All that stuff just sits there and the Operating System initializes it.
Unless you set your system up to initialize it in the BIOS. For instance, you can configure your SCSI card to not do anything with the BIOS. If you're not booting from a SCSI device, that is.
An 'open architecture' has to do with more than the BIOS. BIOS extensions are fairly reactionary. Most other hardware platforms don't need them.
Actually, there's a whole load of rock-n-roll that blasts out of Windows 98 the first time you boot into a fresh install. Don't you remember that 'click here and NEVER HAVE TO SEE OR HEAR THIS AGAIN' box that pops up?
One of the things I am really really really sad that I didn't hang onto in my youth was a vinyl LP that was my dad's from IBM. It was a record from some IBM Symposium and it had the IBM Company song along with some other tunes by a motley group of company 'minstrels' on it.
I do still have an official System 360 ash tray, new in the box, from back then. And an actual wooden 'Think' sign from the late 1950's. Wonder if that would fetch much on eBay yet...
Can you imagine a current computing company coming out with an ash tray that subtlely matches the logo and decor of a computer? I mean, you have to know your System 360 stuff to know that's what it is.
Ed is a "text editor", that is, an interactive program for creating and modifying "text", using directions provided by a user at a terminal. The text is often a document like this one, or a program or perhaps data for a program.
This introduction is meant to simplify learning ed. The recommended way to learn ed is to read this document, simultaneously using ed to follow the examples, then to read the description in section I of the UNIX Programmer's Manual, all the while experimetning with ed, (Soliciation of advice from experienced users is also useful.)
Do the exercizes! They cover material not completely discussed in the actual text. An appendix summarizes the commands.
- taken from A Tutorial Introduction to the UNIX Text Editor by Brian W. Kernighan. USD:12
Life wouldn't be complete without a complete printed paper copy of the BSD Unix manuals....
The RS-232c standard is useful, it promotes interoperability, and layer upon layer of kludges aren't needed, as is the case with USB, for hardware hackers and hobbyists to make use of it.
My ancient HP95LX palmtop has an RS232 interface and supports Xmodem transfers, and hence it talks with any machine. Except for wizzy-whoo machines which follow the 'edict of Gates' and are 'legacy free.'
We used to despise Gates' mandate of 'legacy free computers.' Now it seems it's cool and sophisticated for some in these parts to talk them up.
Unless you were being sarcastic, which I hope was the case.
But to the hardware vendors, having a hard-wired way of differentiating 'workstations' from 'servers' could be a real gold mine. Traditionally, there are many 'server' tasks at sites that get filled by the secretary's old desktop machine.
There's even something of a precedent for using the BIOS to partition machines into different market segments. I once bought a surplus Alpha motherboard that was really cheap at the time, intending to run Linux or NetBSD on it. When it arrived, I found that it had only the crippled BIOS capable of loading Windows NT, and that it was going to be a complicated kludge to get anything else to run. I found someone in Australia who had paid money for DEC's SDK, necessary to recode the BIOS to run a Freenix, but he wasn't willing to share it. So that $150 motherboard, minus the $100 more I would have had to spend to enable it to run a Freenix, became dead hardware to me. And yes, I looked in DEC's catalogue. Back when the motherboard was 'current' hardware they were selling the same exact motherboard with the bios to run Windows NT for a low price, and with the bios to run Digital UNIX for a HELL of a lot more.
My impression has been that with NT and the Freenixes, the BIOS is a bootloader and then it gets the h*ll out of the way. Using BIOS calls as part of the operating system died with MS-DOS, for the most part. So long as the BIOS provides enough hardware abstraction to allow the kernal of whatever OS you choose to run unimpeded, it shouldn't matter much.
Plus, the day of new, new, ever-new hardware is coming to a close. I'm picking up nice Pentium II systems for under $10 at auction these days. Unless you're part of the bloat-of-the-month-club you don't need anything more than what a Pentium II 400 provides, anyway. If you're a gamer, get real- buy a console and be done with it. I'm making sure I pick up a good stock of the current generation of 'obsolete machines' because I don't think I'll ever again have to buy a new machine. Let someone else spend more than $25 on a desktop machine....
Re:A better lifestyle...
on
Beyond Fear
·
· Score: 1
There has been this pathological fear of 'online identity revealing' as long as I've been online. I come from the old BBS scene of the 80's. We had big social BBSes back then and had social events. We, for a time, played softball every Sunday afternoon and then had a barbeque at one guys house.
