drpixie

9 ms·

241.
by drpixie 3y ago
Seriously - something similar to the threaded interpreted code of Forth might be a nice way to implement low level byte-code on a modern machine.
242.
by drpixie 3y ago
(Turns to phone booth, ripping off tie) "Sounds like a job for super Forth!"
243.
by drpixie 3y ago
It would be interesting if a program could define some new instructions by specifying the microcode :) Though that might make context switches rather expensive. And if you're going down that route, perhaps time to think about FPGAs.
244.
by drpixie 3y ago
> Seems irrelevant though - the internet exists. ... but it's not always useful. MSI motherboards doing "secure boot" can't check for key revocation until after they've booted :( Sometimes you just have to rely
245.
by drpixie 3y ago
> Also to my great disappoitment none of those CPUs has checked overflow for arithmetic operation. Yes - we're stuck with the historically convenient collection of operations and types. Our CPUs are amazingly fast but very little sm
246.
by drpixie 3y ago
Absolutely, and almost everything I want to follow has a feed. I run a tiny (Raspberry pi 3) miniflux instance as a web-based reader and have been very happy so far.
247.
by drpixie 3y ago
Just a few years later - Kraftwerk https://youtu.be/eSBybJGZoCU
248.
by drpixie 3y ago
This RISC/CISC debate made a lot of sense when the "guts" of the CPU (decoder, registers, ALU, etc) occupied almost the whole chip (or box of transistors), and when memory and CPU had similar speeds. The situation now is very
249.
by drpixie 3y ago
A smidgen is precisely defined ... just not in relation to another physical quantity. (And any one smidgen is not necessarily the same as another.) The smidgen is defined by its effect. A smidgen of X is the precise amount of X that produce
250.
by drpixie 3y ago
> Computational power ... could be harnessed by anyone with a few thousand dollars and the willingness to learn. That was the "vibe" of early mass computing. Anyone could be involved and anyone could make things better. It was
251.
by drpixie 3y ago
Are there really people who haven't seen it? Sad face :(
252.
by drpixie 3y ago
I have similar compiler history and employ similar practice. Keep things where they logically belong - the parser is for mapping tokens onto syntax, for that it needs tokens! It doesn't matter if lexing is done in a separate pass, or &
253.
by drpixie 3y ago
If I was parsing the input, then parse-time is the time to assign a token-id (I'd use an enum). Then you have a unique identifier and wouldn't need to use the fancy hash function. It's a nice function, takes good advantage of
254.
by drpixie 3y ago
Nice shortcut if you know the input is correct, but seems like it might be prone to failure with carefully constructed input strings. If I can find an input token that produces the same hash as a normal keyword, a program using this might m
255.
by drpixie 3y ago
Thanks - there was talk when Fuschia first became visible that it was to be locked down. The project site does not agree with that.
256.
by drpixie 3y ago
Ha ha - I had the opposite reaction ... "Oh dear, some god botherer has wrapped their special text in a shell script. Why is this on hacker news? #$%$%$@!!" Turns out I quite misdirected myself. Don't agree with all his tric
257.
by drpixie 3y ago
Isn't Fuchsia google's attempt to move phone off open systems, and onto all-google-no-license code? If it becomes a big thing, then only increasingly obsolete open-systems for us :(
258.
by drpixie 3y ago
I'm sure security is part of the reason for the ever increasing locking-down of android ... but I'm more sure it's an excuse and that security FROM the user is a huge factor in google's thinking.
259.
by drpixie 3y ago
Reminds me of being taught hand-cut recursive descent compilers, back in the day. The process was: define what info is required and passed around, then the code becomes (almost) trivial.
260.
by drpixie 3y ago
I was somewhat amazed that a significant university doesn't handle IPv6 - sigh.
261.
by drpixie 3y ago
You seem to be answering your own questions :) Of course, "abstraction lead[s] to less fine control" - at the the lowest (assembly) level, you can do almost anything - and make all the mistakes imaginable. Sometimes you want to ge
262.
by drpixie 3y ago
In general, simple interface is easy for the programmer to understand, and provides less opportunity for system bugs and strange interactions.
263.
by drpixie 3y ago
It is - until you want to do something interesting, then it's all ioctl(), recv(), and the various gather calls! I'd like to see an OS that drops the concept of files. Files are very low level (generally a stream of bytes) - the a
264.
by drpixie 3y ago
Wipe them off, Jim.
265.
by drpixie 3y ago
I was taught "Jack left port" and I know port (wine) is red.
266.
by drpixie 3y ago
And if you like the O'Brien books, read the C.S. Forester "Horatio Hornblower" books - historically & nautically accurate and very readable.
267.
by drpixie 3y ago
Everything I need to run I can run on Linux, either as Linux app or via Wine. The few things I can't run on Linux, they're not that important...
268.
by drpixie 3y ago
Agree. And I pay - with my time and expertise - by using Linux :)
269.
by drpixie 3y ago
Agreed - given a complete and correct test suite! But even so, "correct" code is not necessarily efficient or readable code. And note that writing the complete and correct test suite is likely to be a task of similar difficulty to
270.
by drpixie 3y ago
Sounds like using Pilot or ChatGPT for coding - no understanding of the situation/problem, just add some code that makes the error message go away :(

More ›