- Related Stories
-
Foundation wants more of the PC to be free
February 28, 2005 -
Intel to open code for booting up PCs, servers
June 1, 2004 -
Writing an end to the bio of BIOS
December 30, 2003 -
Phoenix toughens up BIOS
November 26, 2003
(continued from previous page)
replacement, with its own set of drivers to turn on elements of the PC such as the processor, based on EFI. Committing it to open source means others will be able to download it from a Web site called TianoCore.org and use it to make products under the Berkley Software Distribution, or BSD, license. The BSD will allow anyone who uses it to change it and create products out of it. But it does not require they provide the changes they made to others via open source, which provides the means to help companies protect intellectual property.
The effort by Intel creates a framework for a BIOS replacement, and thus could become the basis for free BIOSes. But it leaves the work of writing the code that initializes PC components to the downloader. One licensee likened it to having to build a race car. Intel, he said, provides race rules and the car's frame but leaves licensees to do their own engine, suspension, body work and other elements if they want to enter a race.
'Evil' companies?Stallman argues instead that Intel is not doing enough and BIOS makers are not needed. Instead, he wants information.
"We're not wanting to do anything with the BIOSes from Phoenix or any of the others," he said. "We're not asking them to do anything, any more than we're asking Microsoft to do anything. These (companies) are evil. You can't expect them to do anything just because you ask them to. Our goal is to escape from them."
Thus, the free BIOS effort, as Stallman sees it happening, will essentially bypass traditional BIOS makers and instead focus on appealing to hardware manufacturers. The campaign will ask those companies, including PC makers and motherboard makers, to make available specifications on their products to allow free software writers to create BIOSes for them.
Stallman also dismisses rebuttals that free BIOS would compromise a PC's security, stability or reveal companies' proprietary chip, motherboard or other product information.
"Each one could be saying, 'If the others knew what we were doing, it would help them tremendously.' It might be true in a few cases, but it's impossible in all cases," Stallman said. "They can't all be sitting on secrets that are beyond the ken of their competitors. They can't all be the ones that know more than everybody else."
Moreover, detailed chip and motherboard information will not be required to create a free BIOS, he said. Instead, free BIOS makers would need access to closely held instructions, such as how a BIOS loads and how it initializes various devices inside a PC.
A free BIOS would also help circumvent, if necessary, digital-rights management, allowing people to run any software they choose on their PCs. In theory, the BIOS can be used to aid security technology, as it initializes hardware such as security chips.
Although BIOS makers and Intel say the BIOS' role is limited to helping get those elements of a system up and running along with the rest of it, a BIOS writer could write around them in order to shut them off, if needed, Stallman said.
"DRM is theft," he said. "The idea of the free software movement is you should be in control of your own computer. Treacherous competing (his term for so-called trusted computing) is a scheme to make sure you're not in control."
Ultimately, the free BIOS would emulate software such as the LinuxBIOS-- a free BIOS that's already in existence for Linux, but does not work with a large number of PCs--on a much broader scale.
"It's generally known that free software is very secure and very reliable," Stallman said. "If there's a bug in the BIOS, the only thing that will happen is some part of your machine won't work and that bug would be quite noticeable and it would be fixed, presuming that the information was available."
But that's the rub. Detailed specifications on cutting-edge PC hardware may be tough to come by. The information given to BIOS makers now is granted under nondisclosure and it's not clear whether companies such as Intel, PC makers like Dell, or motherboard makers would reveal even a little bit of information.
"You'd need to know the confidential information about the chips to write" a free BIOS, Insyde Software's Joseph said. Right now, "that info is only available on old hardware that nobody really cares about anymore."
That, however, won't stop Stallman from asking.
See more CNET content tagged:
Richard Stallman, BIOS company, critic, BIOS, Free Software Foundation
Free means the Freedom to use, modify and redistribute the Software under the auspices of its license. This never was, never is and never will be about the amount of money that is charged for Free Software.
care because you don't care. You aren't a
consumer. More to the point, as long as there's
some semblance that things are working "good
enough", you could care less about the BIOS.
