Talk:Influence of the IBM PC on the personal computer market/Archive 1

Archive 1

Purpose

What is the purpose of this list? How is it different from a list of personal computers? Assuming this list has a purpose at all, would the purpose be better filled if this list was confined to List of computers running MS DOS but not otherwise completely compatible with the IBM PC which would be a more interesting list. --Wtshymanski (talk) 23:53, 23 December 2007 (UTC)

What does "IBM PC compatible" mean, anyway? None of the clones supported the cassette port so in that very restricted sense there were *no* IBM PC compatible machines. Most machines didn't come with BASIC in ROM. --Wtshymanski (talk) 23:56, 23 December 2007 (UTC)

The purpose of this list is to show the state of the computer market around the time the IMB-PC came on the marked (and wiped away most of this kind of systems). It also shows that some of these systems, but not a majority of them, had also adopted the 16-bit Microsoft operating system "MS-DOS", but used it in a completely different way than IBM did. Only because IBM dominated the market -their- version of MS-DOS (PC-DOS), and the architecture it ran on, became a de-facto standard.

In a way this article gives an insight of where the market could have gone if IBM had not introduced its PC.

Its a "List of early non-IBM-PC-compatible PCs", because its about "personal computers", that where on the market early, (same time as the introduction of the first IBM-PC), and had a different architecture from it and were not simply its "clones".

It's not "List of computers running MS DOS but not otherwise completely compatible with the IBM PC" because the majority of these systems do -not- use MS-DOS.

This list began as a split off from the main article personal computers . Mahjongg (talk) 16:10, 24 December 2007 (UTC)

deeply flawed, pointless list

NeXT computers came on the market in 1988, hardly 'shortly after' the IBM PC. And the BeBox came out in 1995 almost 15 years after the original PC. The SGI machines are generally considered workstations, not PCs. There's no focus; it's just a list of random computers whose only common point is they're not IBM compatible. Agree with W* above, it would be much more useful as a list restricted to the years from, say, the 8086's introduction and the introduction of the Compaq (or whatever you want to take as the first true IBM compatible) - the Sanyo PC, Columbia PC, any MS-DOS PC that couldn't run Lotus 1-2-3 or Flight Simulator.

I have removed the improper systems from the list. They were probably erroneously placed in the list when it was still part of personal computer, by someone who did not understand what the list was for. This list -is- supposed to be a list of systems on the market around the introduction of the 8086, as you said, and your critique therefore was valid. The only thing is.. you could have fixed the problem instead of complaining how "deeply flawed and pointless" this list is. Mahjongg (talk) 19:39, 29 December 2007 (UTC)

Tried to clarify the reason of existence of this list, hope it helps. I will also try to add more citations and external links soon Mahjongg (talk) 20:10, 29 December 2007 (UTC)

The article says 16 bit processors deeply changed the market and then lists a bunch of 6502 and other 8-bit computers. Why? What is the point of this? Would it not be more meaningful to distinguish "computers that pretty well maxed out at 64 K of RAM" vs "computers that could address more than 64 K but less than 1 M" vs. "computers that could address more than 1 M of RAM". ( I don't even want to point out that the 8088 was an 8-bit processor...says so in the Intel handbook.) I still don't know what the list is trying to say...it's like a "List of cars that aren't Chevys" which would include everthing from Yugos to Silver Clouds.

Stylistically, it's not good to keep repeating "architecture" like a shibboleth - a computer scientist would say "all" these machines have fundamentally the same architecture. All that's happened is trivial changes in the I/O chips and memory maps - sufficient to prevent applications that bypass the operating system from working, but deeply trivial in an "architectural" sense. A word like "architecture" should have a meaningful use in a sentence, and should not be put in as a buzz word to fill space. --Wtshymanski (talk) 14:00, 30 April 2008 (UTC)

