Talk:IBM Personal Computer/Archive 1

Archive 1

Archive 2

Floppy disk capacities/formats on IBM PC

The article states that the 1981 PC had one or two 360 KB 5¼-inch single-sided double density floppy disk drives."

The 360 KB disk was a double-sided disk using forty tracks per side and nine sectors per track (512 bytes/sector * 9 * 40 * 2 = 368640).

There were also double sided 320 KB disks, using eight sectors/track. Also, there were single-sided disks with half these capacities.

I don't know if any size other than 360 was used on the PC. My correction here is only to say that 360 was a double sided disk. Jm546 21:41, 13 August 2005 (UTC)

Yes, the very earliest versions of PC-DOS/MS-DOS used the 160/320K format rather than that 180/360K format. As far as I know, there was no physical difference in the disks, and it was just a formatting issue. I don't personally recall which version it was, but this site (http://www.myoldcomputers.com/museum/soft/pcdos.htm) seems to indicate that it came in with 2.0 and that the very first PC models with DOS 1.0 had single-sided-only drives Nkedel 18:16, 16 August 2005 (UTC)

Character set

"The original IBM PC used the 7-bit ASCII alphabet as its basis, but extended it to 8 bits with nonstandard character codes." Was this EBCDIC or something else? HiFiGuy 20:42, 19 January 2006 (UTC)

EBCDIC is totally incompatible with standard ASCII. Basically, back in the 1980s, every computer company had their own use for the top eight 128 characters (and often the lower thirty-two). So on the PC, the characters from 32-127 were "standard ASCII", but (most of) 0-31 and 128-255 were IBM graphics characters. Standard ASCII defines 0-31 as control characters, and doesn't use the eighth bit. Nate 02:30, 27 April 2006 (UTC)

Not quite right

There's something wrong with this article, or at least the opening section.

I changed the opening sentence "IBM PC is a trademark of IBM"; whether this is legally required or not, it shouldn't go there (if you want to nitpick, it doesn't even refer to the article subject, it refers to the *name* of the subject).

But that was the obvious problem; somehow, the opening section doesn't get across that this was *the* first "Wintel" PC, or the importance of that. I'd like to rewrite it myself, but I'm honestly not sure how to do it; I don't want to get bogged down in a verbose explanation. Fourohfour 13:43, 3 February 2006 (UTC)

Expansion slots

The PC/XT's 8 expansion slots were closer together than the PC's five. The slot spacing on the PC/XT became a standard still used today on all PC compatable systems. This info needs stuck into the XT section somehow. —The preceding unsigned comment was added by 67.136.145.219 (talk • contribs) . Fourohfour 19:14, 25 April 2006 (UTC)

Good Article nomination has failed

The Good article nomination for IBM Personal Computer/Archive 1 has failed, for the following reason:

• Trivia contravenes WP:NOT. I don't actually think the fact there is trivial, so it should be integrated into the rest of the article, and the trivia section then deleted.

• I also think the bullet points at the end of the intro look really bad. You should probably create a section called 'terminology' to put this kind of thing in.
• I don't see why 'IBM PC models is bullet pointed.
• I don't think the character set needs to be included in an encyclopaedia article.
• I don't think the tables under 'IBM PC and PS/2 models' actually tell the reader very much at all about IBM PCs. If you really feel they're necessary, I would recommend some text explaining them, and putting them in collapsible boxes, but I don't think the article would be worse off without them. Worldtraveller 15:03, 19 July 2006 (UTC)

• The bullets are changed for a list of page links of models only. Remaining text is reformatted as paragraphs.
• I zapped the char set table and added a sentence referring to the code page instead.
• The PS/2 table and PS/2 info was removed as the relevant info is on the PS/2 page.
• The trivia section was removed and replaced with a section called Longevity. A note on the continued use of old IBM PCs is included, absorbing the example that was in the trivia.
• The tables under the "IBM PC models" summarize some basic details and seem fine to me. Placed in collapsible boxes.

Michael Daly 21:44, 10 January 2007 (UTC)

Flowery crap

"The 'solution' of Third Parties came as an answer to a maiden's prayer."
I don't know exactly what the wikipedia term for this sort of nonsense is (NPOV?), but it probably wants a bit less of it. —Preceding unsigned comment added by 160.5.98.139 (talk)

This is characteristic of User:Futureobservatory's contributions to date. He puts a lot of personal speculation and analysis into the articles he edits. The sections he has added need to be reviewed by other editors for NPOV. Gazpacho 18:35, 18 September 2006 (UTC)

Please do review my contributions. As should be obvious, they are abstracted from my trade books and are, as such, more populistic. I keep on trying, but I will be delighted if your editors also can knock them into more suitable shape. Futureobservatory 10:37, 1 October 2006 (UTC)

Price

Does anyone know what the original price was? Other PC pages (such as ZX-81) give some indication. I would find it interesting. Google shows that this page did used to have a price, but I can't see any record of it in the history. Thank you.peterl 01:49, 25 October 2006 (UTC)

Straight from IBM's website:

"On August 12, 1981, at a press conference at the Waldorf Astoria ballroom in New York City, Estridge announced the IBM Personal Computer with a price tag of $1,565."

http://www-03.ibm.com/ibm/history/exhibits/pc25/pc25_birth.html

I will add this to the article. Axeman (talk) 02:55, 26 May 2010 (UTC)

This is a prime example of why entries in this site are routinely deemed unreliable - the only surprise being that one would have thought the entry on the IBM PC, the grand daddy of today's PC's, would be somewhat accurate. —The preceding unsigned comment was added by 68.161.172.81 (talk • contribs) .

Of course, what is even funnier, for someone who was around at that time is that IBM had the PC on sale at ComputerLand for $1,565, ( The 166/5 Slot version in July of 1981, shipping with DOS 1.0, and when Aug 12 rolled around, the update to DOS 1.1 was already available, along with MASM, the macro Assembler. have YOU been to Building 12 @ Boca Raton? Hmmm?

And since I am speaking from direct personal experience, That was my older brothers system, When it came time to put together my system, four years later, after both the introduction of the IBM XT, and the conversion of the PC to the 64k/8 slot motherboard, I opted for a "turbo" clone. Also the IBM XT/286 was not an AT but an XT/286. So there were actually 3 models of the IBM PC. ( pay good attention here..) The PC (16k/5 slot ), THe PC (64k/8 slot ), and the PC (256k/8Slot). To actually verify this information, you would need to dig through thousands of IBM Product releases.

I changed my mind. Anyone who wants to see all this information should look into Muller, Guide to Repairing and Upgradeing PCs, 1~4th edition.— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 07:20, 20 February 2011

The IBM XT came only in the 8 slot versions, and there were only three models of that, the XT (64k/8slot), the XT(256K) and the XT/286.

Consider the RAM numbers, if you divide 1MiB by 8, you get 128, exactly 2 64k segments. So there were 16x64k segments.

Ponder what I have written...— Preceding unsigned comment added by 71.135.168.118 (talk • contribs) 03:07, 18 February 2011

Here is the citation for the Ram configurations,

[1]

16Kib Motherboards ( 5 slot, 64K Max ) and 64Kib Motherboards ( 256Kib Max ), the 640 Config? It was had by a 384Kib Short card.

