and out warfare just makes MS look like a scared bully.
Just to clarify something here when it comes to Cuba. If you remember back, it was Cuba that allowed the USSR to put nuclear weapons on the island that is just minutes from our border.
Is it fair to compare the Windows/ACPI/Linux situation to Cuba….I think not.
Dude, read up on your history! Cuba had to make a deal with the USSR because no one else was allowed to buy their main export, sugar. The main reason for this blockade in the first place was the USA. Also, the USSR wanted to offset the huge amount of missiles installed on European soil buy doing what the US had been doing for years, park some fat warheads in your enemy’s backyard.
The USSR agreed to buy Cuban sugar in exchange for the installation of war heads. Because the US had set up a blockade, Cuba was not only forced to concede but was more than happy to. The fact remains that the US was playing the bully long before those missiles where installed but if you want to blame Cuba for what the rest of the world considers rabid antipathy on the part of the US, be my guest. Just don’t be surprised when everybody else has a good laugh at your expense.
Hah, we’ve still got some of the damn things lying around Belgium. They’re still on our soil illegally and our government is *still* denying it after all these years although it’s a public secret they’re there. Each year there’s a demonstration at the army base of people coming to inspect for WMD’s ( http://www.bomspotting.be/ ). They never get in though.
The distance Brussels-Moscow is about the same as Havanna-Washington BTW (around 2000km).
What I don’t get is this rabid anti-Linux attitude from Gates and CO. It’s like they’re reaction out of pure fear or something.
Surely you must? Bill Gates (or whatever person or persons call(s) the shots at Microsoft) can’t stand anyone but him making money. His attitude is he should be the only one to make it, and then he’ll be “kind” enough to give the small change to everyone else.
Reminds me of the US’s response to Cuba for the last 40 years, rabid antipathy towards something they really should not be all that bothered about. It’s not like Cuba have any means of attacking the US and it’s not like Linux is going to take over the Desktop anytime soon.
It’s really not as much about Linux/Cuba as it is about competition/communism. After spending millions trying to bury other computer companies/Communist states in the ground, Microsoft/the US can’t afford to let Linux/Cuba off the hook or people might get antsy.
Notice also that part of the tactic is that as long as you gibber on about intellectual property/capitalize your economy, they really couldn’t care less about people like Novell/China sharing code/torturing its citizens.
I can understand being weary of a competitor but all this out and out warfare just makes MS look like a scared bully.
Look like? That’s exactly what they have been all along.
Very true but I don’t think it’s in their interest to actually shout about it. Everybody knows you can’t compete against a free product but trying to dirty the name of said product, although a time honored business practice, does end up backfiring in almost all the cases I have witnessed.
I’ve always wondered why ACPI doesn’t work reliably on Linux – in my experience and others I know – but it has on Windows for years (with some exceptions of course). I always thought it was because OEM manufacturers install certain custom drivers with their laptop to make sure that ACPI works. But a friend of mine pointed out that you can wipe out the Windows version that came with the laptop and install a generic box version, and ACPI still works – so obviously it’s not the custom drivers that are doing it. Yet it seems to be one area that Linux has such a hard time with. Why is that? (I use Ubuntu myself, and suspend works correctly about half the time on my laptop. I haven’t used Windows in a couple of years, but most of those I work with do.)
Notice that the Linux ACPI implementation is written and maintained by two or three or maybe even more Intel guys.
Intel was who invented ACPI. Still they can’t write ACPI code that works 100% with the hardware made out there. Even Intel itself has released (at least in the past) chipsets etc. that weren’t 100% ACPI compliant.
ACPI is a crappy standard, and vendors only test their hardware against Windows. That makes the Linu implementation harder.
There are two main ACPI “compilers”: Intel’s reference ACPI-CA used by the BSDs and Linux, and Microsoft’s. The Intel compiler is very strict. Microsoft’s is not. Since most motherboard manufacturer’s target Windows, they use the MS ACPI compiler and can be very lax in their compliance. They’ll still have an “ACPI-compliant” board … but it only works with Windows.
The Intel implementation is available for anyone to use, while the MS one obviously isn’t.
The Intel compiler is stricter about enforcing the specifications, but the Linux interpreter (almost entirely written by Intel) will correctly interpret almost all code produced by the Microsoft compiler. Be conservative in what you emit and liberal in what you accept, and so on. The use of the Microsoft compiler really isn’t the problem in most cases.
Or will people still believe EU commission just hates Bill Gates on a personal basis ?
Kochise
No, They want the hundred’s of Millions of Euros in Fines, it’s nto personal, just part extortion, part control freaks
It’s something to consider, Microsoft this something similar back in the days of DOS (DR-DOS and an artificial error to break compatibility between DR-DOS and Win3).
The reason most ACPI-implementations are horrible and don’t work well with non-MS OS’s is that Microsoft has been so friendly to create an easy DSDT compiler (easy because it doesn’t complain about bugs & warnings, perfect for lazy programmers).
