CP/M creator Gary Kildall's memoirs released as free download • Hacker News Board

Re: CP/M creator Gary Kildall's memoirs released as free download

#91

Post by Tor3 » Fri, Jul 18, 2025, 7:31 PM UTC

WalterBright wrote: post #71

acdha wrote: post #61

Earlier quoted context omitted.

None of us know what was said but I have no reason to doubt it based on the reports of his subsequent conversations with lower-level IBM executives. It probably didn’t seem like an especially consequential decision both because neither Gates nor Kildall were especially proven at that time by the standards of a Goliath like IBM and the mainframe guys were notoriously dismissive of PCs (Opel came up through S/360). I’v…

Ms Gates wasn't on the board of IBM, she was on the board of another company. That isn't nepotism. There is no way successful IBM would commit to Microsoft without a thorough vetting. Few remember, but IBM also sold CPM/86 for the PC. Kildall had his chance, and muffed it with the high price.

I remember the very high price of CP/M-86. If that was because of DRI's pricing and not something IBM did, then indeed that made the choice simple, in Kildall's disfavour.

Re: CP/M creator Gary Kildall's memoirs released as free download

#92

Post by WalterBright » Fri, Jul 18, 2025, 7:37 PM UTC

ChuckMcM wrote: post #83

WalterBright wrote: post #77

Earlier quoted context omitted.

Ms Gates was not on the board of IBM. She was on the board of United Way.

The original comment said they were both on the board of United Way. Those are the votes he'd be curating. United Way, at the time, was the largest non-profit in the United States and it's mission was to funnel donations to "deserving" non-profits. Many companies, including IBM, had a payroll contribution you could make to United Way. The 'service' United Way provided was doing the research to avoid scam charities an…

The Gates family was indeed wealthy, but they were nowhere near the kind of wealth that would influence IBM. I'm sure Ms Gates was intelligent, well educated, and quite charming, but she didn't know anything about computers. IBM would have been foolish to let her decide their PC division's major decisions.

I.e. I'm not buying the notion that her influence went beyond simply suggesting they check into what her son was doing.

Re: CP/M creator Gary Kildall's memoirs released as free download

#93

Post by jen20 » Fri, Jul 18, 2025, 7:39 PM UTC

WalterBright wrote: post #71

acdha wrote: post #61

Earlier quoted context omitted.

None of us know what was said but I have no reason to doubt it based on the reports of his subsequent conversations with lower-level IBM executives. It probably didn’t seem like an especially consequential decision both because neither Gates nor Kildall were especially proven at that time by the standards of a Goliath like IBM and the mainframe guys were notoriously dismissive of PCs (Opel came up through S/360). I’v…

Ms Gates wasn't on the board of IBM, she was on the board of another company. That isn't nepotism. There is no way successful IBM would commit to Microsoft without a thorough vetting. Few remember, but IBM also sold CPM/86 for the PC. Kildall had his chance, and muffed it with the high price.

> There is no way successful IBM would commit to Microsoft without a thorough vetting.

As I recall, at the time said commitment was made, Microsoft didn't even _have_ an operating system, and subsequently bought QDOS! Their original deal was for languages.

Re: CP/M creator Gary Kildall's memoirs released as free download

#94

Post by WalterBright » Fri, Jul 18, 2025, 7:42 PM UTC

vidarh wrote: post #90

WalterBright wrote: post #73

Earlier quoted context omitted.

The parsing, even if line by line as necessary, is still compiling.

In 45 years of writing software, I've never before seen anyone call tokenizing a BASIC program compilation. It's decidedly not common usage.

I've been writing compilers for 45 years now. Tokenizing is a big part of every textbook on compilers. To resolve expressions (which are recursive in nature) it would have had to do more than just tokenizing. While this isn't hard at all, it's "parsing" which is also qualifying it as a compiler.

I.e. the basic program was lexing and parsing. It's a compiler. A very simple one, sure, but a compiler.

Re: CP/M creator Gary Kildall's memoirs released as free download

#95

Post by WalterBright » Fri, Jul 18, 2025, 7:46 PM UTC

whobre wrote: post #81

WalterBright wrote: post #76

Earlier quoted context omitted.

> It was IBM who set the price Not what I heard at the time - IBM accommodated what Kildall wanted. Kildall could have sold it separately at a competitive price.

https://www.computerhistory.org/collections/catalog/10271725...

The date on it is 2016. 33 years later. I don't find that particularly convincing.

BTW, I've read a lot of computer histories, and I often find errors in them. This is why professional historians try to use contemporary sources as much as possible.

Re: CP/M creator Gary Kildall's memoirs released as free download

#96

Post by WalterBright » Fri, Jul 18, 2025, 7:49 PM UTC

jen20 wrote: post #93

WalterBright wrote: post #71

Earlier quoted context omitted.

Ms Gates wasn't on the board of IBM, she was on the board of another company. That isn't nepotism. There is no way successful IBM would commit to Microsoft without a thorough vetting. Few remember, but IBM also sold CPM/86 for the PC. Kildall had his chance, and muffed it with the high price.

> There is no way successful IBM would commit to Microsoft without a thorough vetting. As I recall, at the time said commitment was made, Microsoft didn't even _have_ an operating system, and subsequently bought QDOS! Their original deal was for languages.

Gates convinced IBM that he could build one, as he knew about QDOS, and immediately went and bought QDOS as a base to start with. So, yeah, it was a bit of bluster on his part, but he was able to fulfill the contract.