The article tries to give historical perspective to what exactly happened to the computer market when the IBM-PC was introduced, it shows how the "computing landscape" was just before, and just after IBM released it's PC, and how other manufacturers reacted to the release of a very powerful competitors competing system, an act that was widely seen as a turning point in computer history. The competing manufacturers generally reacted either by creating a "similar" system, or by trying to create a system that would be received as "technical superior" to the IBM-PC, even though they might still use the older 8-bit architecture. By this time many 8-bit systems had learned how to break through the 64K barrier, and by many it was felt that 16-bit processors were much too expensive, and you could not use the already existing software. So its also historically interesting to note how not all manufacturers abandoned 8-bit microprocessors immediately. By the way, the "PCs" in the title is meant to mean "personal computers", in the broader sense, not in the specific sense that we now (often) use for "wintel" machines. Regarding the use of the word "architecture", it only occurs three times, so what's the big deal? It's a perfectly fit word for the context, not "outdated" in any way (as you seem to imply). And no, the PCs that used 8086 (or 8088/80186 etc) processors did not have "trivial" architectural changes! Some were as different as you could get while still using a similar microprocessor. As dissimilar, for example, as say a Commodore 64 was to a BBC micro. Almost the only thing these systems shared was their use of an Intel 16-bit microprocessor and their use of MS-DOS. Mahjongg (talk) 14:50, 1 May 2008 (UTC)

OK, I kind of see what you're driving at, and I've revised the text somewhat to reflect what I understand of your intent. I'm still not sure that a list of doomed machines helps illustrate the point. Should not your discussion above be in the personal computer article, outlining the significance of the Cretaceous-style collapse of all the different competing phyla of PCs into only a couple of survivors?

I think the architectural changes are trivial - in the CP/M world there was a *lot* more variability in how peripherals were arranged. However, if you designed an 8088-compatible machine and decided to use a Western Digital floppy controller instead of the Intel 8272 (NEC 765), though you could still run MS DOS, all the games and 1-2-3 would no longer run on your machine. Even slightly changing the addresses and IRQs used by I/O was enough to break applications. I think the point needs to be made that in the PC-semi-compatible world that this kind of difference could no longer be tolerated because everyone bypassed the useful abstratction layers of the operating system and BIOS. But it shouldn't be US editors saying this, we need to find some references criticizing the IBM PC design and the stifling effect it had on personal computer designs. Some of the errors of the original PC were corrected in Microchannel, but it was too late and too expensive for clone-builders to use.

The article should say that it's purpose is to illustrate the shakeout in the personal computer field that took from roughly August 1981 to, oh, say, 1985 or so to stomp out any hope of a non-clone selling in high volume. Again, we need references for this,not just us editors. --Wtshymanski (talk) 15:05, 1 May 2008 (UTC)

Actually, shouldn't an article that is a list just have the list, and not all the narrative? That belongs in personal computer where I see a lot of duplicated content. And we need references, darn it. --Wtshymanski (talk) 23:14, 1 May 2008 (UTC)

Ill try to think of a better title for this article, one without the word "list" in it that explains its function better than this one. The article itself clearly explains it's purpose. As is said before, it was part of personal computer, but was taken out by someone, and made to it's own article. The article could do with a new name, as it's no longer just a list of computers around the introduction of the PC, actually the listed computers are a source of reference to back up the claims in the article, but more direct references would be welcome, although at least one direct reference is given. Mahjongg (talk) 23:31, 1 May 2008 (UTC)

When the article was just the list lifted from its context in the article Personal computers, there were complaints why there was no explanation of the function of the list! It seems the title of the article is truly unhelpful, therefore I now have changed the title to "The influence of the IBM-PC on the PC market", so that the lists are simply a means to an end. With the old title "List of early non-IBM-PC-compatible PCs" readers were too focused on the use of these lists.Mahjongg (talk) 23:43, 1 May 2008 (UTC)

About the differences in "architecture", I disagree again that the differences between early 8086/8088 using systems, (the ones that were designed without the pressure imposed on the market by the IBM-PC) were much smaller than say Z80 using systems. Simply check the 8086/8088 based systems that are listed, and compare their "architecture", it often goes much further than just the choice of FDC chip! Especially the video hardware could be greatly different from what the IBM-PC did. It's only after the introduction of the IBM-PC that these systems grow more similar to each other, and the differences become more trivial. But quickly after that the realisation grows that even trivial differences can mean the system sells badly, as they get bad press for not being "compatible enough", so even trivial differences are abandoned in an attempt to be "100% IBM-compatible". Systems that do not are quickly thereafter "extinct". Mahjongg (talk) 00:18, 2 May 2008 (UTC)

Converting list to "prose"