From that site: "64 KB (The very first ones had only 16 KB), 256 KB max. (then later 640 KB max.)"— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 12:12, 18 February 2011

The article states "After the IBM XT shipped, the IBM PC motherboard was configured more like the XTs motherboard with 8 narrower slots..." No IBM 5150 was ever made with anything other than five slots - not when the XT came out and not ever. The slot size, number and layout never changed on any 5150. This was one of the main features that seperated the PC and XT; the XT had more expansion capability. PC motherboards came in two types: a "16-64k" version and a "64-256k" version; both describe how much RAM was supported on the motherboard. That, and the BIOS version was the only technical difference between 5150 motherboards. Nothing more. — Preceding unsigned comment added by 174.130.84.213 (talk) 16:51, 1 March 2012 (UTC)

Irrelevant item

The current third paragraph (on Lenovo) is irrelevant here - this is about the PC and history, not IBM as such and certainly not about Lenovo or recent events. Even if it were appropriate, it's introduced in a completely dislocated way. It would be best to remove it but it should at the least be rewritten. 216.77.225.52 19:19, 21 November 2006 (UTC)

And redundancy: the trivia about the National Weather Service occurs both in Trivia and in the main body. 216.77.225.52 21:56, 21 November 2006 (UTC)

Was IBM BIOS open or proprietary?

This is discussed at: http://en.wiki.x.io/wiki/The_20th_century_in_review

IBM decided on an open architecture so that other manufacturers could produce and sell peripheral components and compatible software. The ROM BIOS source code was published. IBM did not anticipate that its competitors would find ways to legally duplicate the entire system.

and discussed further at: http://en.wiki.x.io/wiki/IBM_PC

IBM made all specifications for their computer open rather than proprietary, with the exception of their BIOS. As the only impediment to an open system with interchangeable suppliers was this BIOS, it was reverse-engineered by Compaq, and the IBM PC became the first fully open-specification computer system, leading to its current dominance in the marketplace. Riding on this wave of popularity, the operating system vendors for the PC (Microsoft) leveraged their position to become the most powerful software company in the world.

--Flsaisalie 22:50, 16 December 2006 (UTC)

Answered at Talk:Bill Gates. Gazpacho 00:23, 17 December 2006 (UTC)

It was proprietary in the sense that IBM asserted intellectual property rights. My recollection is that the BIOS listing in the technical reference manual not only had a copyright notice, it had a conspicuous copyright notice and, I think, other warnings as well.

It was open in the sense of "not being hidden or secret," i.e. anyone could buy the technical reference manual.

Obviously, Compaq, Phoenix et. al. would not have had to perform complex and expensive "clean-room" cloning of the functionality of the BIOS if the BIOS had been "open" in the sense of being GPL-licensed or anything like that. Dpbsmith (talk) 03:12, 17 December 2006 (UTC)

The article specifically says "reverse-engineered", which is neither technically nor legally necessary if you can just go out and buy a ROM listing.

It was nessessary to BlackBox/Reverse engineer if you want to sell machines with compatible code. -PC_GOD— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 12:31, 18 February 2011

88.217.110.55 (talk) 23:51, 16 May 2008 (UTC)

Now, think about this - even if every page of the listing didn't carry a copyright notice, unless IBM specifically disclaimed the rights, it would still be copyright infringement to duplicate the BIOS from a listing. Just because you can buy Stephen King's latest off the shelf doesn't give you permission to print your own copies. I recall reading that what was done is Team A went through the BIOS listing and identified every entry point and subroutine and what they did, wrote this out as a specification, then handed it to Team B who wrote code that did the same thing. Compaq didn't get sued, so IBM must have been happy that this was sufficient intellectual effort that their ROMs were not just being copied. --Wtshymanski (talk) 01:17, 17 May 2008 (UTC)

It was called Black boxing, Team-A wrote the specifications, Team-B specifically without the knowledge of IBMs code, wrote their own code. Phoenix and AMI both did this, at first Pheonix was more successful, but later on AMI teamed up with more motherboard vendors, and dominated the market. -PC_GOD— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 12:31, 18 February 2011

IBM PCs BIOS was Open in the sense you could read it, and open in the sense that as you could read it, you could easily modify it, and several disassemblers were available that could disassemble the ROM BIOS. but it was proprietary that you could not offer a computer for sale on large scale that infringed significantly on the copyrights which were enforced. It was a mix of both that only the largest computer manufacture with its extensive legal and marketing department could come up with. Its not easy to classify it, but only to observe what the technology and legal organizations were doing. -PC_GOD— Preceding unsigned comment added by 69.232.205.64 (talk • contribs) 04:26, 2 March 2011

Cassette

Did it use a proprietary cassette drive or would any normal tape recorder work? How were tapes loaded? Were there special MS-DOS commands to deal with tape functions? Bastie 06:09, 26 February 2007 (UTC)

The cassette port was only used in Cassette BASIC - the BASIC that loaded from ROM when the computer had no way of booting from floppy or hard disks. The versions of BASIC on the DOS disks (BASIC and BASICA) did not support the cassette. IBM did not offer an 'official' cassette deck, any third party one with the correct connector (common with computers of the time) could be used. Howard81 16:13, 26 June 2007 (GMT)

That DIN connector was not just used with computers' cassette drives of the time, it was common with audio cassette players. IBM anticipated that many people would buy their model 5150 in its cheapest configuration (no floppy drives, no monitor) and connect their preexisting regular cassette players and TV sets. That's not what happened, however. 86.56.41.93 (talk) 04:58, 17 July 2008 (UTC)

A late addition - I recall using the MOTOR() command in GWBASIC (or was it BASICA? 20+ years ago...) under a later version of MS DOS - booted from a Hard Card 10 Meg disk - so I think that some diskette-based versions of BASIC still supported the cassette port. I always thought of that as IBM's attention to backward compatibility - how else would you get cassette programs onto a disk machine? --Wtshymanski (talk) 17:55, 11 January 2008 (UTC)

There was a hack to poke memroy locations to get the program to be copied to disk, but there never was a supported way. ( at a computer lab we worked at, we found the hack in PC Tech Journal, and was able to resque a great many beginners who put their boot floppies in Drive B, and started programming in Cassette basic. ) Disk Basic, and Advanced Basic, could turn the motor on and off, but could not write to the cassette tape, unless... you hacked it.

The answer to your question is that GWBasic did not have a Motor command, and with BASIC.COM and BASICA.COM, the motor command did nothing. Of course BASIC.COM and BASICA.COM required Cassette Basic to be in ROM, and would not run on clones. Wait I have GWBasic right here, Ill test it. ... ... ... GWBasic 3.23 Does NOTHING with a MOTOR command. It accepts it if you pass a number between 0 and 255, anything else gives you an Illegal Function call.

None of the documentation lists either a NON-OP or Diffrences for the Motor command in GWBASIC.EXE —Preceding unsigned comment added by 69.232.192.143 (talk) 08:22, 20 February 2011 (UTC)

The BASIC manual, provided by IBM, lists compatibility for the MOTOR command, as MOTOR is available in 3 versions of Basic, Cassette Basic, Disk Basic (BASIC.COM) and Advanced Basic (BASICA.COM), but not Compiled Basic (BASCOM.EXE)

And just for the record, BASIC.COM and BASICA.COM ran in my clone with copied ROMs :Ppppppt!

The PC BIOS contained some special routines for handling the cassette. Programmers called INT 15h, function 00h and 01h to turn the cassette motor on and off respectively. INT 15h, function 02h was used to read data and 03h was used to write data. (Source: MS-DOS encyclopedia) Retron (talk) 13:18, 11 January 2008 (UTC)

After reading the BIOS listings from the IBM Tech Ref, the calls to Int 15h are all there, but the Int15 routine is NOT documented. It's functions are listed in the DOS programmers guide, for which the MS-DOS encyclopedia essentially reported on verbatim.