The "consumer" in this case consists of hardware
manufacturers and hardware support personnel.
The former have an interest in an open BIOS
because it would speed development and give more
freedom in design. The support people are
insterested because an open BIOS could aid in
developing hardware diagnostic tools, and in
tuning hardware to be more secure, fault
tolerant, or higher performance according to
their specific requirements.
As a paltry applications programmer -- you could
care less and will settle for the environment
the hardware and host environment present to
you. It's the person that deploys your software
that has to make sure the hardware is up to
snuff (and they, in turn, rely on the hardware
vendor to make it so).
But consumer having to spend on PC does.
Closed BIOSes and closed hardware is what ensures livehood of some hardware companies, producing some ****** hardware we have to live then with.
Have you know amount of bugs BIOS has to fix - you as customer will choose more carefully what kind of computer to buy.
And this is especially issue for notebooks - they are not that easy to upgrade or repair compared to PC coffins.
On modern systems, OS depends more and more on BIOS to perform most of the operations. Having standards for BIOS or (even better) open source BIOS will ensure that any one can port OS to given computer. (*)
Lower entry price means more companies can compete in market.
Higher competition inevitably means lower prices for customers.
Simple logical chain. Closed BIOSes in wake of DRM talks become more and more heated topic. Hopefully, asian computer producers are (seems at moment) embrasing idea of open source BIOS. They are very - more than we will ever be - tired of dependency on BIOS makers.
(*) Absence of standard probably one of the reasons why Apple doesn't want to deal with PC hardware. I can only praise both Apple and Sun who came up many years ago with OpenFirmware.org project. Intel's EFI didn't impressed me. Thou better than current nothing.
I want the choice. Whether I exercise it or not is irrelevant. What I'm willing to contribute to have that choice is.
As machines get more complex, they have to be more simple to be useful to the world at large. If people would stop hating on Bill Gates for a minute and listen to the guy when he says people just want stuff to work (Steve Jobs is big on this as well), the world would be a better place.
It's bad enough that today we expect computer users to have updated drivers, now you want them to have the "freedom" to screw around with their BIOS? Can you imagine the acceptance rate of everything from iPods to toasters if people could screw around with them? Two words: Support nightmare.
Stallman has drifted so far away from computer science and market reality and firmly into religion that he's totally out of touch.
I do disagree with the notion that a "closed source" model promotes competition as Intel claims. If I could take my new Dell computer and install a BIOS from HP or Gateway, THAT to me seems like more competition. The need however really isn't there. No matter how nicely tweaked a BIOS is, it is not likely to provide a gain in performance or productivity measureable by the end user.
Stallman has more "non free" software on his computer than he's willing to admit. The fact that it took him this many years to realize his BIOS was paid for is astonishing! The CPU has instructions that are "non free" even though the most common ones (referred to as the core X86 instruction set) are well published.
I don't see where he can gain any win in the free BIOS fight. If you want a free BIOS, create one and good luck getting the manufacturers to use it.
That is a bit old school. Open Source is about "to all according to their needs, from all according to their ability."
Open Source is about community, not Redmond-style cutthroat freemarketeering.
In a sensse Open Source does have a cost--in the time needed to make use of often little documented but powerful software. But the key is that the investment builds on itself, and you know that you need not throw it all away just because some rich guy wants more of one's money.
It's a CIA plot dude. Make sure that you have your aluminum foil hat on at all times so they can't reprogram your mind as we all know it's definitely open source.
Fred Dunn
All previous laws in all other areas may have outlawed certain activities, but never was the principle ever enshrined that the choice to obey or violate should itself be extinguished--until these dark days of copy locks and the DMCA.
And yet this is is exactly the way predatory corporate interests are absuing copyright, in terms of "Activation" and soon hardware-based software and content locks. This is a trend that lawmakers should stop in its tracks before it gets out of hand. Supreme Court guys, listen up -- you gotta stop this nonsense.
