Blurring The Line Between BIOS And OS

Jon Kincade writes "The Register has an article about Phoenix Technologies cME software that allows users on anything from servers to embedded systems to run diagnostics, browse the web and other things without having to boot into a full fledged OS. The primary use seems to be recovery from system crashes. Also, this may explain why the Phoenix browser was asked to change its name a few months ago."

Comments

Finally!

[creative_name]

Command line browsing at it's best!

Look out ASCII porn here come the BIOSonly users of the world ;-)

Re:Finally!

[Anonymous Coward]

Gives new meaning to FLASHing the BIOS!

Re:Finally!

[Tacky the Penguin]

Now, all we need is a computer that boots straight into a programming language.

Uh, wait a minute.......

Re:Finally!

[Anonymous Coward]

I just wanted to post something.

Why do we even have a BIOS anymore?

[Anonymous Coward]

Couldn't we move to some sort of system where it's no longer necessary? Or maybe just a very skeletal one to start the boot process.

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

That would never work.

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

What a silly idea. PC vendors are obviously much too smart to fall for this.

Re:Why do we even have a BIOS anymore?

[El Cubano]

Couldn't we move to some sort of system where it's no longer necessary? Or maybe just a very skeletal one to start the boot process.

I believe that is the point of the current setup.

The BIOS kicks things off, does the self test to verify RAM/HDD/video operational status and if there are no errors, turns it all over to the operating system.

I don't see how it could possibly get more skeletal than that.

Re:Why do we even have a BIOS anymore?

[azmaveth]

Because we need a standard way to interface with hardware from the OS, and hardware vendors keep adding and improving features (darn them!).

The BIOS is what allows motherboard manufacturers to have different chipsets with different memory, disk, etc, controllers. The BIOS translates basic data requests from the operating system into the specific instructions that each chip requires.

I'm all for open standards on hardware, but the fact is that every time a new feature is introduced or an old feature is upgraded, you need a backwards compatible method of accessing it. Thus the BIOS layer.

My $0.02.
-azmaveth

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

A lot of bios these days have all sorts of unnecessary things, like virus checking, etc. Not all hardware needs to be detected by the BIOS.

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

You mean like this?

It supports emacs keybindings, too!

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

I take it you never owned an IBM XT did ya? Mono video, NO hd, just floppies... 640K mem.... Nuthin smaller than that (wanna buy mine?) :)

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

what virus has bios checking? you mean the ones that see if a program writes to the boot sector? that's all they do. it's simply a thing that looks like, "if the BIOS argument of the sector is the boot sector, stop everything happening and put a message on the screen". I don't think those sorts of BIOS options can go around things like win2k multitasking either. (once you start 32-bit mode, ring 0 can only be obtained through the OS.)

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

linuxbios is pretty basic. nothing but what you need. of course you can throw the linux kernel in BIOS space too, but that's just those few people who want super super fast boots. (less than 3-5 second from power on to login prompt)

Re:Why do we even have a BIOS anymore?

[ebassi]

Because we need a standard way to interface with hardware from the OS

Standard way?

That's funny, because my BIOS (a pre-98 Award BIOS) recognizes my primary HDD as a 8GB drive, while the label upon my drive, and my OS, say that it's a 15GB...

BIOSes are used by old operating systems (DOS, for instance). Modern OS rely on their personal hardware recognition. On modern machines, a BIOS is only the crap that makes your machine to boot up in 15/30 seconds, while you could have a 2 seconds boot sequence (cfr. LinuxBIOS).

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

BIOSes are used by old operating systems (DOS, for instance).

And Windows ME, and OS/2, and NetWare -- all still in production.

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

Doesn't the BIOS have a lot of old code designed to support DOS?

Re:Why do we even have a BIOS anymore?

[ebassi]

BIOSes are used by old operating systems (DOS, for instance).

And Windows ME, and OS/2, and NetWare -- all still in production.

I did not say "old" as in "released long time ago": I meant old by design, which OS/2, WinME (to an extent, that is maintaining the backward compatibility with old drivers) and NetWare are, just like DOS.

Re:Why do we even have a BIOS anymore?

[CerealQiller]

I had a couple of those, they didn't have much in the way of a BIOS either. chuckle. Wish I'd kept one, I have this door in the back of the house that slams when I open the front door... also excellent for protecting closet shelves from dust.. so light... and sooo easy to use.

Re:Why do we even have a BIOS anymore?

[Anonymous Coward]

> Because we need a standard way to interface with
> hardware from the OS, and hardware vendors keep
> adding and improving features (darn them!).

That is what ACPI is for.

Re:Why do we even have a BIOS anymore?

[CerealQiller]

why do we need a BIOS anymore...? Visualize this two part exercise. Install an unformatted HDD in a new machine and then leave the OS CD in the cd box it came in... The machine will run, devices will be found, and connections made to them without the OS loaded or even formatting the disk. Up to several pages of data will scroll up your screen beginning with the video BIOS information, memory test,and moving on to HDD, CDR/CDRW/ATAPI/SCSI/floppy devices, and PCI card slot population and device configuration, USB info, IDE RAID ports/devices installed and finally integrated peripherals and NIC BIOS along with any RAM boot settings just before it tries to find the boot server (if it has a RAM boot Eprom on the NIC card) All this activity seems fairly useless to the user and the time it takes is generally annoying, but all that activity is from the BIOS alone, it hasn't even started setting up the control logic in memory because the control logic has to be loaded from the HDD... but to get to the HDD, floppy drive, or other bootable device, the machine has to have some form of control over these devices, starting instructions. Those starting instructions are the BIOS, which loads the control logic (Operating System) into an area of memory from which the computer can assume control of its own functions. The BIOS checks out the hardware resources of the machine at boot, and builds a control table in memory listing the basic resources of the motherboard and essential components of storage and control, it then hands this off to the OS after the OS has finished loading, but these functions are independant of the OS, which OS it will be booting, or the location of the connection to the boot information (HDD, RAID, Server, Floppy, SCSI device, or even a RAM chip. The BIOS doesn't even care if you have more than one OS available. It will look for starting instructions in several places which are configurable in CMOS, and if those instructions happen to be a boot manager, then it will load the boot manager without skipping a beat. That is already a LOT going on without having an OS running yet. Now, disable or FlashFry your BIOs and put the CD in the drive... You now have the ultimate in desk fans, even with the CD in the drive, the only thing that will function on it will be fans and motors and only those which are independant of BIOS control. Why would something like this happen? Because the BIOS is not the Eprom chip on the MB, it is the logic written to the chip. When the machine powers up, that set of instructions starts running, with one simple basic task, to load the OS so that the machine can function at a more complex and user friendly level. The BIOS boots your system, not the OS. The term Boot comes from "Bootstrap" and that from the phrase "raising oneself by one's own bootstraps". This word was coined as the name for a process required to start old mainframe computers in which a crew of several people would spend weeks to months setting toggle switches on a special panel. a row of 8 switches in two groups of four, with an extra switch on the end. The two banks of switches were properly set into position, and then the end switch was hit. This sent current through those switches in the banks which were closed, this current passed through an array of crossed wires with tiny magnetic donuts at the intersections. Called "flip-Flops" or memory "cores" (this is where the term "core memory" comes from) these donuts would magnetically allign with the current passing in the wires by physically changing position, when running this made these machines quiet noisy as the magnetic cores "flip-Flopped". The racks on which the cores and crossed wires were strung were called registers. The connections between the registers, were the busses. If the current was the same polarity as the donut, there was no change in orientation and the result was 0 (false). If the current caused the donut to change orientation (which generated a current pulse in the second wire) then the signal would be 1 (true). The process of flipping the panel switches set up the orientation of the flip-flops as machine code instructions in 4 bit nibbles each paired into an 8 bit byte, basically storing information by tipping over a magnetic washer balanced between two wires. The process of loading machine code manually by toggle switch was mind numbingly tedious and was replaced by a punched paper tape with holes indictating which switches would be closed for each byte, the paper passed through a reader with small spring wires over each column on the strip, if a wire passed through a hole it made electrical contact with a pad on the other side of the strip, which in tern sent current flowing through the wires in the "core" exactly as it would if manual switches had been used, but taking only hours rather than weeks. (and you thought your system booted slow. ;) The BIOS came along after computers became semiconductor based. It is a set of machine code which loads the control code from a storage device, just as the punch tape reader replaced manual boot panel switches.Now keep in mind that magnetic storage computers were huge in size, small in capacity. A 4-banger pocket calculator has more capacity than one of those monsters did. The BIOS on a modern PC replaces what would be a crew of thousands of people flipping manual switches for decades, writing the OS into the memory of the CPU. Think of the BIOS in terms of tiny service engineers manually reading each line of the Operating System from the HDD and flipping switches to set the logic states for each byte of code, one bit at a time. Only after the control logic is set up in memory can the computer function. Thats why we need BIOS.

Re:Why do we even have a BIOS anymore?

[MSZ]

BIOSes are used by old operating systems (DOS, for instance). Modern OS rely on their personal hardware recognition.

And what loads bootsector into memory? Boot fairy maybe?

EVERY system, new or old, including Linux, *BSD, Windoze, OS X etc etc needs to be loaded into memory first, before it can use it's more or less advanced hardware recognition, fancy drivers and so on. LILO, GRUB or NTLOADER need BIOS to read in the OS kernel from disk.

Re:Why do we even have a BIOS anymore?

[HishamMuhammad]

Not actually. Today's BIOSes are filled with "legacy stuff" that is not used by most of the modern operating systems.

The IBM-PC BIOS was designed to cooperate with MS-DOS, providing a lot of low-level services to it. Many will surely remember interrupt-oriented assembly programming. It was basically a lot of DOS calls for filesystem handling and BIOS calls for everything else. For this reason some people even refused to call DOS a proper "operating system".

A lot of posts here try to reminds us about the meaning of BIOS as Basic Input/Output System, and that's the key to the argument against the existance of BIOSes: modern operating system don't need the BIOS to perform basic input and output anymore.

Making a PC without a BIOS, only with a minimal "kickstart ROM" that moved right on to the boot loader, would be perfectly feasible. The problem is that the idea of a PC that can't run DOS is a bad one for many people. (Remember that "can't run DOS" automatically means "can't run Windows 9x/ME").

Re:Why do we even have a BIOS anymore?

[ebassi]

And what loads bootsector into memory? Boot fairy maybe?

Almost true. A boot sector loader, as a matter of fact. It scans the physical place where the OS boot system should be located, and transfers control to it immediately after.

So, since it disappears as soon as the operating system boot manager kicks in, and since it manages only the "preboot" phase, it's not a Basic Input/Output System. (that is, a layer between the very lowly hardware level and the OS).

wow

[The Other White Boy]

thats...actually a brilliant idea. but in implementation, how well would a bios chip doing all this work do? i mean, we've got bios chips nowadays that need active cooling, what would it take once we add features like this?

Re:wow

[Neon Spiral Injector]

I don't think any BIOS chips need cooling. They are just the little flash roms on the motherboard. You are probally thinking of the chipsets.

The BIOS is just code that gets executed by the processor before the OS starts to boot.

Also it was noted that all the extras of this enhanced BIOS will be stored in a "protected" area of the hard drive. I bet a `dd if=/dev/zero of=/dev/hda` will still kill it, just like my restore partition on my IBM laptop.

Re:wow

[Mattcelt]

I agree that it's a wonderful idea. I think this is yet another call for open hardware - especially an open-source BIOS! Heck, the specs are all out there - there isn't anything about BIOS programming that is proprietary or costly, so it just plain makes sense that a group should make an open bios.

Re:wow

[John Hasler]

You've not heard of the LinuxBIOS?
http://www.linuxbios.org/index.html

Comment removed

[account_deleted]

Comment removed based on user account deletion

May?

[The Bungi]

Also, this may explain why the Phoenix browser was asked to change its name a few months ago.

Some backing link, information, you know, proof? Or didja just think "Phoenix" and "Phoenix" and somehow it just "clicked"?

Re:May?

[stankyho]

Well the link to the /. article about "Phoenix, the Mozilla-based web browser, is forced to change name. The new name has not yet been decided, but it is being discussed . The reason is that the BIOS manufacturer Phoenix Technologies dislikes the trademark infrigment. Next week version 0.5 will be released, with a new name." worked pretty good.

Re:May?

[vondo]

Well, the Phoenix people at mozilla.org were asked by the people at Phoenix the BIOS company to change the name. So I'd say this is a pretty good guess.

Re:May?

[The Bungi]

Yes, I know that. But is Phoenix Technologies having a fit over a Mozilla project simply because of the name, or because they were planning to embed a browser in the BIOS? The text of the submission apprently makes the link between what the snazzy BIOS can do and the browser itself, but not necessarily with the name "infringement".

Re:May?

[Anonymous Coward]

Well let's see. From the linked slashdot article:

The reason [for the name change] is that the BIOS manufacturer Phoenix Technologies dislikes the trademark infrigment.

From Mozilla/Phoenix developer Blake Ross's post to Mozillazine , which was linked from the old slashdot article:

We need to rename Phoenix. The kind folks over at phoenix.com have complained about trademark infringement. Anyone have thoughts on a cool, meaningful new name?

So I'd say it was just a guess.

Re:May?

[SquadBoy]

This makes the suit actually have some merit or at least more merit than it had before. Before we knew this it seemed like it was just another case of corporate landsharks going after any and everybody with a similar name. Now that they both have products in the same field it appears that this is what they where thinking about when they decided to sue.

Re:May?

[The Bungi]

Yep, and that's why I was asking about some backing info on why Phoenix did what they did.

Re:May?

[blair1q]

They didn't do it in release 0.5, btw.

They only said in 0.5 (scroll down to items 14 & 15) that they would do it in the future.

The codename for the 0.6 beta is "Glendale", but I'm not downloading it just to see what's inside. Maybe after dinner.

Personally, I think they should change the name to Pheonix and tell Phoenix to go shove an EPROM up their ass.

Re:May?

[Longinus]

Some backing link, information, you know, proof? Or didja just think "Phoenix" and "Phoenix" and somehow it just "clicked"?

Some reading the links, and little information gathering, you know, research? Or didja just think "Slashdot" and "ignorant assumptions" and somehow it just "clicked"?

Re:May?