If you would like to see how BASIC.COM and BASICA.COM aside from they were almost fully compatible with GWBASIC.EXE, There was a program called MOVBAISC.COM,version 5 that copied the ROMSs into an executable file, with BASICA.COM that would allow the interpreter to run without BIOS calls, and it offered a significant improvement, ( all the function calls were from RAM ). As an intended/unintended consequence, it allowed BASICA.COM to run both on any version of DOS, and any machine. You can easily find it, if you look. I found it in about 3 mins. Also available were simulators for both the z80, and S/370 processors that still function on Windows XP/6. 04:32, 2 March 2011 (UTC)

Wow! I found the S/370 processor emulator. There is a repositary of LanPack benchmarks that was able to be assembled on the S/370 emulator, and although its interpreted, because its so small, it fits in cache memory, and really flys. ( I was running S/370 software on a MacBook Pro :) --16:00, 3 August 2011 (UTC)

This? MOVBASIC.ZIP 69.232.193.99 (talk) 02:54, 3 March 2011 (UTC)

We were able to run this on an AT, and it did indeed do what it was supposed to. --107.43.122.169 (talk) 16:00, 3 August 2011 (UTC)

Objectionable renaming

Could we put the article back where it was? The suffix "original" is unnecessary, the original article title was perfectly un-ambiguous. --Wtshymanski 21:41, 11 March 2007 (UTC)

KB / KiB Edit War

This is from WP:MOSNUM

"The use of the new binary prefix standards in the Wikipedia is not required, but is recommended… " "If a contributor changes an article's usage from kilo- etc. to kibi- etc. where the units are in fact binary, that change should be accepted."

It appears that you have to be a contributor to the article to have any say in this choice. A "Drive By" editor can not make the decision. SWTPC6800 00:18, 30 March 2007 (UTC)

This is clearly a matter of momentous importance and worth fighting a major edit war over. Drutt 08:17, 16 April 2007 (UTC)

What a nice article. Pity about the kibibytes, though. --Wtshymanski 17:44, 14 May 2007 (UTC)

Please notice the kibibyte ROMS in the megabyte address space. How many kibibytes fit in t oa megabyte? --Wtshymanski (talk) 13:57, 14 April 2010 (UTC)

As a more than drive by editor of this article, I have changed as many references to the new naming convention as possible. Although I am extraordinarily familiar with the old terminology and its limitations, I much prefer the new terminology in terms of clarity and readability. Note: The war is over, and we lost. Kib it is. Can someone put a link to the KiB page?

Here it is: [2]

If someone is going to list it the old way under the picture, he or she should make sure it's with a big K (to imply 1024 bytes) instead of the little k (which implies 1000 bytes). — Preceding unsigned comment added by 199.96.38.41 (talk) 20:47, 5 August 2011 (UTC)

IBM 5100

Wasn't affordable and thus not a Personal computer. Also it's hardware and the 5150 were not related and the 5150 was a totally new design from the 5100. Alatari (talk) 17:19, 18 January 2008 (UTC)

It was a personal computer in the sense that you didn't need to go through a professional staff of operators to use it - remember, 1975 was *very* different from 2008! "Affordable" is a relative term...in 1975 your department head might not have been able to wrangle more time on the company's mainframe, but could afford $20,000 or so to get your very own 5100 that didn't need to be run through the corporate data center. --Wtshymanski (talk) 17:41, 18 January 2008 (UTC)

I agree. Historically, personal computer and personal computing has been in regards to a computer designed for single person or personal use and direct interaction, as opposed to a mainframe or mini and timesharing environments. Likewise as Wtshymanski said, such systems would not require a staff to operate, including submission of materials for programming (i.e. punch cards to a staff who then puts it in to a card reader). The definition of a personal computer has never revolved around afforadability - only usability. --Marty Goldberg (talk) 21:19, 18 January 2008 (UTC)

Right. I get the impression many Wiki editors have never had to use a keypunch and stand in line to submit their decks. --Wtshymanski (talk) 16:40, 19 January 2008 (UTC)

If Steven Jobs can call the Cube a Supercomputer, then IBM can call the IBM 5100 a personal (albet luggable) computer.

FYI shall I put up the List of IBM 51x0 series computers? —Preceding unsigned comment added by 69.232.207.6 (talk) 02:57, 23 February 2011 (UTC)

Note on the Keyboard

Imho the article fails to take notice of the problems which arose by using the IBM standard keyboard in non-English countries which naturally used letters which were not available on the English-only standard keyboard (remapping of keys with TSRs).

I hope that someone can enhance the article with that. Unfortunately I'm not an expert in this field, so I won't dare.

Alrik Fassbauer (talk) 16:41, 21 February 2008 (UTC)

The "problem" you mention is not specific to the IBM 5150 and didn't require TSRs for remapping (unless you consider the MODE command a TSR, which is probably technically incorrect). The issue of different character sets was addressed via code pages. Problems still arose if programs assumed the use of CP 437 on international systems which used different code pages, e.g. CP 850, but in any case, code pages weren't the problem per se but an (imperfect) solution. Code page issues also aren't really particular to the IBM PC/XT keyboard. This information probably has no place in this article (there already are other articles on code pages and DOS; and keep in mind that DOS wasn't the only OS one could run on the 5150). I do agree that people who are basically only speculating should probably (apart from simple copyediting, etc.) refrain from contributing technological particulars to this article, as half-knowledge is likely to do more harm than good. 31.16.129.11 (talk) 03:18, 26 October 2011 (UTC)

Font

Was there any specific name given to the font? If so is it used today? Was it like times new roman or arial or sans serif? That's pretty important since I'm assuming there was only one font. 69.251.82.213 (talk) 18:39, 1 June 2008 (UTC)

The exact font depended on which video card was used; the monochrome display adapter had more pixels in each character cell than the Color Graphics Adapter. A Hercules card could display a lot of different fonts using software. I don't recall the font having a name. It was never used for printing, only for the screen display - dot-matrix impact printers of the era had their own wretched caricatures of fonts. --Wtshymanski (talk) 21:48, 3 June 2008 (UTC)

IBM did not give a name to the font, until much later. They named it the EGA Font, Differing from the VGAFONT commonly found on windows machines.

The Hercules monochrome adapter, could emulate both the MDA and the CGA, as well as have a few more modes, and a few more fonts.

They even released both a font program, and software that had the ability to use fonts in RAM. VGA explicitly added this ability.69.232.205.64 (talk) 04:59, 2 March 2011 (UTC)

"Here is the skinny on the later fonts. The EGA uses the same font as the MDA. The MCGA and VGA use the same new font, with letters in a 7x10 matrix. These letters are extremely close to the MDA/EGA fonts, usually with an extra line added to each letter. (MDA/EGA usually adds an extra two lines to each CGA letter.) If the column width is 8 pixels, the letter will look normal. If the width is 9 pixels, then the extra pixel column will duplicate whatever is in the 8th pixel column." [3]

—Preceding unsigned comment added by 69.232.192.143 (talk) 08:33, 20 February 2011 (UTC)

Monitor

This is a forum thread, but maybe this info, particularly the info at post number 4 and 5, could be incorporated into the article? 86.56.122.190 (talk) 08:42, 22 June 2008 (UTC)

Its good information, and actually much more reliable than the authors of the article here, but still a few more points of confusion:

After the IBM XT was released, the IBM PC was using an 8-slot motherboard, exactly based on the XT, but with a 63W Power supply, and with C1.10 Cassette BASIC.

And Diablo Valley College had at least 60 of these types that could use the IBM Network. But if you left the floppy door open, you got casette basic.

The final revision of the XT ROMS supported 720kiB floppies, but the Bus/controller could not support the throughput to format or read HD 1.2/1.44 Floppies ( No 15 Sectors per track )

A few aftermarket cards added a faster disk controller, and buffering so that the PC could read HD floppies.