Everybody would freeze up with paranoia if anybody said anybody elses' real name at the barbeque.
It was and is pathetic. It seems to hearken back to a fear of computers that people still harbour.
I think you mean that most of the recent prevalent worms are written in VBA, and prey only on the popularity of Microsoft products, which results in a large guillible population of naive users.
Most decent viruses are written in assembly language. There's no way to code something tiny that resides in a boot sector, or latches onto the front end of an.EXE file in a bloat macro language.
Windows 3.1 supports Real Mode. The thing lacking in 3.1 is support for a Hercules Graphics card. But there was a long time when I ran an XT with an original IBM EGA card in it, which has excellent Windows support running in 'monochrome' mode. There are jumpers on the IBM EGA card to set it to drive the old 9 pin monochrome monitors. You get nice graphics, better than a Hercules card. But enough 8088 arcana from the deep dark past...
'Minimun processor requirements' are what Microsoft specifies for what they consider an 'acceptable level' of performance.
For instance, I run Office 2000 on an old laptop. It's a 486DX-2 50 laptop. That's well below what Microsoft specifies as the 'minimum' hardware to run Office 2000 on. It works fine for light word processing and spreadsheets. I'm not insane enough to try to create or run big Powerpoint apps on it.
The hardware was much faster under Linux+X than under Windows 3.1. Yes, X on an 8MB, 25MHz PC.
I first ran Windows 3.1 on an XT. It was a 10MHz 8088 with a Hercules card. It was responsive enough to be useful. I upgraded to a 286 after awhile (I think it was a 12 MHz.) Eventually I upgraded to a 386, because then I could (cooperatively) multitask.
I still have Windows 3.1 on an old laptop (a 386SX with 4 megs of RAM). Believe me, Linux and X run far more bloated on such hardware. I have NetBSD and X on an SE/30 to this day.
I don't know why you think it's amazing that X would run on an 8 MB 386.... You definitely didn't run any of the current bloat desktops on it...
But you fail to consider that the PC had multiple chips for the BIOS and that it cost no more to pair them for 16 bits than it cost to make them sequential for 8 bits.
True for the IBM motheboards, but I seldom, if ever, saw an XT clone motherboard with more than one ROM installed. There were sometimes one ROM and a row of empty sockets in early boards.
You're mixing up address bits and data bits.
My semantic error. I meant 'latching and routing' data bits, not 'decoding.' And you're right that often it's an 8 bit data path to I/O.
I am no expert at all on wireless networking. Hell, I only switched to twisted pair a few years ago when I finally got rid of all the 10base2 coaxial ethernet on my home LAN. But wouldn't this networking interfere with the 802.11 people? I can see roving gangs of kids with GameBoy Advances encroaching on wireless network users. I see a bandwidth conflict looming, in an area of spectrum where everyone is unlicensed and supposed to cooperate.
But for embedded systems (yes, they do exist for x86 processors) you had to have a 16 bit wide data path, meaning either two regular EPROMs or the non-standard 16 bit EPROMs. That adds cost. Further, decoding a 16 bit data path is considerably more expensive, this was particularly the case back in the old days when the popular peripherals, i.e. 8250 and 8255, had an 8 bit data path. It's a legacy concern, but it made the 8088 significantly less expensive. I've read that IBM originally wanted to go with the 68000 for the IBM PC. How different the world would be if they had....
I used to love good old alt-F4. You walk up to the computer of someone who's severely dependent on their mouse to get around in Windows. Hit alt-F4 a few times and everything they had open is closed. *biff* *boom* *biff*
Often they go into a panic, gripping their mouse for dear life.
Apple could run the iMusic service at a substantial loss. It's about selling the Apple brand, and it's about selling Macintosh PCs. Releasing it only on the Mac was an important part of that strategy.
Man pages are considered to be deprecated, even considered a bad thing, by that faction of the GNU movement that promotes the info system of documentation.
Part of the reason that I strongly favour NetBSD over Linux any more is that besides the fact that it's a unified whole, and the base userland is buildable from a coherent source tree, the Man Pages are coherent, consistent, and fairly complete. Plus the whole config and system is compliant to documentation that's been established and solid for over a decade. The old BSD 4.3 Manual Set (seven printed volumes) that I got ahold of several years ago is actually relevant.
This is different from the sort of crapshoot conglomeration of assorted parts that almost any Linux distro represents.