[The Bungi]

I don't know Mr. Longinus. How about you read the rest of my posts in this same thread and you impart some more of your impressive wisdom? Perhaps my "ignorant assumptions" just are not "clicking" for you, hmmm?

Here's one from ZDNet.

[Anonymous Coward]

Phoenix Technologies is hoping PC manufacturers will latch on to its newest software to help solve PC problems and improve data security.

Phoenix, which creates BIOS software for many of the largest PC manufacturers, branched out this week with the new CME, or Core Managed Environment, software suite.

Where the BIOS (basic input/output system) provides a bridge between a PC's operating system and its hardware, CME will create protected areas on a PC's hard drive that can host sensitive data or applications that alleviate common problems, allowing the computers to run even if the operating system is damaged, the company said.

Phoenix, which will sell the software directly to PC makers for an undisclosed price, becomes the latest in a series of hardware and software makers trying to make PCs easier to use and more secure.

Intel, Transmeta, Via Technologies and Microsoft have recently launched new security initiatives. IBM has also been offering special data recovery software and a security chip in its new PCs. Collectively, the companies are seeking to better protect sensitive data, owned by companies or individuals, against thieves.

Although the security offers--including Intel's "LaGrande" technology, Transmeta's newest Crusoe processor and Via's Padlock--are built into chips or, in the case of Microsoft's Palladium project, into the operating system, Phoenix's CME will reside in a protected area on a PC's hard drive.

CME applications are intended to protect and recover PC users' data and to help the PC itself repair damaged software or connect to the Internet to download updates, the company said in a statement.

Phoenix will also offer versions of CME for embedded devices, such as industrial equipment, consumer electronics and servers.

Re:Here's one from ZDNet.

[Anonymous Coward]

Please note that's a troll posting. He replaced AMD with transmeta, moderators need to read cut and pasted articles without moding. For that matter, stuff that's accessable elsewhere should not be pasted as it is, rather it should be linked.

Re:Here's one from ZDNet.

[Anonymous Coward]

"For that matter, stuff that's accessable elsewhere should not be pasted as it is, rather it should be linked." Dude. How long have you been on Slashdot?

Re:Here's one from ZDNet.

[Anonymous Coward]

it sounds like DRM is here. time to go out and buy all the trustworthy hardware you can before none of it is left. don't confuse my use of "trustworthy" with THEIR use of trustworthy.
it sucks that i will be limited to 3 ghz for the rest of my life. :/

BIOS and its definition

[very]

Basic Input Output System

If you can do more advance operation through "BIOS", then the "BIOS" is no longer BASIC.

Therefore, it should no longer be called BIOS (Basic IOS)
Call it Embedded Operating System (EmbOS).

Just a thought!

Re:BIOS and its definition

[SN74S181]

Maybe they should quit fooling around and have the sort of robust firmware that Sun has in their SparcStations. You can do a hell of a lot with a SparcStation at the bios prompt. I believe it includes a Forth engine. Correct me if I'm wrong.

The BIOS on PeeCee motherboards has always been anaemic. And it's shocking how paranoid everyone in this discussion is about that changing.

Re:BIOS and its definition

[brunnock]

And what about SCSI?

Re:BIOS and its definition

[MamasGun]

Apple has the same thing too...it's called Open Firmware. It's pretty cool.

Re:BIOS and its definition

[(H)elix1]

Therefore, it should no longer be called BIOS (Basic IOS)
Call it Embedded Operating System (EmbOS).

Or EmacsOS... It seems it can do most everything else, why not the BIOS too?

Re:BIOS and its definition

[Anonymous Coward]

Give me a Real System Monitor over a BIOS any day.

So it's supposed to be "thin". We need to redefine what is "thin" as the hardware capabilities increase!

Look at the FORTH interpreter on a Sun. Why can't we have something like that? We settle for BIOS on PCs because it represents more-or-less what we've always had, incremental development since the 8088 PC.

Re:BIOS and its definition

[Anonymous Coward]

> Therefore, it should no longer be called BIOS (Basic IOS)
> Call it Embedded Operating System (EmbOS).

What about
Bastardized Incarnation of an Operating System?

Re:BIOS and its definition

[Anonymous Coward]

More likely they will re-define the acronym as Built In Operating System

Re:BIOS and its definition

[pmz]

Maybe they should quit fooling around and have the sort of robust firmware that Sun has in their SparcStations. You can do a hell of a lot with a SparcStation at the bios prompt. I believe it includes a Forth engine. Correct me if I'm wrong.

You are mostly correct. The only thing is that it isn't a "bios prompt" but a "boot prompt" or an "ok prompt". Additionally, enabling full diagnostics at the boot prompt and rebooting sends a rediculous amount of info out of serial port A. It is very useful. After seeing that, working with PCs is painful, simply painful.

Re:BIOS and its definition

[mcSey921]

BIOS -- Basic Intergrated Operating System?

Well it's kind of needed now isn't it?

[Telastyn]

Now that computers aren't coming with floppy drives, there better be something that users can do crash recovery from if the machines won't boot from cd....

Personally I'd like the choice of recovery cds/floppies.

Re:Well it's kind of needed now isn't it?

[Iguanaphobic]

BIOS support to boot from USB. Then you could use the USB pendrive as a floppy disk replacement.

Re:Well it's kind of needed now isn't it?

[Flossymike]

Is it that new? How about the linux bios? www.acl.lanl.gov/linuxbios

I think I would agree with what seems to be the majority view here ... more complexity leads to more potential areas of failure, however the idea of having a relatively low level OS that you could boot into to examine hardware / obtain drivers could be very useful. Sounds more like a good option, but not for everyday work.

A bit like etBIOS then

[Bubble King]

A bit like the offerngs here?

Wow...

[cybermace5]

It's something we've all wanted: a way to run your computer when the OS is hosed. Happened to me last night, actually. Booting off a floppy or CD-ROM is slow when you're in the tweak-reboot-tweak-reboot phase.

Also, it would be great to be able to get those drivers and updates, or even HOWTO's while fixing the computer.

Now, what I'd like to see added into this: The ability to instantly switch into the BIOS system. Then you could bring up a HOWTO, switch back and forth between your BIOS and struggling OS.

Re:Wow...

[blackmonday]

Problem is you're booting off a chip, not the hard drive, so where would you save your drivers, assuming you could download them? Maybe a firewire or USB drive, or the network? Sounds complicated. Just use another computer.

Re:Wow...

[cybermace5]

How much is a 128MB USB keychain drive these days? CompactFlash?

How much do you think it would cost to put 128MB of flash directly on the motherboard (or simply provide a CompactFlash slot next to the BIOS)?

Re:Wow...

[SlightlyMadman]

How much do you think it would cost to put 128MB of flash directly on the motherboard (or simply provide a CompactFlash slot next to the BIOS)?

You can already add one yourself, for about $25. It works just like any other bootable IDE drive.

http://store.ituner.com/ituner/emstcfl.html

Re:Wow...

[cybermace5]

That's pretty good.

I was just looking at DiskOnChip prices for a single-board computer I'm using, and it wasn't pretty.

Re:Wow...

[blackmonday]

But all these things need drivers, software, etc. Isn't plugging in another hard drive just as easy?

Re:Wow...

[fishbowl]

>Problem is you're booting off a chip, not the
>hard drive, so where would you save your drivers

You're not thinking "next gen arch".

You neglect the NVRAM which is ALSO on your new chip. Along with the hard crypto that identifies YOU and lets the OS Vendor control when and how you install the OS. I cringe at the thought of who "OS Vendor" is, but I would like to see the effect it would have on the market, if everyone were actually required to pay retail for the popular OS; I mean, if everybody who used it actually did pay. I wonder if "elminating piracy" would actually work against the industry by lowering the volume? I wonder if, instead of hypocrisy, we would start seeing real momentum toward alternative computing solutions? I really do sometimes wish that everyone who runs Windows had to pay for it, or it plain didn't work... No in-between, no legal gray areas. I'd feel much better about the price I paid then. As it stands, the price does not accurately reflect the market. (I have similar views regarding pay-TV. If I didn't *know* that so many were getting it "free", I might not care so much about the price. As it happens, I don't believe the price represents anything, therefore I can't buy it.)

Re:Wow...

[JohnFluxx]

That $25 for the adapter. It's another £30 for the memory ($50 or so I think that is)

Re:Wow...

[cybermace5]

*confused look*

Well, as the actual BIOS of the computer, you'd expect it to have most devices well under control, right? About the only thing you'd have to worry about is a non-integrated network card or modem.

Microsoft.

[$$$$$exyGal]

... its possible to imagine people in a hurry booting up their Windows PC and accessing the Net using cME without touching Windows.

So Windows crashes, and you can't get it to come back up. No problem! You just boot up into your BIOS, send the built-in web-browser to support.microsoft.com, and then your set. Except one problem: Microsoft decides (accidentally) to send your BIOS browser really unusable HTML. The end result is you can't diagnose your Microsoft problem because you aren't using a certified Microsoft product.

--sex

Re:Microsoft.

[Anonymous Coward]

The end result is you can't diagnose your Microsoft problem because you aren't using a certified Microsoft product.

No, but it probably works a lot better when you
try to go to your hardware vendor's site.
(i.e. the guys who actually put the browser there).

Now whether or not your vendor is giving you adequate support for your windows woes is another thing..

Of course, imagine the incentive for Microsoft to make their website work with this browser when somone like Dell comes and asks them to because Dell has a ton of customers complaining that the browser is useless on MS site.

Re:Microsoft.

[Anonvmous Coward]

" The end result is you can't diagnose your Microsoft problem because you aren't using a certified Microsoft product. "

That's interesting? What, did everybody forget that in that unlikely scenario, there's plenty of places around the web (like Google Groups...) to get support for Windows or any other MS product ever made?

That's like modding me up for saying "Well, the McDonald's next door is closed. I can't eat."

Re:Microsoft.

[The Bungi]

But perhaps you can still log into Slashdot and engage in some karma whoring.

Re:Microsoft.

[t0ny]

Why on earth you need to go to support.microsoft.com when your junky packard-bell locks up is beyond me. Maybe you should go to your vendor's site. Plus, your computer probably wont boot because you messed up the settings.

And by the way, microsoft doesnt make motherboards. you are thinking of intel. just because MS maintains a hardware list doesnt mean that is the only stuff authorized to work with Windows, it just means it passed MS's testing. Doesnt Linux have to have specific computer parts to work as well? How about (hahaha) Apple?

Re:Microsoft.

[Pharmboy]

So Windows crashes, and you can't get it to come back up. No problem! You just boot up into your BIOS, send the built-in web-browser to support.microsoft.com, and then your set.

Dammit! And I'm using the new BORK edition BIOS....

Re:Microsoft.

[Anonymous Coward]

Well, obviously, Microsoft should supply the BIOS code as well ($for$a$fair$price$). Really I'm surprised they don't already. Surely Windows PCs could be more consistent and reliable if they did?

I'm being facetious, but why *hasn't* Microsoft attempted to control PC BIOS firmware? It would have seemed a natural thing for them to do, given their propensities. Is it just fear of antitrust legislation?

Re:Microsoft.

[Anonymous Coward]

you again!

Re:Microsoft.

[Anonymous Coward]

Actually, controlling the BIOS, like apple does, would remove all fears of a monopoly. They are only a monopoly on PC hardware because they do not control the hardware. Apple and Sun machines were removed from the monopoly equation because they were single purpose (Single OS) machines.

Apple is a complete monopoly on their hardware but no one complains when they shut down cloners or software developers for their platform... Because it's *their* platform, hardware and software. Same with Sun.

I suspect that MS will in fact within 5 years be licensing their own BIOS for PC hardware essentially making the platform and the OS a single unit and at that point they can do whatever the hell they want to with it.

You're right though, it's suprising they didn't do it 5 years ago.

Re:Microsoft.

[CerealQiller]

Oh Joy... Just when I thought that a system level vulnerabilty was as bad as it can get. If Microsoft made automobiles, they would be made in such a way that someone could steal your tires while the car was moving. If you crashed, which could happen at 10 mph on the Bonneville Salt Flats, none of the doors would work and your windows would all turn blue. On the freeway, someone could suddenly appear in your passenger seat, remove your stearing wheel and empty your glove box of Lic and reg, checkbook and spare keys and you might never even see them do it. I now quite enjoy my beta model Linux, though it took some getting used to. My car never crashes any more, though occasionaly the radio disappears, or one of the mirrors falls off. At least I don't have to reinstall the engine because someone sent a mangled email to my rear view mirror.

Terrific! (spit)

[Anonymous Coward]

More crap on _our_ disks that we have no control over. Yet another thing I have to avoid when buying a new computer.

Re:Terrific! (spit)

[kfg]

Ah, but they fooled you this time. It isn't on your disk, it's on your motherboard.

Surprise!

KFG

Easy to tell the difference between the two

[Anonymous Coward]

An OS crashes..

whereas you'd hope that a BIOS doesn't..

Tack on too many functions and what's the diff?

[cryptochrome]

It won't boot up too fast if it's weighed down by too much stuff.

Clearly, it should stick only to vital system functions (especially repair, perhaps online) and perhaps user functions (email, calendar, [help] browsing), and leave the rest to the real OS.

Re:Tack on too many functions and what's the diff?

[Reziac]

It's trying to be the WinXP of BIOSs... a BIOS, and a stripped-down version of every app or util that can be stuffed in sideways, all of which add potential to, um, "interact" in unpredictable ways.

I've always disliked Phoenix BIOSs, way back well before they ate Award... buggy, weird limits, poor feature set, general lack of user options ... and they're doing nothing to improve my opinion.

In fact when I motherboard-shop for a serious system, one of my primary criteria has become "Does it have an AMI BIOS?" If not, I move along.

And the world over....

[Rooked_One]

Windows users rejoice!!!

Re:Bad

[Anonymous Coward]

But I thought BIOS stood for "Built-In Operating System". And everyone knows the web browser is a fundamental component of any OS, built-in or otherwise. I'd almost go so far as to say the two should be tied together.

Gee...copying EFI, huh?

[devaldez]

A BIOS vendor is copying the EFI Specification ...too bad it is several years overdue...

While I won't suggest Intel has it right per se...it makes sense as a method by which a BIOS can extend functionality...

Big stuff filtering down..