Pacific-Data made a 80Tpi 720kiB HD floppy as an external drive, but again, only 8 sectors per track. ( HD number of tracks, LD number of sectors )

Although you could technically count Bernoulli boxes as Floppy disks...they plugged into a hard disk interface. ( 5MB and 10MB ). ----~~ —Preceding unsigned comment added by 69.232.207.6 (talk) 08:13, 25 February 2011 (UTC)

First personal computer

The article speaks about first personal computer being the Palo, but there was 1949 the Simon what is first personal computer, it fits perfectly to the personal computer category. http://www.blinkenlights.com/pc.shtml I think we need to edit the article about this and then the "PC" article to mean only personal computers and not IBM PC. There is so many overlappings of these multiple articles that would need to be cleaned up based to historical and technical facts Golftheman (talk) 08:46, 6 May 2009 (UTC)

That is bretty much correct, the first personal computer was Simon and first PC was IBM's PC. But there is lots of confusion that PC means exactly same as personal computer. The IBM's project was Project Chess name and they were building first personal computer of IBM what they called "PC" and it was marketed as "IBM PC". Too many just thinks that PC's were before IBM PC but that is just wrong. 62.165.184.109 (talk) 12:37, 31 May 2009 (UTC)

Why does the term Personal Computer have to apply to electronic discrete digital computers? ----~~

IBM PC

The IBM PC was as much in the 1980s the notion of what a computer ought be, as things like a Wintel machine is today. The point is that the cloning efforts on PCs was targeting the IBM machine. By the end of the 1980s, even IBM was no longer making genuine IBM computers.

Much of this is in the transition from 8-bit computers to 16-bit machines. The small PCs and many of the early game consoles and portable computing was still largely 8-bit machines. When users were allowed to access programs directly, it was by means of BASIC.

Because of anti-trust settlements, IBM had to out-source most of the parts for it, including the operating systems. However, they did retain the BIOS, with its built-in BASIC. ROM-BASIC never got past version 1.1, even though BASIC grew to 3.4 or 4.0 (they're the same).

The IBM PC became popular, because it was well supported, and because it had a growing software market (eg Lotus 1-2-3). It thus becomes a suitable thing to target. Since most of the parts were freely available, it was the BIOS that was reverse engineered, eg by AMI, AWARD, and Phoenix. It is these BIOSes that were sold to third-party manufacturers like Compaq and Dell.

In the 1980s, the subject of the computer articles attention was IBM-DOS, and what one might do with this. MS-DOS was largely a side issue. DOS 4 was the last version of DOS in the 1980s. It was IBM's test bed for ideas that would ultimately make their way into modern operating systems, like OS/2 and hence Windows NT.

By 1990, computing had largely changed. New users were being introduced to computers, and they were sufficiently powerful to displace game-consoles like the Atari, etc. Machines typically shipped with DOS menu programs, that users could configure to launch Word processors, games, and other applications. DOS was largely avoided, even before Windows.

Digital Research was already in the OEM market, and had just released their retail DR-DOS 5. Windows 3.0 had been hugely successful for Microsoft, and from here they decided to ditch the arrangement with IBM, and go it alone. Computers had largely shifted away from the enthusiast, and being main-stream, programming in either BASIC or its derived COMMAND.COM language had largely disappeared.

IBM had stopped making IBM-Machines, and had joined the rest of the clone-makers.

It leaves some legacy, that would filter through to current: IBM-compatable, IBM formatted floppies, PS/2 keyboards and mice, still are with us. But cloning IBM Machines is a thing of the past. --Wendy.krieger (talk) 09:54, 13 November 2009 (UTC)

True enough, but why did you post this on the talk page?--Anss123 (talk) 01:59, 14 November 2009 (UTC)

Also, true to this day, a PC Compatible can still only have a maximum of 4 primary partitions. -PC_GOD— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 12:46, 18 February 2011

Debatable

Once the IBM PC became a commercial success, the product came back under the more usual tight IBM management control. IBM's tradition of "rationalizing" their product lines, deliberately restricting the performance of lower-priced models in order to prevent them from "cannibalizing" profits from higher-priced models, worked against them.

Oh, really? Has anyone seen an analysis saying that IBM would have done more to improve shareholder value by allowing low-end machines to poach market from higher-cost (and higher-revenue-yielding) machinery? The purpose of IBM is not to make computers, but to make money. It's not clear to me that IBM's strategy wasn't optimal; notice that they got out of making desktop PCs when the price point sank, then unloaded laptops when those got cheap. Let's see what the business schools say, not the hacker columnists. --Wtshymanski (talk) 15:06, 9 December 2009 (UTC)

Your talking about the IBM Corporation, that until 2005, had lost the largest amount of capitalization known to mankind? In 1992, IBM's 4th Quarter loss topped 5.4 billion.:

"IBM has been unable to manage the rapid technological changes in the computer business, especially the dramatic shift away from large, expensive computers based on proprietary designs and toward cheap, generic and ever more powerful personal computers." LA Times.

Its so clear that for once even the LA Times got it right. but ... lets look on...

from Wikipedia:

"However, IBM soon lost this early lead in both PC hardware and software, thanks in part to its unprecedented (for IBM) decision to contract PC components to outside companies like Microsoft and Intel. Up to this point in its history, IBM relied on a vertically integrated strategy, building most key components of its systems itself, including processors, operating systems, peripherals, databases and the like. In an attempt to speed time to market for the PC, IBM chose not to build a proprietary operating system and microprocessor. Instead it sourced these vital components from Microsoft and Intel respectively. Ironically, in a decade which marked the end of IBM's monopoly, it was this fateful decision by IBM that passed the sources of its monopolistic power (operating system and processor architecture) to Microsoft and Intel, paving the way for rise of PC compatibles and the creation of hundreds of billions of dollars of market value outside of IBM."

Looks like IBM failed to both produce profit, and shareholder value, by both cutting off the beginning and rise of the Personal Computer. They, and this is in hindsight, would have created enormous shareholder value had they purchased both Microsoft and Intel, for... CASH ON HAND.

But desipte these shortcomings and more, they held on, and despite their mistakes, are still in business. —Preceding unsigned comment added by 69.232.205.64 (talk) 09:42, 2 March 2011 (UTC)

IBM Closes PC plant in Boca Raton

I was at Rodime in Boca Raton directly across the street from the IBM personal computer plant. We manufactured hard drives for their PC and the Apple. At the time they were using what they called pc-dos. DOS had been around for many years as we had used it on Qantels minicomputer years earlier. It should be noted that IBM said "there was no future in the personal computer" and closed their plant.This was,I believe in 1983, and left the 8088 open to any source. IBMs' error opened the way for Intel and Bill Gates as the 8088 spread rapidly and allowed other companies to make sound, video and other boards for this computer. I hate hearing the word IBM clone as they gave up their rights when they closed the plant. At the time they were the business computer company and sold their mainframes for millions of dollars. I believe they felt they could squash the personal computer with their public opinion thus eliminating a threat to their mainframes. When this did not work they came back a few years later with what they called microchannel hardware which was a gigantic flop and died within a year. This information should be included in the "IBM Personal Computer" Wikipedia page.

Ref.: Rodime PLC., Intel, Microsoft, Qantel Business Computers —Preceding unsigned comment added by Mastersro (talk • contribs) 19:51, 19 December 2009 (UTC)

The above paragraph is utter nonsense. Without going into detail, the Boca Plant was moved the Research Triangle Park, NC and that happened later than 1983. The Mexican plant was not built to replace Boca and did not supply the US. It was was brought about because of Mexican pressure. The rest of it is rubbish too. Terry Ex-IBM PC Company 86.134.79.51 (talk) 16:12, 25 July 2011 (UTC)

From this source: http://en.wiki.x.io/wiki/Timeline_of_x86_DOS_operating_systems

IBM was still selling IBM PCs/XTs and ATs up until the introduction of the PS/2 systems, the higher end which had microchannel architecture, the 50, 60, and model 80.

from IBM: [4]

It was in 1987 that IBM moved their manufacturing to Mexico.