The first principle of a free society is that the governed consent to be governed. There can be no consent when there is no choice in so doing, as is the case with the DMCA, secret BIOS systems, and so forth.
Lawmakers need to take action soon to modify copyright and patent law to specify that no lock mechanism, whether it be software- or hardware-based, may have the effect of preventing the user from using software fully, modifying software fully, or using hardware fully, or modifying hardware fully, and that further prevents the use of license terms that control or direct the user in ways unrelated to the basic licensed use itself.
I do not want open source BIOSes. There are enough things to troubleshoot besides having a client's 13 year old try to flash with the latest BETA bios. At least the manufacturers can be held responsible for a bad BIOS.
I think the open-sourcers will get in over their head on this one. All OSes depend on BIOS standards to operate properly and a bad BIOS can cause major problems, even if it POSTs.
Oh and I want the blueprints and a wrench so I can make some changes... for the benefit of everyone.
Get Real!
If you don't like protected software, don't USE it. Make your own (of course it will be better, right) and give it away free for the benefit of us all.
I happen to agree that their is no reason to force BIOS maker to open up their source. I don't think every piece of software has to be open source, but I find nothing wrong with open source.
What I do find wrong is Licencing. Open source has too many variations and a lot of closed source companies have very greedy ones. For example, Microsofts OEM licence states something to the effect that if you replace your computer (even due to it being hit by lightning) you have to buy another copy of Windows. Of course you can always licence it to your mouse, but if you replace that then you have to buy a new copy of windows. Yeah I know I am picking on Microsoft, but they just have lots of good examples. Of course with dual core procs Oracle may be the next best to pick on.
SIGNED:PHYSICIAN THOMAS STEWART VON DRASHEK M.D.
As for a free BIOS, I think it would be a welcome addition. Today's pre-fabricated PCs come tied to their original hardware and replacement parts can only be bought from the original PC maker. BIOSes are no different. With each new year, BIOSes include less and less configuration options, so yes, consumers are loosing control of their PCs. DRM is the industry's answer to piracy (which they think all end users are guilty of). These DRM methods which are built into hardware will soon be hacked (didn't they say at first that DVDs couldn't be copied?) and there is nothing the companies can do to stop it. They shouldn't be afraid of open source BIOS, open source has been around for 14 years or more and propietary companies haven't lost their market. If they don't allow open source BIOS to exist, and let those users who care to "hack" or modify their machines with it, they will resort to illegal means of doing so. Why not let there be an open source BIOS, and let few users who have to know-how to use it, do their thing legitimately? I don't think that whatever losses they have will be more than what they will spend in filing lawsuits against those who "hack" their precious DRM.
Buy a HP or Dell and you will find a gimped BIOS, and for good reason. It scares me to even think of the generic computer user having open access to bios functions, when they have a hard time just operating the computer without breaking it.
And, talking about "a user-controllable free BIOS being important" in enabling the porting of Operating Systems to PCs, as well a all the various functionalities and about open-source codes... Question: Why is it that IBM has not yet decided to make OS/2 Warp Open Source; additionally, how much of the remaining unknown APIs (Application Programming Interfaces) for MS Office will Microsoft will be willing to share with the open-source communities. Re: http://www-pam.usc.edu/volume4/v4i1a3s2.html Hence the inherent computer software application "Functionalities" for Economic And Financial Assessment.
why don't stool-man start its own pc manufacturing company aka GNU Computers, and give away free computer to everybody. like opensource, drink open source, use only gnu pcs. ha ha ha ho!.
http://russ.johnsonville.net/default.aspx?Page=Blog
All he is asking for is what Intel gave to Pheonix. Enough information to write a BIOS. How is that a bad thing?
With regards to open-source BIOS, I guess it is valid to be concerned about controls, standards, security and ethics. Although I am not contending that proprietary BIOSes are controlled, secured and ethical, we can assume that they are already in place -- given the amount of investment needed to develop and maintain BIOSes, businesses are likely to fund only what works and what sells.
