[VLM]

I suspect some statistical weirdness going on in the precise formulation of the survey question.

I know for a fact 30M commodore 64s were sold in the 80s in the USA. Not all commodores, not all home computers, just the classic model C64, 30M units sold. That's in a country that used to only have 250M people, so in theory 12% of Americans as of 1990 had purchased one specific model, the C64, leaving only 3% for all other models combined, which seems very unlikely.

Some of those probably went to schools not homes, although schools were owned by Apple II in those days...

My suspicion is many of those were unused in basements and closets, or the question was phrased weirdly like "have you purchased a computer in the last three years" or "used a home computer in the last month" or they defined "home computer" to be "not an IBM (office) or Apple (school) product" or something like that.

[simonh]

Wikipedia says between 12m and 17m were manufactured, in total.

Trammel claimed 30m based on a remembered estimate of rough sales numbers per year, but the only estimate that’s based on objective evidence - serial number analysis - is 12.5m.

https://web.archive.org/web/20160306232450/http://www.pageta...

[VLM]

Yeah interesting. Maybe the 30M figure comes from 6502 shipments. I don't have the tab up anymore that was claiming 30M+ shipments.

Here's an interesting discussion link. Merely being on wikipedia doesn't mean its correct that site is a hive of disinfo in general:

https://www.pagetable.com/?p=547

This site even mentions the peculiar 30M figure.

I would tend to believe the linked site's serial number analysis result of exactly 12.5M. The americans did something like that to the germans in WWII, it turns out a remarkably small totally random sample of sequentially assigned serial numbers is enough to very accurately predict the highest number sold. Assuming very random sampling, which is never truly random, of course.

Doesn't change the overall outcome, however, when there's a stat that a small segment of an industry is "about" the size of what's claimed to be the entire industry, something's off in the numbers.

A mere 12.5M sold remains 5% of the entire USA population at that time, and honestly, having been there, almost everyone I knew had a PC clone or some apple product, usually a mac. The number must be larger than 10%. "a computer" was required at college ... ed.gov claims there are 19.4 M college students in the USA right now and google claims 332M people in the USA right now, so about 6% of the population are in college right now, so back in 1990 guess "around" 6% of the population was required to own a computer just to attend higher ed ... the claimed 10% seems like an incredibly low number.

[randallsquared]

> "a computer" was required at college

In 1990? No way. I was taking college courses (for HS credit) in 1990, and the first I heard of a requirement to bring your own computer was years after that.

Another thing that skews the numbers is that my household during the 80s had two Color Computers, a C64, and an Amiga 500, but no one else in my social circle had anything more general than a Nintendo or Atari console.

[SoftTalker]

I was in college in the latter half of 1980s. Computers were very much in use at the time for writing papers, and some classes required them for other work. Students were not expected to actually own one, however -- they would go to labs around campus which were basially just rooms full of PCs loaded up with all the common software.

Computer Science courses generally just required a remote dumb terminal, such as a VT100 or ADM 3a. These were available in a few rooms around campus as well.

[UncleSlacky]

In UK universities at least, circa 1990 we had computer labs/rooms which were home to dumb terminals, or if you were lucky, Atari STs running a terminal emulator. Some of us brought our home computers to university, they were mostly Atari STs and Amigas - I actually lugged my Amstrad CPC6128 and colour monitor to campus, but never used it for course work as there was no networking in the dorm rooms. Some new-build rooms had it by 1990, and a few people had PCs linked to it by then, but only used as terminals to the mainframes and minis on the campus network.

One person on my course caused a minor stir by bringing a "laptop" to a lecture to take notes in about 1991 - the keyboard was so noisy that they only did it once!

[ghaff]

I went to grad school for business in the mid-80s. I had a computer but I doubt there were more than a handful of other people in my class who had one. While I was there they went from a very limited computer lab in the basement with a few Macs, a Lisa!, and some DEC terminals to a much bigger lab of 286s off the library.

[dhosek]

I was a freshman in nerd school in 1986. Out of the 8 frosh on my hall, only 2 had their own computers (I was not one of them). I don't think any of the seniors in the singles on the other side of the hall had computers either.

[simonh]

I graduated in 1990 and owning a personal computer capable of running a C or Pascal compiler useful for academic work was an impossible dream. I was lucky enough to have an Amiga, but a C compiler such as Aztec C cost serious money back then. Open source existed but GCC started out on Vax and even by 1990 I think it only ran on very expensive systems like Unix minis and workstations.

[NateLawson]

As late as the mid-90s in California, some kids showed up at university without a computer. There were PC and Mac labs on campus that were open pretty late, as well as VT100 terminal labs open 24/7 (though these were only used by most students to check email between classes and were on the way out).

All engineering students had a computer, though.

[bombcar]

Even in 2000 computers weren’t required - the school had a computer lab you could use if needed, and it was only a few years before that they had begun requiring essays and papers to be printed from a computer.

[7speter]

My sister had a “word processor” for college in the mid 90s, it did spreadsheet stuff and had a floppy drive and a monochrome monitor, maybe those count as a computer too?

[wooosh]

Wouldn’t it be more than one person per computer though, seeing as it’s a "home" computer that would likely be shared?

[tingletech]

when I went to school, each school usually had only one or 2 apple ][s -- but we had a whole classroom of PETs in 7th grade that got replaced by a classroom of C64 in 8th grade. I saved up and bought a TRS-80 CoCo 2, otherwise there would not have been a computer in our house.

[7speter]

Maybe the military made secret clusters akin to the ps2 and ps3 distributed systems they made in the 2000s

[Transfinity]

Maybe. 15% sounds about right to me. Back in 1990 computers were expensive, bulky devices, required specialized knowledge to operate (Apple's ad copy notwithstanding), and had limited use outside of a few specific domains.

[netsharc]

Kinda amazing that nowadays anyone can go to a store and get a "super" computer that fits in their pockets in the form of a smartphone. Want to video call someone across the world? Buy one of those, setup an account for the Apple/Google store (yay walled garden), install WhatsApp or your messenger of choice, and ring ring!

Funny how many of the ads say "Call us for more information about our offerings!" instead of a "For more info: www....".

[ethbr0]

Call-out: we're also talking about mostly un-networked computers here.

1989 is the very beginning of "the Internet might be useful for general purposes by non-scientists." And incidentally, the same year BGP was dreamed up. https://en.m.wikipedia.org/wiki/The_World_(internet_service_...

So most computers were running boxed retail software with extremely limited hard drive data space (if present at all).

[loloquwowndueo]

Computers were networking years before the rise of the Internet. BBSes were a thing in the 80s.

[ethbr0]

We're talking about mass market adoption. Most people (by which I mean not-HN) weren't using acoustic-coupled modems over device-specific ports to dial into BBSs over long distance lines.

Computer control of modems using standardized Hayes(+) command sets over RS-232 serial ports (1981), the Bell breakup (1984) and subsequent competition, and increased modem manufacturer competition breaking price-per-baud standards and increasing affordability (~1988) were required enablers to take it mass market.

It took until the early 90s that you could (a) buy a PC at a reasonable price, (b) buy a modem at a reasonable price (that was guaranteed to work with your PC), (c) dial into an ISP's local number... from anywhere in the US.

Yes, some people lived in NYC, Chicago, SF, etc. and had access to knowledge and BBS infrastructure at local rates earlier than that, but most people didn't.