"ESD Boca Raton laboratory becomes responsible for all U.S. hardware development for PS/2 products."

(ESD or Entry Systems Division was the division that designed and built the PC— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 12:59, 18 February 2011

IBM PC cassette recording signal

This were deemed to technical, but here it is anyway, should anyone need it for reconstruction: The technical reference for the IBM PC 5150 specify that the WRITE-BLOCK routine turns on the cassette drive motor and transforms each byte into bits where a (1) bit corresponds to a 1.0 ms timer period, (0) bit corresponds to 0.5 ms (1000 - 2000 bit/s). First 256 bytes of "11111111" is written. One synchronization bit "0". A synchronization byte of 0x16. 256-byte blocks of data and a 2-byte CRC is written until all data is transferred. Electron9 (talk) 04:51, 25 December 2009 (UTC)

References

5150?

Does anyone know why or how the number "5150" was chosen for the first IBM PC? Inside joke? Coincidence? —Preceding unsigned comment added by 173.58.251.147 (talk) 20:55, 22 June 2010 (UTC)

It followed the 5100 nomenclature. The 5100/5110/5130 were all Entry Systems Division products,

So, the IBM PC was the successor to those machines, hence the 5150. Later it became the police code for insane.— Preceding unsigned comment added by 69.232.192.143 (talk • contribs) 13:01, 18 February 2011

Right, so the IBM System/23 was model 5130?? One site says it was 5322. – Wbm1058 (talk) 00:54, 17 September 2014 (UTC)

Citation NOT needed|date=June 2009

Removed Citation for IBM 801 Being an order of Magnitude more powerful than the 8088. At the time, in 1980, the ( If you had bothered to read the IBM 801 Article ... ), 801 Processor was benchmarking at 15Mips, vs the 8088 was benchmarking at 0.33 to 1 million ( from the 8088 article, ) thus the statement was an UNDERSTATEMENT. 15Mips is 15 to 45 times more processing power, hence it was chosen as both the processor to develop for IBMs workstation effort ( the IBM RT ) and its Workstation Line ( the RS/6000 ) and its eventual dominance of workstation Processors. Changed Wording of article to say 'More than an order of magnitude' :) Citation is Wikipedia's own articles on the 801 and the 8088. The ROMP processor that was the 801 spinoff had a 5.6 Mips benchmark, which would be 6 to 17 times faster.

Benchmarking IBMs AT against the IBM RT, the RT, WITH THE PROPER COMPILER, was easily 6~10 times faster. Didn't some tech magazine have benchmarks on it?--107.43.122.169 (talk) 18:20, 3 August 2011 (UTC)

Yes. The IBM TECH Journal, had an article called 'A Significant Departure' which ran benchmarks of the RT against the PC co-processor card. THe article appeared in December of 1986. — Preceding unsigned comment added by 69.38.208.214 (talk) 22:12, 23 April 2012 (UTC)

More bitspam...DOS 1.0 NOT superior to AIX 1.0

Who wrote this? "the operating system many years more advanced than the DOS operating system from Microsoft[original research?], that was finally selected" The Original Operating system AIX was a unix/BSD derivative, the type of which eventually inspired GNU/Linux, so how much research does it take to say that the operating system had features which became popular 20 years later? Compaired to DOS 1.0 which was CP/M like? The AIX operating system had directories, processes, multi-tasking, a Virtual Resource manager that would show up as a feature of Windows NT, in a few years? The IBM Tech Journal in its article on the IBM RT, called 'A Significant departure' called the AIX Operating system, a technological leap, bringing the power of minicomputer software to a desktop computer.

DOS was so primitive that it barely qualified as an operating system. There was no Memory Management in DOS. AIX is a sophisticated operating system and it is foolish to compare DOS with AIX. Sam Tomato (talk) 21:08, 16 December 2013 (UTC)

Nonsense about AIX

Why is AIX even mentioned in this article? AIX could hardly have been a candidate OS for the IBM PC in 1980, when it was first released with the IBM RT in 1986. If IBM had been considering a Unix OS for the PC, Microsoft's Xenix (released in 1980) would have at least have been a potential candidate (though any Unix, including Xenix, would have required more powerful hardware). AIX certainly wasn't. Shalineth (talk) 15:21, 15 April 2011 (UTC)

I have a copy of Xenix for an IBM PC XT/286. Young people would be surprised how small an operating system can be without a GUI. A GUI consumes huge resources, comparatively. Xenix worked well with a 20 MB hard drive and 2 MB of main memory. Sam Tomato (talk) 21:17, 16 December 2013 (UTC)

PC/IX (Personnel Computer Interactive Executive) was IBM's first UNIX implementation and was announced by IBM on the PC XT in 1984. So UNIX obviously didn't need more powerful hardware. MS Xenix was announced for the IBM PC AT concurrently with the hardware later in 1984. Source: I was TS for PC/IX in the UK. Terry July 2011 — Preceding unsigned comment added by 86.134.79.51 (talk) 15:57, 25 July 2011 (UTC)

AIX is mentioned in comparison to selection of Operating Systems. AIX is ATT System V based, and SCO/Microsoft Xenix was Also System V based, but, I would agree, from first hand experience that They require a processor that supports virtual memory. --71.202.7.188 (talk) 18:23, 6 July 2011 (UTC)

Xenix on the 8088, had virtual memory without the hardware, and because it lacked the hardware for memory managment, and the disks were slow, when you had to swap, the term 'glacial' was used. The use of a second hard disk improved the situation, if it was dedicated to swapping. Xenix 2.1 came in XT versions and AT/286 versions. A SABRE program trainer, which required Xenix, literally sped up 10 times with virtual memory in the hardware. ( It ran in 286 mode completely ). You dont need a screen to use a computer, but it helps a lot. You dont need hardware virtual memory to run UNIX, but it helps tremendously. ( There were stories about running a tiny unix on floppies... SCARY...) --107.43.122.169 (talk) 16:16, 3 August 2011 (UTC)

The Facts, and er.. this article...

"The IBM PC's ROM BASIC and BIOS supported cassette tape storage."
According to the IBM PC Tech Reference, and the BIOS listings therein, its the COS or Cassette OS that supported Cassette Tape Storage.
The BIOS called Cassette BASIC if no bootable media was found, ( i.e. if your floppy door was open, or the Hard disk could not boot ),
Cassette Basic, using Int19, ( Documented in the DOS Tech Reference ), called the Cassette OS to actually write data to the tape port on the back, where hopefully it was recorded.
( The error handling was not exactly well through out... )
What I am looking for now as proof is a Dis assembly of the ROM for Int19... ----~~

More on the Facts... Ahem...according to InforWorld...( somewhat of a popular press newspaper for computers at the time...

"In spite of the sale of hundreds of thousands of MCA machines, and the announcement of EISA architectures, Computer Manufactures exhibiting at Comdex will be strongly supporting the future of the 386- and 286-based architecture AT architecture - ISAISA architecture is 16-bit

For what it is worth, an original IBM PC (at least the XT/286 model) could boot from a keyboard. Obviously it would have to be a special keyboard. I know because I found the code in the BIOS. Sam Tomato (talk) 21:21, 16 December 2013 (UTC)

11111 days old

Just thought I'd point out that the IBM PC will be 11111 days old on the 12th of August 2011, that's binary 31 years I guess.

Cheers to IBM!

Whitehatnetizen (talk) 04:01, 2 August 2011 (UTC)

NEC V20 CPU.

