Having actually used GEM - yes, I am a fossil, and was configuring 3274 cluster controllers for IBM in 1984 - I can say that I really wish GEM had continued. It was beyond its' time and incredibly easy to use. A pox on that which replaced it!

"The PC world might have looked very different today had Kildall's Digital Research prevailed as the operating system of choice for personal computers."

Perhaps in a parallel universe people wonder at the might-have-been of poor old William Gates' Microsoft, whilst lambasting the restrictive tyranny of the Sinclair Digital Research juggernaut. Perhaps there is another world where Microsoft's graphical operating system - funded in part by a local showman, and called SteVision - is fondly remembered by the few hundred people who bought it. They fondly the curious one-button pointing device that came bundled in the box, and dream of another world.

"But then Kildall was motivated by technical excellence, not by the need to dominate his fellow man."

We would all like to think that had we been at the helm in the old pc pioneering days, things would have evolved better. However once someone is in a position of power, greed is often an irresistible temptation and Bill succumbed to it. His company's decisions have cost humanity billions of man hours in poorly documented interfaces, buggy and slow software, and using it's monopoly to block more innovative products from entering the market.

Today, as a software developer I am hugely disappointed with the "progress" that the PC market has made, ubiquitous or not. Running a current version of MS word on a modern today is slower than running an old version on a machine of that era. Opening documents takes several seconds. Starting the OS takes half a minute while dos with windows 3.1 was faster than this. Microsoft has no excuse for this. Unfortunately optimization and efficiency rank at the bottom of MS's priority list.

Crediting MS/Bill for single handedly creating the PC revolution is incredibly naive (it's not entirely his fault). I still think the author was correct that if market forces had selected someone driven by technical excellence instead of Gates, the computing world today could be very much better.

Things might have been better, but they could have been worse too.

The UNIX and "hacker" communities had zero interest in providing computing to non-technical regular people. The Apple/Microsoft window interfaces were called "point and drool" in the MIT hacker dictionary, they had complete contempt for that approach. So be thankful that at least Apple and Microsoft focused on what businesses and average people wanted to do.

It is also possible that we would just have closed hardware and "thin clients" today, that nobody can program. We might keep all our data on big IBM servers. Again, the PC software revolution and free innovation on the web might never have happened.

There are a lot of people who want to "fix" the way personal computing happens, and it is not at all clear that the consequences would be better.

CP/M, and for that matter, all the other contemporaries, ran on numerous configurations of minimal size. With great ease. Compare that to the current MS offerings. Vista won't run on most of the machines produced a year ago, or at least not well. In July of 2006, the average laptop came with 512 MB and a 60 GB drive.

It's a never ending story. Bigger hardware allows for sloppy coding. Sloppier coding require even bigger hardware. And so it goes. Where's Atrau?

Apple steals from Parc, MS steals from DRI, and did we really get better products? Windows is OK, but I'd never use a Macintosh (too much pretense). I can't argue the fact that Steve Jobs and Bill Gates are superior egomaniacs in bringing products to market, but maybe we the rest of the people got screwed in the process.

Quite right. I was there; I remember asking Gary about that - especially the bit about flying!

:-)

See my memory of the occasion in a recent blog entry at NewsWireless - http://www.newswireless.net/index.cfm/article/3539 is the link.

The piece is a bit of a cheek on my part. I have lost an awful lot of back numbers of various pieces I wrote, back to Electronics Weekly in 1974, and including PCW NewsPrint starting 1977, right through to PC Mag and ZDNet. Precious little of that is archived on the Web.

If any readers have bits of nostalgia stored on their shelves, it would be VERY MUCH appreciated if they could mail me (go to NewsWireless.net and hit "contact") with the date and a scanned page from the magazine? I'll try to acknowledge anybody who submits a memory which makes it into publication...

Guy Kewney

There is an often told story that Gary Kildall typed in an undocumented CP/M command on a DOS machine and got the CP/M response.

Can anybody shed any more light on this?

I'm always surprised, during these nostalgia-fests, that nobody (but me) brings up the origin of the DOS filesystem. IIRC (from a MSFT publication), Patterson used the "Disk Basic" filesystem, originally designed, and possibly coded, by BillG himself, instead of the CP/M filesystem. The Mac, again IIRC, also used a scheme similar to FAT, with every file a linked-list of blocks. CP/M used what would today be called an "extent based" system, which can require a bit more work on write, but usually leads to less fragmentation and faster reads.

