|
What Every Computer Scientist Should Know About Floating-Point Arithmetic Coming in at a mere 45 pages, this should be top of your summer reading list.
|
# ¿ Jun 16, 2019 19:59 |
|
|
# ¿ May 17, 2024 17:15 |
|
It's a context free language. Emulate a pushdown automaton and move on with your life.
|
# ¿ Jun 18, 2019 04:28 |
|
The number of people who really understand monads is far smaller than the number of people who sound like they do.
|
# ¿ Jun 19, 2019 01:41 |
|
Elegant error handling with the JavaScript Either Monad This is a good introduction because it doesn't spend a lot of time explaining what monads are in general, and it shows you how to use one to solve a real problem. The error handling methods that you get out of functional programming are good and people should be paying attention to them.
|
# ¿ Jun 20, 2019 04:47 |
|
Qwertycoatl posted:They talk about a growing range of needs they can't get from existing libcs, but don't say anything at all about what those needs are. I wonder why they want to reinvent this particular wheel. Their library developers need something to do.
|
# ¿ Jun 26, 2019 21:32 |
|
strlcpy is the preferred alternative to strcpy. It's not part of the standard, but it's generally available.
|
# ¿ Jun 26, 2019 23:01 |
|
Arsenic Lupin posted:And of course Google, back when it was just Google, wasn't a new idea. There were search engines before Google, but building a search engine based off PageRank was a genuinely new idea. That's why they cornered the market.
|
# ¿ Jun 29, 2019 19:25 |
|
Arsenic Lupin posted:We really need a "but ephebophilia!" smily. No. No, we don't.
|
# ¿ Jul 13, 2019 02:59 |
|
https://twitter.com/mike_conley/status/1149845391153737728
|
# ¿ Jul 13, 2019 15:32 |
|
https://twitter.com/rickhanlonii/status/1149868394881015808
|
# ¿ Jul 14, 2019 16:23 |
|
Functional programming is much easier to learn if you don't have to unlearn imperative programming first. I think the transition from functional to imperative isn't as bad, but I'm not so sure about that.
|
# ¿ Jul 17, 2019 04:19 |
|
Qwertycoatl posted:My "fav" was a code mis-execution cpu bug that went away when run in the debugger There's nothing quite like sidestepping a complicated and horrible crash at application shutdown with a sleep(500).
|
# ¿ Jul 18, 2019 15:06 |
|
US schools are slowly coming around to the idea that CS and SE should be different programs, but I think it'll still be another decade or so before there's consensus, and after that it's still going to take a while to switch programs.
|
# ¿ Jul 19, 2019 00:41 |
|
Ola posted:"Professor" is supposedly a synonym for "teacher" in most western higher education institutions, but most professors sadly think that teaching is the tax they have to pay in order to fund their ridiculous pet project, so they spend zero effort getting good at it. This is a fairly common misconception but it's very wrong. In theory, a professor's job duties are roughly 40% teaching, 40% research, and 20% service. In practice, this varies considerably by university. In a major research university, grants (i.e., money for research) are a very large part of the university's income, so someone who's bringing in that money can get away with a lot, and someone who isn't will be shown the door in short order. At least in the US, there are institutions where teaching is much more important, but even the professors there are expected to contribute to their academic field. ultrafilter fucked around with this message at 02:30 on Jul 19, 2019 |
# ¿ Jul 19, 2019 02:14 |
|
University classes last for one semester at most. That means that there are some pretty hard limits on how large a project you can have attached to a single class, and that places some limits on what you can effectively teach. Yes, you can talk about version control, and testing, and object-oriented design, but those are all things that don't really show their value until you're working on a big piece of software. Those are things that you really can only learn on the job. Theory probably isn't all that relevant to most jobs, but it is relevant for anything you want to do at scale, and the best jobs are moving in that direction. In addition, it might be the case that your job would be made easier by a piece of theory that you don't know. No one's going to be teaching basic science to their employees, so university really is the ideal place to cover that. In short, you can't get a complete education from a single source. You need to learn some things in school, and you need to learn other things elsewhere.
|
# ¿ Jul 21, 2019 19:06 |
|
leper khan posted:I don’t know how to get through the political mire to slow everyone down enough for us to move faster though. If you figure it out, let me know.
|
# ¿ Jul 21, 2019 23:18 |
|
redleader posted:see also: names, dates, times, writing (may god have mercy on the unicode committee) The single biggest book I've ever seen was a hardcopy of the unicode 4.0 standard. I can only imagine that it's gotten worse.
|
# ¿ Aug 1, 2019 14:47 |
|
New code can be legacy code if it's written badly enough.
|
# ¿ Aug 2, 2019 14:48 |
|
If illuminated manuscripts were good enough back then, they're good enough now.
|
# ¿ Aug 12, 2019 01:29 |
|
At one point the standard was that the number on the progress bar should be what percentage would be done after the current step. Some people never moved on.