[Richard_at_work]

As usual, this just seems to be a case of stuff origionally designed for big expensive systems coming down to lower levels. IBM servers tend to have this functionality, hell on our F50 at work you can dial in using a modem when the AIX OS isnt running, the bios/firmware will take care of you. The AlphaPC164 i was jsut given suprised me by having a almost full unixlike OS as its firmware (SRM).

This has happened with SCSI, raid, SMP etc so it doesnt suprise me to find a BIOS that does more than normal, and in many cases it is a bonus, depending on wether it does certain things. I use serial consoles a lot, and would love to have a better way to talk to the computer at a really low level, without resorting to expensive hardware.

Re:Big stuff filtering down..

[meme_police]

Precisely, previous posts on this subject lead me to believe Slashdotters have never heard of OpenFirmware or similar. OpenFirmware and OpenBoot are big reasons I stick with G4s and Sun Blades for home use.

Re:Big stuff filtering down..

[AsbestosRush]

I was going to mention this, but it appears that someone has already beaten me to it.

I must say, that I do miss OpenBoot when I'm not using a piece of Sun hardware. Makes a lot of things a lot easier, esp in initial build and disaster recovery.

Re:Big stuff filtering down..

[spinlocked]

The AlphaPC164 i was jsut given suprised me by having a almost full unixlike OS as its firmware (SRM).

This is what I love about Sun hardware. You have a complete FORTH interpreter in the OpenBoot PROM, you can 'cd /' and 'ls' to get a list of devices, you can 'test net' or 'watch-net' to diagnose network issues, 'probe scsi' to find SCSI devices, etc. etc. A few months ago I wrote a little bit of FORTH which would make the chassis LED blink, so I could find a particular box in a rack full of kit. Marvellous. The lights-out managment features on the newer boxes allow you get to the PROM through the console line when the machine is switched off - well, enough of the PROM to let you power up the machine remotely - this was fantastic for one datacentre that I've worked in, you had to give security 24 hours notice if you needed entry to the facility.

Re:Big stuff filtering down..

[glenstar]

While I love my Sparcs, and think OpenBoot is really neat, I recently did an upgrade to Solaris9 on an old SS5. I went out and bought a SCSI CDROM to install from, popped open the case, installed the drive, shut case and rebooted. Nothing. After much, much, *much* digging, it turned out that the CDROM was hardwired to SCSI ID 5 and the Sparc expected its CDROM to be at 3. Damn! So, I dropped into OpenBoot and tried to make it be. Anyway, to make a long story short, I got it to work after much swearing and smacking of the poor Sparc and learning that the name of my new CDROM was actually: sbus@1,f8400000/esp@4,40000/sd@5,0:a (or whatever it was).

I mean, *that* is friendly! ;-)

Re:Big stuff filtering down..

[AsbestosRush]

Yah, that is the one thing that I never really "got" about the Sun hardware. Try having 15 of those on one box. Made my head spin when I started messing with our E450 that I had loaded out.

Too bad that box didn't make it home with me when they laid me off. Something tells me that a piece of equipment that large would have been missed, tho :D. That and I'm too darned honest. Even returned the bag of network cables that I borrowed for lan parties.

Snow Crash

[Anonymous Coward]

Looks like Neal Stephenson was right after all when he made the mistake of saying "BIOS" meant "Built-In Operating System" in Snow Crash.

It's not a BIOS chip.

[FreeLinux]

It is software that is stored on a "protected" portion of the hard drive.

Basically, it is like the system partition on Compaq and Dell systems but, it will have some additional tools that they don't have. eg. a browser.

Re:It's not a BIOS chip.

[John Hasler]

> It is software that is stored on a "protected"
> portion of the hard drive.

A _really_, _really_ bad idea.

Re:It's not a BIOS chip.

[The Troll Catcher]

Indeed it is - can you say "BIOS virus", anyone?

Can't wait till that starts happening...

Re:It's not a BIOS chip.

[TheMidget]

A _really_, _really_ bad idea.

Especially if you use that computer to do your taxes...

Re:It's not a BIOS chip.

[unitron]

Especially since one of the first things you do when troubleshooting a motherboard is disconnect every drive except maybe the floppy.

Re:It's not a BIOS chip.

[Hast]

CIH was around in '98, so the basic concept has been round for quite some time already.

Re:It's not a BIOS chip.

[The Troll Catcher]

Indeed it was - but this seems a bit easier to pull off than CIH, no? All you need to do is write to the disk, not re-flash a BIOS chip.

Next step, OS in ROM

[Anonymous Coward]

Wouldn't it be innovative? Imagine it, the full OS loaded from a chip, just power on and you get the OS in a couple of secs. Woah! Trully new idea!

Wait... all those 8 and 16 bit computers...

Re:Next step, OS in ROM

[Anonymous Coward]

Actually, this was done back in MS/DR-DOS 5 with external ROM cards.

Re:Next step, OS in ROM

[Anonymous Coward]

Not worth it, it's easier and cheaper to load it from disk.

Re:Next step, OS in ROM

[Anonymous Coward]

I had an ADAM computer that booted into a "typewriter" mode, but if you booted from a floppy that had "SmartBasic" on it, then you had an OS. I had the 40-column SmartBasic, with other enhancements, and that was a good enough Basic to do quite a lot of programming. I subscribed to some ADAM newsletters, and got ahold of some nice programs, that I could later modify or use parts thereof to do what I needed done. Very nice setup, quite unlike the Commodore 64 that had that OS built in.

hard disk?

[Anonymous Coward]

"Resides in a protected area of the harddisk".
I think we all know what this means! Track 0 anyone? This could be interesting for TurboTax, and all the other horribly crippled applications forced on consumers nowadays.

Re:hard disk?

[Mancomb Seepgood]

Wrong (that wouldn't exactly be a protected area, now would it?) .. try Host Protected Area

M.S.

Microsoft BIOS XP(tm)

[Anonymous Coward]

Will Microsoft stand by and let some lowly BIOS vendor threaten their control of the desktop from below?

I predict we will soon see official "Microsoft WinBIOS XP" as standard equipment on all OEM PCs that ship preloaded with Windows. They will probably "encourage" the OEMs to include this BIOS in favor of Phoenix or any other versions. Of course there will be "accidental" glitches caused by Microsoft security "enhancements" that render such PCs incapable of booting Linux or any other OS. They will be able to easily sell this to the public under the guise of increased security from dangerous criminal hackers and some bogus benefits to gaming or media playing. You heard it here first.

file system access

[binaryDigit]

Usually the two most critical items needed to help a problem system is file system access and then some basic editing tool. If this bios can come with

1 - your choice of file system driver (ntfs, ufs, whateverfs)
2 - a raw sector editor
3 - a simple text file editor

That would be a godsend. A tcp/ip stack with telnet/ftp would also be very useful, but I could live without that.

Re:file system access

[Amazing Quantum Man]

Maybe they'll get Microsoft to fully document NTFS!

Re:file system access

[Minna Kirai]

Not very useful, I don't think.

Most people who have the expertise needed to make meaningful changes to system config at the sector level (or even the /etc level) will already have another method of writing to the hard disk.

The fix-it guy could slam the drive into a computer dedicated for that task, or insert a boot CD to take precedence over the hard disk. In either case, the repairer then has access to a full suite of personalized software and online documentation to help him do the repairs right. Much more powerful than anything the BIOS could provide.

Re:file system access

[binaryDigit]

Most people who have the expertise needed to make meaningful changes to system config at the sector level (or even the /etc level) will already have another method of writing to the hard disk. The fix-it guy could slam the drive into a computer dedicated for that task, or insert a boot CD

Yes, in an ideal world all these would be true. However there are always times when a) you don't have a bootable cd (at least one that allows you to do anything other than install/etc) b) you don't have another machine that you can take down to plop the offending hd in. The latter can often happen if you're offsite and the former if someone "borrows" your bootable cd, etc. The whole point of this is to allow you to get something done in a worst case scenerio. All of the above has happened to me, and I'm not even an IT guy (though because of my knowledge I get placed into situations where I have to do IT stuff). Being able to tweak something without having to find/gen boot disks or tear apart another machine (that you hope you have, fine for a pc, not so fine if it's something that doesn't lend itself to having a duplicate of, say a large RS/6000).

I'm not talking about power, I don't want a full suite of stuff. I just want something that I can rely on to get the often times simple tweaks done.

floppy replacement

[g4dget]

I think this sure beats BIOS upgrades and OS installs via floppies--if they do it right. Imagine being able to just point the BIOS at an IP address in order to download the initial disk image via HTTP, or being able to use the BIOS to partition the disk and copy an initial disk image from a USB drive. Right now, trying to do the equivalent by, say, booting a Linux rescue disk can be hard--most Linux rescue disks still don't know how to deal with USB devices, and network boots are a pain to set up and don't (generally) work through firewalls or HTTP. This is particularly nice for lightweight portable devices that may not have much in the way of drives or interfaces, but they will have networking and possibly USB.

If they do it wrong, however, it might be a nightmare of DRM, spyware, and commercial apps sitting in weird disk partitions. That, we definitely don't need. I don't want my machine reporting to Phoenix every time I boot, for example.

I hope, however, that Phoenix will be cut out of the loop. Something like the Linux BIOS or OpenFirmware make a whole lot more sense to me as the basis for this.

Re:floppy replacement

[raynet]

Hehe, that would be great.. i can see it already:

boot from: http,cdrom,a,c

Re:floppy replacement

[Anonymous Coward]

You can already do OS installs with PXE (Preboot Execution Environment) -enabled NICs, which are pretty common these days. I use PXE to install Linux on all sorts of machines from various hardware vendors, thanks to pxelinux from H. Peter Anvin, part of the syslinux package. Since most distros' install disks are based on syslinux, substituting pxelinux is a cinch.

There are also proprietary software packages that will let you install OS images using PXE, but that's harder to maintain in the long run than a routine that uses the OS installer and then makes some minor, easy-to-track modifications, IMHO.

With a little help from a DHCP and TFTP server, my systems build themselves, do some custom reconfiguration, and start offering the appropriate services right after I stick them in the rack.

Re:floppy replacement

[Anonymous Coward]

Sure, there are lots of ways of doing OS installs via NICs. But PXE and TFTP require a lot of setup and aren't good protocols to use over the Internet. Some combination of DHCP and HTTP just seems better.

What we really want is being able to type, say, "install.debian.org" into the BIOS and have the rest happen automatically.

Re:floppy replacement

[Flick]

Interesting...

http://www.eetimes.com/sys/news/OEG20030218S0039

Re:Bad

[Anonymous Coward]

is this a troll? BIOS means basic input/output system...

Re:Bad

[jonnythan]

Basic Input-Output System.

I assume you were kidding, but you never know ;)

Re:Bad

[Stigmata669]

BIOS stands for Basic Input Output System, and as for the idea of enahancing the BIOS kindof defeats the purpose of having a bare bones way of controling the hardware, however if it is implemented well, it could provide a good intermediate level between the BIOS and a full OS for low level diagnostics, but i don't see a need for web browsing.

my 2c

Remember the Commodore 64 & Amiga!

[MagikSlinger]

I remember when all OS's came on ROM. :-)

Re:Remember the Commodore 64 & Amiga!

[Anonymous Coward]

I do too. I can also remember that there were few bugs too. I was always amazed at their ability to write so much code with so few problems (at release time).

Re:Why don't we just get rid of the OS?

[Anonymous Coward]

As long as the BIOS can boot up emacs, who needs an OS. Let's cut out the middle man.

Re:Bad

[Anonymous Coward]

No, BIOS stands for "Basic Input / Output System"

Re:Bad

[Milican]

Ahh very sly Daniel-san. Much like how DVD changes from Digital Video Disk to Digital Versatile Disk.. if the acronym don't fit.. change it!

JOhn

OS on a floppy

[mrklin]

Are there already a # of OS that fits on a 3.5" 1.4MB floppy disc? http://www.woalf.uklinux.net/ Also, as soon as more MB manufacturers supporting booting from USB devices (thumb drive, for example, which comes in all variety of sizes even at 1GB+)m would this features be pretty useless?

Re:Bad

[azmaveth]

I agree.

The whole point of a "Basic Input/Output System" is for, well, basic I/O. It was meant to be a thin layer between the OS and the hardware.

While I agree that the technology can be improved upon, I don't think this is the proper direction to take.

I much prefer the route that Gigabyte has taken with their DualBIOS. If there is serious enough trouble with the OS, just boot to a CD with recovery tools on it. If there's a problem in the BIOS, you now have a spare. I don't see the necessity of a TCP/IP stack in the BIOS.

My $0.02.
-azmaveth

BIOS improvements?

[Dark Lord Seth]

This info might be rather outdated, considering I do not have any access to more then one modern BIOS. However, when are these things going to become a bit more standard? It seems like every BIOS is vastyl different from any other BIOS. I mean, if you take away the most logical stuff such a different vendors and motherboard capabilities, there are no real standards. Some BIOSes allow you to select where to boot from by giving you a limited set of capabilities designated by standard C:, D:, etcetera, while other BIOSes (even made by the same vendor) give you a much wider choise or even better, allow you to manually select a boot order using devices, not logical station name. (boot from Primary IDE, Master... Like that.) Even worse, some mobos I had didn't even allow slightly more exotic choices such as SCSI/IDE controllers, RAID (SCSI/IDE) controllers and network booting...

I'd like the possibility of having more then three choices and to be able to delect from which device I boot, listed by IDE connection. Also, which is mighty handy, some sort of boot menu which can be called during boot, allowing you to select where to boot from. My current computer has it, if I want to boot from CD, I just have to select it... Very handy.

Boot-ROM would be better

[Mr_Tulip]

I think a basic shell with ftp would be great on a mobo, but only if it resides on a boot-rom, and can be 'switched on' from the bios.

Pheonix are going to use your HDD to store their shite, so it doesn't really solve any problems if your HDD is pushing up daisies.

Alpha SRM console

[bstadil]

Old news, it is only the PC world that is taking another belated clue from the high end.

This nifty thing has been available on the Alpha machines for 10 years or so.

Re:Alpha SRM console

[TheRaven64]

Old news, it is only the PC world that is taking another belated clue from the high end.
Are you allowed to say that kind of thing on /.? I thought that any hint that x86 wasn't the One True Architecture, and everything else was either obselete was heresy (well, I've been flamed / modded down every time I've suggested such a thing).