Anyway, the very early MSFT/QDOS connection is, to me, more interesting than the notion the BillG was just wandering down the street and stumbled across the source of much of his fortune. :-)

I've probably got a copy of GEM still somewhere. It worked really well, especially as the Windows alternative at the time was the tile based Win1.0 (anyone recall that?).

I think the FAT file system comes from RT/11, the old DEC format. The one complaint I had with CP/M was the random floppy disk formats, but then one of the first programs I ever bought for the PC was a piece of software that allowed me to read and write such disks so 'no harm done'.

Where MSFT really went off the rails was with networking. I worked on their early MS-NET stuff and to say that 'they didn't have a clue' would a very nice, and somewhat flattering, way to describe them. It didn't stop them from charging large sums of money up front for the stuff. They really didn't get networking at all until Win3.1 and I think that was more to do with trying to screw Novell than encourage interconnectivity between computers. (Internet? What's that?)

Who invented the railway sleeper car?

Most people will say Pullman.

Google Philip Berlin, Chambersburg, and Cumberland Valley Railroad.

Now compare that with Kildall, Gates and DOS

@Lou Gosselin - Yep, Bill is such a greedy person that he gives oodles of money away to good causes via the Bill and Malinda Gates Foundation and is basically leaving all his money to the foundation when he dies. Ruthless businessman probably. Greedy, probably not.

Another comment, generally offered: Acorn could have been the PC that everyone had on their desks. First company to make a RISC chip, kick ass multi-tasking windowing operating system. They never made it out of the UK with the Archimedes, my understanding is that Apple were all set to sue the shit out of them if they released RISC OS in the US, this pretty much killed them outside the UK.

Yep - it existed, ran really well on the then current hardware . I hacked the version that came with the Amstrad PC to work on my (VGA I think) Toshiba (luggable) laptop. Was really quick and resource light compared to Windows 3.1. I used to use it with DeskPress desktop publishing software and nothing on Windows came close for functionality or performance for years. Oh the things that might have been, eh?

You mention that: "in GEM, a portable graphical environment which would undoubtedly have brought the GUI to the low-cost PC desktop years before Microsoft's Windows"

It is unfortunate that you forget the past. GEM "did" bring the GUI to the low-cost PC desktop. But, you might say that Apple gave the "coup de grace" to DRI with their GEM Lawsuit. DRI was forced to change GEM significantly and pay lawyers fees to Apple. GEM was never the same after that and even worse, Apple's victory took much of the market from under the GEM system and DRI. Blame Apple. Not to hail Microsoft, but once they had the money for the lawyers, they were the only ones that took Apple all the way to the U.S. Supreme Court and won (except for the garbage can). It was not IBM nor anyone else, it was Microsoft; and a monster was born. You might say that the field had been weakened (all other predators eliminated) by Apple and once they were out of the "picture" (sorry for the GUI pun) the MicroSoft infectious disease took advantage of the situation and devoured the host (we).

True, GEM on the PC stinked, at least in comparison to the version on the Atari ST's due to the Apple lawsuit. I tried it and couldn't believe how stripped down it was.

Too bad, at least the Atari version was much better than the early crappy versions of Windows.The final blow was Microsoft making Windows incompatible with DR-DOS.

Not hardly.

If you squint _really_ hard you might see a family resemblance between FAT and the original PDP-11 DOS format. But RT-11 used resolutely contiguous files, the "Reductio ad Absurdum" of extent-based file systems. FAT was really good at one thing: cramming a usable and faster than cassette filesystem into very little code. You might be thinking of the external (command line) appearance of the two (or three, or more) file systems of the time, as they pretty all used base.ext naming. Some (typically DEC, but some IBM) bothered to squish the filenames, presumably on the grounds that "If we aren't going to let you used mixed case, at least we could use those bits for a few more characters in the name." QDOS (and I believe CP/M) did not. Harder to code, I guess.

