Jeffrey P. Bigham ๐ฅ๐ฅthe two hardest problems in computer science are: (i) people, (ii) convincing computer scientists that people are the hardest problem, (vi) asynchronous programming, and (iii) off by one errors
Lightning Bjornsson* 
@jbigham That's a buffer overrun. You only allocated two itm_t, but you pushed four.
Erica
William Herold@bigratsgoboom @jbigham and after dates, caching.
Zeborah@bigratsgoboom @jbigham Timezone glitches are the *worst*. You report something broken, support desk says it's fine, you try again and sure enough it works. Only later do you realise it's broken in the morning and works in the afternoon. And still, "Oh it works for us," say the developers in GMT+0.
Mathias โ๏ธ๐บ๐ถ 
@jbigham f u c k ๐ซ
Angle๐@jbigham Funny thing about that, I'm having trouble with an off by one error right now. XD
Jason Maltzen@jbigham As a long-time programmer, I can confirm this is the single most difficult problem in computer science by at least O(n^2).
John Kellden@jbigham It's a three-body problem: computer, interface, people
Andrew Saffrey@jbigham I hate two things in life: (1) numbered lists (2) hypocrisy
Philip Guo@jbigham (vi) asynchronous programming, and (iii) off by one errors, the two hardest problems in computer science are: (i) people, (ii) convincing computer scientists that people are the hardest problem,
Mark Atwood@Cmastication
The solution to the problem of people is to give them what they demand. Good and hard.
Barry Schwartz ๐ซ@jbigham Dependent and linear types in ATS help with off by one errors. For instance, not passing typechecking if you missed an array element.
Dan Hon@jbigham you forgot the second hardest problem, naming people
@jbigham (this is also why parents who are software engineers never come back from parental leave on time)
@danhon yeah, seems to be mostly arguments between partners, where one is adamant that it's reasonable to name their kid "Jimmy'); DROP TABLE Students" and the other partner is adamantly horrified that this is the person they chose to procreate with
๐ญแดษดแด๊ฑ ๐ฎแดแด แดส ๐ฌสส ๐ฟสแด ๐ฐแดสแดส@jbigham () integer overflow, (i) people, (ii) convincing computer scientists that...
Des Dev@jbigham that statement is a bit wise.
Tofu Golem@jbigham
So many fencepost errors and rounding errors whenever I have to deal with programmers. No idea why.
So many fencepost errors and rounding errors whenever I have to deal with programmers. No idea why.
wb x64@jbigham you forgot to either number this list starting at (0) or say the list had either 3 or 5 elements, so that the off by one error would be legit ;)
Joshua K. Farrar ๐ @jbigham asynchronous programming got easier when i learned about kinds and abstractions over execution context ๐ณ
hallvors@jbigham Good one! The three hardest problems, maybe?
bananamangodog@jbigham I see what you did there. Do I get the job?
Pseudo Nym@jbigham I thought the there hardest problems were naming things, and off by one errors.
myrmepropagandistWhy are people a hard problem? Just use my open source people emulator.
With a little work anyone can implement as many people as they need by simply entering the correct parameters to model the type of person your application requires.
Daniel Howard ๐ป@futurebird @jbigham I hear Kubernetes does this.
ConsoleWitch@jbigham I LOL'd
plinth@jbigham (v) naming things
Joana de Castro Arnaud@jbigham You forgot (iv) Naming things.