The DSDT (Differentiated System Description Table) is like the index of the a BIOS’s ACPI functions. Now it so happens to be that MS’s DSDT compiler generates non-100%-ACPI compliant & bugged code which only Windows can understand and work with.
Intel has a free DSDT compiler that does work 100% compliant, why are the OEM manufacturers so bloody stupid,you get one for free, why buy MS’s?
One can load a custom fixed DSDT table into the kernel at boot time (see http://acpi.sourceforge.net/dsdt/index.php) for more info. They have several fixed DSDT’s available as well and there’s plenty of documentation to learn the AML language for fixing your own table.
I got my laptop with ACPI working that way
“One can load a custom fixed DSDT table into the kernel at boot time […] They have several fixed DSDT’s available as well and there’s plenty of documentation to learn the AML language for fixing your own table.”
FreeBSD allows loading a custom dsdt since… version 5? Simply insert
acpi_dsdt_load=”YES”
acpi_dsdt_type=”acpi_dsdt”
acpi_dsdt_name=”/boot/my_custom_dsdt.aml”
into /boot/loader.conf and test.
“I got my laptop with ACPI working that way
“
I was lucky to disable ACPI loading on my “old” laptop and have APM running instead. Impossible today, because APM does not exist anymore.
I remember that, when I got my Toshiba Centrino laptop in 2003, I had to make a custom DSDT in order to have any sort of ACPI working in Linux.
Later on, I just have to add a workaround in the kernel. Guess what that was?
The trick was to change the Operating System string in the Linux acpi.h to be “Windows XX”. As long as it’s a 10-letter phrase with the starting word “Windows” on it the laptop would happy comply with the Linux ACPI implementation.
Coincidence? I don’t think so.
Conversation between my daughter and me concerning our new computer. I have removed all access to IE and replaced it with Firefox.
Me: Hey honey look at our new computer
My Daughter: Wow, daddy can I try it?
Me: Sure
My Daughter: How do I get to the internet there is no “E”.
Me: Aww honey don’t be like that.
If I were to put IE on my daughter’s (9 years old) computer, her first question would be “Dad, where’s Firefox?”
This document is 8 years old. Why is everyone getting into a frenzy over it? This technology will be outdated soon.
Um, because of blatant anti-competitive behavior by Microsoft?
Because it lifts a tip of the veil showing how MS tries to make it harder for their competition. In ways that are neither legal nor fair…
(This particular thing makes my really pissed off – in case of IE, you could say it’s just crappy software, they didn’t INTEND to make it harder for non-IE users to browse the web. But here, it is pretty clear they INTENDED to make it harder for non-windows users to suspend their laptop… Which is, imho, a pretty mean thing to do, and a reason to dislike them – I have such a laptop…)
This document is 8 years old. Why is everyone getting into a frenzy over it? This technology will be outdated soon.
True, this document provides a Window (sorry, couldn’t resist the pun) into how Microsoft operated 8 years ago.
But do you really think anything about them has changed since then?
And remember, this is for posterity, so be honest.
“outdated”????
Dude, ACPi is here to stay (sadly). There’re new revisions of the ACPI spec that 90% of the mainstream hardware is going to implement, and etc. I’d say that this is quite important.
ACPI is a massive specification, covering a huge range of different pieces of functionality. It’s not just a replacement for APM – it replaces several other existing specifications as well. And, like most large specifications, implementations have bugs. Until relatively recently, the Linux implementation was almost certainly less conforming than the Windows one.
It doesn’t help that the specification itself is unclear in places. Recent HP BIOSes have tended to be implemented in ways that are valid interpretations of the standard, though only likely interpretations of the standard if you’re on drugs. They worked with Windows, but failed in bizarre ways on Linux. After a while people figured out that it was entirely legal, and recent kernels (2.6.18 or so) handle it.
There’s no need to attribute any sort of malice or excessive incompetence to the differences between Windows and Linux in this respect. Software has bugs. Implementations of standards have bugs.
Not that I disagree with you, but what then is your interpretation of Gates’ words? Doesn’t it at least have the appearance of malice to you?
Yes, it has the appearance of malice. It’s also written after Windows 98 was released (the version of Windows with the buggiest ACPI implementation, to the extent that it’ll crash if anything reports a temperature of 15.8 C or lower) and after the Microsoft ACPI compiler had started shipping and causing various chunks of buggy code to end up in ACPI tables (that can basically be worked around in entirely trivial ways, so, well).
Since then, Windows has ended up more spec-compliant and ACPI tables have got more compatible. It’d be nice to blame Linux’s ACPI problems on Microsoft, but the fact remains that most of the problems we’ve hit in power management have been bugs in the Linux kernel – things like not reprogramming hardware correctly when resuming, executing ACPI methods in the wrong order, not executing some ACPI methods at all, hardware vendors not documenting their own deviations from the specification, that sort of thing. Like I said, this stuff is hard, and people have only really started working on it in the last couple of years.
“One thing I find myself wondering about is whether we shouldn’t try and make the “ACPI” extensions somehow Windows specific.
