Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by titzer on Monday September 25, 2023 @11:28PM
(#37653956)
To get great software, you need a small team and give them extremely clear requirements of what it should do with an assload of extremely clear testcases.
[ Parent
| on HN ]
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by sasaf5 on Tuesday September 26, 2023 @12:24AM
(#37654347)
And lots of time to tune things and solve all little problems, and manually implement things when abstractions hamper performance.
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by fsloth on Tuesday September 26, 2023 @03:16AM
(#37655602)
This is a fairly generalizable rule in any engineering, unless you really are running the Apollo program (then when divide-and-conquering should probably remember this rule).
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by hoseja on Tuesday September 26, 2023 @06:22AM
(#37657144)
This is running into the no-code problem of clear-enough requirements being as complex as the code, via Kolmogorov.
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by taneq on Tuesday September 26, 2023 @06:39AM
(#37657289)
The best software comes from a small team of top level devs who are also top level subject matter experts on the software’s domain. This doesn’t happen very often.
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by robbs on Tuesday September 26, 2023 @09:26AM
(#37658958)
It doesn't have to be a small team. See "They Write The Right Stuff".
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by 2rsf on Tuesday September 26, 2023 @10:06AM
(#37659533)
And even they make mistakes, software contains much more moving parts than hardware.
https://space.stackexchange.com/questions/37110/what-was-the...
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by tekla on Tuesday September 26, 2023 @11:25AM
(#37660742)
Wow, thats a hell of a HN thing to say.
Re:SR-71 Blackbird Astro-Nav System (a.k.a. R2-D2) worked by...
by iudqnolq on Tuesday September 26, 2023 @06:26PM
(#37666616)
> Use of pointers required a written, approved, variance from the SW standards board. All the code was formally peer reviewed by at least 6 people. More complicated code would be reviewed by 20+ people - in the same room.