Aside from the slight processing speed boost, the other reason to swap an 8088 for a V20, or an 8086 for a V30, was the NEC CPU also included a Zilog Z-80 compatible CPU in the same package. There were Z-80 emulators for running CP/M which could detect the presence of the NEC CPU and use the real Z-80 instead of emulating it. Thus a PCjr with an NEC V20 could run CP/M software faster than a 80286 with the same program emulating the Z-80. One of those emulators was called 22-Nice. The company also produced a companion program called 22-Disk to enable a PC to read and write many different CP/M diskette formats. Bizzybody (talk) 06:52, 12 August 2011 (UTC)

Almost correct. It did not include a Z80, but it did effectively include an Intel 8080. CP/M only requires an 8080 to run. It will run on a Z80 but does not specifically require it unless a computer supplier has used Z80 code in the BIOS part of their CP/M implimentation (such as Commodore did for the Commodore 128). I B Wright (talk) 16:20, 24 February 2013 (UTC)

add a picture of the computer while it's working?

a suggestion: add picture of how the screen looks like when the computer works and a section that explains what you can do with this computer.

i know nothing about old computers (since i was born more that a decade after this computer was introduced) so i can't contribute to articles from this kind-- Someone35 (talk) 14:36, 12 August 2011 (UTC)

It's hardly possible to come up with a representative screenshot that shows what things look like when the computer works, especially since way back then, there wasn't yet a ubiquitous GUI on these machines, so most DOS applications were fullscreen and had their own baked-in user interfaces (many of which were very different from each other). Probably the most "representative" screenshot possible would be a DOS command prompt, maybe showing a completed dir command. But even that doesn't explain what you can do with this computer. So basically, what you're asking is impossible. If you want to learn about the original PC, start by learning DOS. If you can't afford buying an old PC and original software, you could use the free DOSBox emulator to install the likewise gratis FreeDOS on your existing PC. Your local library may still have many old computer books about the old IBM-compatible PC and DOS days. And you can google for more information. There is plenty of information out there about the olden days (despite the fact that these computers preceded the WWW by more than a decade). But just a single screenshot will not really help you in your learning task. Good luck.

Also, there are plenty of people who can contribute to this article, so it's actually better if you don't try to contribute to this article as long as you don't know about its subject matter. (Well, you can still do basic copyediting, but that's it.) And there's no shame in refraining from contributing here. We're all ignorant about almost everything. Find out what it is that you do know about, and contribute to those articles. 31.16.107.239 (talk) 10:56, 18 October 2011 (UTC)

Best thing would be to show a screen shot of a spreadsheet, which was the second most common useage, back in the day. — Preceding unsigned comment added by 69.38.208.214 (talk) 22:04, 23 April 2012 (UTC)

Modulator requirement

Regarding this edit:

Isn't it the case that NTSC sets didn't need an extra box (and that, if they needed a modulator, then that was built into all or most TV sets)? Also, isn't it the case that some extra converter/modulator boxes were available that could allow connecting the CGA card composite port to a PAL TV set? Can someone say this authoritatively, preferably with a citation?

31.16.107.239 (talk) 10:33, 18 October 2011 (UTC)

No. NTSC sets needed a modulator to change the composite signal, to modulate it into a channel, that could be recieved. ( the same was true of game systems of the day. ). The RCA plug on the CGA card, could drive composite Monochrome, or Color monitors, and PAL was not available. ( it could not produce the frequency. A friend of mine overseas was using a lab that orderd the computers, and monitors, but had to wait 4 or 5 months for the NTSC sets to arrive, while the PAL sets had to get shipped back. That type of multiple frequency was not available until the VGA cards with programmable RAM/DACs were availible. ) — Preceding unsigned comment added by 69.38.208.214 (talk) 22:02, 23 April 2012 (UTC)

IBM JX 5511

Shouldn't the IBM JX be added to the models section? It definitely belongs to the IBM PC line, just like the PCjr (which the JX was based on). Even if it wasn't sold outside of Japan, Australia and New Zealand, it's worth a mention. (This was also the first IBM PC derivative to feature 3.5" drives.) - 82.80.137.100 (talk) 03:56, 5 December 2011 (UTC)

Well, yes. It should. How did you remember this one. I thought that the BIOS was in Japanese though? and I thought it came out soon after the 3.5" drives were available both with DOS support ( Dos 3.2 ), and the IBM XT/286. ( That is where I saw them. ) — Preceding unsigned comment added by 69.38.208.214 (talk) 21:57, 23 April 2012 (UTC)

The BIOS was in 8088 machine language, it's text/strings were in Japanese, as well as the character set and keyboard, so it was a software modification, not necessary a new technology or model, but I believe it passes the test for notability so I would vote for inclusion. BUT we need a citation. — Preceding unsigned comment added by 173.200.183.146 (talk) 03:21, 28 August 2012 (UTC)

Can be any colour as long as it's big blue.

The article starts with a black and white photo (even though the caption states the particular IBM PC config depicted sports a CGA, therefore colour monitor). A bw photo is so unjust for the lead of this important article!

IBM PC changed the world more than wheel or fire or wheel on fire (vehicular mobility) did. There will never ever be anything comparably important in the future (save for affordable space warp drive travel, which scientists find unlikely).

Surely WP can pay better tribute to the PC. Why not put the Chaplin clone ad imagery in the lead? 82.131.210.163 (talk) 11:51, 27 April 2012 (UTC)

Macintosh only other platform with significant market share ????????????????

The article must be specific about the timeframe to which this statement pertains. Post turn-of-the-century only. Please see reference The final sentence of this article's introduction requires revision. Wikkileaker (talk) 20:13, 18 July 2016 (UTC)

References

Citation needed question

I'm a newbie, but I have some questions.

I wanted to remove the "citation needed" for the price of a used 5150 as I have found various sites that sell operational 5150s in the approximate range the author cited.

How would I go about this? Or would it just be better to remove the mention of prices for used systems? — Preceding unsigned comment added by ThorbyTech (talk • contribs) 07:19, 10 November 2016 (UTC)

real cause for IBM PC designed to Open Standards

According to my memory of events 36 years ago, the actual reason IBM designed the PC to open standards was the antitrust lawsuit filed by the Federal Trade Commission regarding IBM's business practices and resulting monopoly of the mainframe market. They didn't want to take a chance on being accused of trying to corner the microcomputer market, or scaring the stockholders. Take a gander at this cite. As you all can see, this suit was not dismissed until 1982, whereas the PC design activity took place 1980-81 for its 1981 rollout.

I strongly suggest someone further investigate this and make the appropriate edit(s). (sorry, I'm busy) I never bought this BS story about some Apple longhairs steering the business decisions of a bureaucratic behemoth like IBM. An extremely staid behemoth such as IBM was. Thank you for your consideration.Wikkileaker (talk) 20:18, 23 August 2017 (UTC)

References

I concur, although it's four years later, and I'm still not sure why Wikkileaker has no time to do this, "too busy" yet put citations here. If you aren't willing to do the work, don't bother here! Compaq was one of the first to encourage "Open Standards" as described in "Open: How Compaq Ended IBM's PC Domination and Helped Invent Modern Computing" by Rod Canion 2013. It was announced in September 1988 by a consortium of PC clone vendors (the Gang of Nine) as an alternative to IBM's proprietary Micro Channel architecture]] (MCA) in its IBM PS/2 series. The Gang of Nine was the informal name given to the consortium of personal computer manufacturing companies that together created the EISA bus. Rival members generally acknowledged Compaq's leadership..." The members were: AST Research, Inc., Compaq Computer Corporation, Seiko Epson Corporation, Hewlett-Packard Company, NEC Corporation, Olivetti, Tandy Corporation, WYSE Zenith Data Systems. In comparison with the AT bus, which the Gang of Nine retroactively renamed to the ISA bus to avoid infringing IBM's trademark on its AT Personal Computer. The above is mostly from other Wikipedia articles, which I specifically linked. I'm not "too busy" to check those references. I'm going to leave this unsigned because "too busy", and trolls might find time to retort with snarky comments, and I have no time for that particular type of exchange