I am concerned that when converting these lists to "prose" a lot of important information will become lost in an attempt to only mention "the most important" systems to keep the prose short. But part of the function of the currents lists is to convey the impression of how many systems of each kind there were on the market at that time. Mentioning just a few of them will defeat this, and will rob the reader of this important information. If all these systems actually had articles, a partial solution would be to assign each system to one of the tree categories so the category pages would become alternative to these lists, but unfortunately many systems do not have an article, or an article of their own. I think that listing the systems is still the best solution. Mahjongg (talk) 23:09, 4 May 2008 (UTC)

Either the article should be prose or it should be a list. If it is to be prose, the lists should be done away with and the writing should be significantly refactored and sourced (at the moment large chunks of it appear to have been drawn from personal expertise). If it to be a list, it should establish its notability through a minimal amount of prose and reference to other articles, but again there needs to be some trimming unless it is supposed that the 50% of examples which do not currently have articles are to gain articles at some point.

Either way, the article should make a decision. It should not be the current chimera of several long paragraphs drawn from personal experience together with three huge lists of systems of which about 50% aren't even notable enough to have gotten articles by this point.

I should note that the reason I tagged the article was because over the last three days you appear to have been attempting to move the article away from list format into prose. If this wasn't your intention, I don't really understand the direction the article has taken since that point. Chris Cunningham (not at work) - talk 23:23, 4 May 2008 (UTC)

It is prose that tries to make it's points clear by providing some lists. Originally these lists were part of personal computer, and tried to get the same messages through as the current article.

In short these messages (points I tried to get across) are/were:

• before the coming of the IBM-PC on the market, the range of systems that experimented with different architectures and "plus points" were much more diversified than afterwards. The first list proves how much diversity was in the market.
• For a short time some manufacturers made computers with the same wide range of architectures, but followed the lead of IBM by adapting the 16-bit Intel architecture. Almost without fail, these 16-bit Intel based systems used MS-DOS (and sometimes also CP/86) as their Disk Operating systems of choice, using BIOS calls to interact with MS-DOS, just like in the CP/M era.
• in these first years, being "MS-DOS compatible" did NOT mean the systems were IBM-PC like, they just used the same microprocessor. The second list shows that these systems run MS-DOS, but are in no way IBM-PC compatible!
• Many other manufacturers simply ignored the IBM-PC at all, and made wholly incompatible systems, with other (non 16-bit intel) CPUs, which consequently could not, nor wanted to, run MS-DOS. The third list shows that a remarkable large number of manufacturers simply ignored IBM, and continued to do their own thing.
• Because of the fame of IBM a large body of software was quickly written for the IBM-PC, but for technical reasons most of it did not work on non 100% IBM-PC compatible systems, and a working "abstraction layer" (like windows) did not exist yet, so there was much more software available for 100% compatible systems, than for systems that only used MS-DOS, or that were completely incompatible, like those in the third list. As a consequence these systems could not compete in the business market. Only systems that were bought for their games and multimedia features still thrived for a few years, until "clones" became a valid competitor, through the emergence of cheap VGA copies and sound cards which made clones more competative.
• Finally, the main message is, that in just a few years, the market workings engaged by the appearance of the IBM-PC turned a market full of diversity into a monoculture where manufacturers no longer tried to compete by designing a superior system, but were confined to just making a valid "clone", any attempt to be "better" was punished by the market. On the other hand, IBM "validated" the market, and it grew to much larger proportions than before, it also broke the "8-bit is good enough, 16-bit is too expensive and there is no software available for it" limbo the market was in.

It became quickly obvious that without the context out of which the lists were ripped their use became wholly unclear, so that I added prose to explain their purpose, but the name the "ripper" had given the article still influenced the understanding that people had of the real purpose of the article, so recently I gave the article a more proper title, and that is when things speeded up a bit.

So yes the article -should- be converted to prose, but I'm unclear how to do it without giving up the "proof", in the form of the lists, that shows the state the market was in. Perhaps you can split the lists off to three "sister" articles, that only contain a list each, with a link back to the main article, and give each article (list) only a small and hopefully short description of what the purpose of the list is, like "List of computers that appeared on the market less than a few years before the appearance of the IBM-PC", "List of non IBM-PC compatible, but MS-DOS using systems that appeared on the market shortly after the appearance on the market of the IBM-PC", And "List of computers appearing on the market shortly after the appearance of the IMB-PC, but that ignored its architecture and went their own way". The main article can then list (in prose) only a few of the most important examples, and perhaps give an indication of the number of such systems, like several dozens etc etc.