As for the multiplicity of "disk formats" under CP/M, that was (intended, AFAICT) a feature, not a bug. Under CP/M, the hardware manufacturer wrote the BIOS, and it was supposed to (usually did, in my experience) abstract the hardware sufficiently for the file-system to treat it as a black box. I presume that "things would have been nicer" if CP/M forced everybody to use the same oddball disk controller, right down to bits in controller registers, just to have basic Disk I/O work. Oh, wait, I guess that _was_ the wave of the future. There are those who feel that, had the IBM-PC had a better BIOS, the device-independence folks had come to expect from CP/M would have spread its cancer of non-conformity to the PC, The we would never have the opportunity to use computers that had a hard time keeping up with a 56kbps modem because to put a _working_ FIFO in the UART would break so much software. Well, it drove the price down. There's that.

@Fraser- Notice BG did not create the Gates Foundation until after he had been richest man in the world several years running and would surely never have to drive himself or wash his own clothes.

The BMGF is more likely a legacy for posterity than because he's just such a nice guy at heart. Much like Andrew Carnegie, this will help future generations think, "my what a noble man" rather than "robber baron".

I don't know what Bill Gates will be remembered for 50-60 years hence but I doubt it will be for the worlds most popular operating system. That bloated mess with backwards compatibility for 16 bit code in a gigabit world will live only in science fiction nightmares thank god.

"greed (grēd) Pronunciation Key

n. An excessive desire to acquire or possess more than what one needs or deserves, especially with respect to material wealth:"

I'll admit I don't know a thing about Bill G's personal life. However if MS is an extension of his character then it isn't a far stretch to conclude that Bill is an greedy monopolist.

As for donations the amount should be measured with respect to one's disposable income. I believe the average income of everyone in the US (including the likes of Bill who drive up the average) is $30K (not confirmed). Compare that with god knows what Bill's roped in over the years as legal salary as well as other fringe benefits and then put his donations in proportion.

Just because he has too much money to spend on himself doesn't make him a generous man. But that's my opinion.

but I really feel that this quote is relevant (Mark 12:41-44) :

And Jesus sat over against the treasury, and beheld how the people cast money into the treasury: and many that were rich cast in much.

And there came a certain poor widow, and she threw in two mites, which make a farthing.

And he called his disciples, and saith unto them, Verily I say unto you, That this poor widow hath cast more in, than all they which have cast into the treasury:

For all did cast in of their abundance; but she of her want did cast in all that she had, all her living.

Now then, what is this I hear about Bill G being a Good Man ?

As someone who used to program at that level, I realised immediately that the IBM PC and MSDOS had put computing back 20 years. A pathetic design from IBM compounded by a botched OS leaving the rest of the world struggling trying to overcome memory map limitations, lack of real interrupts etc, everything we had already sorted on commercial UK non IBM PCs. I blame it on Charlie Chaplin who IBM used in the advert, which convinced so many Americans to buy the IBM PC because Charlie thought it was OK so it must be good! Once they had the market share and the software package development followed that was the death knell for everything else.

...is a big tax dodge thingy to protect his assests. He only has to give away 5% of the money it makes, while the rest can be held in a sort of trust for his family.

http://www.latimes.com/news/nationworld/nation/la-na-gatesx07jan07,0,6827615.story?coll=la-home-headlines

You don't need to have been a "fossil" to have used GEM - I'm 28 and it was the main UI on my Amstrad 1640 when I was a kid, along with an installation of Windows 1. I remember GEM was years ahead Windows at the time. I remember it had a really good DTP package as well, "Timeworks" or something...

Yup, I'm 27 and the Goldstar 286 I was bought in 1990 came with MS-DOS 4.1 and GEM pre-installed.

CP/M and it's applications runs very well on my Sinclair Spectrum +3 with 3" 180K disk drive!

I'm 25 and we used to have GEM on the family 386. On the Amstrad we had before that we used CP/M. And yes, I remember a DTP package we used on GEM, can't remember name though.

Most if not all modern Operative Systems have deliberately copied or have been "inspired" by DEC's 10-11ish interactive OSes.

TOPS10, RT-11, RS-TS all those operative systems made the history of the command line and I seem to remember reading that Mr. Kildall's first real commercial incarnation of CP/M (1.3) used to spot those similarities as a features not as a rip-off.

PIP (Peripheral Interchange Program) is the classical example.

Nevertheless CP/M and RT-11 are two wonderful exercises of effective minimalism: they did their job and they did it well (they do, actually, as far as I am concerned if I want to have some fun I switch on either my Altos 8000 or my 11-23).