Until version 5, MS-DOS was never, ever, available as a retail product. MS-DOS was something you bought bundled as an OEM product with a Personal Computer from one of the clone manufacturers. MS-DOS as a stand alone product never existed on a store shelf until the 'MS-DOS 5 Upgrade' product appeared. And even that was only an 'upgrade' product, unusable unless you already had DOS installed on your machine.
People who built their own machine, different from people who bought a clone box from Compaq or AST or any of the other clone and white-box vendors, had to buy PC-DOS, the MS-DOS cousin, from IBM, to run on their machines, or make off with an OEM copy of MS-DOS and use it illegally. PC-DOS was available as off-the-shelf software from IBM.
When I built my first PC clone, I did it by going to a swapmeet, and buying an 8088 XT clone motherboard, disk controller, an MDA display card, a couple floppy drives and a used copy of IBM's PC-DOS 3.1. I read that PC-DOS manual cover to cover (it was TERRIBLE documentation, in hindsight). I didn't buy a monitor because by then I was salvaging bare CRT chassis out of dumb terminals, reverse engineering the Horizontal, Vertical, and Video signals and grafting on a 9 pin cable on my own. I had one of those on my BigBoard (CP/M-80 clone) and since my dad had the IBM PC Technical Reference Manual it was trivial to hook up my own kludge monitor. It was years later before I had a monitor hooked up to my PC that actually had an enclosure.
The PDP-8 was first a Minicomputer based on small scale TTL gates and Core memory, but it was later implemented as a single chip microprocessor. In fact it was produced by Intersil, and the part number was IM6100. Harris also produced a version of the 6100 processor. It's a 12 bit data buss chip that runs the full PDP-8 instruction set. There were popular personal computers made up using them, one was called the Intercept.
I've got a whole bunch of Harris 6100 chips new in the tubes from the vendor and am working on a project to kit them up and sell them to hobbyists. It's a cool chip for a hand-rolled project, it has an easy clock (unlike the Intel-based chips which need a 1/3 duty cycle clock signal) and can be casually 'breadboarded.'
The 8088 was and is a cheaper version of the 8086. It's 'cheaper' in that it has an 8 bit data bus, so your memory subsystem is only 8 bits wide. The chip itself wasn't appreciably cheaper, though.
Slashdot Mirror
All that stuff just sits there and the Operating System initializes it.
Unless you set your system up to initialize it in the BIOS. For instance, you can configure your SCSI card to not do anything with the BIOS. If you're not booting from a SCSI device, that is.
An 'open architecture' has to do with more than the BIOS. BIOS extensions are fairly reactionary. Most other hardware platforms don't need them.
Actually, there's a whole load of rock-n-roll that blasts out of Windows 98 the first time you boot into a fresh install. Don't you remember that 'click here and NEVER HAVE TO SEE OR HEAR THIS AGAIN' box that pops up?
One of the things I am really really really sad that I didn't hang onto in my youth was a vinyl LP that was my dad's from IBM. It was a record from some IBM Symposium and it had the IBM Company song along with some other tunes by a motley group of company 'minstrels' on it.
I do still have an official System 360 ash tray, new in the box, from back then. And an actual wooden 'Think' sign from the late 1950's. Wonder if that would fetch much on eBay yet...
Can you imagine a current computing company coming out with an ash tray that subtlely matches the logo and decor of a computer? I mean, you have to know your System 360 stuff to know that's what it is.
- taken from A Tutorial Introduction to the UNIX Text Editor by Brian W. Kernighan. USD:12
Life wouldn't be complete without a complete printed paper copy of the BSD Unix manuals....
Screw 'legacy free computer'. To hell with it.
The RS-232c standard is useful, it promotes interoperability, and layer upon layer of kludges aren't needed, as is the case with USB, for hardware hackers and hobbyists to make use of it.
My ancient HP95LX palmtop has an RS232 interface and supports Xmodem transfers, and hence it talks with any machine. Except for wizzy-whoo machines which follow the 'edict of Gates' and are 'legacy free.'
We used to despise Gates' mandate of 'legacy free computers.' Now it seems it's cool and sophisticated for some in these parts to talk them up.
Unless you were being sarcastic, which I hope was the case.
But to the hardware vendors, having a hard-wired way of differentiating 'workstations' from 'servers' could be a real gold mine. Traditionally, there are many 'server' tasks at sites that get filled by the secretary's old desktop machine.