Re:Alpha SRM console

[Anonymous Coward]

People are probably jealous because they are all too miserly to save up for a decent box.

Re:Alpha SRM console

[Richard_at_work]

Heh, i dont know about others on slashdot, but recently ive found myself coming to the limits of what is "fun" with the x86 arch, and have been looking for other architectures to play with. Thats the reason i currently have 2 RS/600s, a Alpha, numberous suns, SGI systems etc etc. Its fun to see jsut how different systems are. I have a apple airport coming, and am going to have a bash sticking linux on it.

Re:Alpha SRM console

[bstadil]

Heh, i dont know about others on slashdot, but recently ive found myself coming to the limits of what is "fun" with the x86 arch, and have been looking for other architectures to play with

Same here. I bought a 500Mhz Alpha a few month ago just to play with. That is why I ran into SRM. When I first saw it I thought how clever and wondered why nobody had thought about putting this on a standard pc/ x86 machine.

Coming from a windows world a few years ago I had the same feeling when I tried Linux / Unix the first time.

There must be a lot of other stuff out there that could usefully find its way to "normal" workstations.

Re:Alpha SRM console

[drsmithy]

Same here. I bought a 500Mhz Alpha a few month ago just to play with. That is why I ran into SRM. When I first saw it I thought how clever and wondered why nobody had thought about putting this on a standard pc/ x86 machine.

They most likely have, and dismissed it as economic foolishness.

See, the PC plaform was evolved, not designed. And, as such, backwards compatibility is simulatenously the most important aspect of the platform, and also the most difficult to sustain, with kludge upon kludge being bolted on just so DOS 3.3 will still boot on the latest and greatest 3GHz P4.

I'd love to see someone like Dell make a PC with Open Firmware instead of the standard BIOS. Trouble is, it'd need to have special OS distributions just so they could boot on the platform.

Then, of course, you'd have to deal with the Luddites. The same people who complain very loudly whenever anyone talks about removing old, worthless-to-most-people legacy crap (like floppy disks, PS/2 ports, serial ports, parallel ports, ISA slots, etc etc) that has been completely replaced by better things and is only really needed so - you guessed it - DOS will still boot on their 3GHz P4. Judging by the reaction here to the previous article about Dell starting to not include the floppy by default (amongst others), I'd imagine the average /. poster will be jumping up and down and screaming about potentially losing that ability.

There must be a lot of other stuff out there that could usefully find its way to "normal" workstations.

There is. Trouble is manufacturers have learnt from bitter experience that "better" does not necessarily translate into "more popular" - particularly when people are still insisting (and advising other customers to do the same) that they get serial ports and floppy drives.

Re:Alpha SRM console

[unitron]

"See, the PC plaform was evolved, not designed."

I realise that it's two different ways of saying the same thing, but, where the PC platform is concerned, "mutated" just seems to capture the spirit of things better than does "evolved".

Dual BIOS

[vondo]

Gigabyte's dual BIOS is quite nice. Especially so when they release an updated BIOS that crashes LILO.

Basically, in the BIOS you can select to run the backup version, which, if you're smart, is a backup of the last known good BIOS.

Re:Dual BIOS

[karlm]

They realy should have the backup version be non-flashable, preferably in a ROM. You'd want the backup BIOS to actually call the upgradable BIOS by defult. I've seen a machine hang during a BIOS update. It isn't pretty. There are also some viruses that attempt to overwrite the BIOS. It would be great to have an absolutely unbreakable BIOS. In the backup mode, you might have some bugs in obsure features and might lack the most modern features, but you most likely retain something that can boot at least a recovery floppy. More importantly, it'd be really hard to lose the ability to flash the flashable version of the BIOS.

Basic

[phorm]

BIOS=Basic Input Output System
If you tag a bunch of O/S stuff in there, suddenly you're getting to be a lot less basic... so maybe it would be an IIOS (Integrated Input Output System) or AIOS (Advanced Input Output System).

However: There were days when machines had no hard-drive and came with 2 floppy drives and a build-in GWBasic (we had these in Elementary). If we're talking very basic OS, with maybe the option to mount into a greater OS, it wouldn't be bad. Integrating a complete and functional OS into future BIOS is not a good idea. But really, an OS comes on a hard-drive... if you fry something, you can format the drive, or replace it. Frying a BIOS OS would be a lot worse.

Oh dear. Is it terminal?

[Zaiff Urgulbunger]

Excellent! I've so been waiting for a simple web terminal. Years ago I was supporting a system with users on green screen terminals. Then those terminals got replaced with Windows(TM) PCs 'cos someone thought they needed them so they could access the internet. Or something.

Anyway, Windows PC's are a nightmare to support. Terminals are easy. So for corporate use, I think a web terminal would be great (I did think of suggesting using a DreamCast, but then thought better of it!).

Simple. And plus, it means they should be cheaper because there's no MS tax. So thats nice too.

Re:Oh dear. Is it terminal?

[Anonymous Coward]

Look up the Linux Terminal Server Project! (Google for it!) Any PXE based (lots of later model Intel and Dell stuff) system can remote boot without even a seperate rom. There are several sites promoting general network boot linux using just a floppy disk to store the network driver.
There are lots of options too! From remotely booting a full system to a ramdisk or strictly a terminal only thing.
I've been playing with Knoppix on my work network and have just about got it working with remote boot but I think there is another Windows DHCP server hiding out somewhere. (I only spend an hour or so a night) Best part is Knoppix dosen't cost you any changes to your PCs hard drives. When you remove the disk, your regular install is there.
I like the idea of automatically booting to some type of shell with basic internet for just this purpose. Imagine flipping the switch, loging in, and getting a full Gentoo install with a completely bare system!

Re:Oh dear. Is it terminal?

[Animats]

If you want a pure browser terminal, you can try the free version of QNX, and bring it up in diskless mode. That's enough to run a browser. It's the same system that ran the i-Opener, but not as locked down. You can run Mozilla or the lightweight Voyager browser. Voyager has much better response that Mozilla, but hasn't kept up with what Microsoft has done to HTML.

Before Microsoft wrecked web standards, this was a viable option. There ought to be a stateless i-Opener like device in every hotel room in the developed world.

Re:Oh dear. Is it terminal?

[Zaiff Urgulbunger]

I like the idea of automatically booting to some type of shell with basic internet for just this purpose. Imagine flipping the switch, loging in, and getting a full Gentoo install with a completely bare system!

Coolio!

Re:Oh dear. Is it terminal?

[Zaiff Urgulbunger]

There ought to be a stateless i-Opener like device in every hotel room in the developed world.

Agreed. Not just for hotels though - I think the general public of any country should expect to have access to information via public access terminals. And it makes sense that everytihng about these terminals should be open, including, I guess, the hardware design!

Palladium

[t0ny]

OH MY GOD!!!!! Pheonix is looking to monopolize the content of all our computers!!!

Somebody has to stop this!! Where is Steve Jobs? Where is Linus Torvalds? They need to GPL our freedom to open-source! Somebody hand me a GNU!!

Re:Palladium

[Anonymous Coward]

"Somebody hand me a GNU!!"

Will this do?

Macs had this on the Mac Classic

[ad0gg]

Boot hitting Open Apple control and some other buttons would cause the computer to load up off a system stored in the rom. It could read the drives on the scsi chain, disks and even had Apple share support to connect to an appletalk network.

Re:Bad

[Pharmboy]

BIOS stands for Basic Input Output System, and as for the idea of enahancing the BIOS kindof defeats the purpose of having a bare bones way of controling the hardware, however if it is implemented well, it could provide a good intermediate level between the BIOS and a full OS for low level diagnostics, but i don't see a need for web browsing.

Admittedly, I'm over my head here, but can't you have a complex BIOS that gets out of the way when the OS boots, or acts as a mini OS when the real OS wont load?

I mean most support for computers is online now, and its kinda hard to log in to "dell.com" if the damn thing won't boot.

Why can't the bios be both? for instance: IBM used to have a BASIC interpretor in bios (286 and pre), but it didnt get used unless the system didn't find an OS. It didn't get in the way.

Or

[sulli]

"Visual Basic Input Output System"

In other news

[worst_name_ever]

In other news, Microsoft today announced the availability of Bios XP Service Pack 3, available as a 900MB download from www.microsoft.com or on two handy CD's for only $19.95 plus shipping and handling.

IIS on Linux: The proof.

[Marcus Brody]

Hi just had to report my latest findings to the slashdot community. Looks like there is now a port of Microsoft-IIS to Linux, as run by goatse

http://uptime.netcraft.com/up/graph?site=www.goats e.cx

Blurs the lines somewhat: Quite the opposite of running a real server without having to boot into a fully fledged OS.

Re:IIS on Linux: The proof.

[Anonymous Coward]

The site www.goatse.cx is running Microsoft-IIS/5.0 on Linux.

Wow. The troll speaks the truth.

Back in the day

[kfg]

we used to buy the magazines, go to the stores and log on to the net to be eagerly greeted with hot new tech that made us want to immediately scrap the $10k worth of goodies we'd just bought the week before.

Now it seems that every day brings "innovations" that seem designed to further my intent to hang on to what I've got forever if I can manage it.

What's wrong with this picture?

KFG

Re:Back in the day

[buysse]

You, sir (or madam), are getting old. It is a natural process, please do not be afraid and keep your hands in the car at all times.

Enjoy the ride.

Re:Back in the day

[kfg]

I've been getting old all of my life. I'm actually used to it. Since what I've always wanted to be when I grew up is a grumpy old man I'm certainly not afraid, and can't wait until I get there.

I'm afraid, sir, that it's quite impossible to keep your hands in the car at ANY time if you're driving a Lotus 7, if you wish to maintain anything vaguely resembling control.

The ride, however, is quite enjoyable.

KFG

Re:Back in the day

[unitron]

It has a lot less to do with getting old than it does with a desire to avoid getting hosed by new stuff that's worse than the old stuff and designed more to extend the manufacturer's control over you than your control over the equipment.

Re:Back in the day

[Anonymous Coward]

Getting old, as in 'more experienced'. Yeah, probably.

This is a bad idea!

[linuxghoul]

When i am doing a fresh install of WIN2K and REDHAT on my comps (a once-in-6-months exercise) i make sure the machine is not even physically hooked up to the net untill i have a software firewall configured, up and running. I keep the firewall packages on CDs, along with config files, and ONLY AFTER these are setup, i go online to install other things, update drivers, etc (usually, only for win2k, not for redhat, which is trivial to bring back to my customized setup)...

having TCP/IP built into the bios, with no firewalling support, and no possibility of frequent/safe upating, no easy way to check for "being" owned is a very bad idea. Also, Phoenix being a popular bios manufacturer, there will be a lot of worms targetting this bios tcp/ip stack.

I dont see a single genuine advantage of having all this crap in the BIOS anyway. I mean, if u hose ur drive, and need to go online for some critical information/software before u can bring ur comp back up, just keep a KNOPPIX cd handy. I personally think BIOS shud be thinning down even further, given none of the modern OSes really use most of the services, and the BIOS mostly just gets in their way. All the bios shud be capable of, is to bring up the OS, and then let the OS configure everything. It wud be so neat to have the OS kernel setup all the hardware, the powersaving policies, everything when it starts up. Of course, the best is to just have the OS kernel as the bios!! just throw this anachronism completely out. (yeah for ppl whos fav os is not linux, sumthing else might need to be worked out ;) ...

Ghoul

Re:This is a bad idea!

[Anonymous Coward]

don't worry, linux bios can boot linux, windows, plan9 and sometimes *BSD :)

it has plenty enough for everyone...

Re:This is a bad idea!

[nuggetman]

Because as we all know, there are millions of people just waiting for you to do that fresh install so they can compromise your data (which before installing your patches would consist of system files).

Re:This is a bad idea!

[tunah]

When i am doing a fresh install of WIN2K and REDHAT on my comps (a once-in-6-months exercise) i make sure the machine is not even physically hooked up to the net untill i have a software firewall configured, up and running.

having TCP/IP built into the bios, ... is a very bad idea

Well, if you're "not even physically hooked up" while installing, then how exactly is someone meant to break in when you're fiddling with the bios? Not to mention that a software firewall is neccesary (excepting trojans, which won't install on the bios) if you have open ports, which the bios won't, or if there is a problem with the tcp/ip stack, which a software firewall won't help with.

BTW, why are you reinstalling OSes every 6 months? Windows 2000 is a lot better at not getting gummed up like 9x used to be, but perhaps you might want to reinstall anyway, but redhat? You DO know that you don't have to reinstall both at once, don't you?

Re:This is a bad idea!

[linuxghoul]

Well, i am talking about my desktop machine(s) here. Win2k is not bad at all, one of the few MS products i actually like, but as i end up downloading and playing around with a lot of software, i just find it a good idea to start afresh every once in a while. As far as reinstalling redhat is concerned, that is done to keep up with the latest releases. I know, not required, just something i have always done in my 8 years of linux use.

My point about the tcp/ip stack in bios and the firewall had to do with the fact that i don't even let my OS tcpip stack onto the net unprotected.

Add to that the fact that there would be no way to run a firewall over a BIOS tcp/ip stack (yes, they may include a firewall there as well but which then needs to be configured for my specific use) and the fact that the only point of having a tcpip stack is to get online, you will eventually have to hook ur comp with that (BIOS based) tcpip stack upto the net, and u see the problem.

Re:This is a bad idea!

[BigBadBri]

Relax, dude

This stuff *isn't* really in the BIOS - it's Phoenix selling the idea of having a LILO in BIOS so an OEM can install Phoenix's own software on the HDD (read: boot device).

This is no different in essence from hitting F10 on a (properly installed) Compaq server or WS - boot to different partition, run OS of choice.

It's a load of marketing flim-flam, designed to confuse the easily muddled.

HP/Compaq Servers and SUN

[EggMan2000]

HP/Compaq Servers and SUN have some similar things on their servers now. Although no web browser, and not technically BIOS. The Remote Insight Lights-out (RILo)boards provide great disaster recovery tools. Compaq An oldie but goodie is SUN Solaris PROM. Much more useful than BIOS.

I know people clammor for the good ole days of BIOS and CMOS, but now we just need to accept that these new products are going to be better.

How is this better than a "repair" partition?