|
# ¿ Aug 19, 2019 14:54 |
|
The animated gif progress bars make people think that things are getting done faster despite the complete lack of information.
|
# ¿ Aug 19, 2019 20:31 |
|
https://twitter.com/dataandpolitics/status/1163893697718525955
|
# ¿ Aug 21, 2019 01:24 |
|
Is there anybody who does mostly PRs? I'm not even sure how that would work.
|
# ¿ Aug 25, 2019 01:10 |
|
canis minor posted:My previous company - the myth: "we're doing Machine Learning to track which routes are profitable, so that our network of connections adapts itself to demand"; the reality - meet Bob, Bob will download a database as a CSV, open it on a custom purchased machine that has loving water cooling and all the fancy specs to handle the amount of data that has to be imported into Excel, make alterations to what the routes should be on given days and how should they be priced, then will upload it to the server.
|
# ¿ Aug 29, 2019 00:35 |
|
An unsupported claim isn't an argument.
|
# ¿ Oct 3, 2019 05:17 |
|
Git Koansquote:A Python programmer handed her ~/.gitconfig to Master Git. Among the many lines were the following:
|
# ¿ Oct 5, 2019 01:19 |
|
https://twitter.com/noop_noob/status/1183234844534304769
|
# ¿ Oct 14, 2019 01:11 |
|
What every computer scientist should know about floating point arithmetic
|
# ¿ Oct 16, 2019 15:35 |
|
Silicon Valley pioneered self-driving cars. But some of its tech-savvy residents don’t want them tested in their neighborhoods.quote:Karen Brenchley is a computer scientist with expertise in training artificial intelligence, but this longtime Silicon Valley resident has pangs of anxiety whenever she sees Waymo self-driving cars maneuver the streets near her home. Read the whole article. It's good.
|
# ¿ Nov 8, 2019 01:07 |
|
A lot of Americans are anti-public transit because it benefits poor black people.
|
# ¿ Nov 8, 2019 03:04 |
|
Algorithm Aversion: People Erroneously Avoid Algorithms after Seeing Them Errquote:Research shows that evidence-based algorithms more accurately predict the future than do human forecasters. Yet when forecasters are deciding whether to use a human forecaster or a statistical algorithm, they often choose the human forecaster. This phenomenon, which we call algorithm aversion, is costly, and it is important to understand its causes. We show that people are especially averse to algorithmic forecasters after seeing them perform, even when they see them outperform a human forecaster. This is because people more quickly lose confidence in algorithmic than human forecasters after seeing them make the same mistake. In 5 studies, participants either saw an algorithm make forecasts, a human make forecasts, both, or neither. They then decided whether to tie their incentives to the future predictions of the algorithm or the human. Participants who saw the algorithm perform were less confident in it, and less likely to choose it over an inferior human forecaster. This was true even among those who saw the algorithm outperform the human. So we would probably be OK with a change that made human drivers only kill 100 children per year, but if that same death toll comes from a self-driven car, it's a problem.
|
# ¿ Nov 8, 2019 17:03 |
|
quote:The Holy Roman Empire was neither holy, nor Roman, nor an empire. In terms of the project's actual goals, it needs some work. But as is, it's perfect.
|
# ¿ Nov 9, 2019 20:11 |
|
Yeah. I thought about switching a long time ago but that was before the scipy stack was compatible, and that was a dealbreaker for me.
|
# ¿ Nov 23, 2019 01:57 |
|
Doc Hawkins posted:the current target of my mild ire is nullthrows, a completely unnecessary external dependency that doesn't even do its job well Isn't the only thing it can print "Got unexpected null"?
|
# ¿ Nov 27, 2019 01:03 |
|
Magissima posted:Haskell can technically be written with semicolons (and brackets) instead of indentation but I've never seen it used in the wild. It's useful in the REPL though. That just looks so wrong. https://twitter.com/zkat__/status/1204188153008906240 ultrafilter fucked around with this message at 06:30 on Dec 10, 2019 |
# ¿ Dec 10, 2019 06:22 |
|
https://twitter.com/_taylorswope/status/1205252714680045568
|
# ¿ Dec 13, 2019 16:08 |
|
rt4 posted:"Frontend" is such a weird description. It's as if there's normal programmers and then a group of people who somehow only know Javascript. That's not wrong.
|
# ¿ Dec 19, 2019 16:14 |
|
https://twitter.com/MStrehovsky/status/1215331352352034818 It's pretty interesting but absolutely qualifies as a coding horror.
|
# ¿ Jan 10, 2020 02:09 |
|
https://twitter.com/vamchale/status/1215787244633628672
|
# ¿ Jan 11, 2020 01:58 |
|
|
# ¿ May 17, 2024 17:15 |
|
https://twitter.com/jxxf/status/1219009308438024200
|
# ¿ Jan 20, 2020 07:20 |