There's even something of a precedent for using the BIOS to partition machines into different market segments. I once bought a surplus Alpha motherboard that was really cheap at the time, intending to run Linux or NetBSD on it. When it arrived, I found that it had only the crippled BIOS capable of loading Windows NT, and that it was going to be a complicated kludge to get anything else to run. I found someone in Australia who had paid money for DEC's SDK, necessary to recode the BIOS to run a Freenix, but he wasn't willing to share it. So that $150 motherboard, minus the $100 more I would have had to spend to enable it to run a Freenix, became dead hardware to me. And yes, I looked in DEC's catalogue. Back when the motherboard was 'current' hardware they were selling the same exact motherboard with the bios to run Windows NT for a low price, and with the bios to run Digital UNIX for a HELL of a lot more.
My impression has been that with NT and the Freenixes, the BIOS is a bootloader and then it gets the h*ll out of the way. Using BIOS calls as part of the operating system died with MS-DOS, for the most part. So long as the BIOS provides enough hardware abstraction to allow the kernal of whatever OS you choose to run unimpeded, it shouldn't matter much.
Plus, the day of new, new, ever-new hardware is coming to a close. I'm picking up nice Pentium II systems for under $10 at auction these days. Unless you're part of the bloat-of-the-month-club you don't need anything more than what a Pentium II 400 provides, anyway. If you're a gamer, get real- buy a console and be done with it. I'm making sure I pick up a good stock of the current generation of 'obsolete machines' because I don't think I'll ever again have to buy a new machine. Let someone else spend more than $25 on a desktop machine....
There has been this pathological fear of 'online identity revealing' as long as I've been online. I come from the old BBS scene of the 80's. We had big social BBSes back then and had social events. We, for a time, played softball every Sunday afternoon and then had a barbeque at one guys house.
Everybody would freeze up with paranoia if anybody said anybody elses' real name at the barbeque.
It was and is pathetic. It seems to hearken back to a fear of computers that people still harbour.
Something about:
Or do I have the meme wrong?
I think you mean that most of the recent prevalent worms are written in VBA, and prey only on the popularity of Microsoft products, which results in a large guillible population of naive users.
.EXE file in a bloat macro language.
Most decent viruses are written in assembly language. There's no way to code something tiny that resides in a boot sector, or latches onto the front end of an
Windows 3.1 supports Real Mode. The thing lacking in 3.1 is support for a Hercules Graphics card. But there was a long time when I ran an XT with an original IBM EGA card in it, which has excellent Windows support running in 'monochrome' mode. There are jumpers on the IBM EGA card to set it to drive the old 9 pin monochrome monitors. You get nice graphics, better than a Hercules card. But enough 8088 arcana from the deep dark past...
'Minimun processor requirements' are what Microsoft specifies for what they consider an 'acceptable level' of performance.
For instance, I run Office 2000 on an old laptop. It's a 486DX-2 50 laptop. That's well below what Microsoft specifies as the 'minimum' hardware to run Office 2000 on. It works fine for light word processing and spreadsheets. I'm not insane enough to try to create or run big Powerpoint apps on it.
There is no such thing as X Windows, except in the mind of the ignorant. It is called 'The X Window System' or 'X'.
'X Windows' is what illiterate journalists call it, and the people whose main experience with X is reading their inaccurate prose.
The hardware was much faster under Linux+X than under Windows 3.1. Yes, X on an 8MB, 25MHz PC.
I first ran Windows 3.1 on an XT. It was a 10MHz 8088 with a Hercules card. It was responsive enough to be useful. I upgraded to a 286 after awhile (I think it was a 12 MHz.) Eventually I upgraded to a 386, because then I could (cooperatively) multitask.
I still have Windows 3.1 on an old laptop (a 386SX with 4 megs of RAM). Believe me, Linux and X run far more bloated on such hardware. I have NetBSD and X on an SE/30 to this day.
I don't know why you think it's amazing that X would run on an 8 MB 386.... You definitely didn't run any of the current bloat desktops on it...
That would be 'USB over Ethernet'
There's no need to cram in a protocol requirement where any protocol would do.
But you fail to consider that the PC had multiple chips for the BIOS and that it cost no more to pair them for 16 bits than it cost to make them sequential for 8 bits.
True for the IBM motheboards, but I seldom, if ever, saw an XT clone motherboard with more than one ROM installed. There were sometimes one ROM and a row of empty sockets in early boards.
You're mixing up address bits and data bits.
My semantic error. I meant 'latching and routing' data bits, not 'decoding.' And you're right that often it's an 8 bit data path to I/O.
