First portable computer?

[Eudimorphodon]

So how would you establish the eras of computing and where in those eras would you place the Osborne and how did it compare to other portables of that era? There's going to be some overlap or eras too I'd imagine.

I'm not sure that "era" means anything here? Terms like "era" or "generation" suffer from severe imprecision when applied to diverse groups/population, and things can be sorted according to any criteria that appeals to you. And the question was what was the world's first "portable computer", not what was the first object that fits this very specific arbitrary definition of what a "computer" is... or what "portable" is, for that matter. If it's a digital electronic Turing-complete programmable device (IE, not a fixed-function calculator) that's self-contained and small enough to be reasonably transported by a single person (it's probably fair to limit this to devices that aren't much heavier/unwieldy than other portable productivity tools like a sewing machine or portable typewriter) it would seem to fit the bill, and therefore the fact that an Osborne has a "generic operating system" while a TRS-80 Pocket Computer/Sharp PC-1211 doesn't is immaterial.

How about this one:

Honestly, if we *are* ruling out things like numeric-only calculators, gunnery computers (FWIW, the FADAC also weighed 200 pounds), etc, I think the MCM/70 might be the best answer to this question. The prototype was being demonstrated in 1973, the same year as IBM's SCAMP, the production version came out slightly before IBM's 5100, and, importantly for the framing here, the MCM/70 is based on a microprocessor, IE, the 8008, not the multi-chip array required by IBM's PALM processor. The prototype version of the MCM/70, named the "Executive", was actually squeezed into a briefcase and could run, if fairly briefly, on batteries:

(Photo circa August 1973.) In short, other than the one-line screen and being a *lot* heavier the prototype MCM/70 was almost the same computer as the Epson HX-20, often considered the first "notebook" computer, from 1982.

The main ding against the MCM/70 is the full version cost almost ten thousand dollars (a stripped down 2K version with no storage was still $4,700), so you could make a coherent argument that anything that costs that much doesn't count as a "personal" computer, but, well, it's definitely portable. (If somewhat less portable than the prototype.) If you want to argue about what the first "affordable" portable personal computer was... I dunno, maybe there's a case to be made for the AIM-65:



(Note that a simple two-piece plastic case to hold it all together was a few bucks more, $375 is naked.) It has both a printer and a 20 character alphanumeric display included, so it's not a "numeric-only calculator", nor is it limited to octal/hex pad interaction like a KIM. For well under a thousand bucks you could expand one of these enough to run BASIC in a decent amount of memory, and, again, easily passes the "portable as a typewriter" test.

I'm not going to completely leave the Osborne in the cold here, it does seem to be the first mainstream "portable business productivity computer" to go into large-scale production; the overall design for "sewing machine" portables seems to have been kicking around for at least five years or so, but it seems to have taken about that long for the stars to align to make possible to produce at a reasonable price point. I would just say it's important to stress that it wasn't an "innovative" machine in terms of technology; it's a computer that could have been built in 1977, but it would have cost $5,000 instead of $1,795 and the applications that made it such a hit when bundled at that lower price (WordStar, et al) hadn't been written yet. The aforementioned Pocket Computers that came out the year before the Osborne were *far* closer to the bleeding edge of technology (LCD displays, CMOS semiconductors, battery and power management, etc) which ultimately lead to our modern machines. From that standpoint CP/M luggables were already dinosaurs they day they came out.

 

[johnx993]

We have some great contenders here. AIM-65 looks good as the 'mass market' entry, and the MCM/70 for 'publicly available'.
But I should mention that defining computers as using text/strings essentially eliminates many of the early machines (I'm thinking Altair, Mark-8, etc.) from even being considered computers. These were originally front panel binary machines. Even the 4K MITS BASIC omitted string functions.

 

[snuci]

The main ding against the MCM/70 is the full version cost almost ten thousand dollars (a stripped down 2K version with no storage was still $4,700), so you could make a coherent argument that anything that costs that much doesn't count as a "personal" computer, but, well, it's definitely portable. (If somewhat less portable than the prototype.)

I also think the definition of "portable" changed over the years. The MCM/70 was portable in that it could be taken from one area to another being carried by one person and it could sit on desk. Most computers were minis or mainframes at the time.