If seems unfortunate if we do this work and get our partners to do the work and the result is that Linux works great without having to do the work.
Maybe there is no way Io avoid this problem but it does bother me.
Maybe we couid define the APIs so that they work well with NT and not the others even if they are open.
Or maybe we could patent something relaled to this.”
Edited 2007-04-13 14:51
That is typical Bill Gates thinking. That guy has a criminal mind.
If you don’t want to share something, then you don’t share it. You copyright it, or patent it. If you own it, that’s your right.
But, making people believe that something is a open protocol, while in fact is not, is a deception. It is a false advertising of some kind, and I think is illeagal in some countries.
“That guy has a criminal mind.”
Being a greedy a-hole isn’t a crime.
Ah, a Lotus fan, I see.
http://lwn.net/2001/0704/a/ac-acpi.php3
Linus Torvalds on ACPI:
Modern PCs are horrible. ACPI is a complete design disaster in every way. But we’re kind of stuck with it. If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce. ( http://www.linuxjournal.com/article/7279 )
The fact that ACPI was designed by a group of monkeys high on LSD, and is
some of the worst designs in the industry obviously makes running it at
_any_ point pretty damn ugly. And the fact that MB vendors don’t test it
with anything else than Windows (and sometimes you wonder whether they do
even that) doesn’t help. So hell yes, it’s ugly, and nasty.
( http://www.ussg.iu.edu/hypermail/linux/kernel/0507.3/2331.html )
And something I myself found on LKML:
EFI is this other Intel brain-damage (the first one being ACPI).
To get the vendors on board, they would have to make it worth their while. What people forget about the fiasco with webstandards is this; Netscape was the *first* vendor to start implementing and extending the webstandards, the problem is that these never really caught on.
Microsoft on the other hand provided extensions to the webstandard by providing developers with features which they wanted/needed to make their development easier, rather than just a ‘screw you’ which was Netscape’s only motivation – which by the way, never stopped Netscape from implementing those features just as Internet Explorer implemented the mirrad of extensions which Netscape made themselves when they were top dog.
How this relates back to Windows, Microsoft and ACPI – Microsoft would have to offer these vendors something which overly helped them but covertly helped Microsoft in return.
The bigger issue isn’t the grand unified conspiracy theory between hardware and Microsoft but the blatantly obvious, there are a lot of crap vendors with crap programmers who couldn’t code their way out of a paper bag if their life depended on it.
With that being said, Fedora Core 6 claimed that my ‘ACPI’ had a bug, and yet, Ubuntu and SLED 10 had no problems with it – so alot of the bugs are also distro specific as well.
To me this is just a symptom of a larger problem…
Why are companies (who have vested interests and care more for shareholders than the industry itself) allowed to make standards and specifications that should apply to everyone in that industry (including their competition)?
We need unbiased standards committees, who fully intend to improve technology for everyone.
We’ve got self appointed standards committees formed from companies that share common biases, who fully intend to improve sales for themselves.
ACPI is “co-developed” by Intel, Microsoft and Toshiba. If you’ve got a Toshiba laptop with an Intel chipset and CPU running a Microsoft OS, then ACPI probably works flawlessly.
Some people say we cant blame MS for lousy acpi stuff in linux? thats highly wrong.
Microsnots dsdt compiler on purpose messes with the the linux specific stuff, and also stuff which is generic for all OS’s. And how do i know this? you may ask. I will tell..
1: download intel acpi tools.
2: decompile your own dsdt.
3: “obtain” MS acpi compiler
4: make some nice linux specific things, some things that are generic, and some MS specific things(doing all the same)
5: compile using MSFT compiler.
6: decompile and observe how it has deliberately(or because of very coincidental bugs) destroyed and messed with linux specific parts, and some places even turned generic stuff into win-specific stuff..
7: compile original with intel
8: decompile that compiled with intel, and observe how it has NOT been destroyed.
9: conclude the most likely cause, especially with this new mail we have where bill gates is so kind as to state he wishes linux not to work, and are fine with actively seeking to sabotage us.
put it simple, bill gates should be put in prison, and tortured every day for his crimes against technology and humanity. His assets ceased, and microsoft shut down, or atleast forcibly restructured.
19,20d18
< put it simple, bill gates should be put in prison, and tortured every day for his crimes against technology
< and humanity. His assets ceased, and microsoft shut down, or atleast forcibly restructured.
Good to see I’m getting modded down for not agreeing with the EU’s actions against MS. My statement was not offensive, I think it was inline with the general topic, but I was modded down because some people don’t agree with me, but nobody replied to it. Perhaps they don’t have any good arguments?
Stating that EU’s trial is “part extortion, part control freaks” isn’t exactly what I call a good argument either…
Kochise
I seem to remember a time when Mr. Bill was alleged to have said “DOS ain’t done ’til Lotus don’t run.” Even if it ain’t true, it still speaks to a corporate attitude, perceived by the community, that apparently hasn’t changed any.