[ColGraff]

Personally, I'm happy enough just using a bootable disk to fix WinXP (yeah, I know), if it gets hosed. It actually doesn't happen very often anyway. But if for some reason the OS was hosed and using bootable media wasn't an option (perhaps security concerns, for example), how would this be better than just have LILO and a dedicated "repair" partition with linux and a bunch of repair tools on it? This way, as another poster commented, you aren't introducing unneccesary complexity into BIOS. and in fact, this seems very similar to what Phoenix is doing - the ZDNET article mentioned that the bios would have special hard drive partitions with recovery apps at it's disposal. So it seems like the only thing we're doing here is giving BIOS the ability to operate apps on a recovery partition directly, instead of using LILO+linux. Unless your boot record is screwed as well, what's the advantage to this? And if your hard drive is that trashed, doesn't it make more sense to stick the drive in a machine that also has a working HD, copy what you can, and reformat? For that matter, if your hard drive is getting pooched so often that you NEED dedicated bios support, then you should replace it. At least,that's my opinion. IAACTAAHBMMIPS (I Am A Computer Technician As A Hobby But My Major Is Political Science)

Re:How is this better than a "repair" partition?

[DakotaSandstone]

> how would this be better than just have LILO and a dedicated "repair" partition with linux and a bunch of repair tools on it?

"What, grandma, your AOL isn't working? Ok, do this. Launch LILO and go into your alternate Linux boot. Ok, then..."

Phoenix is not targeting the /. crowd. They're targeting OEMs who want to reduce customer RMAs by giving them a "known good" OS installation they can self-repair their systems from and such.

Also, the ATA "host protected area" is 100% protected from an OS gone awry. What if your hosed OS decided to start writing into other partitions, including your repair partition?

All right!

[Anonymous Coward]

So when are they going to start getting hacked, ads, viruses, and spyware?

Re:Bad

[Anonymous Coward]

The parent post was a joke, of course (did no one else notice the bit about "tying" browser and OS?). But you raise an interesting point. Is the BIOS acronym really descriptive anymore? It seems as if current PC OSes don't use it beyond the most basic boot, where DOS and CP/M actually used it for, well, I/O services to talk to the hardware. Isn't much of the BIOS 16-bit code anyway? I thought that (and bugginess/poor performance/Windowscentricity) was kind of why Linux and FreeBSD eschew the BIOS routines after the barest, earliest part of the boot cycle.
Really, isn't it Basic Bootstrap Services nowadays?

IEEE 1275

[pretoris]

I've always wondered why IEEE 1275 / OpenBIOS / OpenFirmware never caught on. IMO it is a much better and much more powerful alternative to the closed and aging BIOS found in most PC's. People are always complaining about "Closed" operating systems but don't bat an eye that their BIOS is closed...

Re:IEEE 1275

[Some Dumbass...]

People are always complaining about "Closed" operating systems but don't bat an eye that their BIOS is closed...

Probably because their favorite OSes don't use the BIOS at all after booting... and don't get rebooted often.

And if your BIOS is still too closed for your tastes, there's always the LinuxBIOS project.

Re:IEEE 1275

[toddhisattva]

OpenFirmware caught on with lots of platforms, like Sun and Apple, just not the one platform seemingly stuck forever in the early 80s ;-)

Re:IEEE 1275

[drsmithy]

(As I posted elsewhere...)

See, the PC plaform was evolved, not designed. And, as such, backwards compatibility is simulatenously the most important aspect of the platform, and also the most difficult to sustain, with kludge upon kludge being bolted on just so DOS 3.3 will still boot on the latest and greatest 3GHz P4.

I'd love to see someone like Dell make a PC with Open Firmware instead of the standard BIOS. Trouble is, it'd need to have special OS distributions just so they could boot on the platform.

Then, of course, you'd have to deal with the Luddites. The same people who complain very loudly whenever anyone talks about removing old, worthless-to-most-people legacy crap (like floppy disks, PS/2 ports, serial ports, parallel ports, ISA slots, etc etc) that has been completely replaced by better things and is only really needed so - you guessed it - DOS will still boot on their 3GHz P4. Judging by the reaction here to the previous article about Dell starting to not include the floppy by default (amongst others), I'd imagine the average /. poster will be jumping up and down and screaming about potentially losing that ability.

Re:IEEE 1275

[demon]

You _could_ implement a loader module that does a compatibility chainload for older operating systems. There's nothing stopping that. With the right code, an OpenFirmware for PCs could still boot DOS and Windows the traditional way, and also support more modern booting arrangements that forego all the real-mode crap.

Could this be how Palladium BIOSes will update?

[user32.ExitWindowsEx]

Could this be the first version of a 'Auto-Updater' for future Palladium-enabled BIOSes?

Halt and Catch Fire

[yerricde]

there isn't anything about BIOS programming that is proprietary or costly

What about talking to the motherboard's chipset? Many chipsets have settings that if accidentally triggered could make the motherboard HCF. Of course, the official BIOS is careful never to trigger those settings, but just randomly poking at the I/O registers could do Bad Things.

Re:Halt and Catch Fire

[Mattcelt]

True, I hadn't thought about that. How closely do the chipset manufacturers keep the specs? Do Intel and the others charge for access to chipset specs?

Re:Halt and Catch Fire

[Anonymous Coward]

The licenses needed to allow access to chipset specs and APIs are completely incompatible with any kind of open source development. Another issue is that, in order to have a BIOS available soon after the release of a motherboard, you'd have to sign NDAs to get advanced specs and demo hardware. An Open Source project obviously couldn't operate under the terms of an NDA.

I also think it needs to be noted that BIOSes have to be customized for individual motherboards, the part of the BIOS that is specific only to the chipset tends to be provided by the chipset manufacturer anyway.

Re:Halt and Catch Fire

[Anonymous Coward]

The licenses needed to allow access to chipset specs and APIs are completely incompatible with any kind of open source development.

Actually, Intel are very liberal with their chipset documents. I've just downloaded their PII4X chipset specs, and they have thousands more. You don't even need to sign up.

AMD have a sign-up process for OSS developers (Still waiting to hear back from them...) and VIA has a silly sign up process where it claims you can get access to documentation, but only for their IDE and Ethernet, and in reality is a link to some Linux patches.

So although AMD and VIA are much rah-rah'd here on Slashdot (Even by me), Intel are far better in this regard.

BeOS

[Bernie Fsckinner]

acronym for Basic embedded operating system

Maybe not...

[Anonymous Coward]

Sure you can use a OS on a floppy, but I don't think that they are a lot less easier to set up and get running then a BIOS with this stuff built in.

Taking linux example, unless there you have the offical drivers or your hardware is supported, it can be a time consuming task that these BIOS could handle quite easily and quickly.

BIOS is just a bad way to market it?

[DJStealth]

What if they called it an OS on a chip?

Make it separate from the BIOS; but possibly on the same EEPROM chip; it will only load up if you hold down the F12 button (or something) when you boot.

Re:BIOS is just a bad way to market it?

[Flossymike]

Would be very useful in the job I do, most of the day I determine whether a customers harware is knackered, or it's just Windows playing up ... again

Re:BIOS is just a bad way to market it?

[red_dragon]

Make it separate from the BIOS; but possibly on the same EEPROM chip...

It's been done. Some Tandy 1000-series PCs (like the 1000SL and the 1000RL) had DOS and DeskMate in ROM, and the BIOS would boot into that if it couldn't find a disk to boot from (or/also if the user pressed a particular key, IIRC). It appeared to DOS as an extra disk, usually D:. This extra disk, besides containing IBMBIO.COM, IBMDOS.COM, COMMAND.COM, and a few core DeskMate files, also contained basic DOS utilities like FDISK, FORMAT, and CHKDSK, in case the user needed them to fix up a borked disk.

Re:BIOS is just a bad way to market it?

[beerits]

The Macintosh Classic also had this feature. To boot up using the os burned into the rom you held command-option-x-o at startup.

NetBooting Macs

[Anonymous Coward]

That was the whole mac OS actually! System6 I think. Macs no longer have a complete operating system in firmware, but it is still more complex than PC bioses. The CLI is based on the FORTH programming language, and is used for mostly (in my experience) setting variables equivalent to the ones PC users set in their bioses. Apple has been a good cantidate to add this sort of functionality before, but I suspect they haven't for the abovementioned reasons. They've actually been going the other way! You can boot OSX over a network, does that imply that there is a full TCP/IP stack present???

the line is obvious

[Anonymous Coward]

It's between the "B" and the "OS"

thank you, you may laugh now!

aww fuck it. I need karma:

THIS IS TEH DRM AND PALADIUM SUXXX0RZ!!

there you go. Mod me up, bitches.

Re:the line is obvious

[BrainInAJar]

Don't you think if you needed karma, you *WOULDN'T* post AC?

*shrug* I dunno... just makes no sense to me to karmawhore when your posts don't affect your karma

a better sig for you

[Anonymous Coward]

Score: -1, Pretentious, whiny, and contrary for the sake of being contrary.
Karma: Crappy (see above)

question on the OS - BIOS relation

[Maimun]

After the BIOS hands the control of the
machine to the OS, to what extent is the
BIOS used, if at all? I mean, userspace
code cannot circumvent the OS -- if it tries,
the process gets killed by the OS. AFAIK, there
is no such a relation between the OS and the BIOS:
if the OS tries to circumvent the BIOS and talk
directly to some device, it does not get killed.

So, the BIOS is not a layer below the OS,
right? I am talking about real OS's, not DOS
or 'doze 95.

Re:question on the OS - BIOS relation

[stripes]

After the BIOS hands the control of the machine to the OS, to what extent is the BIOS used, if at all?

Some of the APM stuff is done by the BIOS in a way that is very hard for an OS to interfere with, or at least to alter other then preventing it. Some OSes use the "BIOS" (actually video card extensions to it) to change vid card modes. An OS could use it to do disk I/O, but then you can't do async I/O which is why no modern OS does it that way.

Sun uses their equivoent of the BIOS (OpenPROM I think) more, it actually does the console screen I/O (and use to suck because they read it out of 8 bit wide ROM with no caching, but that is over with now). Sun didn't make it palatable to do disk I/O or network I/O via OpenPROM though. "New World" Macs uses the same OpenPROM stuff Sun does, but I don't think they use any of it after the boot anyway.

if it tries, the process gets killed by the OS.

Not exactly true. There isn't a way to try in most cases. You can't try to alter another processes memory because you can't express that except via an OS construct, and that ones doesn't kill you if it doesn't work, just returns an error code. Sometimes the response is more severe then an error code -- a signal. But almost never is your process just killed...not in any normal Unix (I think OpenBSD will panic for some security breaches, like a second chroot in the same process hierachy...but it might just do the equivoent of kill -9).

a simple text file editor

[kentyman]

If this bios can come with...a simple text file editor

How many floppies would it take to reflash your BIOS with the newest GNU Emacs 83.2?

"Bootable Interactive Operating System"?

[Speare]

Just because BIOS currently stands for "Basic Input/Output System," doesn't mean it can't be coopted for some new meaning.

I'm glad to see PCs getting something similar to the tftp and other bootprom tools which good minicomputers had. Something that will let you build a machine from nothing, or fix or salvage data devices on a damaged system.

Simple?

[kentyman]

If this bios can come with...a simple text file editor

Not even Emacs comes with one of those. ;)

Xbox

[yerricde]

why *hasn't* Microsoft attempted to control PC BIOS firmware? It would have seemed a natural thing for them to do, given their propensities. Is it just fear of antitrust legislation?

No. Instead, it's because Microsoft is still working on it, using Xbox as a test platform.

Hmms...

[Chexsum]

Also, this may explain why the Phoenix browser was asked to change its name a few months ago.

So... I wonder when they are changing it and what it will be? :\

correct link

[kib]

just adding a correct link to the english page for KNOPPIX

am downloading that tool as we speak .. looking good so far :)

relation?

[minddog]

When we talk about the BIOS, we talk about firmware written for a piece of hardware to talk to software. Now OpenFirmware from Sun's labs had caught on, but didn't catch any waves. (IMHO)The pc market doesn't go for something thats going to make us be able to install a new version of our firmware and create adapters that can be easily programmed to interface with the that firmware, a plentiful abstraction layer. Apple, Sun, Compaq, and a few others have implemented it, but they aren't going to package this into a pc motherboard. It's enough to get 20 pcb pc motherboard manufacturers to follow the same chipset. If a company like NVidia or VIA adopt, or even our chip manufactures, bring the idea back to the drawing board, we might have a chance at seeing firmware upgrades in our hardware be able to increase performance out of those old boards. With an open standard, WE, US, The People, will be able to manage, extend, manipulate, integrate the firmware into things it has never been supposed to do before. Honestly, Solid state devices are the future, we're just not there yet, but we will get there. Anyone out there could be breaking out their breadboard right now building an implementation of OpenFirmware or alike. The first step to this must be a partnership formed by hardware and software manufacturers to develop such a standard. I don't think IEEE1275(which was denied) got anywhere because the community wasn't open enough at the time. Now that Free Software has made its dent into the economy, I think its time they jump on the band wagon. Thats my 2cents for today =)

Dead HDD

[bninja_penguin]

Phoenix's CME will reside in a protected area on a PC's hard drive.
I laugh as I read this. At this very moment, I have a system on my bench with an HDD that just let the magic smoke out of it's chips in a great gout of fiery stink!!! No chips, no spin, no way for their "protected area" of the HDD to do anything but stink.
I love the smell of burning chips....It smells like
PROFIT!!!

Re:Dead HDD

[TheMidget]

Never had any problems like this. But then, I usually do my tax declaration on paper...

the McDonald's next door is closed!

[taliver]

Now that's worthy of a PANIC if I've ever heard one.

Re:the McDonald's next door is closed!

[Anonymous Coward]

And /that's/ reason to kill one's self if I ever heard one!

Re:Bad

[Anonymous Coward]

But I thought BIOS stood for "Built-In Operating System".

Nope, it's actually "Boots Into Operating System."

(I know, I know...)

