my multimedia gallery. In those days search engines were quite raw, and you could usually see the exact query that a user had typed to be routed to your Web site.
So imagine not my disappointment, but that of the searcher, on arriving at an artful picture of a huge paperclip given that their search query was for "huge vacuum penis expander".
In response to a thread on HackerNews about "C++17 constexpr everything" and someone saying "Code generation is widely used, at least in automotive. Comparison of hard-to-read and hard-to-debug advanced template/constexpr machinery vs code generated by standalone tool that is easy to read and easy to debug would not be taken seriously" I responded:
Both approaches have their problems, but resolving to compile-time constants with simple expressions is NOT "hard-to-debug" if done with care. As ever, tools can be abused, and real life can astonish.
Example: for the credit dept of a now-ex investment bank many moons ago we had a set of blessed (including with the correct correlations) random numbers pre-computed and baked in via code generation.
I discovered that I could actually generate numbers faster at run-time with highly-tuned code because of the high cost of paging in the large-precompiled numbers array across the network.
Yes, that really was astonishing, and meant that we could actually generate new numbers each time with the right characteristics faster than pre-generating, and vary those characteristics, which made our Monte Carlo algorithms happy. This for the derivatives/credit desks of Lehman Brothers...
I find it amazing that the Casio fx-100 pocket calculator with which I did my A-levels* is still useful for me today, eg to check investment numbers for my business and compute energy consumption figures circa 2018.
I don't think that I've changed the batteries even 10 times over that 30+ years.
*There was some suspicion that 'my' fx-100 was swapped inadvertently with that of one of my fellow A-levelers at some point, and I thought that we then scratched our names in the cases to avoid further mishap, though I don't see any evidence of such marks right now!
In response to seeing again the infamous We can't send mail more than 500 miles story, I recall something similar!
We had 10x faster file transfers in one direction than other at a large oil company's campus. After weeks of effort and hiring some very expensive test gear and a person to operate it, turned out to be kinked fibre causing large (NFS) packets to fail frequently in one direction.
Many years ago I was working on a small island in the Med, staying in a friend's flat in a village in the back of beyond. There was no wired broadband. GPRS (early slow mobile data) was the new kid on the block, and the (near) monoploy mobile telco wasn't selling access to it. In fact, it was free if you know how to use it. But reception in my friend's flat turned out to be far too patchy for streaming, which killed my habit of listening to Radio 4 in the bath!
Luckily I knew someone who worked in the telco and mentioned this to him. After a few days he suggested that I try streaming Radio 4 again, and, well, it was a sea change. Rock solid. Turns out that in the absence of any paying customers the tech guys had had some fun and electronically pointed all the antennae in reach at my friend's flat, so I had probably the best mobile data on the whole island. And perfect streaming of Radio 4's dulcet tones in the bath!
(Soon after, the telco's CFO had an epiphany and realised that this service might be worth something, and started charging an eye-watering rate per MB, but found some real customers prepared to pay it. No more R4 for me!)
Near the start of my time in the City, as a tech admin, someone up high pronounced that there would be company-wide pagers for tech problems. For a long time nothing happened and then I was told that there would be two, one for my then desk (Swaps) and one for the whole London office as I recall. So I set up our desk one to send me useful messages, which it occasionally did. Carrying the London office primary pager was apparently worth a whole extra day's work every week. And no one knew how to send it messages. YES PLEASE!
Many uneventful page-free weeks passed, possibly partly because my colleagues insisted on showing me how the Hutchinson-Wampoa pagers would not survive being dunked in a pint of beer on a Friday night. And then I was told it was hardly my fault if I couldn't respond. Still rather more important was that no one yet knew how to send that primary pager a message!
Until a couple of pints into another Friday night, the pager actually went off. With some very inscrutable messages. And phone numbers that when I rang were nothing to do with the London office. And no one answered my calls back to my desk. Weird and worrying.
It seems that there was no crisis in the office: Hutchingson had simply decided to liven my evening by routing random other pager messages to me, possibly intended for bouncers.
I didn't get to keep that pager for much longer, but for the odd Friday fright it was definitely worth a free day's pay per week!
Site content copyright Damon Hart-Davis 1996--2017 unless otherwise stated.