Re: CP/M creator Gary Kildall's memoirs released as free download

#97

Post by vidarh » Fri, Jul 18, 2025, 7:56 PM UTC

WalterBright wrote: post #94

vidarh wrote: post #90

Earlier quoted context omitted.

In 45 years of writing software, I've never before seen anyone call tokenizing a BASIC program compilation. It's decidedly not common usage.

I've been writing compilers for 45 years now. Tokenizing is a big part of every textbook on compilers. To resolve expressions (which are recursive in nature) it would have had to do more than just tokenizing. While this isn't hard at all, it's "parsing" which is also qualifying it as a compiler. I.e. the basic program was lexing and parsing. It's a compiler. A very simple one, sure, but a compiler.

Yes, but tokenization on its own is not compilation any more than whiskers are a cat just because a cat has them.

"Nobody" uses it that way, and language is defined by use.

Re: CP/M creator Gary Kildall's memoirs released as free download

#98

Post by wvenable » Fri, Jul 18, 2025, 7:58 PM UTC

WalterBright wrote: post #44

rbanffy wrote: post #25

Earlier quoted context omitted.

> BASIC compiler Interpreter - an entirely different kind of animal. Microsoft didn't get a BASIC compiler until much later. > He helped Wozniak implement a version of BASIC supporting floating point numbers. No. He sold Apple a BASIC, then used it as leverage to prevent Apple from making a BASIC for the Macintosh. > Ballmer was the biggest businessman in the bunch. He suggested cutting Paul Allen's family off when A…

Um, it is necessary to compile a program before being able to interpret it. I don't know how early BASICs were implemented, but the usual method is to compile it to some sort of intermediate representation, and then interpret that representation. D's compile time function execution engine works that way. So does the Javascript compiler/interpreter engine I wrote years ago, and the Java compiler I wrote eons ago. The…

Early BASICs didn't compile a program before interpreting it. The interpreter read the code as written and executed it step-by-step. There was some tokenization; keywords were turned into single or double bytes and that was literally done when you pressed enter on the keyboard. Your source code was these actual tokenized bytes. On the Commodore 64, you could type the tokenized versions of keywords instead of the full keyword as a shortcut. Even numbers were not transformed into bytes ahead of time.

This was used to save memory -- there wasn't much room to hold both the source code and an intermediate form. But also it wasn't that necessary, with the keywords tokenized and the syntax so simple that there wouldn't have been much savings in space or performance.

Re: CP/M creator Gary Kildall's memoirs released as free download

#99

Post by canucker2016 » Fri, Jul 18, 2025, 7:58 PM UTC

ChuckMcM wrote: post #83

WalterBright wrote: post #77

Earlier quoted context omitted.

Ms Gates was not on the board of IBM. She was on the board of United Way.

The original comment said they were both on the board of United Way. Those are the votes he'd be curating. United Way, at the time, was the largest non-profit in the United States and it's mission was to funnel donations to "deserving" non-profits. Many companies, including IBM, had a payroll contribution you could make to United Way. The 'service' United Way provided was doing the research to avoid scam charities an…

Paul Allen states his version of the IBM selection of OS for their IBM PC in his autobiography, Idea Man.

see my comment at https://news.ycombinator.com/item?id=43591941 for the long version.

Hopefully short version.

IBM went to Microsoft (MS) for languages for the new PC. IBM asked if MS could provide an OS. As per unwritten agreement, MS told IBM to go to Digital Research (DRI) for an OS.

Whatever happened at DRI, IBM didn't get a licensing deal for an OS. No OS meant no need for MS languages. When IBM complained to MS about not getting a licensing agreement for CP/M for IBM PC, one of the MS people suggested Tim Patterson's CP/M clone. IBM was outsourcing everything to keep away the IBM bureaucracy, so they told MS to handle everything.

When MS asked IBM how they wanted to pay for the OS, MS gave several options including 1) per copy royalty, or 2) flat rate (which turned out to be $40K). For easy accounting, IBM chose 2). MS asked to be able to license OS to others. IBM said yes. MS didn't really care about how IBM paid for the OS, their bread and butter was languages. DRI wanted to be paid per copy of CP/M-86.

DRI still didn't have a retail CP/M-86 for IBM PC at launch time. By the time they did ship CP/M-86, charging much more than PC-DOS1.0, Lotus 123 would launch within a few months running on PC-DOS. By the time DRI lowered the price for CP/M-86, they were way behind in market share.

Re: CP/M creator Gary Kildall's memoirs released as free download

#100

Post by wvenable » Fri, Jul 18, 2025, 8:03 PM UTC

WalterBright wrote: post #74

eichin wrote: post #57

Earlier quoted context omitted.

> necessary to compile Um, no? your experience is probably at least two decades after the time period in question.. The more advanced versions of, for example, the TRS-80 BASIC (part of this "microcomputer BASICs that all share a common set of bugs") did no more than tokenize - so, `10 PRINT "Hello"` would have a binary representation for the line number, a single byte token for PRINT, then " H E L L O " and an end-o…

Tokenizing it and interpreting the token stream is still a compilation process. Even if it re-tokenized it each time it executed a line.

Tokenizing is a necessary but not a sufficient task for compilation. I could tokenize this comment to efficiently store it in a database but that would have nothing to do with compilation.