GaB

Let's not forget that Kildall and Patterson both borrowed freely from DEC RT-11 just as, a few years later, Apple and Microsoft both borrowed a great deal from Smalltalk when developing their graphical operating systems. Couldn't have been as many lawyers around in those days. In any event they all got fair exposure and the marketplace picked the winner.

"I blame it on Charlie Chaplin who IBM used in the advert, which convinced so many Americans to buy the IBM PC because Charlie thought it was OK so it must be good!"

Maybe it was a "liberal" backlash against the House of Un-American Activities Committee's dislike of Charlie 8-)

True - I'm of the same age and had it installed on my first PC, the trusty Amstrad 1640... Upgraded to a 186 with winblows installed a few years later and regretted it for years to come....

Could also run lemmings on those trusty ole 1640's unlike the next gen 186's (it even ran better on the 1640 than the 286's) blame power hungry windows?

That reeks of train spotting. I thought that wasn't allowed at El Reg!

no mate, it reeks of reading a bluffers guide that mentioned it.

Pullman did not invent the sleeper car (even though he is creditied with it). He just took the idea, made it better, and marketed it better.

It depends what you were doing.

If you were into magazine publishing back in the 70s you might have used a XEROX WIMP system for page layout and editing.

If you were one of the lucky professionals in the mid 80s, you may have used an Apollo, and Indy or a SUN 3. All these had WIMP and user access to the GUI application builder. They were 32bit, had multiple autonomous processors for different functions like Disk, Graphics, Ethernet, Floating Point, etc so they were very fast. The 19"+ portrait screens were large enough to cut and paste between two A4 pages. One could populate the screen with graphical icons representing application functions that when dragged on dropped onto the application window would apply the function asking for any required parameters. So incredibly easy. Of course the size of the screen meant that to be productive, one used pop up slide and select menus, as mouse travel to a drop down menu took uneccessary time. Menus could be "pinned" if needed. One can see this reflected in the design of the GIMP interface.

In my experience todays Windows WIMP is neither as cognitive or productive as were these mid 80s workstations that closely followed the XEROX design parameters. Stomfi

davcefai, that claim comes from Jerry Pournelle, who claims that he personally saw Kildall type the command and wrote it down. He has been asked several times to produce the command and has not done so.

The Evans book mentions Kildall using such secret command to catch other companies copying CP/M, but not Microsoft. Kildall wrote a unpublished memoir shortly before he died that was very bitter about the whole affair, and if he had such damning evidence he certainly would have produced it. There is no indication that he did.

There is no "vindication" of Kildall here nor any finding that revises the history of CP/M and DOS in the least, just a finding that Paterson filed a sloppy, emotional lawsuit, which would be obvious to anyone reading it.

If you want to know the full story, it's all explained in detail, with sources, in Wikipedia's articles on Kildall and 86-DOS.

GEM is still available (now open source), from http://gem.shaneland.co.uk/

I think the FreeDOS distribution includes it as well

All WIMPs arise from work done at PARC. They first invented the GUI and then invented the pointing device that "enabled" that GUI - i.e. the mouse. Their version went on to become the X11R3 standard that was used on a lot of early PCs/microcomputers by those who were bothered to "fit" the X11 free source code into their flavour of x86 unix !!

As mentioned above, Apollo, Sun and others used various flavours of the X11R3 on their systems and was highly sought after by those who had to work in graphics !!

When you pressed Ctrl-k, after a few seconds a menu popped down and you could see to press 's' to save. And the text tried to appear like what would be printed out.

And many applications had forms to fill in that allowed you to tab back and forth between fields.

Lotus 123, had a menu that popped up after you pressed the '/' command key. With 123 you pointed to the cell and did stuff to it. Wasn't 123 based on VisiCalc of 1979?

In 1981 many Unix style workstations had windows and icons.

A lot of different people developed the ideas that grew in to the WIMP/GUI.

I think it came from Xerox.

The OS is important, but it is waning.

Instead the Internet is what really matters.

People will be doing what they do on PCs now on the Internet through a myriad of devices.

The paradigm has changed and that is a good thing.

Less politics and more innovation. That is the great advantage that the Web has over Windows.