Host Protected Area Feature Set

[DakotaSandstone]

Starting with the ATA-5 spec (IDE disk drives conform to this), drives can implement a "Host Protected Area." This lets the BIOS OEM put executable code (or anything) on a hard drive, and then make it untouchable by an OS by physically reducing the size of the drive (and the OS cannot undo that size reduction command)

That's actually not a bad idea, and has at least some potential for Good, IMHO. With storage as cheap as $1 per gigabyte, an OEM moving their 650MB Recovery CD into this area seems like a great idea (not great if you want to swap in a new hard drive yourself, but they're not targeting power users)

However, Phoenix seems to be under the impression that just because they have this space, they should use it. A web browser and TCP/IP stack? Someone explain to me how this is good, and how the target "non power user" will be able to make sense of this technology without RMA'ing their computer if it gets that hosed.

Re:Bad

[Anonymous Coward]

It's not hard to keep the added complexity out of the boot path. The way I would do this (they never ask me) is to have the brower, etc only launch from a BIOS menu. If you never run it, it's just taking up ROM space, no biggy.

how low can you go?

[nxxuzzzvvnn]

As almost everything with x86/ia32 technology this is just another braindead cheapo option.
I mean, sure this will need another primary partition? And why the heck they just don't implement OPENFIRMWARE 'coz thats all the bios
you'll ever need?
I really hope the guys from the different open/free bios projects get their things right, up and running. I can't stand the trash anymore!

TurboTax (was Re:hard disk?)

[NaDrew]

Somewhat OT, but you brought it up. I just finished installing Win2K-Pro as a guest OS in VMWare, running on Win2K-Server. I'll be installing TurboTax in the guest OS tonight and will post in some appropriate place about how it goes. I don't expect there will be any problems, but you never know.

more important: connect to the BIOS via network

[g4dget]

The only reason to get a KVM these days is to be able to fiddle with the BIOS. After the OS has been loaded, I can connect to it over the network easily anyway. Serial port BIOSes are one solution, but they still require a network-to-serial-port interface, as well as a serial port.

So, are there any BIOSes that can be accessed via HTTP or TELNET? That's perhaps more important to me than more local functionality.

OpenFirmware not just for Suns...

[kwoo]

Actually, newer Macs use OpenFirmware. I have it on both of my Macs, and they're both over two years old.

For a little bit of fun, hold down CMD+OPT+O+F at the boot chime. This will put you into the PROM, which is scriptable in Forth.

Given that board test and driver suites are written in OF, I don't see any reason at all why a web browser would be difficult. Text-only, perhaps -- but not terribly difficult.

There is also a project I noticed one day on Freshmeat that I think was called Retro Native Forth, for the IA32 arch. I wonder if that could be molded into an OF-like role on that platform?

Re:OpenFirmware not just for Suns...

[Satai]

For a little bit of fun, hold down CMD+OPT+O+F at the boot chime. This will put you into the PROM, which is scriptable in Forth.

As I recall, isn't the FreeBSD bootloader scriptable in FORTH? Is OpenFirmware a completely separate development from the FreeBSD bootloader, or is Forth just particularly well-suited to that time of application?

Re:OpenFirmware not just for Suns...

[red_dragon]

OpenBIOS is working on an OpenFirmware-compliant firmware for PCs and other platforms that don't already have such firmware (Alpha, x86-64, and IA-64, according to their page).

Re:OpenFirmware not just for Suns...

[karlm]

or is Forth just particularly well-suited to that time of application?

I have no special knowledge of the FreeBSD bootloader. However, Forth was actually originally designed as part of a "system-on-a-chip" CAD system. Forth bytecode is extrememly compact, the VM is very light, and the language itself greatly encourages code reuse. (I forget the statistics, but Forth programs on average are very much characterized by many short functions that get used over and over.) Forth doesn't ncessarily lend itself to fast native binaries on modern superscalar CPUs, but the BIOS needs to wait for the HD to spin up anyway. A 200% faster language for the BIOS won't cut your boot time by 66%. In summary, Forth is simple and compact and designed for embedded applications. This is why it finds its way into bootloaders/firmware.

Re:OpenFirmware not just for Suns...

[Anonymous Coward]

Even though offtopic, the FreeBSD bootloader does in fact support FORTH. I don't know the specifics, but at least two of the files in my /boot directory are written in FORTH.

Phoenix vs. Phoenix

[drakewyrm]

Phoenix probably did not officially announce the connection between their spiffy BIOS and the similarly-named Mozilla project. To do so would have announced their project before it was more than vaporware. Sometimes companies slip up and make *ethical* business decisions...

To understand what happened and why, look at the two related events: the lawsuit and the BIOS release. Add some logical speculation, observation, and assumption: Phoenix Technologies wanted to name their brower the Phoenix Browser (speculation). Phoenix Technologies did not (AFAIK) sue the Project Phoenix, the capital of Arizona, Phoenix Contact, or any immortal flaming birds(observation). Phoenix Technologies is staffed by humans(assumption). Humans tend to be jealous and illogical(observation).

Stir until mixed. Bake until done. Remove conclusion from head. Mine looks like this: Phoenix Technologies planned to release a browser under the name 'Phoenix Browser', but someone beat them to the name. Rather than pick a different name, they convinced themselves that they had more right to the name than the other guy. Then, they send hordes of marauding Vikings to pillage and plunder. Or something like that.

Was there a connection? Probably. Will they admit it? Probably not. Am I done ranting? Reply hazy; try again.

Re:Phoenix vs. Phoenix

[Anonymous Coward]

Trademarks relate to rather broad areas of products, for example "Computer Hardware and Software related".

I couldn't just call my spiffy new spreadsheet "Linux", because Linus hasn't currently announced a spreadsheet program.

So, I think if you decided to call any software program "Phoenix", you'd get the same letter, whether or not Phoenix Tech was working on a similar product.

Why doesn't microsoft cash in on this?

[ninkendo84]

It makes me wonder why microsoft hasn't started making BIOS's a long time ago. It would be a good way to implement some embedded software, which seems to be the goal of the whole .NET strategy. I can see microsoft BIOS's that are specially meant to handle Windows' API's, and maybe could further their monopoly by only being able to launch microsoft operating systems.

Then again, would they really be called microsoft anymore if they did that? It'd be more like microhard.

Re:Bad

[fishbowl]

"Admittedly, I'm over my head here, but can't you have a complex BIOS that gets out of the way when the OS boots, or acts as a mini OS when the real OS wont load?"

Sun firmware contains a FORTH interpreter that "can" do "anything". It offers all kinds of access to hardware, and can even be used to tune the OS after boot. There's a symbolic debugger, facilities for running programs that don't need an OS, and tty support. If you wanted/needed to be clever, you could do *anything* from here.

Except for the fact that PCs follow certain design factors (decisions made in the early-mid 1980s), there is no reason we couldn't have something similar in the PC world.

Not new

[acceleriter]

Phoenix has been working on a spyware web-enabled BIOS for years. Now they've just moved the network stack into EPROM, it seems.

It'll be a cold day in hell that I run any machine capable of connecting to the net without my telling it to explicitly.

Re:Not new

[jkrise]

"It'll be a cold day in hell that I run any machine capable of connecting to the net without my telling it to explicitly."
And yet, you use Windows? Putting a net stack on the BIOS does not automatically imply the system will ocnnect to the internet- LAN maybe, and you ought to have a firewall. The utility of having a network stack on the BIOS far outweighs the risks.

Re:Not new

[base3]

Fair enough--but for those without the savvy (or desire to manage yet another box) to run a real, separate firewall, this opens the door to all kinds of abuse.

And we all know that firewalls are best at keeping people out, not keeping sensitive information in. I doubt most firewall rulesets would block a port 80 connection to some innocuous site outside of Phoenix's known IP allocation . . .

It could work...

[ilctoh]

Now what would happen if we had a beowulf cluster of these? Seriously, you buy a computer with a power cord and ethernet connection. Have it boot off a server open-source NFS and such, and you're set. No boot roms, floppy disks, ethernet drivers, etc.

Not as cool as it sounds...

[Slack3r78]

I read the headline and thought "Wow, that's the coolest thing I've seen in a while." Then I read the article and it took only the words "stored on a protected portion of the hard drive" for me to become severely disapointed. This would be awesome if it were embedded as a ROM, but as a HD partition, it's usefulness is questionable at best. In short, if your HD dies, you're still up the creek, much like you would have been without this technology.

Re:Bad

[ebassi]

Admittedly, I'm over my head here, but can't you have a complex BIOS that gets out of the way when the OS boots, or acts as a mini OS when the real OS wont load?

"Old" Alpha servers from Digital used to have a "BIOS" that had a shell and supported even TCP/IP connection; i't was used for diagnostic and recovery purpose when the machine was rebooted, something went bad in the process, and maybe the sysadmin isn't even in the building...

On a standard workstation or on a PC, this doesn't strike me as useful: what good would it do, when I do have a boot floppy lying around?

Even having a BIOS isn't that useful anymore: OSes do better hardware detection and resource assignment (you OS does them every time it starts, unless you are using DOS) than any BIOS.

Phoenix

[zoomshorts]

So Phoenix thinks it OWNS the name? They soon forget the mythology about the bird that rose from it's own ashes! They forget the city in Arizons named Phoenix. They are a JOKE. I would tell them to stuff it up their asses in court! And then I would strip them of the name on general principals.(sp) Then I would extort all their earnings for trying to trademark a common name that has been in use for CENTURIES before they TRIED to lay claim to it! Then I would shoot the board of directors and all their kinsfolk. Gotta stop the DNA before it spreads!

Re:Phoenix

[SirSlud]

I pity the potential contributor to society who spends his cycles trying to defend ones right to use a name which may cause confusion in a small market space.

You wanna rail against dirty corperate tactics, of which we both know there are many?

a) find a real dirty corperate tactic (shouldnt be too hard)
b) spend your valuable time fighting it

But honest to god, dude, you are welcome to open a sofa store called Apple .. you just can't open a computer company called Apple. Thems the breaks. It actually helps people less knowledgable than you not get rooked or confused. Your anti corperate ire, which I tend to have lots of myself, could be far more usefully directed than getting all irate when somebody has a semi-legit case with trademark dilution .. I mean, they didnt even try and put Pheonix the browser out of 'business', they just wanted them to change their name.

No biggie, so chillax and fight the battles worth fighting.

Re:Phoenix

[Anonymous Coward]

You're pretty stupid. 99% of companies are named after people or things that already exist- Paramount Supply, Ambrosia Software, Apple Computers, Dell, Hewlett-Packard. Phoenix has been given a trademark to use the name "Phoenix" to sell their products, and if any other group tries to distribute a product carrying the Phoenix name that could compete with Phoenix's products, they can and should stop that group. Seriously, get a clue, and don't act like you're 12 years old.

Re:Phoenix

[Anonymous Coward]

Uh, you do know that ambrosia doesn't actually exist, right? Not to mention the phoenix...

Re:Bad

[WhaDaYaKnow]

I totally agree this is bad.

I understand Phoenix is trying to protect their business, but really, the days of the BIOS as it is should be over.

The BIOS is a legacy piece of crap that serves practically no purpose, but to boot the OS.

The services provided for the "Input/Output" go largely unused, mainly because the majority are 16 bit services and no modern OS has a way to call them in the first place (well, without a high latency). Even the 32 bit services go largely unused,- PCI for example is practically always implemented by a driver that does direct IO vs. calling the BIOS.

In other words, non of the most prominent operating systems call the BIOS for services such as RS-232, IDE, LPT, Video, you name it, after the apprioriate drivers are loaded.

The REAL purpose of the BIOS should be: initialize the hardware up to a point so that it can boot the OS. This means memory initialization, some timer and interrupt related stuff and whatever code is required for the boot devices (I personally think IDE and Ethernet are the most important, but I can see that USB and SCSI are important to a lot of people)

After that the BIOS should load the OS image and be done.

Don't think I'm making this thing up; I've actually implemented a boot loader that completely eliminated the need for a BIOS and it was very fast; ready to boot of the harddrive as soon as the harddrive was spin up (e.g. 3 seconds!)

LinuxBIOS is doing something similar.

Anyways, sorry for this little rant without any proper links or so, but I gotta go to be in time for Apres-Ski happy hour! :-)

MOD PARENT UP!!!

[ebassi]

I completely agree with you.

LinuxBIOS is doing something similar.

And I'd like to flash out my BIOS with it, if only I'd have a little time...

Re:MOD PARENT UP!!!

[BlueUnderwear]

And I'd like to flash out my BIOS with it, if only I'd have a little time...

... and a spare BIOS chip, when it will go wrong...

Re:MOD PARENT UP!!!

[ebassi]

And I'd like to flash out my BIOS with it, if only I'd have a little time...

... and a spare BIOS chip, when it will go wrong...

Some risk certainly makes it more interesting to me... ;-)

And if it goes wrong, I'd have an excuse to change my old, crappy, motherboard, so it ain't that bad, after all... :-)

Re:Bad

[Pharmboy]

On a standard workstation or on a PC, this doesn't strike me as useful: what good would it do, when I do have a boot floppy lying around?

I don't use floppies, and most of my computers don't have floppy drives. Many computers don't come with floppies anymore.

Even having a BIOS isn't that useful anymore:

It is if you want the OS to be able to talk to the hardware. Without the bios, it can't.

And the OS doesn't do better hardware detection in my experiences. Remember, the OS detects the hardware THRU the bios, by comparing it to a list of known hardware.

Yea, if you use only hardware that is older than the OS, then the OS is great at recognizing it. And I prefer to have more control over the resources. No OS, especially Windows, can anticipate all the crazy stuff I may be trying to do. Maybe I want to have 6 pci video cards or modems. (Yes, some of us do crazy stuff like that.) Windows doesn't handle that very well, even now. With a more sophisticated bios, i could debug and tweak the way it was seen by the OS.

Paranoia: DRM?

[suwain_2]

"Resides in a protected area of the harddisk".

Did anyone else just have flashbacks to the whole TurboTax thing that wrote to 'protected areas of the hard drive'? This technology sounds cool, but I'm a bit concerned: it almost sounds like the beginnings of BIOS-level DRM (digital rights management), sugar-coated with some seemingly-good features.

It has cool potential, but at the same time, scares me somewhat at what it could theoretically be used for.

Re:Paranoia: DRM?

[Anonymous Coward]

No it doesn't.

Re:Paranoia: DRM?

[edgecase]

Secure booting is the only link left out of the chain in the DRM OS (which MS has a patent on). By putting a public key in the BIOS, and only allowing files signed by the matching private key to boot, the circle could be closed. Truly scary, having 0 control over your machine.