Open-source BIOS seems to open up many possibilities beyond the limitations of the proprietary BIOSes available today. In fact, open-source BIOSes can be powerful BUT, looking forward, they call for greater responsibilities besides just developing and publishing them.
If the BIOS links software to hardware and vice versa, then opening the BIOS may introduce inventions that allow software to have greater access and control on the hardware and/or hardware to have greater access and control on the software. Sounds good... Looks dangerous...
Hopefully, implementors and users of the open-source BIOS are responsible enough to control and standardize what gets manufactured.
As for "What about the Processor Microcode???" It was Intel's' opening of the code for the 4004, 8080, and 8085 that allowed them to steal the market from Fairchild, National Semiconductor, Motorola, and the rest. I learned Intel Machine code from Intel engineers who taught at the college I went to in the early 80's. All those new engineers knew Intel code and that was the processors they designed for. It was a true stroke of brilliance on the part of Intel.
Those who badmouth the free interchange of information are just a bunch of "Johnny-come-latelies" who wouldn't know a COM port from a crouton.
In any case, having access to microcode would not provide much value to anyone writing external programs and thus couldn't have anything to do with Intel's success.
...when, the hardware will ONLY allow 'approved software' to run? ("Trusted Computing")
...when, network-routers will ONLY ALLOW 'approved computers' on the internet? ("secure" networks)
...when, the PRIMARY-PURPOSE of the computer, itself, is to CONTROL consumers? (DRM)
...when, the "closed" proprietary-BIOS is directly-linked, and CROSS-DEPENDANT upon proprietary-software? (OS-specific APIs)
...THERE IS NO CHOICE.
Nor, are we actually talking about "secrets". The fact is that, today, anyone can easily gain access to (or legally reverse-engineer) any level of the PCs-hardware, ...which is the way it should be (and why the PC-market has been so robust). ANYONE can write 'extensions' or 'software' that will run on ANY PC. But, much of the "Trusted Computing" architecture’s implementation is specifically designed to eliminate that FREEDOM.
The simple reality is that many of the technical-specifications of "Trusted Computing" primarily revolve around "DRM" and 'proprietary-solutions LOCK-IN. In other words... CONTROL (and choice of functionality) BEING IRREVOCABLY REMOVED FROM CONSUMERS/USERS.
A 'closed', inaccessible, BIOS is nothing more than an action similar to an automobile-manufacturer's 'WELDING the hood shut' (and claiming that it is for motorist's, and the independent car-mechanic's, own good).
Consumers DO care (if not about the technical-specifics, ...then they care about the direct results upon their ability to make choices) I know, I have spent years providing service, and explaining such things, to them.
And, I am telling you that, ...IF THEY HAVE A CHOICE, ...consumers ALWAYS choose to have CONTROL FOR THEMSELVES.
And, yes, I CAN write a 'device driver', a 'BIOS function', program an 'ASIC' and burn a 'ROM'. But, once the PC-hardware and the BIOS itself is 'completely locked-down', I will have no more choices available to me, than the "...average people...", to which, I provide solutions, and training.
But I can only see / use 3.5G.
This is because the STUPID BIOS reserves the last 512MB of address space addressable in 32 bits for its own use.
You don't notice this LEGACY FEATURE if you only have say 2GB. But when I upgraded to 4GB - there it was.
Of course on a 64-bit-capable system such as mine, the BIOS should be a 64-bit program (I bet it is on Itaniums, because they can't do 32-bit, but I like the ability to do that as AMD processors offer).
Would I be up to fixing the problem? 35 years in the business of systems programming says: go slowly. Maybe I could reduce the reserved amount while staying 32-bit. Moving to 64-bit requires a development environment that I really don't have right now.
Anyway should the BIOS be open-sourced?
