$4,000 wasn't exactly cheap. They must have been extremely well-heeled hobbyists I suppose front panel switches were an improvement on APL.

> $4,000 wasn't exactly cheap

Maybe you misread it?

$4K was for a tricked-out box, something that might just about have been able to run CP/M -- and you'd need a separate serial terminal as well, of course.

That's the point. All you needed for 4K BASIC was the base model ($439) and the next-up-from-the-smallest RAM board, and something to load it from, like a paper-tape reader.

Not quite a tenth of the price but in the ballpark of it.

CP/M: north of $5K

BASIC: under $1K.

Easily misread, given that nowhere in the article does it actually state what the *base* price of the system is - you go almost directly from quoting that tricked out price to then mentioning how much it costs to add the 4KB expansion required to run BASIC, so like the good Dr, I too misread it the first time through as implying a combined cost of just north of $4000 to get something able to run BASIC...

> nowhere in the article does it actually state what the *base* price of the system is

Fair point. I guess I should have. I didn't because the article isn't about the Altair hardware, or the company behind it. It's about the first deal of the as-yet-unnamed Gates/Allen partnership. But point taken.

Add in the teletype unless the base model also included the video card although that would have included the paper tape reader. Allow for inflation to get the current equivalent. It was a little later that I was using an S100 based system. I can't remember the overall cost but it was for use as a scientific work station bought for the lab in which I worked - as a mid career scientist that wasn't something I could ever have afforded as a hobby.

) ⍞ THE 5100'S SCREEN WAS TOO TINY *UNLESS* YOU WERE WRITING IN APL.

) by itself is invalid. ) ⍞ is no better.

⍎A←'⍎A'

Did it come with a keyboard with just a 0 and a 1 on it ?

Yes it did. They are called panel switches :) Up is (usually) 1 and down is (usually) 0. Unless you mounted them horizontally orientated :)

YMMV

AAC

No, it did not come with a keyboard, only front-panel switches and lights.

My dad added a keyboard, a paper tape reader, a tape drive, and later a 5.25" disk drive to his.

In 1976, when ICL had an office on 5th Ave and a 1904 somewhere downtown, every few weeks I used to nip over the 39th on Broadway to visit the Computer Store and admire the those MITS boxes with their rows of switches and red LEDs in the sidewalk display. However, if I wanted to play with anything, I always went inside to diddle with an SWTPC box, which was usually up and running

with green screen, keyboard and printer attached and powered up.

At the time I was working out at Longbeach programming an ICL 2903 in COBOL for a toy manufacturer.

On my return to London in 1977 I found that those trips to the Computer Store hadn't been wasted because the small London company I worked for had bought a somewhat bigger SWTPC system Uniflex OS-9 OS on 6809 microchips and capable of supporting 4 terminals and a couple of printers. We sold a reasonable number of these systems, which were quite popular with small family businesses, running software written in the Sculptor 4GL, which I liked a lot.

Out of interest, does anybody else remember using this hardware and software?

I only did that for 2-3 years before I switched back to designing and writing more complex systems running on ICL 2966 boxes, all written in COBOL and using IDMSX databases.

COBOL? You must have become rich during the Y2K mess.

Sadly, or not, ICL 1900 kit sailed through the Y2K fiasco without a glitch. That's because all ICL 1900 computers held dates as a 23 bit "days since 1900" integer, stored as a signed 24 bit word. This means that the 1900's 'Y2K moment' will occur somewhere around 22982 AD.

All date conversions between the external dd/mm/yy format and the internal binary form were handled by a standard set of subroutines maintained by ICL.

As a result, Y2K was a pretty much non-event for anybody maintaining code running on a 1900 because these systems almost always expected dates to be input and displayed using the dd/mm/yy format. As a result fixing the problem was relatively trivial since it only required any programs we'd written that accepted and displayed or printed human-readable dates to be changed from using dd/mm/yy to using dd/mm/yyyy.

From memory, about the only program we ran that needed anything more than simple adjustments to date I/O formats was a payroll system we ran for a company that still had pensioners born in the 1890s on its payroll.

The thing get irks me is that Gates made a deal over a product that didn't even exist. I would almost call this fraudulent and I'm not sure if there are laws against this nowadays.