Also of note, the production MCM/70 also has an internal battery pack. It's more or less an internal UPS with a short runtime (not mentioned how long). At one point I was going to replace the batteries and see how long it would run with them because the lead acid batteries it has are still available. But in my case they have long since been dead so the originals have been disconnected and sealed to not cause any battery damage. It was pretty pricey to test it and probably never use them again.

According to the MCM/70 brochure:

POWER - FAILURE PROTECTION
Method: Internal Batteries
Transient Power Loss: System continues under battery power
Extended Power Loss: Initiates orderly shutdown; workspace saved ; automatically reloads and continues when power restored.

 

[Eudimorphodon]

I also think the definition of "portable" changed over the years. The MCM/70 was portable in that it could be taken from one area to another being carried by one person and it could sit on desk. Most computers were minis or mainframes at the time.

Certainly. Not very many people would call an old sewing machine luggable “portable” today, so far as that goes. (Even at the time a lot of Osbornes never went anywhere, people bought them because as a package deal with all the included software they were a cheap date for a business computer.)

The old enough fogies here might remember how they used to sell “portable” 19” color TVs that were just a heavy as the tabletop models, the only distinction was the portable had a handle molded into the case. For laughs I looked it up and the 1978 vintage Compucolor II (13” color display with a floppy drive built in, detached keyboard) only weighed about ten more pounds than most luggables, so… I dunno. Definitions seem like they could be pretty slippery here.

 

[KE6JAG]

Wiki (https://en.wikipedia.org/wiki/Portable_computer) has a article on portable computers. But some of them are portable only because they were mounted big military trucks. And yes they have dates.


Tom
KE6JAG

 

[cj7hawk]

I'm not sure that "era" means anything here? Terms like "era" or "generation" suffer from severe imprecision when applied to diverse groups/population, and things can be sorted according to any criteria that appeals to you.

Well, even if you said that, you defined the "era" for the question as "microprocessors" which isn't what I was thinking, but it's a reasonable enough demarkation for the era.

Without something like that, you end up with this;

 

[Eudimorphodon]

Well, even if you said that, you defined the "era" for the question as "microprocessors" which isn't what I was thinking, but it's a reasonable enough demarkation for the era.

Once upon a time the term "computer" referred to someone employed to sit around all day using their built-in three pound-ish ball of nerves combined with external memory (pencil and and paper) to calculate things like navigational star charts and artillery range tables, so if we don't bother applying any brackets at all the first "portable computer" was invented between around six hundred million and a billion or so years ago. (Although I suppose even this might be making the unfair assumption that a "brain" needs to be a distinct specialized organ. Even the simplest single-cell animals that are capable of motion generally possess rudimentary ability to make "calculated" responses to external stimuli, and in some cases even seem to be able to "learn" from experience, which implies some kind of memory.)

But I should mention that defining computers as using text/strings essentially eliminates many of the early machines (I'm thinking Altair, Mark-8, etc.) from even being considered computers. These were originally front panel binary machines. Even the 4K MITS BASIC omitted string functions.

That's certainly a point, but... given how open-ended this question is it sees reasonable for this context to limit "computer" to something at least remotely in the ballpark of something a modern reader would recognize as one. Obviously this a slippery slope. Prior to the late 1970's most people who didn't work in the field would probably vaguely imagine a "computer" as being a wall of refrigerator size boxes bristling with reel-to-reel tape decks, switch panels, and oscilloscope screens tended by lab-coated wizards, and in that context a portable computer sounds like something Spock would wave around and consult after discovering a redshirt was just devoured by a living carpet(*). But we live in the post-home/personal computer explosion world, in which the word "computer" to most people means a thing like what you're using to read this forum today, IE, a thing that has a keyboard and display device capable of processing "human" language. Fair or not that would pretty thoroughly exclude mechanical devices along with digital computers that are programmed with punch cards, binary front panels, and hex keypads. There's definitely still some gray area in there when you consider "programmable calculators"; realistically there's not a lot of difference between an HP-41 and a Radio Shack pocket computer (and someone skilled in programming the HP machines will be happy to talk your ear off about how much power the HP's language gives you once you put in the effort of learning it compared to the limited BASIC of the pocket computer), but one of them is still going to feel more like a "computer" than the other if you just hand them to a newbie.

