Computer Science

« first day (908 days earlier) ← previous day next day → last day (3517 days later) »

10:32 AM

Oh my, the answers to this are mostly horrible:

Q: Language for teaching basic programming

I'm interesting in teaching programming to middle school students. I'd like a programming language with the following criteria: Simple - pared down to the absolute minimum needed to support sophisticated programming without too much code. As such, for this language, I'm not interested in poin...

Even though the question is just so "good subjective" -- the OP lists some reasonable objective criteria -- the answers tend to ignore them.

Should we act on that at all?

FrankW

Downvote the bad ones?

Raphael

@FrankW Definitely that.

We could use this and similar questions (I think we had some in the last few weeks) to motivate a new, stricter policy on recommendation and/or list-style questions.

I vividly remember @Gilles predicting this, vehemently speaking out against allowing list questions. Too few people agreed back then, iirc. List questions do have some appeal; they often "answer" "relevant' questions and are low-hanging fruit.

6 hours later…

Wandering Logic

4:50 PM

@Raphael "Tend to ignore," but two of the answers, in fact, address all three criteria head-on. cs.stackexchange.com/a/27641/7459 (the one OP marked "correct") says "Javascript", and points out that it is "forgiving for new programmers", has lots of libraries for doing graphics and networking and that all the leading browsers in fact have very usable IDEs and debuggers built in.

cs.stackexchange.com/a/29546/7459 points out that Python (which the OP mentions as fulfilling the first two criteria but not the third) actually does have at least one easy-to-use, useful IDE/debugger.

Are those two answers not what you are looking for?

Raphael

5:31 PM

@WanderingLogic In the Javascript one, I missed "debugging" -- I'm not sure the "browser" item implies all the question asks for. (It does if you use FF, afaik.) The second one may be better than I gave it credit for, now that you say it; I missed that the OP cleared Python on all counts but the debugger.

Gilles

6:11 PM

Parkinson's law of triviality

Parkinson's law of triviality, also known as bikeshedding, bike-shed effect, or the bicycle-shed example, is C. Northcote Parkinson's 1957 argument that organizations give disproportionate weight to trivial issues. Parkinson observed and illustrated that a committee whose job is to approve plans for a nuclear power plant spent the majority of its time with pointless discussions on relatively trivial and unimportant but easy-to-grasp issues, such as what materials to use for the staff bike-shed, while neglecting the less-trivial proposed design of the nuclear power plant itself, which is far more...

everybody has an opinion on which programming language to choose

this thread has pretty good answers, for such a topic (which is not to say that they're good in any kind of absolute terms)

Raphael

@Gilles Do you mean this type of question, or us meta-talking about the issue? ;)

Gilles

@Raphael the question

Raphael

@Gilles True. I'd paint the rather precise requirements posed by the OP responsible.

@Gilles I assumed so. ;)

Gilles

It looks like Operating Systems is going to shut down for lack of traffic. I am going to ask for the applied CS questions to be migrated here.

vzn

lol @ "Parkinsons"... hadnt heard that & thought had heard all the major/classic criticisms of bureacracy! reminds me of meta around here. at least its segregated on stackexchange... & lately rather inactive on the cs groups... actually maybe a good sign... no news/good news...

re the programming question. it seemed to be forgotten/ dormant/ mostly harmless until a new answer. its +6 showing group interest. see questions that get good votes but unpopular answers are more of a systemic failure of stackexchange. anyway it has a +7 answer which is more than the question. so overall, think "theres nothing to see here, move along"...

it is a rather rare question in the sense of a comparison of languages over some requested criteria which seems quite ontopic. does somewhat go against bias of "avoiding implementation-specific stuff" wrt CS theory.

sort of bummer about OS group biting the dust. dont really like to see area51 proposals fail because it involves significant group effort. wonder at times if the overall stackexchange cross-site expansion is starting to "mature"/ level off/ plateau.

re meta there is an old stackoverflow ceo blog "meta is murder" lol

Raphael

6:30 PM

@Gilles How many questions are we talking about, migrated over which time period?

Gilles

@Raphael there are 46 non-closed questions, of which maybe 1/3 would be on-topic here

Raphael

Oh. That's not a lot. (The migration should still probably be spread out over, say, a week?)

How many closed questions were there?

Gilles

@Raphael I think that's out of the question. The questions will be migrated during the few minutes before the site closes.

@Raphael 9 (excluding deleted ones)

Raphael

@Gilles Okay. That means that we'll get about as many question as we usually get in a day all at once. Not nice, but manageable.

@Gilles 55 in total, that's... indeed not much. Where did all the committers go?

Gilles

@Raphael the questions won't appear on the front of the sort=newest list, since they'll be days old

Raphael

6:34 PM

Okay, that makes sense.

Gilles

they'll probably come up in the active list because they'll need to be retagged

1 hour later…

Gilles

8:03 PM

Q: Trouble knowing which AI techniques to use

I've been reading Artificial Intelligence: A Modern Approach, which describes many techniques in AI. My problem is that often I don't know when to use a given technique. This can be for 2 reasons: Other techniques seem to be as good as or better than the given technique in every way. For exampl...

artificial-intelligence

ok or too broad?

Anthony

8:23 PM

@Gilles Could you help me out with an algorithm question?

Gilles

@Anthony I'm not an algorithm specialist. Better ask on the site

Anthony

Thanks.

Wandering Logic

8:40 PM

@Gilles too broad, but it could easily be turned into two reasonable questions (one about when to choose variable elimination vs. enumeration for Bayesian inference (not that I know what either of those actually is), one about tradeoffs between random-restart vs. simulated annealing.)

the bayesian inference question might be better on cross validated.

and I doubt any of us really know the answer to the hill-climbing / simulated annealing question.

Maybe that would be better on scicomp?

« first day (908 days earlier) ← previous day next day → last day (3517 days later) »

all rooms