ABSOLUTELY
"when, the "closed" proprietary-BIOS is directly-linked, and CROSS-DEPENDANT upon proprietary-software? (OS-specific APIs)"
also,
...when, the hardware will ONLY allow 'approved software' to run? ("Trusted Computing")
Question: Assuming that the network (banking and financial services) must safeguard "sensitive client" information... should there not be restrictions as to who will be permitted to access data on such networks. There is an old saying that goes, When in Rome do as the Romans do. Astronauts and cosmonauts must know that there are rules to be followed to ensure a successful re-entry back to earth's atmosphere when returning from outer space. I am for stamdards, control and experimentation. A good position that could be taken is to let the open-source community do what they want to do and let the who want to bring astronauts safely back to earth or protect clientele data as with ("Trusted Computing") adhere to strict standards and control.
Once again, "OS-specific APIs" make quite a lot os sense in that; and, to repeat:
Functionality, Functionality..............
Short Session on the Economic Rate of Return (ERR)
Here is an Outline.
A. Nature of the ERR - shadow prices as proxies to
opportunity costs.
B. Purpose of the ERR - economic viability
C. Methodology for arriving at the ERR
1. Financial computations (IRR) as inputs
2. Apportioning variables to shadow prices
a) theory: all inputs and outputs
b) practice: 1) important costs and
benefits; and,
2) most likely distorted items
and so on, and so on.....
D. Interpretation of the ERR
1. If ERR > IRR
2. If ERR < IRR
As Keith remarked "Open Source is about "to all according to their needs, from all according to their ability" That much I can discuss with the community for now. The point that I am attempting to make is this , where do people in general want to go! If it is outer space then all the resources would have to be gotten together, inclusive of at least a skilled captain; just so, in the case of arriving at an economically viable international project all the resources would be needed - computer operating system and application, along with the multi-disciplinary analytical capabilities of the particular individual/s to be involved. Just how well developed is the computer industry (Free and Open Source as well as propietary + shared-source code) to satisfy the above-mentioned functionalities. Questions: Think Windows, Linux, MS Office, OpenOffice... Are ready with the requisite "OS-specific APIs"? Think Again! Who has funded/will fund the developments? What have been/will be the important costs and benefits? Does Open Source have a cost?, and how much!
Can I have a firmware download... No, your series of model 8690 Mfr Yr 2000-1 was not intended to go to Corporate Network users !!! Sorree... so why does it have a NIC at all? Why can't I reprogram the BIOS ? Click.
- Wouldn't this render support impossible?
- by Christopher Hall April 6, 2005 12:34 PM PDT
- One of the things that makes Linux such a turnoff for the masses is its lack of "concreteness." That is to say, once someone gets their hands on it, absolutely anything can (and probably will) go wrong, with very little chance of support. I grant that most of the people who use Linux have quite a good degree of competence when it comes to the inner workings of an OS (that's my perception, anyway), but I think if I gave open source software to everyone in my family and told them to go to town, when they came to me with questions on their specific configurations, I would probably be lost in the dark.
- Like this Reply to this comment
-
-
- How is Linux not supportable?
- by pcLoadLetter April 6, 2005 2:39 PM PDT
- No matter what distro you choose, you can get support for it. Go to a software store and pick up a copy of SuSe or Red Hat and support is pretty much what you are paying for.
- Like this View reply
Processing -
- Just re-install your BIOS
- by zaznet April 16, 2005 12:13 AM PDT
- If the company could not support the system with a BIOS provided by another vendor/programmer then they would just tell the user to re-install the original manufacturers BIOS.
- Like this
-
Showing 1 of 2 pages (118 Comments)In my opinion, one strong selling point for closed-source is the fact that it's supportable en masse. I'm not familiar with how you would support something that is so highly customizable, but perhaps someone can fill me in.
If you choose to make your own alterations to the code, you obviously won't get much, except maybe in newsgroups. But, someone with no knowlege of how to change the code, is going to be doing that. They are likely going to get a retail box or buy a little support after they download, so your point is pointless.
That's as much as double clicking an EXE.