It's also a ominous sign of the shady business practices Microsoft has been engaged in ever since, i.e. borderline illegal.

People have a making deals over vapourware since the dawn of time.

Nope. According to the MSFT timeline linked in the article, Allen & Gates wrote their BASIC during January 1975.

They completed it on 01-FEB-75, shipped it on 01-JUL-75 and signed the licensing deal on 22-JUL-75.

So I don't see it as a product that didn't exist.

HOWEVER, what I do see as shady is 01-MAR-75: "Paul Allen joins MITS as director of software."

So they wrote BASIC, Allen joined MITS, then Allen signs the licensing with Gates. Hmmmm... that seems a bit off.

That MSFT timeline looks screwy to me. For instance, the entry that you referenced:

February 1, 1975

Bill Gates and Paul Allen complete Altair BASIC and sell it to Microsoft’s first customer, MITS of Albuquerque, New Mexico. This is the first computer language program for a personal computer.

By all accounts, Allen didn't demonstrate the BASIC interpreter to Ed Roberts of MITS until March. Again, according to various accounts, they worked intensively to develop the software in time for the meeting. I really doubt that they finished on February 1 and then sat around for a month! In fact, Wikipedia says that a "preliminary" version was completed in March 1975.

Even assuming that Allen joined MITS before the licensing agreement was signed, at a small company like MITS the licensing decision wouldn't have been his, or at least not his alone. And, since Allen demonstrated the software for Ed Roberts, Roberts would have known of Allen's involvement with Microsoft. So, there wouldn't have been an undisclosed conflict of interest.

> That MSFT timeline looks screwy to me.

You... you're not... you wouldn't be suggesting that _MIcrosoft would LIE_ would you?!

*Falls off chair*

And Paul Allen wrote the boot code on the plane flying over

There was no fraud, or anything close to fraud. It wouldn't have been fraud unless they failed to deliver a product as contracted for which they had already been paid. But, in fact, in first face-to-face meeting with MITS, Paul Allen demonstrated a working BASIC interpreter.

But, I can see where that wouldn't have been apparent from this article. Using words like "pitch" gives one the impression that Allen/Gates were selling an idea instead of a functioning product.

".......selling an idea instead of a functioning product."

That definitely hit a nadir with Windows 11, then.

On his own blog Gates said that when they made the deal with MITS they didn't have a product, not even a single line of code had been written. I doubt that the customer was aware of this.

I quote: "We coded for the two months to create the software we said already existed" (gatesnotes.com)

On July 23, 1975 Microsoft gave the world the first BSOD.

"B" being black? It would have been odd if there had been colour displays available[*] for such a micro back then :-)

* there quite possibly were, but not for sale to hobbyists and not running on 8-bit home computers.

Also in 1975, launched at the Paris Motor Show in October 1975

[1]https://en.m.wikipedia.org/wiki/Lotus_Esprit

Did Musk ever get one of the 007 Spy Who Loved Me models modified to a proper transforming car/submarine, or was that just another failed project?

[1] https://en.m.wikipedia.org/wiki/Lotus_Esprit"

From what I’ve read in wonderful books such as ‘Hackers’ by Steven Levy, the BASIC interrupter was much sought after, but rather on the pricey side. So, not unsurprisingly, people started copying the paper tapes, and before long, there were multiple copies of the BASIC showing up at places like the Homebrew Computer Club meetings.

Which prompted Bill Gates to write his legendary open letter to computer hobbyists, in which he accused most of them of being thieves, and berating the hobbyists for openly stealing the fruits of his labours (3-man years of programming, as he put it - although it’s a pity he didn’t use the phrase ‘blood, sweat and tears’).

The letter was dated February 3, 1976. So it was really only a matter of months between the release of his BASIC, and its widespread pirating. I think that was a really big kick in the balls for Bill, and would shape his business attitude, going forwards. No more Mr Nice guy. And the rest, as they say, is history.

I don't recognise everything in the photo, but there is the famous triumvirate of Tandy/Pet/Apple. Was it a foreshadowing of things to come that the Apple was relegated to back of the photo? :-)