In the late 1990's I worked at a large bank's Online Retail Banking website running Netscape's iPlanet webserver on Unix (AIX). Being ~25 years ago, I'm a little foggy on the technical details but at a high level we started getting complaints from customers running IE having trouble connecting to our website. The reps from Microsoft blamed iPlanet, said it worked fine with Microsoft's IIS webservers, and had almost convinced the bank's senior management to simply replace iPlanet with IIS as it would be the trouble-free webserver for use with the most popular browser, IE.
Nobody on the tech team wanted that so we launched into a major tracing & debugging effort and eventually found that a change in IE caused it to start doing the SSL handshake slightly different if it thought it was connecting to a non-IIS webserver. Netscape provided a patch and we were able to keep iPlanet on our beefy Unix servers instead of migrating to a farm of IIS servers on tiny Windows servers (they weren't all that powerful in the 90'). This was about the time that the DoJ was going after Microsoft for non-competitive practices. I recall that someone on our team sent an email to DoJ telling them of our experience, but I don't think they ever got a reply.
It's a good option purely for cosmetics but don't rely on it for any kind of serious obscurity since it's trivially reversed. I've used it to great effect in the past to encode multiple integer values like start/end ID, sort order (0/1), etc. for cursor-based pagination. But that's only because there's nothing secret in those numbers. Just purely for convenience.
I once heard of some fancy data science that inferred that the company was the supplier's major customer from the gaps in the sequential ids they saw when using that supplier's API. That was useful in negotiations.
And so the company that did that data science realised they too were susceptible to exactly the same 'attack'. So they created a system to obscure the ids they were themselves exposing to their customers, using some cheap cut-down tea64 encryption iirc. My memory is it never went live, though.
Not that fancy (or at least not that novel). In WWII the allies used sequential serial numbers on destroyed and captured tanks to have a frighteningly accurate picture of German industrial output.
This is a common misconception. hashids is _not_ an encryption algorithm. It's just an encoding with a tiny bit of obscurity layered in.
hashids.org:
> Do not encode sensitive data. This includes sensitive integers, like numeric passwords or PIN numbers. This is not a true encryption algorithm. There are people that dedicate their lives to cryptography and there are plenty of more appropriate algorithms: bcrypt, md5, aes, sha1, blowfish.
Literally spooling message-filled tape onto a tape reel! I never even thought about that.
On the topic of message queues, I wonder if anyone ever designed a mechanical, asynchronous hardware IPC mechanism based on tape. Imagine: an endless-loop tape that stretches between a "sender" tape-drive on one machine, and a "receiver" tape-drive on another machine; with each tape drive locally tensioning the tape as it passes through, but letting it hang slack outside the drive; with two tape-width buffer boxes in between the two drives, for taking up the slack; and where each tape-drive has a strain gauge on its input-end capstan.
With this design, you'd never need to seek the tape (backward, at least) — the receive head would just read the tape forward at whatever speed it could until it ran out of slack in the tape (i.e. blocked on read); and the send head would write the tape forward at whatever speed it could until it ran out of slack (i.e. blocked on write.) The tape itself would be a literal, physical ring buffer.
Western Union built a store-and-forward message switching system in pretty much exactly the way you describe (but with paper tape rather than magnetic): https://en.wikipedia.org/wiki/Plan_55-A
Analog audio delay/echo machines work like this. They could have multiple read heads and you could move them relative to each other and adjust the speed of the tape to get various cool effects.
I recall one of the computing magazines of that era, possibly the Dec 77 issue of BYTE you mentioned, published the entire code in BASIC for a space-oriented game. Inputs were simple letter or letter+number commands and outputs were an ASCII drawing of a sector of outer space (complete with * for stars). You would shoot at an alien then skip to another sector before getting counterattacked.
My high school buddies and I desperately wanted to play this game but ... none of us had a computer! So, sitting in the cafeteria at lunch/breaks in our bell bottom pants, we walked thru the code line by line recording the values of variables with pencil on paper. This was when BASIC only allowed 1 or 2 character variable names and only had GOSUBs instead of proper functions. It was slow going at first but we eventually got to understand what sections of the code was doing and could replace line by line with, oh, its doing this again.
I'll have to look for that Dec 77 BYTE so see it that was the magazine. I didn't know it at the time but it was probably very formative for my future IT career.
What a great story! I remember also "hand running" programs at times.
While the Dec 77 issue of Byte[0] had Star Trek on the cover it didn't contain a BASIC listing. Could it have been either the May/June 1975 issue of Creative Computing[1] or one of the compilations (BASIC Computer Games[3])?
Nobody on the tech team wanted that so we launched into a major tracing & debugging effort and eventually found that a change in IE caused it to start doing the SSL handshake slightly different if it thought it was connecting to a non-IIS webserver. Netscape provided a patch and we were able to keep iPlanet on our beefy Unix servers instead of migrating to a farm of IIS servers on tiny Windows servers (they weren't all that powerful in the 90'). This was about the time that the DoJ was going after Microsoft for non-competitive practices. I recall that someone on our team sent an email to DoJ telling them of our experience, but I don't think they ever got a reply.