I am no expert at all on wireless networking. Hell, I only switched to twisted pair a few years ago when I finally got rid of all the 10base2 coaxial ethernet on my home LAN. But wouldn't this networking interfere with the 802.11 people? I can see roving gangs of kids with GameBoy Advances encroaching on wireless network users. I see a bandwidth conflict looming, in an area of spectrum where everyone is unlicensed and supposed to cooperate.
It's not something you do to strangers or mere acquintences.
It was coolest of all back in the days of Windows 3.1 when you could flip them all the way back to a DOS prompt in a matter of a second or two.
But for embedded systems (yes, they do exist for x86 processors) you had to have a 16 bit wide data path, meaning either two regular EPROMs or the non-standard 16 bit EPROMs. That adds cost. Further, decoding a 16 bit data path is considerably more expensive, this was particularly the case back in the old days when the popular peripherals, i.e. 8250 and 8255, had an 8 bit data path. It's a legacy concern, but it made the 8088 significantly less expensive. I've read that IBM originally wanted to go with the 68000 for the IBM PC. How different the world would be if they had....
I used to love good old alt-F4. You walk up to the computer of someone who's severely dependent on their mouse to get around in Windows. Hit alt-F4 a few times and everything they had open is closed. *biff* *boom* *biff*
Often they go into a panic, gripping their mouse for dear life.
Your mistake was installing Windows XP. Windows 2000 doesn't have the registration headaches, and it has everything that XP offers that is useful.
Apple could run the iMusic service at a substantial loss. It's about selling the Apple brand, and it's about selling Macintosh PCs. Releasing it only on the Mac was an important part of that strategy.
Man pages are considered to be deprecated, even considered a bad thing, by that faction of the GNU movement that promotes the info system of documentation.
Part of the reason that I strongly favour NetBSD over Linux any more is that besides the fact that it's a unified whole, and the base userland is buildable from a coherent source tree, the Man Pages are coherent, consistent, and fairly complete. Plus the whole config and system is compliant to documentation that's been established and solid for over a decade. The old BSD 4.3 Manual Set (seven printed volumes) that I got ahold of several years ago is actually relevant.
This is different from the sort of crapshoot conglomeration of assorted parts that almost any Linux distro represents.
Until version 5, MS-DOS was never, ever, available as a retail product. MS-DOS was something you bought bundled as an OEM product with a Personal Computer from one of the clone manufacturers. MS-DOS as a stand alone product never existed on a store shelf until the 'MS-DOS 5 Upgrade' product appeared. And even that was only an 'upgrade' product, unusable unless you already had DOS installed on your machine.
People who built their own machine, different from people who bought a clone box from Compaq or AST or any of the other clone and white-box vendors, had to buy PC-DOS, the MS-DOS cousin, from IBM, to run on their machines, or make off with an OEM copy of MS-DOS and use it illegally. PC-DOS was available as off-the-shelf software from IBM.
When I built my first PC clone, I did it by going to a swapmeet, and buying an 8088 XT clone motherboard, disk controller, an MDA display card, a couple floppy drives and a used copy of IBM's PC-DOS 3.1. I read that PC-DOS manual cover to cover (it was TERRIBLE documentation, in hindsight). I didn't buy a monitor because by then I was salvaging bare CRT chassis out of dumb terminals, reverse engineering the Horizontal, Vertical, and Video signals and grafting on a 9 pin cable on my own. I had one of those on my BigBoard (CP/M-80 clone) and since my dad had the IBM PC Technical Reference Manual it was trivial to hook up my own kludge monitor. It was years later before I had a monitor hooked up to my PC that actually had an enclosure.
The PDP-8 was first a Minicomputer based on small scale TTL gates and Core memory, but it was later implemented as a single chip microprocessor. In fact it was produced by Intersil, and the part number was IM6100. Harris also produced a version of the 6100 processor. It's a 12 bit data buss chip that runs the full PDP-8 instruction set. There were popular personal computers made up using them, one was called the Intercept.
I've got a whole bunch of Harris 6100 chips new in the tubes from the vendor and am working on a project to kit them up and sell them to hobbyists. It's a cool chip for a hand-rolled project, it has an easy clock (unlike the Intel-based chips which need a 1/3 duty cycle clock signal) and can be casually 'breadboarded.'
The 8088 was and is a cheaper version of the 8086. It's 'cheaper' in that it has an 8 bit data bus, so your memory subsystem is only 8 bits wide. The chip itself wasn't appreciably cheaper, though.