Posts by Arthur the cat 3378 publicly visible posts • joined 18 Aug 2009
We were able to get permission to take taxis around London when visiting our HQ as "the motors on the tube will corrupt our important floppies".
The earlier version, and the reason(*) I got taxis between London rail stations, was that tube train motors would corrupt 12 inch "washing machine" disk packs.
(*) To be pedantic, it was the excuse - the reason was that lugging 12" disk packs on the Tube, especially in the rush hour, was a bugger to do.
One week I went over and put a small piece of Sellotape across the optical sensors on our their mice.
The Sun optical mice with specially patterned pads were great fun - turn the pad through 90° and they became very unreliable and nobody really thought about pad orientation as a problem (until they'd been had a couple of times).
tagline made me think it was going to be some overenthusiastic hole punch work...
I knew someone who needed hard sectored disks in a hurry and tried to add extra holes to soft sectored disks with an office hole punch. It worked better than I thought it would. (One out of six attempts stored enough data for what he was doing, long enough to get proper disks.)
… made by many. As an example, I present this 1969 SF novel by Jack Vance.
one house that was a repossession had just about everything removed, power sockets, radiators - the works
My first house (bought in the 70s) was a repossession and had everything left behind, including a load of rubbish that needed a skip to get rid of and a lot of definitely NSFW photos taken by the previous owner who was a professional photographer.
That test was done on Solaris in 2011 with the then current Java version.
11 years is a very long time in computing. An awful lot has been done to Java in that time and a single measurement from 2011 isn't really relevant today.
You might care to read this on GC improvements from JDK 8 to JDK 17 which covers changes from 2014 to 2021. You now have a choice of four different garbage collectors, so you can pick the one best for your work load.
as the memory must be swapped to disk even if unused
As someone else remarked, that hasn't happened in decades. Unmodified memory has no need to be written out so isn't. Unused memory probably doesn't even have swap space allocated for it.
A single active object per page can thwart the paging to disk.
Which is why many/most GCs compact to reduce VM usage. You really need to read up about the latest GC technology or you'll confuse yourself with long dead misconceptions.
GC-languages consume 100% more memory than reference-counted languages
No, it depends on the garbage collection technique used(*). You're probably thinking of a simple dual space collector. Even that doesn't use twice as much real memory except in very worst case, only twice as much virtual memory. Using ephemeral spaces reduces that considerably. From memory, it's possible to reduce GC space overhead to ~10% at the cost of extra GC work and still be practical.
Oh, and Rust isn't a GCed language. It can be, but it was specifically designed not to require one (or any other runtime environment).
(*) For full coverage of the topic see The Garbage Collection Handbook: The Art of Automatic Memory Management and its web site.
Of course the climate is changing, its been changing since the planet was formed.
Sure, but rate of change matters as well. One of the fastest natural shifts in CO₂ we know about is the Azolla event in which atmospheric CO₂ dropped from 3500 ppm to 650 ppm in 800,000 years. That's a ΔCO₂ of 3.56 ppm/kiloyear. In my lifetime CO₂ has gone from 313 ppm to the current 417 ppm, a ΔCO₂ of 1.53 ppm/year. That's 430 times faster than an unusually fast natural event and is definitely not a normal terrestrial process at work.
Arguably paranoia is the first requirement for anyone in the computer industry. We're all fighting against Resistentialism.
Client was a provider of subsea control systems for offshore oil and gas installations…
The more I learn about the North Sea gas infrastructure, the more I see the need to switch to renewables.
Back when the land side pipeline to Bacton was being installed the welders were very happy the fields had a copious supply of magic mushrooms growing in them.
For the curious, is it possible to say vaguely what could be on a met office supercomputer that would be quite so sensitive?
Not Met Office myself, but it used to be part of the MoD, and they work(ed) on the assumption of everything having sensitive data on it just in case.
Kagome refers to the Kagome or trihexagonal lattice. And yes, it's used in Japanese basket making, but there's also a lot of interesting physics involving kagome lattice shaped things.
So naturally, I had to write a C compiler in it, just to prove to myself that I could.
OK, that's perversion above and beyond the call of insanity(*). A major tip of the hat and enough alcohol so you never do it again.
(*) I've only written a calculator in it. I'm bloody jealous.