External links modified

Hello fellow Wikipedians,

I have just modified 5 external links on IBM Personal Computer. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20071217120019/http://www.tmworld.com/article/CA187350.html to http://www.tmworld.com/article/CA187350.html
• Added archive https://web.archive.org/web/20080627010207/http://www.howard81.co.uk/upload/vcf/xt/newxt/xt_mainboard2.JPG to http://www.howard81.co.uk/upload/vcf/xt/newxt/xt_mainboard2.JPG
• Added archive https://web.archive.org/web/20110724180117/http://www.arl.wustl.edu/~lockwood/class/cs306/lecture/l16.html to http://www.arl.wustl.edu/~lockwood/class/cs306/lecture/l16.html
• Added archive https://web.archive.org/web/20110728124053/http://www.ctv.es/pckits/tutorial.html to http://www.ctv.es/pckits/tutorial.html
• Added archive https://archive.is/20130506223315/http://www.itdictionary.org/term/IBM_PC.aspx to http://www.itdictionary.org/term/IBM_PC.aspx

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 00:59, 10 November 2017 (UTC)

History of interrupt selection

Does anyone happen to know why the DOS designers ignored Intel's comment "The next 27 interrupt vectors are reserved by Intel" and started using interrupts at 8? It's slightly before my time. See also Talk:Intel_8259 Number774 (talk) 19:22, 15 April 2019 (UTC)

Total rewrite

With all due respect to past editors, who I'm sure did their best within the constraints of the time and energy they had available, this article desperately needs a total nuke and pave.

The article can't decide if it's about the IBM PC or the entire PC series, and consequently if you want to know how the PC - the IBM 5150 - is designed or what role it played, it's nearly useless since you have to filter out acres of chaff about other computers. We have articles about PC clones and about all the other IBM PCs that we can link instead of putting it all here.

The entire history section, while extremely well sourced - perhaps *over*sourced - is not at all encyclopedic. It reads partly like a romantic editorial about IBM's come-up in the industry, but mostly like a book about this specific subject, including every name of every player, major or minor, that anyone could find, tons of punchy quotes, and so on.

Everything is far too verbose. Commentary on the long tail of the PC de facto standard can be reduced to a mention since we have multiple whole articles on that topic and the greater PC compatible concept. I am working on a rewrite.Gravislizard (talk) 06:56, 3 October 2020 (UTC)

I have completed a rewrite. It is an absolute hatchet job from any perspective, which I think is unavoidable here. Trying to whittle *down* this article is extremely hard given just how much is here and how much of it depends on everything else, so I rewrote it nearly from scratch, pulling elements and refs from the current version. I'll cop to this version having fewer citations, but everything in here is provable, it just requires more time and effort to get fully sourced which I will try to do soon.

I deleted 90% of the history section, nearly all of the names except for the biggest players, all of the suspenseful dramatization, and cut it down to just the bullet points. It's possible at this point that some of the info is not supported by the references - I copied over what I could, but the visual editor chokes on many of these refs making copying them over tedious, so I picked what looked juiciest for each assertion. Also, there were so, so, so, so, so many refs, far more than I think we needed - four or sometimes as many as six to support a single minor event.Gravislizard (talk) 00:44, 4 October 2020 (UTC)

As the author of the bulk of the former History section, I don't disagree with the appropriateness of the abridgement. I do think that the former text is appropriate in a new History of the IBM PC article.

I do disagree with your pejorative description of it as "romantic" and "suspenseful dramatization". Rather, I disagree with your claim that well-written text is inherently unencyclopedic; please feel free to cite examples that broke WP:TONE. If there are too many citations remove some, but this is a better problem to have than the opposite. You have not yet restored all appropriate and necessary cites for the text. Ylee (talk) 03:55, 7 October 2020 (UTC)

I'll say first that I apologize for being dismissive of your work, which I'm sure was in good faith. I frankly assumed it was the work of many separate people adding bits here and there until it became what it was, not the intentional work of one person with a specific goal. Either way I simply should not have said it like I did. I'm still adjusting to the experience of editing the work of other people I don't know, and that made me nervous about whether I *was* stepping on someone's careful efforts, even though I felt it was justified, and I blurted out some uncalled-for comments. Next, I feel like this is largely a subjective issue, so my strongest argument is that when I read the article as it was, I didn't feel like I was reading a neutral statement of facts. That said, I've reviewed WP:TONE and I believe I'm on the right track, if indirectly.

I read that we should "avoid bombastic wording [..] pull quotes, journalese, and headlinese [and] persuasive writing," and WP:QUOTE makes a point that using many quotes can quietly undermine NPOV. The issue I saw was sort of a combination of these - each division of the history section starting out with a pull quote felt like a book or featured magazine article to me, setting an emotional tone for the next section, and the heavy use of quotations containing "bombastic" verbiage seemed to me to overwhelm the encyclopedic tone of the article. While the quotes themselves were not inherently invalid, nor was there necessarily a problem with the original writing in between them, I felt it just leaned too heavily on unfiltered individual opinions.

For instance: "Some IBM employees opposed IBM entering the market. One said, "Why on earth would you care about the personal computer? It has nothing at all to do with office automation"." MOS:QUOTE says to use quotes for, "emotive opinions that cannot be expressed in Wikipedia's own voice," but I think this could have been paraphrased, e.g. "Some IBM employees opposed IBM entering a market outside of office automation." It can still cite the same source, but the tone would not carry the feeling of anger that the quote did. We could say that this *is* a uniquely emotive opinion, but I think this requires some consideration of cultural context - 1980s businessmen probably spoke very emphatically about almost every topic, and I would wager that this man's opinion can be safely condensed to "opposed" without losing the gist of it.

The same goes for a lot of magazines, which tended to level either vitriol or outrageous praise at business decisions, because they were paid to have strong, firm opinions that made their readers feel reassured that they were getting a conclusive opinion on a subject. "IBM has devoured competitors like a cloud of locusts" is an exciting way to say that the market had swung heavily in their favor, but we can just say that, instead of quoting someone saying it in a more bombastic way. Magazine writing is certainly essential as a measure of the general public opinion, but I think using much of it verbatim can tend to make what was probably a fairly tame series of business decisions and market reactions feel like a fireworks show.

Finally there's quotes that perhaps look like facts at a glance, but might not have been. "Cary agreed about the culture, observing that IBM would need "four years and three hundred people" to develop its own personal computer; Lowe promised one in a year if done without traditional IBM methods." Was the quoted individual exaggerating for effect, or making a realistic estimate? It's a third-hand quote, so I'm not sure, and my take is that it's a safer bet to decouple from the individual statement, that was maybe expressed in a moment of frustration, and say e.g. "a typical development cycle for IBM could last several years and require a large team," which is still generally supported by the thrust of the quote without presenting it as a likely fact. This is a judgment call based on how you read the tone of the source, but like I said, I'm erring on the side of caution.

I hope my point of view looks more reasonable now. Regarding the missing cites, I was having some trouble validating them because the visual editor broke a bunch of refs, and now that I've figured out how to fix those I'll go back through and correct the damage ASAP.Gravislizard (talk) 07:13, 7 October 2020 (UTC)

WP:QUOTE does not recommend what you claim. Its advice against overuse of quotes is primarily for possible violation of fair use. It does say "Using too many quotations is incompatible with the encyclopedic writing style", but should be read in context with the next point, "Quotations shouldn't replace plain, concise text. Intersperse quotations with original prose that comments on those quotations instead of constructing articles using quotations with little or no original prose"; in other words, intersperse short quotations with original text. The previous text does this.

Your paraphrase of the "office automation" quote is flawed. The point of the author who quoted the IBM employee was that to claim that a personal computer has nothing to do with office automation is completely wrongheaded. This was already apparent by 1982-1983.