Script Kiddies

[Pyrosophy]

Yes, very bad, because you know if this happens one day you'll see:

Click here to flash your bios.

And you can count the number of seconds on your right hand that will elapse before that link fries your motherboard.

Re:Script Kiddies

[dublin]

Actually, on most any machine whose BIOS supports PXE booting (pretty much all of them, since it's an Intel/Microsoft standard) it's already possible to re-flash the BIOS remotely, among other equally invasive things.

That's really what it was made for: PXE stands for Pre-boot eXecution Environment, remember???

Re:Bad

[ebassi]

I don't use floppies, and most of my computers don't have floppy drives. Many computers don't come with floppies anymore.

Right. That's when bootable CDs kick in.

And the OS doesn't do better hardware detection in my experiences. Remember, the OS detects the hardware THRU the bios, by comparing it to a list of known hardware.

Bzzzt. Wrong.

Practically, all hardware interaction is done, nowadays, with drivers that can access directly to the barenaked metal stuff inside the case.

A BIOS should only prepare the OS to boot up from a physical medium, i.e. a drive, a flash card, a ethernet connection. Given that need, that is fulfilled in about 3/5 seconds, everything else is only unneeded crap, since modern OS can handle it much better, from the programmer's point of view, and from userland.

And I prefer to have more control over the resources.

That's operating systems and userland applications' stuff, not BIOS'.

Windows doesn't handle that very well, even now.

I'm really sorry for you, and whoever is using a retarded operating system that isn't even good on retrieving the information it needs directly from the hardware, instead on relying on possibly buggy and/or incorrect information from a piece of crap that survived beyond its use.

With a more sophisticated bios, i could debug and tweak the way it was seen by the OS.

Sounds like you need a better OS, not a better BIOS. ;-)

Oh GREAT!

[MonopolyNews]

Now we get to hear how IE "can't be removed from the BIOS".

BIOS = Linux + Emacs

[axxackall]

Emacs in BIOS? That's the most insightful idea I've heard today!

Seriously, the biggest inconvinience of all BIOS' I saw is inability to extend them. But if BIOS will have Emacs as UI - no problems! Few brackets and you have new wonderful function.

UI of Emacs is not overbloated (compared to GNOME/KDE), but still powerful enough. With a choice depending on available memory, it can be compiled with or without X11.

However, which Emacs? GNU/Emacs is more compact and faster, but Xemacs has a great package management subsystem.

Wait a minute. UI is good, but not everything. Who will be responsible fo hardware? Of course Linux. So, that will be sort of embedded Linux with Emacs as UI. Cool! I want PC with such BIOS!

Big

[axxackall]

In good BIOS "B" stands for "Big". That's right. Big Intelligent Operating System.

Amusing

[ty_kramer]

While Microsoft moves more and more application functionality into the OS, Phoenix moves more and more OS functionality into the BIOS.

Re:Amusing

[unitron]

"While Microsoft moves more and more application functionality into the OS, Phoenix moves more and more OS functionality into the BIOS."

At this rate the BIOS will all be in the power supply in a year or two.

Of course with all the bad electrolytic capacitors out there lots of us already have power supplies that have been "flashed".

Re:This is a bad idea! (offtopic)

[linuxghoul]

(I KNOW, COMPLETELY OFFTOPIC, but just sumthing i felt like saying)

i used to think like that, untill i installed my first firewall. The log showed more than 20 intrusion attempts in the first 5 minutes of the firewall starting up. Also, when u cant trust your own OS (or "your" software) to not phone home (only a problem on windows), u better have your firewall up and running before ANY network activity.

Its not as if someones waiting...there are just too many scripts, scans, worms out there...and the primary purpose of these is not to hose ur data..they cudnt care less...what they want is, in nearly all cases, ur net bandwidth, to:

1) run IRC bouncers
2) host warez (ur diskspace is also the target here)
3) launch DDOS attacks
4) any kind of "proxy" that may help in anonymizing any later "serious" cracking activity

and depending upon how "responsible" you are, u may not have problems with 1/2/3 but 4 could end up getting u into very serious trouble. I personally dont want to be a part of the next attempt to bring down EBay.

also, in my case, being at a UNI, and having my machines on 10Mbps, AND with the huge UNI net pipe , my machine is definitely a lucrative target.

Even otherwise, i just dont **LIKE** anyone access one of *my* machines in ways i did not intend...you maybe different.

I think people lacking the will to properly secure their boxes should not be allowed on broadband. Dialup is ok...i think dialup users only waste any crackers time :P

Seriously, my suggestion is for you to give a software firewall a shot. The first half an hour should be enough to convince you of the need. (and no, the built in forewall in XP doesnt count.)

I have myself since graduated to a hardware firewall for real security, with a software firewall only required to prevent your own "untrusted" programs from calling home.

Ghoul2

Re:Bad

[Spoing]

The BIOS is a legacy piece of crap that serves practically no purpose, but to boot the OS.

Let's hammer that home, shall we? Instead of a minimialist BIOS setup, Award and Asus have decided that the following features are more important than conextual help on the P4PE (rev. 1.03) board;

1. Audio POST messages; yep, it speaks the error messages.

2. Graphical boot logo; yep, an advertisement for Asus.

3. 6 types of power on options:
   1. AC power loss restart,
   2. wake/power up on external modem,
   3. power up on PCI card,
   4. power up by PS/2 keyboard,
   5. by mouse,
   6. Automatic power up each day at a specific time,
   7. Automatic power up by a specific day of the month with alarm

4. ...yet, no boot option for USB or firewire devices; no boot off of that USB pen or external CDRW.

   That's not all...

5. Turn off some of the ports (game, serial, ...) and the system either won't boot or it will disable other seemingly unrelated ports.

6. Automatic boot into BIOS setup after boot. This is supposed to reset CPU timings if the system crashes due to overclocking. Please. This time around, unlike the old versions of this misguided feature, it is possible to disable it.

7. ...speaking of disabling things, the boot sector virus detection scheme never made any sense to me. How many years has it been since there has been any boot sector viri -- yet the damn thing has to be turned off much of the time.

Tom's Hardware gave this one a thumbs up?

True but

[ColGraff]

Granny will find it too intimidating to do almost any system maintenance task - she's not the target for this software, any more than she's the target for Linux+Lilo repair partitions. And grannie, as well as most desktop and a lot of server users, is probably running windows. Windows cannot write to Linux partitions, so far as I know.

like SPARC EPROM?

[Anonymous Coward]

That's what I want. I'm sick of ancient PC bioses. Gimme programmability (not necessarily Forth, but something), ttyS1 as console if no vid card, disk chain scanning, and especially net booting with DHCP, TFTP/NFS/Coda, etc. That'd be what I want.

Or commodity UltraSPARC/Sun hardware. Either one. Sun, if you can make it affordable, I'll buy it! Mmmm. 64-bit goodness....

Re:Bad

[drayzel]

Those six types of powering up a machine are actually a pretty good thing. I've used the Automatic power up each day at a specific time as an alarm clock more than once. And it is also pretty nice for the work place.

Powering up a machine by using hotkeys on a keyboard is a nice feature for us big fat lazy guys that don't want to bend down to the level of the computers power button.

The graphical boot logo is ment more for OEM's (HP uses a lot of ASUS boards), and geeks that like to tweak this feature.

Although I like these features I must point out that the latest bios flash for my ASUS P2BLS seems to have caused some issues with powering down either via the power button or software when wake at a specific time is enabled, it reboots 30% of the time rather than shutting down.

As for the boot sector virus settings... I saw it used daily when I was doing phone support for Windows ME. We had a lot of techs that would explain the mysterious freezes that occured during setup as a "BIOS anitvirus" and tell them to call the BIOS maker. It made a great excuse and the cutomer can rarely verify this... ever tried calling AWARD? Good luck Joe Sixpack! That place sure sucked.

~Z

here's some ASCII pr0n

[SHEENmaster]

Lameness filter encountered. Post aborted!
Reason: Please use fewer 'junk' characters.

Umm, remember the 1980's?

[Anonymous Coward]

We already saw this in the 1980's with ROM
BASIC, machines with programs included in ROM
(HP, among others had laptops like this)
and a bare, bootable DOS ROM in IBM PS* machines. I don't understand why
customiz.exe wasn't included in the ROM, but I think the graphical shell was included there.

Re:Bad

[Tailhook]

For God sakes! A browser in the firmware? What for?!

Look. DEC did this right years ago. If you don't know, find a crummy AlphaStation on EBay or something. They're next to free.

You can put a serial cable in the back and when the 10+ year old computer notes the lack of a keyboard and/or video subsystem on boot (because its headless, like most significant computers) it will send a frigging prompt out the serial port. From there you can run basic diagnostics, dump a device list, pick a boot device, etc. You can bootstrap the machine from nothing to full installed and running OS without the use of a "local" keyboard or monitor. I don't think you even need to have a processor installed to get, at least, the prompt. It uses a small, inexpensive and independent CPU!

Sun, HP, etc. I'm sure they all have similar.

Over *here* I have a VA Linux A1000. I got this for cheap during VA's fire sale before they spaced the hardware biz. One of the last machines out the door from VA. What is this machine's solution to the bootstrapping puzzle? A proprietary connector on the back (where?!) attaches a proprietary little black box (poorly made and rather difficult to replace, I gather) that provides keyboard/mouse/VGA connectors. IT'S A 1U BOX! I'm supposed to leave this flimsy little device permanently attached to the machine in a rack? What crap!

What I want is someone (say Phoenix, perhaps?) to create a BIOS for Intel/AMD based motherboards that provides all the basic features of traditional PC BIOS (minus that pointless energy saver thing) configuration through a serial port, with the option of allowing the OS to assume control of that same serial port and thus achieving complete, end-to-end, power-up to OS bootstrapping fully headless. I have no doubt that every cotton picking Intel/AMD motherboard with a built-in serial or USB connector is FULLY CAPABLE of doing this today. All it would take is a tiny bit of inspiration. Why on Earth has no commodity motherboard manufacture thought to do me this trivial thing? I'll pay extra. A LOT extra.

Yeah, I know, buy "good" hardware, the Unix folks already do this. Yeah, I know, some weirdo vertical market board maker has just the thing hiding behind some link. My point is this; allowing a serial port, instead of keyboard+VGA, to perform BIOS config and bootstrapping is trivial to implement. There is no technical reason this should not exist on cheap, common peecee hardware.

Re:Bad

[thogard]

The REAL purpose of the BIOS should be: initialize the hardware up to a point so that it can boot the OS. This means memory initialization,
some timer and interrupt related stuff and whatever code is required for the boot devices
Can someone tell me why the bios needs to look at all 512+ meg of memory?

Why does it play with the floppy drive or CD rom at all if its told to just boot C:? It should just do what it needs to do to load the OS. If that means waking up the disk controller, and reading in a few sectros without starting up the screen, great. I don't need the screen... I need the os loaded in the machine and I'm assuming it will reset up everything else again.

Huh?

[XNormal]

The whole point of a "Basic Input/Output System" is for, well, basic I/O. It was meant to be a thin layer between the OS and the hardware.

I always thought BIOS stands for "Built-In Obsolete Software".

Re:Bad

[fiftyfly]

Tom's Hardware gave this one a thumbs up?

Dude, didn't ya get the memo? Tom's recommends pretty darn everything that doesn't actually burn down the lab. Anything to server another 2 doz add impressions.....

BI + OS do not mix

[TimMD909]

>Complexity is mostly bad. Complexity in software
>leads to bugs. Complexity in BIOS breaks
>the "Basic" part of BIOS.

>Better BIOS - I am all for that. BIOS with the
>the "OS" standing for "Operating Systems" is not
>a good thing IMHO.

Not to mention the voodoo surrounding BI-sexuals. Don't get me wrong, they freak me out. I hear they think differently. The only exception is with highly attractive females, commonly documented under the title "lesbian porn".

Whoa, I got off track. What I really meant to say was, a BI-sexual Operating Sytem scaries the be-geeze-us outta me.

btw, anyone know how to spell what is pronounced the same as "be-jesus"?

Re:Bad

[Isle]

You stole my comment! ;)

The alpha SRM and mini-debugger kicked ass, if you fucked up you OS you could just TFTP a new one from the network or retrieve it from floppy or cdrom.

Re:Bad

[Anonymous Coward]

Remember, the OS detects the hardware THRU the bios, by comparing it to a list of known hardware.

Since linux kernel configuration presents one with options to enable workarounds for chipset-specific hardware bugs (see ide drivers) I've always assumed this to be completely untrue. IIRC, projects like the LinuxBIOS can exist because the BIOS is unnecessary for a modern operating system. Under Linux, it's only value is in telling LILO's bootsector what device it has been loaded from.

I would love to hear more on this subject from any actual gurus in the house. Some random poking around on Intel's site a couple of days ago left me with the impression that chipset documentation was freely available. Is it really true, as has been suggested in a previous post, that BIOS code accesses chipset registers/IO ports which are hidden from the public?

BIOS is a x86 PC-only concept

[olethrosdc]

I am sorry, but this is absolutely true. The BIOS only exists as a concept on the PC. I like to give the Amiga OS as an example in this case.
(It was a single-user Multitasking, 32-bit OS, in late 80s to mid-90s. Last significant version was 3.1 in 1992 iirc, though 3.9 was released in 2001).

Let's take my A1200. What did it consist of? Well, it had the CPU, a custom chipset on the mainboard with DMA-accessible 2MB of RAM and an expansion slot. As you will see, there was hardware support for some things and then everything else was part of the OS:

Hardware-wise a number of protocols were supported, most importantly the mouse and AutoConfig. The mouse was simple. But AutoConfig was the plug'n'play of the 80s. (Introduced with Amiga3000 I think). It scanned the expansion bus (Think PCI bus). No, I should not say scan. It makes everything seem like windows scanning for hardware. It just sent a couple of signals on the bus and any devices that were there responded to acknowledge their presence, then the AutoConfig would ask them some stuff, like, if they had any ROM.

Now, what was interesting was that the A1200 itself had a (512k?) ROM, which contained basically the Kickstart and a large part of the OS (the kernel, disk operations and basic windowing system). The main part of the kickstart was the bootloader. Now, I don't see anything related to BIOS here. The sequence was basically

autoconfig->kickstart->kernel->?