(*) FWIW, the "Spock" definition is actually kind of a useful one for what counts as portable computing *today*; these days you really can hold up your tricorder, IE, cell phone, to something, "scan" it, and store the collected data, upload it for analysis, etc. Of course most of the time data flows the other way, which wasn't something we saw very often on Star Trek; you'd think by the 23rd century they'd have good enough connectivity to use their tricorders as direct terminals to the ship's computer? I mean, their communicators usually worked, just slap a modem in that thing...

If we use a looser version of the Spock rule, IE, a "portable computer" is something you should actually be able to *use* on the go rather than needing a power outlet, which you're unlikely to find on a hostile planet, then I would think it's fair to say that most of the candidates we've mentioned, including the Osborne, *fail*. Even if *theoretically* you could lug along a lead acid battery big enough to power your SCAMP for any significant length of time you're still going to have to sit down and set up to use it. If we make that a requirement then I think we're back to arguing if a programmable calculator counts, and if it doesn't, does a Pocket Computer count, or do we have to consider giving the crown to the Epson HX-20? (Although, again, that prototype version of the MCM/70 is looking pretty solid here...)

 

[johnx993]

The Osborne had a portable battery pack and a dedicated port for it.
But lugging around all that required someone who worked out regularly.

The concept is certainly a moving target (through time).

 

[krebizfan]

Since the era of first portable would have been the 70s, I went with the military definition for a portable device at the time. Does nearly everything the full size predecessor did, weighs less than 75 pounds, had a convenient carry case, needed minimal setup. The carry case could be specialized luggage like most typewriters had or could be built into the design like a lot of industrial and medical equipment.

Weight was why I removed the Olivetti designs with built-in teletype from consideration. I don't know the MCM/70 enough. Was the internal display intended to be the primary output or was it like the Olivetti design just for system status?

The Osborne and Kaypro battery packs seemed most useful to sales people since the demo could be moved from the conference room to an office without stopping. The battery wouldn't last much longer than that.

 

[snuci]

Weight was why I removed the Olivetti designs with built-in teletype from consideration. I don't know the MCM/70 enough. Was the internal display intended to be the primary output or was it like the Olivetti design just for system status?

No, the display is for general output. I typed in the game "horse" where the one line showed each horses position during the race. Further to the carrying case, I don't have the MCM/70 case but there was an auction that had an MCM/70 in portable case and it looked like a typewriter style case of the era. Here is a picture from that auction.

 

[1944GPW]

If we are to include fiction here (re the Spock reference) then it seems fit to mention that an episode of Thunderbirds specifically mentions International Rescue's "portable computer".
This was Episode #4 'Sun Probe' which aired in 1965, predating Star Trek.
Built by Brains (of course) it forms part of the plot when he mixes up what suitcase it is contained in and instead loads Braman the robot aboard Thunderbird 2.

 

[Eudimorphodon]

Since the era of first portable would have been the 70s, I went with the military definition for a portable device at the time. Does nearly everything the full size predecessor did, weighs less than 75 pounds, had a convenient carry case, needed minimal setup. The carry case could be specialized luggage like most typewriters had or could be built into the design like a lot of industrial and medical equipment.

There's a slippery slope buried in here, which again goes to the definition of portable. There were outfits very early on that were happy to sell you specialized suitcases that strapped a TRS-80 or Apple II or whatever to a baseplate with all the cabling intact, allowing for easy transport and very rapid setup. IE, just set the thing on a table, pop off the lid, and hook up to a TV/monitor that was either already onsite or you brought with you in another box. Pick a small monitor like those 9" Sanyos that were often sold with Apple IIs and the total weight isn't much different from an Osborne. (Heck, it might be easier to carry because you can balance the load in both hands.) This is what people did before the luggables appeared and would definitely qualify as "portable" compared to the alternatives.


(FWIW, it looks like an Apple II and a pair of disk II drives weigh about 20 pounds combined, the case weighs 17 pounds, and the best figure I can find for a Sanyo VM-4092 seems to be around 10 pounds, so... 50 pounds for this setup, all in one box. A bit of a backbreaker but still less than 75, about the same as the IBM 5100.)

Of course if you take this to the most absurd extreme you could argue that if you had access to a circus train and a crew of sufficiently burly roustabouts you could take ENIAC on the road with you. Just be sure to allow a little extra time for setup when you get where you're going.

 

[Eudimorphodon]

If we are to include fiction here (re the Spock reference) then it seems fit to mention that an episode of Thunderbirds specifically mentions International Rescue's "portable computer".
This was Episode #4 'Sun Probe' which aired in 1965, predating Star Trek.

