Hacker News Reader

Comment by magic_hamster | Hacker News Reader

This was a great read. Some commenters discussed the nature of the interview, and compared it to today. But personally I get this feeling that this was still in a time when software was developed mostly in bubbles. Knowing how to code was far less obvious back then than it is today, and knowing the right stack meant you could get hired on the spot. Also, I don't know of it's nostalgia, the way it's written or something else, but it feels like a lot of engineering stories from that period have a certain sense of pioneering or discovery, while it's definitely far reduced today.

View on HN

can confirm all this, and more.

It wasn't until the late 2000s when programming became easy enough for the mainstream and you started to see "coding camps," upwork and the like.

In the mid-90s, you had crazy demand for development but needed to be a brain surgeon to get hello world to work, let alone a website to be remotely reliable.

There was no automated testing, let alone CI.

Source control was sometimes used, sometimes not. There were LANs but there was no "web" let alone SaaS apps - there were maybe 1,000 websites and Yahoo! listed them all in a list. It was like Product Hunt, but even smaller.

You almost have to wonder how anyone learned to code? and you'd be right for asking: it mostly happened in top-20 colleges and a handful of companies. I remember showing PhDs about PKZIP and having them not believe it was possible to compress data without losing information - I had to literally show them the (rough) algorithm.

Truly, it was a time of magick.

(I don't miss it: the pay sucked, people treated "programmers" like crap, it was often impossible to reproduce issues, and of course you could only work in an office with dedicated hardware. Compared with today, it felt like the stone age of software development.)

If the demand was high, why was the pay so low?

The work wasn't critical… yet. It was mostly exploratory, a way to soak up an R&D budget, or get R&D tax credits.

It was harder to find the opportunities and therefore leverage multiple job offers.

Jobs were a lot more local, too. People didn't move around as much. In general, not just in software. You also couldn't just spend a couple minutes and get a ballpark for what employers were paying in another city, let alone get a decent sense of what a bunch of markets looked like—it'd take some real effort. Reduced mobility and information being much harder to come by kept salaries lower, I expect.

Was it? In 1995 I was shocked to discover what the “computer guys” were getting paid, and decided I ought to be writing code (etc) as my official job instead of just the thing I do within my other job.

Compared with other types of labor, yes.

Compared with management, no.

> I remember showing PhDs about PKZIP and having them not believe it was possible to compress data without losing information - I had to literally show them the (rough) algorithm.

They were right; that is a well-known and trivial-to-prove theorem about lossless compression.

Lossless compression works because you only apply it to very particular types of data. It doesn't and cannot work on general data; that's why we have specialized compression algorithms for every different kind of data.

I took graduate compression in the mid-90s.. JPEG was already invented and we studied that.. there was some handwaving but it was not as primitive as some comments make it out to be today

JPEG is lossy

of course -- JPEG includes means to adapt to the image data patterns, however, which is mentioned in these threads as architecturally important to useful lossless compression, too. (did you really think that 'JPEG is lossy' is more than a 'gotcha' reply?)

They were not concerned about the compression ratio - they were concerned about data integrity/loss !

I did a CS degree in the UK in the late 80s. The introductory language was Pascal running on a Prime Mini Computer. ANSI C had just come out, but none of us had computers. You had to book time in a lab. The Aztec C compiler we used was on a floppy disc and horrendously buggy. All the coding we did was minimal exercises.

I didn’t even hear of object oriented programming until the 90s, and actually learned how to write useful programmes in Perl and C on the Sun machines at my first job. I learned Unix from scratch from the man pages. It was a different era.

I wrote stuff in the mid 90s and I am pretty dumb

Today you need 11 frameworks to even write a hello world

There's a recent novel by Tamara Shopsin called "LaserWriter II" [1] that gives off that "sense of pioneering or discovery" feel for all ~200 of its pages. I highly recommend it if you're looking for a quickish read.

[1] https://us.macmillan.com/books/9780374602581/laserwriterii

I think this is just an artifact of the pioneers/total body of engineers: having the numerator stay relatively unchanged but with the denominator exploding.

Reply on news.ycombinator.com