pdw

10 ms·

91.
by pdw 3y ago
It's amazing that (if I understand it correctly) the entire security of the system rests on the user not being able to execute arbitrary code. One wrong-code bug in the (privileged) compiler backend and it all falls down.
92.
by pdw 3y ago
Algol 60 also allowed whitespace in variable names. But they had a solution to avoid Fortran's confusion: keywords had to be specially marked. https://en.wikipedia.org/wiki/Stropping_(syntax)
93.
by pdw 3y ago
I'm not familiar with Sail, but the reason for separate logical operators is that you want short-circuiting behavior. For example, in: (a != NULL) && a->x You don't want to right-hand side to be evaluated if th
94.
by pdw 3y ago
Here's my transcription of that book: https://2k38.be/misp/ compiler.pas can be compiled with a modern Pascal compiler, but the resulting compiler cannot compile itself. I don't know if that's caused by
95.
by pdw 3y ago
Similar: Xlib and libX11.
96.
by pdw 3y ago
For reqwest, using the feature "blocking" will enable the reqwest::blocking API.
97.
by pdw 3y ago
The guy knows about emulators. They don't help if the software itself is lost.
98.
by pdw 3y ago
Didn't IPX always run on top of Ethernet? I thought it was pretty much a direct derivative of Xerox PARC's research network protocol.
99.
by pdw 3y ago
More accurately, before Microsoft cared about the Internet.
100.
by pdw 3y ago
Some APIs require the use of setenv. For example, you need to setenv("TZ", ...) if you want to use multiple timezones.
101.
by pdw 3y ago
That's weirdly similar to the alchemical _Axiom of Maria_: "One becomes two, two becomes three, and out of the third comes the one as the fourth." Though the only thing that proves is that mystics all over the world could cou
102.
by pdw 3y ago
It was all extremely silly. Debian took a different approach: before 2005, they put all crypto packages in a separate "non-US" archive, hosted in the Netherlands. American developers weren't allowed to upload there. That way,
103.
by pdw 3y ago
Gandi was sold earlier this year. The new owners do not have a good reputation.
104.
by pdw 3y ago
I'm more curious how they squared this with Apple Records.
105.
by pdw 3y ago
Nah, email and classic Unix use entirely unrelated date formats. The RFC 822 format looks like "Thu, 31 Aug 2023 20:53:11". Unix / ctime(3) uses "Thu Aug 31 20:56:52 2023".
106.
by pdw 3y ago
Almost everything that claims to implement ISO 8601 actually implements RFC 3339 instead.
107.
by pdw 3y ago
It's frustrating when articles like this don't mention the version of the software tested. Because that does matter. From the PostgreSQL 14 release notes: "Allow hash lookup for IN clauses with many constants (James Coleman,
108.
by pdw 3y ago
Now I'm confused. Why would you want duplicate entries in your command history? I get that on an overloaded system in the 80s it might have taken a perceptable amount of time to filter out duplicates. But on a modern system you surely
109.
by pdw 3y ago
Apologies, it's a note, not a footnote. But notes in ISO standards aren't normative either. This is a general principle of ISO standards, see https://share.ansi.org/Shared%20Documents/Standards%20Activi... : &
110.
by pdw 3y ago
You're quoting from a footnote. The footnotes are not part of the normative text, so how you choose to interpret it is irrelevant. The actual definition of undefined behavior given by § 3.4.3 is: "behavior, upon use of a nonportab
111.
by pdw 3y ago
Trapping on integer overflow was completely standard before C came along. Fortran and Pascal compilers did it. (Lisp compilers transparently switched to bignums :)
112.
by pdw 3y ago
This is why I hate it when people describe C as "portable assembler". The Usenet comp.lang.c FAQ was already thirty years ago warning people not to write overflow checks like this.
113.
by pdw 3y ago
Integer overflow is still undefined behavior in C23.
114.
by pdw 3y ago
Or the mistake is that we're running different trust domains on the same CPU core. All the various mitigations have already been very costly (something like a performance loss of 10% on current-gen CPUs?). Getting rid of speculative ex
115.
by pdw 3y ago
Well, according to the article Kermit's performance problems have been fixed :) "Although ZModem came out a few years before Kermit had its performance optimizations, by about 1993 Kermit was on par or faster than ZModem."
116.
by pdw 3y ago
We don't. To be honest, I didn't realize PyPy supported Python 3. I thought it was eternally stuck on Python 2.7. So the good: It apparently now supports Python 3.9? Might want to update your front page, it only mentions Python 3
117.
by pdw 3y ago
OpenBSD is not a fast system.
118.
by pdw 3y ago
That could have been a mainframe, but I think factories are much more likely to be using AS/400 aka IBM i. That runs on regular IBM Power servers these days.
119.
by pdw 3y ago
This is an interesting overview of mainframe vendors and how many customers they might have: https://arcanesciences.com/os2200/app1.html Most of the "dwarves" are now using emulation...
120.
by pdw 3y ago
I half expected to see this instead: https://www.youtube.com/watch?v=lqfdoPAxiVk

More ›