They are huge and take their time compiling, especially LLVM. The optimisation can be superb and is getting better all the time but it takes masses of time and space. Arguably 90% of optimisation can be done using just the techniques Frances Allen listed in her 1972 paper “A catalogue of optimizing transformations".
I've always laughed at the old joke; "A FORTRAN programmer can write FORTRAN programs in any language" but LOL, it's a fact.
Oh god yes. I've seen FORTRAN in Lisp:
(BEGIN
(SETQ X ...)
(SETQ Y ...)
...
(RETURN ...))
[And why does code markup not keep indents but add blank lines?]
To mangle an old trope:
You have a problem. You decide to use m4. You now have two problems and a maintenance nightmare that will prematurely age you.
[Speaking from experience. You may know what it's doing when you write it but the moment someone else touches it (including you 6 months later) you're in for a world of pain.]
Sorry, have I ended up in the Journal of Literary Criticism by mistake? I was looking for The Register. I'm sure it's round here somewhere. You must have come across it, full of raucous weirdos wearing faded T-shirts with incomprehensible slogans on them, arguing about God knows what.
If a Lisp system hasn't got tail call optimisation these days it's not a real Lisp.
In C you can get (the effect of) single function tail recursion with a simple loop, but the multifunction form is a swine. I've seen some horrible trampoline code written just to get that effect. I suspect this is because it makes debugging interesting - "A calls B calls C calls D so why does my stack show A calling D directly and what were B & C's arguments?"
And before that, nightmare territory of having to program directly at the per-bit level, by "literally" throwing the switches.
There was a stage before switches. I never did it but an older friend had the experience of "programming" hardware by plugging brass and ferrite slugs into a pinboard that then got inserted into an array of induction coils. Brass (non-magnetic) was 0, ferrite (magnetic) was 1.
All (untyped) languages aspire to the condition of Lisp.
Historically I'd say that Lisp was the test bed in which we as a profession worked out exactly what the hell it was we were actually doing when we designed and used computer languages. If you look at all the debates which came up in the Lisp community - scope, dynamic vs static binding, downwards funargs, upwards funargs, closures, single vs dual namespaces, continuations, macro hygiene, message sending, purely functional vs stateful programming - they reflected design and implementation choices that occur in all languages, but which could be tested rapidly in Lisp because it is such a simple and flexible language. As such it's in many ways the Ur-language from which all other languages have sprung even if it's always been a minority sport in real world use.
"For fun and profit" is an old trope(*), much used. Pr1me's manual for their text formatter used to have examples involving the "Raniburger Corp" handbook "Frog breeding for fun and profit".
(*) And much debated, see Stack Exchange's thread on it.
The scariest ever and now dead OS from the 80s was the thing that ran on Symbolics Lisp workstations.
Genera was wonderful, a really good OS that I loved(*) with a proper space cadet keyboard too. Shame about the hardware's need for cooling though, not so much fan noise as fan jet noise.
(*) Happy fun times back then, I had a Sun workstation, a Symbolics and the non-existent Smalltalk machine(**) on my desk(s).
(**) See other comment for details.
This article is the equivalent of the era of bubonic plague(*)
In some cases it's more
Your father's OS. An elegant system for a more … civilised age(**).
Multics, Burroughs' MCP and Genera all had features that today's mainstream OSes lack or do badly.
(*) First known bubonic plague outbreak was ~3000 BCE, it's still around today. Long era.
(**) ObXKCD.
But directors should be broad minded
I think what RFC 2119 says about "MUST" and "SHOULD" is appropriate here.
As I said, keeping your work force happy is probably the best thing for the shareholders in most cases - for my start up it was vital because for software companies the workers are the company, there are very few physical assets. My point was that the union was demanding boards ought to prioritise workers above shareholders which they can't do.
BT have a choice to prioritize between its "shareholders or workers"
Nope. Legally BT management is required to act for the best interests of the shareholders and nobody else, and has no choice in the matter.
Now, it's a perfectly reasonable argument that shareholders' best interests are best served by keeping a happy work force that doesn't bugger up the business by striking, especially if the business is making good profits, but shareholder capitalism needs the management to answer to the business owners, not "stakeholders". If the unions don't like that, we're a mixed economy, so they're always at liberty to set up a union led workers' cooperative competitor to BT.
[And of course I'm not even touching on the Principal/Agent conflict inherent in a system with hands off shareholders and an executive management board.]
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
