I was looking up dijkstra quotes and i found a good one i don't remember seeing before:
My point today is that, if we wish to count lines of code, we should not regard them as "lines produced" but as "lines spent": the current conventional wisdom is so foolish as to book that count on the wrong side of the ledger.
Dijkstra also appreciated my current predicament in advance:
I mean, if 10 years from now, when you are doing something quick and dirty, you suddenly visualize that I am looking over your shoulders and say to yourself "Dijkstra would not have liked this", well, that would be enough immortality for me.
incidentally, dijkstra would not have liked this
The question of whether Machines Can Think... is about as relevant as the question of whether Submarines Can Swim.
@dysfun that's probably the least damning thing he'd have to say about the current situation, but yeah
(the current situation being clearly a long way from even facing the one where "we might be biological machines, what if we pattern other machines on us?" is even relevant)
@ireneista @dysfun that and the pileup of "being silly with sf" got taken seriously when it crossed a certain threshold, which looks suspiciously like it also aligns with either personal growth or already deeply held convictions (beware anybody who gets out of the military without some) meeting real world circumstances in which they needed to be expressed
(unlike me, my dad has finished the series)
So great ā thank you! I had to dig up the source:
Dijkstra, E.W. āIntroducing a Course on Calculi.ā EWD1213, Department of Computer Sciences, University of Texas at Austin, 30 Aug 1995.
https://www.cs.utexas.edu/~EWD/transcriptions/EWD12xx/EWD1213.html
It kinda breaks my brain to learn that Dijkstra was teaching at UT while I was a student there. I could have theoretically passed by him in the halls, not having a clue who he was.
I really wonder where he taught (on campus) and where his office was. It'd give me a clue whether or not I might have theoretically run into him. š
cc: @dfloyd888
You might find something in the UT CS Dijkstra archive :
āFor over four decades, he mailed copies of his consecutively numbered technical notes, trip reports, insightful observations, and pungent commentaries, known collectively as āEWDsā, to several dozen recipients in academia and industry. Thanks to the ubiquity of the photocopierā¦ the informal circulation of many of the EWDs eventually reached into the thousands.ā
Those are neat!
I looked at a couple, but they weren't any more specific than "Department of Computer Science, UT Austin."
Tried searching for an old syllabus of his online, but haven't found anything yet.
I might try the wayback machine next. ;)
Found his old homepage:
https://web.archive.org/web/20001210141800/http://www.cs.utexas.edu/users/UTCS/report/1997/dijkstra.html
And found the directory!
https://web.archive.org/web/19990420195154/http://www.cs.utexas.edu/users/gem/facultydirectory.html
His office was Taylor Hall 3.102
No idea where exactly that was in the building, but that building was my haunt! (It held the computer lab)
Its nice to think that just maybe, some day in late 1992 Dijkstra saw a scruffy Persian kid peering into a VT320 and took note. š
That'll be part of my headcanon until I die š
Congratulations! Now whenever you are doing something quick and dirty, you can truly visualize that he is looking over your shoulders and say to yourself "Dijkstra would not have liked this" š
Haha, well, I don't do programming for a living so I'm mostly off the hook, as I mainly just write shell scripts for fun, and try to make them DECENT š š
Hey now, take that back! š¤£
From what I understand, Dijkstra did not like anythings.
@dysfun Once Upon A Time we had a rule that "you can ignore the coding (etc) standards if what you're writing is less than 300 lines and will have a lifetime of less than 48 hours".
And then the use of source code control systems became universal so it was no longer possible to guarantee a lifetime of only 48 hours so that rule evaporated.
Me, I'm one of the nine women who can have a baby in one month.
I love that. We've all had that boss.
Anything to do with the Month Man prophesy?
I once wrote a 40 line encryption program.
In a dick swinging contest, my mortal enemy wrote the same code in two lines.
I know which one I would rather maintain...
š
@n_dimension @ipd @Uair @dysfun I'm reminded of the joke or urban legend:
There was a programming competition: write a Quine - that is, a program that outputs its own source code - and make it as short as possible.
And somebody submitted the empty string.
@f4grx @bms48 @Uair @dysfun use is kinda relative here, if you just want a thing you can build that does the thing its useful if you want to actually understand what its doing or change stuff not really, your effectively having to just reverse engineer it;
immediately 'static recompilation' comes to mind here, where the code technically works but is not really readable at all
also i beleive 7zip is coded in this kind of way ^^
@dysfun This part caught my eye, which I'm not sure has aged as well:
> Unfathomed misunderstanding is further revealed by the term "software maintenance", as a result of which many people continue to believe that programs āand even programming languages themselvesā are subject to wear and tear. Your car needs maintenance too, doesn't it? Famous is the story of the oil company that believed that its PASCAL programs did not last as long as its FORTRAN programs "because PASCAL was not maintained".
@dysfun I totally agree. Shorter solutions are (to a certain point) much more elegant and maintainable. There's a lack of ability in many developers, either before implementation (when considering a problem) or after (when refactoring) that they can't see the short, clean solution.
This gets worse with LLM coding agent use, because an LLM defaults to creating more tokens, and is in itself unable to create "subtractive outputs". So you need to often ask it to re-generate, telling the model things along the lines of "wtf? This whole thing should fit in under 150 lines of code. Discard and recreate the entire file."
Mileage still varies extremely. But at least the LLM is happy to try. Producers/managers usually stop you from instructing human coders to rewrite what they just submitted "working" well enough.
@dysfun I like to think of more code as more technical debt
so increasing the number of lines increases technical debt
now, this does not mean you're doing it without reason. sometimes it can be necessary.
but you should avoid doing it unless you need to
which is why people who use LLMs baffle me
@dysfun
Measure by outcome, not effort spent.
This could apply to many things (takes a long hard stare at the definition of GDP).
gaytabase
Philippa Cowderoy
alcinnz
Irenes (many)
Matt Panaro
Landa 
Greg Lloyd
R.L. Dane 
šµ 
DHeadshot's Alt
Daniel Gibson
rk: itās hyphen-minus actually
Tim Ward āšŖšŗš¶ #FBPE
Poloniousmonk
Luna Lactea
ideaPDish
WulfyāSpeaker to the machines
Li ~ Crystal System
Dr. Bruce Simpson
F4GRX SĆĀ©bastien
Leeloo
Pxl Phile
LabToast
Sundew
scrottie (he/him/them)
Simon Welsh
otto@openbsd
three tigers in a trench coat
solo
Cian O'Connor
Schroedingers_Cat