"IBM has devoured competitors like a cloud of locusts" is an appropriate way to convey the Apple magazine's author's purpose in using such language: Stating how quickly and efficiently the PC had taken over the market, *and* to warn to other people in the Apple ecosystem of this. MOS:QUOTE advises that "Quotation should be used, with attribution, to present emotive opinions that cannot be expressed in Wikipedia's own voice". Your paraphrase does not adequately communicate the author's warning or sense of urgency, and uses an unwanted metaphor (How can market share "swing"?).

You ask regarding the Cary-Lowe exchange, "Was the quoted individual exaggerating for effect, or making a realistic estimate?" *We don't know*. We do know that that's what Cary said at the time. Putting his statement in quotes makes clear that we are representing Cary's own words, as opposed to introvertible fact; nevertheless, the viewpoint of IBM's Chairman and CEO is relevant and important. To try to "decouple" this because it was "maybe expressed in a moment of frustration" is OR.

To reiterate, there is nothing in WP:TONE discouraging compelling text if the text is representative of the citations and does not engage in OR. There is nothing in WP:QUOTES or MOS:QUOTES discouraging use of quotations that are succinct, interspersed with the editor's own text, and accurately communicate the tone and intent of the cite. Ylee (talk) 19:37, 7 October 2020 (UTC)

Clarification: As stated, I have no objection to abridging a section of a long article into a stub and pointing people to a longer article on the topic of the section. I've done this myself, and some time ago gave some advice to a new editor who took a portion of the History section and moved it to History of IBM where, I agreed, it made more sense. Most of your "stubbing" is accurate and relevant, and the missing citations can be restored. But I think you are working under a misunderstanding of what makes an article "encyclopedic", and what the MOS advises on the topic. Ylee (talk) 19:43, 7 October 2020 (UTC)

Looking over your edits in IBM PCjr further causes me to think that you fundamentally misunderstand the MOS. There you have basically removed almost every quotation in the History section without doing much other abridging (even the current IBM PC#History needs more cutting down if a separate History of the IBM PC article is created). The one subsection you really abridged, on the keyboard, is the subject that people still best remember the PCjr today for; for example, you removed the entire discussion of even privileged third-party developers like Sierra being amazed by the keyboard because they were never shown it before the debut. The article now has WP:COMPNOW issues where it didn't before, and language like "pain point" violate WP:TONE and is unencyclopedic more than anything in the previous version of that article, or this one. Ylee (talk) 20:13, 7 October 2020 (UTC)

More on the PCjr keyboard subsection. IBM wasn't the first to release a poor-quality keyboard; that goes back to the original Commodore PET 2001's calculator keyboard. But Commodore in 1977, Atari in 1978 with the 400, and Radio Shack in 1980 with the original Color Computer a) didn't have as many good and bad precedents to work off of, and b) had zero privileged third-party developers working on software before the computers' release. IBM in 1983 had a history of excellent keyboards going back to the Selecric in the early 1960s, the existence of rivals' poor-quality keyboards, and third-party developers like Sierra that, had their prototypes included the chiclet keyboard, would surely have warned Boca Raton of it being a terrible mistake. Your abridgement of the section communicates none, zip, zero of any of this.

Your user page preemptively explaining why you remove so much text from old tech history articles mentions lack of citations as an issue. You cannot claim that with these articles or sections. Ylee (talk) 22:56, 7 October 2020 (UTC)

A problem I feel here is that your responses seem to violate WP:SYN, which is, oddly, not a problem I saw in the article to begin with. You're describing conclusions that would be synthesis if they were written, but they weren't actually written, and they're apparently obvious to you but - I can attest - they are not obvious to everyone. Sierra would have told IBM not to make the PCjr chiclet keyboard? Probably, but do you have a reference for that? I checked the refs surrounding Sierra's reception of the keyboard, and I don't see it. Did I miss one? We aren't allowed to draw this conclusion unless it's in a citation, however interesting and accurate I think it is.

I don't understand "IBM wasn't the first to release a poor-quality keyboard[...]" Was that in the PCjr article? I just reviewed the current and old revisions and I can't find anything like that. I don't know what you're getting at.

I also do not understand what you're saying here: "The point of the author who quoted the IBM employee was that to claim that a personal computer has nothing to do with office automation is completely wrongheaded" - I'm having trouble parsing the grammar of this, but I think you're saying that the employee was upset about personal computers not being useful for office automation even though they were. Is that on a later page of the book? I couldn't find it. The author did not seem to comment on the employee at all, besides stating that he opposed the entry into the personal computer market. And again, you didn't actually state this in the article - there was nothing along the lines of, "Some IBM employees opposed entry into this market[...] based on the belief that the PC would not be useful as an office automation machine." That would be WP:SYN or OR if you had written it, but you didn't even write it, so I don't understand why you're saying it now.

Overall the point I'm trying to make is that I feel you're going way out of scope on these subjects in a way that works great in a book, but is inappropriate in an encyclopedia - and no, I do not think that I misunderstand encyclopedic tone. The MOS and WP guidelines do not describe literally everything, they expect us to apply judgment and read the room. Articles like Nintendo Entertainment System and Macintosh depend far less on both literal quotes and detailed context for virtually every point made, and overall read in a much less bombastic fashion. An understanding of these products and their role in culture is enhanced by including a certain amount of context and literal quotations, but when we get to the point of stating that IBM did not permit salespeople to drink during lunch, or reporting the specific enraged reactions of person after person after person, I don't think it's unreasonable to say that it's going a bit too far. I do not see this level of dependence on explicit quotes or verbatim descriptions of reactions or opinions in almost any article that I read on here. Gravislizard (talk) 00:37, 8 October 2020 (UTC)

Why are you expecting me to provide citations in a Talk article? I agree that any text depends to some degree on expecting the reader to understand what he is reading based on the context. The problem is that you are removing the context. If the article says that the PCjr has a poor quality keyboard that surprises everyone in the industry and the media, including third-party developers like Sierra, the implication is that if Sierra or another such company had been aware of the chiclet keyboard before the debut it might have persuaded IBM of its mistake. There is no need to state this; rather, let the reader come to such a conclusion on its own. (In any case, this is all stated in the Jimmy Maher article that is cited in this instance; that Sierra, so important to the PCjr's pre-debut software development, was completely excluded from key aspects of the PCjr hardware.)

When you write "I don't understand 'IBM wasn't the first to release a poor-quality keyboard[...]' Was that in the PCjr article? I just reviewed the current and old revisions and I can't find anything like that. I don't know what you're getting at" you are overlooking the Dvorak "$99 el cheapo computers" quote and the Tandy executive quote (specifically citing the Color Computer's keyboard).

Regarding "I think you're saying that the employee was upset about personal computers not being useful for office automation even though they were. Is that on a later page of the book? I couldn't find it. The author did not seem to comment on the employee at all, besides stating that he opposed the entry into the personal computer market": Oh, come now. When, in an article discussing the great success of the IBM PC and recounting how it got there, the author quotes an IBM employee opposing the company entering the PC market because "a PC has nothing to do with office automation", there is no need for the author (or Wikipedia article) to say "Boy, what an idiot!" It is obvious to any reader at the time the citation was published, or any Wikipedia user reading the quote today, that the IBM employee was the one who was laughably wrong. Again, you have removed the entire context, while claiming that I a) violated WP:SYN in a Talk section and b) this somehow taints the article itself.

I have no problem with you removing things like the IBM ban on employee drinking if you think that that is extraneous/excessive. But your mass removal of almost all quotations—again, 100% cited, in context, and relevant—and other text based on a tendentious reading of the MOS is unwarranted. Ylee (talk) 01:01, 8 October 2020 (UTC)