I concur that the article, as is, now severely lacks sources, unfortunately these are difficult to find, but should be there. So whenever I find some time Ill try to source the different claims in the article better. Mahjongg (talk) 01:11, 5 May 2008 (UTC)

Well, the basic problem is that the lists themselves are unsourced. If we can find a reference which lists them, there's no need to expand them all here. Not all groups of information are notable enough for articles, and I'd argue that in this case:

1. The prose can stand alone as an article without the lists
2. The lists cannot stand alone as an article

This leads me to believe that the best thing to do would be to find a reliable off-wiki resource which lists all these systems, to use that as a reference in writing the prose, and to then trim the lists to highlight the article's key points. The lists will still be accessible from the resource,m but won't be sitting here as arbitrary groups of data.

Oh, by the way: thanks for all your work on this. Chris Cunningham (not at work) - talk 13:28, 5 May 2008 (UTC)

Well I tried to implement your suggestions, as far as was possible. a similar list is simply not available on the Net (but all the separate details are), so I added references as best as I could, and removed the lists and added prose to attempt to conform as best as I could to Wikipedia standards, without losing essential information. Mahjongg (talk) 11:34, 6 May 2008 (UTC)

Great work! This is a much more compelling read now. Chris Cunningham (not at work) - talk 11:46, 9 May 2008 (UTC)

article is a mess

needs serious copyediting for clarity and flow —Preceding unsigned comment added by [[User:{{{2}}}|{{{2}}}]] ([[User talk:{{{2}}}|talk]] • [[Special:Contributions/{{{2}}}|contribs]]) 14:02, 26 August 2008 Petchboo

Thats your opinion, but I do not agree at all and I suspect exterior motives (you simply might not like the presented facts and their implications). It seems to me the tags are just harassment. But go ahead, and "fix the clarity and flow" of the article. I won't go on a revert-war over the tags, just because they seem over the top for me. In the past I have heavily copy-edited this article already, and after that most co-editors seemed pleased with it. Perhaps another editor can give his opinion on the need for all these tags. Mahjongg (talk) 15:18, 26 August 2008 (UTC)

Five tags is perhaps OTT, but there's certainly a lot of work left before we're ready to go nominating for GA. I'm not personally opposed to the tags, although I would prefer that the editor who placed them left a more thorough explanation. I'll try to have a look over the article and see if I can suggest more specific cleanup in the near(ish) future. Chris Cunningham (not at work) - talk 15:32, 26 August 2008 (UTC)

Yes, I agree, but I don't think every article needs to be of GA quality. But perhaps this [1] edit (the "advantages" section) has something to do with it, I think its a little bit too much non NPOV too. I don't like the tone of it myself, nor its language. Mahjongg (talk) 15:44, 26 August 2008 (UTC)

Entirely unsourced Just undo it. As for being GA class, I consider it a basic criterion for article-worthiness that a subject can have a GA written on it. Chris Cunningham (not at work) - talk 20:15, 26 August 2008 (UTC)

Not clear what jou want to undo? The tags, or the "advantages section". I agree with Wtshymanski that almost nobody seems to be interested in history, just in the "winners". Why should every article be of GA standard? Considering that just one in 527 articles are considered to be "Good Articles" according to WP:GA not much of Wikipedia would remain if we took your "basic criterion" at heart. Now, considering that the user that placed the tags vandalises pages himself, and had several of his articles quickly deleted, it seems he just did it out of spite, but I agree that he has some points. Mahjongg (talk) 22:29, 26 August 2008 (UTC)

I see you removed the "advantages section", good I tried to salvage it, and in principle I think what it said wasn't much off the mark, but it simply wasn't very encyclopaedic. In truth its better its gone. Mahjongg (talk) 22:50, 26 August 2008 (UTC)

We need this article or something like it. But it's so very hard to find references. Editorials in PC magazines go on and on about bark and leaves, and never once step back to look at the shape of the forest. And good luck finding a book taht doesn't focus on trivialities of mastering one application. We may have reached the limits of what folklores and rumor can teach us about the development of the personal computer. --Wtshymanski (talk) 21:10, 26 August 2008 (UTC)