Another interesting thing that hapenned around the same time that the kernel was loaded was what happenned to the ROMs that other devices that were connected on the bus declared. For example, my A1200 had a SCSI card addeed. The drivers for the OS were actually on a ROM on the SCSI card. When AutoConfig asked the card, it said it had a ROM. The same type of rom filesystem was used for the SCSI ROM and the Motherboard ROM. So, basically the ROM was looked at via the ROM filesystem and any libraries in there (the driver was just a shared library and shared device) were added to the system. Very simple.

So, I dunno if you would call AutoConfig a BIOS in itself. Or if you would call the ROM FileSystem part of a BIOS. In any case, the FileSystem concept was part of the OS, which used it to access all kinds of devices. The AutoConfig was an extremely simple protocol that could be done with a minimal amount of hardware. I think the specs are less than 20 pages in the Amiga Hardware Manual.

Anyway, all I see here is the hardware, some *standard protocols* that were implemented on the hardware itself, the bootloader and then the operating system itself.

BTW, Linux systems just ignore the BIOS completely, don't they?

C64 crash bug

[pommiekiwifruit]

There was a crash bug in one of the versions of the C64 rom... something to do with line extensions (which lines were 40 chars and which were the second half of a virtual 80 char line) and pressing the delete key when you were on a certain line boundary. I think that was fixed in later versions of the ROM though (they also changed the colour put in colour RAM when the clear screen character was printed). And don't forget Microsoft wrote some of the base code (which Commodore altered).

The PC needs other features

[yalla]

The PC doesn't need a graphical OS with all that fancy nicknack. The PC needs:

* A serial console! I don't want to connect a VGA monitor to the machine just for tuning the bootdevice
* Get rid of all that INT13 C/H/S booting crap
* The BIOS needs real SCSI support, no more SCSI-BIOS remapping
* Realmode should be dumped completly! When you power on the PC it should be in protected mode immediatly

Everything else like network support is neat, but can be optional.

Just my 2c, Alex.

Re:The PC needs other features

[Anonymous Coward]

As long as you're adding a serial console, why not go all the way and allow telnet/ssh in through tcp? A serial console assumes that you've at least got something physically attached to the machine, and unless you've got your building wired with rs232 lines running about, it only becomes marginally more convenient to attach a laptop computer than it does to attach a monitor. With TCP, you could literally be anywhere and still be able to work on the machine when it goes down.

That said, tcp on bios would probably only work effectively if the network were on the motherboard, or there were an open set of basic level standards for communicating. You don't want to have your bios need 100 different nic drivers in it just to boot. Also, the nic would have to remember it's last known IP address, which isn't too big an issue, as nics that support wake-on-lan already do this.

Re:The PC needs other features

[yalla]

Good point. Maybe some better bootloader as well so that you don't have to write one on the harddrive? I would like to see some sort of OBP for the pc. Booting off a kernel from the net, cool. Some basic filesystem support and an editor to repair a damaged system. If it could do ftp/ssh you could even upload patches to you system or download data from a crashed system.

Yeah, that idea is really sexy and enlighting.

Alex.

Re:The PC needs other features

[edgecase]

www.linuxbios.org everything you listed and more.

It's been done

[Ed Avis]

In PC and server environments, Phoenix cME enables the creation and management of a secure "host protected area" (HPA) of the hard drive, where applications reside. The first of these are Phoenix's own FirstWave apps, which help diagnose and recover PCs if the OS goes tits-up, third party developers can write their own aps for storage in this "tamper-proof"(actually tamper-resistant) area.

This sounds very much like the 'reference partition' used with second-generation (1991-1995) PS/2s. The older models had a 'reference diskette'... but it's too much to ask nowadays that manufacturers could write a diagnostic program to fit in a mere 1440 kilobytes.

Re:Bad

[Ben Hutchings]

There are BIOSes that support a serial console, e.g. in Intel's ISP series of 1U servers.

Re:Bad

[danalien]

it does all that checking "to see if the hardware is OK/allright/nothing is damaged". This is especially good to do when it comes to the memory, cause you don't want to use the bad-sectors of the memory when you are in your OS do yoU?

AIOS? then?

[danalien]

Advanced Input Output System?

Silly Brits...

[Kadagan AU]

The first of these are Phoenix's own FirstWave apps, which help diagnose and recover PCs if the OS goes tits-up

heh... he said tits... heheh...

Not entirely new.

[pmz]

UNIX workstations have had amazing firmware-based diagnostics for at least a decade (the limit of my experience with them). It's called the OpenBoot PROM. Plug a serial console into the back of the machine and reboot, and you'll see.

Setpatch

[CausticWindow]

All Amiga system rom calls are patchable with a simple setpatch(). The program setpatch runs at bootup and for some versions of the roms it patches a lot of system calls before everything works like it should.

Re:Bad

[ReelOddeeo]

The BIOS is a legacy piece of crap that serves practically no purpose, but to boot the OS.

Oh, I beg to diff.

The BIOS has many valuable purposes.

• Branding
• Advertising
• A game or web browser could keep you entertained during the lengthy bootup of whatever latest Microsoft monstrosity you are booting today.
• Protect the computer from you

In short, the BIOS adds a lot of value to the computer. (Value for corporations, to the benefit of corporations.)

C'mon these BIOS programmers need to justify their existance. A BIOS maker can't just ship the same BIOS year after year without improvements. Just look at car companies. Each year there is a new model. What changed? The bumper. The hood ornament. Etc. Or as another example, just look at Microsoft. What changed this model year? The chrome. How things are organized. The value of this? Training revenue. OS Envy creates sales sizzle. What do you really expect a BIOS maker to do? Not start adding flashy useless features? It is just a natural outgrowth that once the product cannot be enhanced in very useful ways, they will concentrate on technical challenges like audio error messages, boot-time browsers, etc.

Just imagine that the boot logo is not only animated, but actually has downloaded advertising each time you boot up? Don't tell me there isn't value in this.

If the ad is seen each time you boot up, it encourages closed, smoke-filled, back-room meetings like this:

Phonyix: Okay, GillBates, we'll give you a 5% kickback on ad revenue if you make the computer crash.
GillBates: But I've already got that feature implemented.
Phonyix: Okay, we'll set up a schedule. If the computer crashes once per hour, your kickback is 5%. If the computer crashes 2 times per hour, then 5.28%. If more often, then we'll work out a graduated schedule.
GillBates: How about some innovative thinking. Suppose our software connects to your .NET enabled server and obtains a schedule of how often you want it to crash? An agent service in the machine can connect to our server and balance the need to crash and generate ad revenue, with the need to not crash if the revenue is not high enough.

C'mon guys. This is progress. Imagine back in the early days of computers, like the original 1981 IBM PC. Computers would never have provided the incredible economic opportunities that they do today. This is progress and innovation.

.NET enabled BIOS

[ReelOddeeo]

A .NET enabled BIOS can ensure that you see a fresh, non-stale advertisement each time you boot.

A .NET enabled OS can ensure that you see the ad frequently.

Re:Bad

[jandrese]

Not that any bios memory scan has ever caught an actual memory bug. Those things are about the most worthless scans the bios does.

Time for a Legal Precedent

[Black Copter Control]

Sometime, someone running around threatening unfounded lawsuits if people don't do what they want should get sued for extortion -- because that's really what this is. People are being shaken down daily for what, in any other context, would be called 'protection money'.

The obvious problem with this, of course, is that if such a lawsuit won, it couldput a dent in the activities of the lawyer cabal.... and lots of commercial lawyers wouldn't like the idea. (something of a catch-22 problem).

Re:file system access... boot loader?

[HishamMuhammad]

I think your idea does make sense, as I've been through this "hell, where's my boot CD?" many times. But I think it would be better put at the boot loader, since you probably won't be able to rescue a drive that does not reach the boot loader by editing a text file.

I love GRUB's ability to access the file system and 'cat' files. I always wondered why it is not able to write to them. R/W support is possibly much harder to implement, I think, and maybe would not fit in the restricted space the boot loader has available. But then, if you asked me way back if I thought if R/O support for file systems in the boot loader was possible, I'd probably say "no way" too.

Can you say TCPA?

[HerbieStone]

This would be perfect for TCPA wouldn'it? Having a connection to the internet to update Keys and download/upload stuff before even a bootloader had a chance to do stuff on it, would be a nice thing to have to control a box, wouldn't it?

Just a thought

BIOS support for DOS?

[Chromium_One]

More like the other way around...

DOS uses support routines contained in BIOS ROM to access hardware. Your standard PC BIOS contains stuff like harddrive, floppy, video, printer and (horribly primitive and broken) serial port access routines. "Current" OS's pull some system config information from what BIOS discovers during boot, but more or less ignore BIOS completely after that.

solving problems with problems

[nmp]

Because EFI has its own filing system that lives on a reserved part of the hard disk, it can become the standard home for a whole set of utilities that have always had an awkward fit with the BIOS

I my experience any thing with removeable parts or that moves, has a high chance of failing. With the current system trouble shooting is simplified somewhat by systems living in their own hardware. If systems start being shared across hardware components the risk of errors rises at at a rate inversly proportional to the ability of humble tech to resolve problems in a reasonable amount of time.

Imagine the tech support responses you would get then!

.: My machine wont boot

:: Throw it away and purchase a new Intel(tm) Machine

Welcome to Planet Mirco$oft/Intel, We own where you want to go today.

This worries me...

[Sentry21]

From the article:

Because EFI has its own filing system that lives on a reserved part of the hard disk, it can become the standard home for a whole set of utilities that have always had an awkward fit with the BIOS: things like disk partitioners, multiple OS boot controllers, system backup and restore, will be natural EFI applications.

I hope they're referring to the boot track, but this still worries me:

1) They're not referring to the boot track. They make a separate 5 meg partition at the start of the first drive. This gets obliterated whenever a stupid/inexperienced user/tech repartitions, and the whole machine crashes to a halt.

2) They're referring to the boot track. DRM schemes from e.g. that tax software mentioned the other day on slashdot (too tired to look) gleefully overwrites part of it, and your computer will not boot - or even load the BIOS.

3) It uses the boot track, but everyone suddenly gets religion and treats the boot track like sacred space. No one ever dreams of overwriting it. Intel makes up with AMD. George Bush makes up with Saddam Hussein, who then shows his good faith by helping Bush improve his English. The world is as it should be. The BIOS works exactly as designed, but is absolutely useless when you try to boot a diskless machine or your hard drive gets formatted/replaced, and you're screwed anyway, but who cares, because the world is happy again.

The designers seem to be hoping for option #3. Unfortunately, the sarcasm I've painstakingly inserted seems to be the most likely part of the whole paragraph, since relying on the hard drive is even stupider than relying on the BIOSes we have now, because hey, people change their hard drives, and LILO changes the boot track, and heck, DRM changes the boot track, and guess what, we all get screwed.

--Dan

There goes any security

[Anonymous Coward]

So.. if you can use your BIOS to browse the web, that in itself is a security issue, because the user isn't authenticated in any way to the machine (because there's nothing to authenticate against)

You could always secure it with a password, but it's child's-play to remove a BIOS password if the machine can boot to any OS.

Then they tell you that you could use the browser to download critical drivers etc. Even if they implemented a basic browser that rendered 80% of the pages out there okay, then this question pops up.

Where do you save the files you just downloaded ?

In order for this to work, the "BIOS" would have to have write access to some partition on the disk that you could also access from your normal OS.

If they intend to save files to your regular partitions, then it would need to understand the file structure of those partitions, and how to write to them whist ignoring/avoiding/bypassing any file system security restrictions that are in place.

If they write to a "common" area that is saved as say FAT16 that everything can understand, then that's 2 partitions gone from your partition table (one for the "bios" software and one common area)

If they save to the same partition that the BIOS is in, then they risk having it altered/tampered/damaged.

And what if it's a disk driver that you're trying to download ?

I think they're really crossing the line between embedded-OS and "BIOS" here.

Re:Bad

[cerberusti]

Why does it play with the floppy drive or CD rom at all if its told to just boot C:?

The purpose is to load the operating system, not boot "C:". The operating system may be on a floppy disk or CD. installing an operating system on a new hard drive should become a decent challenge ( =

And how do you recover from a failed harddisk?

[edgecase]

Put in the recovery CD? Please.

Re:Bad

[cybrangl]

I think I woudl have to agree with you here. Now what I WOULD like to see, is a slot where you can put a firmware OS in. Granted this may be a simple boot/limited functionality OS, but it would allow for a distinction between where the BIOS should be and the OS starts. This chip could do what the Phoenix is stated to do without the fea of corss-corruption.

Re:Bad

[CerealQiller]

when, not if, some kiddie finds a hole in this illconceived poster child for the DUH foundation, BIOS will come to mean "Bend It Over and Squeeeeeeel." When will they learn that features are secondary to function and stability and that security is paramount. The fastest way to screw up security, function and stability is to tack on internetable widgets. When some aluminum foil beanie-head writes a cute set of bells and whistles and bundels it up with mass manufactured computer software/firmware so that the average user can't even disable it, secure it, or remove it, the author is essentially building a bomb which someone will end up dropping on the entire internet. UPNP's SSDP Discovery Service is a stellar example of what happens when bells and whistles attack. What possessed MS to believe that it would be a good idea to build a system level back door into XP that starts by default, while leaving default disabled, the puny "firewall" they put in (which at least forces the attack on this service to be focused and deliberate for it to succeed, just makes my brain hurt with the logic of it. UPNP service was supposed to make home and office automation simple and easy. Like anyone is really gonna shell out an extra $500 for a refrigerator that can plug into their computer and run amok with a credit card, automatically ordering 30 head of dairy cattle delivered to their third floor walk-up apartment each time they run low on milk. "Windows couldn't find Dairy|Milk|Skim|32 Oz... Ordering malnourished skinny cows at best price-break supported by your Total Net Worth. At least the UPNP vulnerability was only a system level exploit.

Last Post!

[alpg]

If the designers of X-window built cars, there would be no fewer than five
steering wheels hidden about the cockpit, none of which followed the same
prinicples -- but you'd be able to shift gears with your car stereo. Useful
feature, that.
-- From the programming notebooks of a heretic, 1990.

- this post brought to you by the Automated Last Post Generator...