FWIW, I don't think they ever actually call the Tricorder a "computer" on Star Trek? To modern eyes it performs a lot of functions that would lead naturally to assuming that it incorporates a computer, but there's definitely some evidence that it isn't, or at least a fully functional one. Take "The City on the Edge of Forever", where it's revealed that despite apparently being able to *record* data at incredibly speeds it apparently requires an external memory/processing module to review it.

It makes very little sense today, obviously, but I suppose we need to cut the writers some slack. We may not ever get transporters or warp drives, but the fact that computer technology of today often seems outclass what they thought the 23rd century would be capable of was probably a legitimately surprising development.

 

[1944GPW]

Surmising here, but perhaps Gene Roddenberry knew that Star Trek's audience would have found a tiny hand-held box being able to record and also process to give meaningful results stretching their credulity a bit too far, bordering on magic.
Thus uploading to a mainframe computer, ship's computer or some external device etc. (it's been a long time since I watched 'City on the Edge of Forever' - great episode!) would have kept the Tricorder within the bounds of 1960s people's understanding of the limits of portable/wearable technology....?

 

[krebizfan]

Miri had the smaller biocomputer that got moved to the planet able to produce a solution on its own even if everyone involved would have preferred using the ship's computers for the task. Seems rather self-evident that a larger computer could do more than a smaller computer.

Except for speech recognition and FTL, the 60s series did a good job of making most technology seem to match what would happen in the next decade. The future but not complete magic.

 

[MiaM]

View attachment 1307961

Tangent: It's funny and/or sad that a LA company at the time had to travel all the way to the Bay Area / SF to get a shot of their product with modern public transit (BART) in the background.

 

[johnx993]

FWIW, I don't think they ever actually call the Tricorder a "computer" on Star Trek?

The book 'Making of Star Trek' (co-authored by Gene Roddenberry) p.174
states the Tricorder is a 'sensor-computer-recorder'.
Three main functions, so I guess that's where the got the 'Tri'.
-J

 

[johnx993]

Thus uploading to a mainframe computer, ship's computer or some external device etc. (it's been a long time since I watched 'City on the Edge of Forever' - great episode!) would have kept the Tricorder within the bounds of 1960s people's understanding of the limits of portable/wearable technology....?

Yes, and the medical data in Miri was also sent up to the big computer on the Enterprise for further analysis.

But what's amazing about the CotEoF episode is that Spock uses a row of vacuum tubes to build the 'mnemonic memory circuit'!
10 tubes, IIRC. So does that make the Tricorder a 10-bit machine? Or is there some parity/checksum business in there?
-J

 

[johnx993]

Not knowing anything about the Tricorder memory system, and knowing 10 bits ain't diddly squat for storing streamed data from a time vortex, I can only surmise Spock brought data in serially a block at a time from the raw recorded file to be corrected/amplified/whatever by the vacuum tube circuit then stored in another file to be viewed later. This would jibe with the fact that it took several hours to 'process' the data.

The typical use of a Tricoder on a landing party is sense and record. When they beam up to the ship, I'd imagine debriefing and analysis of the recorded data.
So it would make sense there is only minimal computing power for use on the surface- as a supercomputer with the combined knowledge of the galaxy is just waiting upstairs.

 

[Eudimorphodon]

But what's amazing about the CotEoF episode is that Spock uses a row of vacuum tubes to build the 'mnemonic memory circuit'!
10 tubes, IIRC. So does that make the Tricorder a 10-bit machine? Or is there some parity/checksum business in there?

Standard flip-flop memory requires a couple triodes; they did make dual triode tubes, but if you're stuck using stuff from the 1930's equivalent of Radio Shack you might be needing two tubes per bit? And of course that's not including the logic you'd need in front of it to route the set/reset signals.

Without invoking "magic" like Spock having built some kind of positronic brain inside that Jacob's ladder my theory is that the coils of wire *around* the ladder are actually a serial delay-line memory and the tubes are sense amplifiers. Maybe it works by "snapshotting" a frame at the ludicrous I/O speed that the data was recorded at and recirculates it while it's being read back out at a rate the processor in the tricorder can actually grok?

Of course if this was TNG-era trek we'd just handwave it being tachyons or something.