Computer Science

vzn

01:46

0

Q: Blum's speedup thm in big-O format?

is there a way to state Blum's speedup thm in terms of Big-O (Landau) notation?

vzn

02:01

0

Q: Blums speedup thm showing unclassifiable complexity languages?

do any of Blums thms prove that there exist decidable languages that are unclassifiable anywhere in the time hierarchy? in other words, asserting they (mentioned in the proofs) are computable within any time $O(f(n))$ for some time-constructable function $f(n)$ leads to a contradiction?

complexity-theory

Raphael

16:33

Candidate for closing as "dump"? I'm hesitating because it will reject the migration.

0

Q: Finding all paths with lengths in a fixed interval in sparse graphs

What is the most efficient way to find all paths of length M to N in a large sparse graph? Some general information: Graph has 30,000 to 50,000 nodes Average number of edges per node ~ 10 M=4, N=7 Graph has cycles

algorithms graphs shortest-path

Juho

What do you mean with "rejecting the migration"? As in, what happens if the question is closed?

Raphael

16:46

@Juho Migrated questions are rejected when closed, that is the migrating sites sees "they did not want that". Plus, the question is locked and can no longer be edited (although we can clear that, apparently).

Juho

I see

Juho

17:14

@Raphael It seems like a dump to me as well. After all, the OP has stated no own effort

Artem Kaznatcheev

@Raphael sorry about that migration, it was very borderline in my mind, but I thought one of the comments (that has now been inserted into the question) did specify it well enough to be non-throwout

Patrick87

17:58

@Raphael Let's consider the linked question, for instance. Is it a question that a random person off the street could answer? Probably not. Has the specific question been asked before? Unlikely, though it's possible that a substantially similar question already does exist, in which case it's not bad, just a dupe. Would a reasonable person assume that an expert in computer science could provide insight to the method of resolution? Seems clear. Now, it's possible that an argument could be made

for moving this somewhere else, or asking for clarification on what the problem is. However, there could also be an argument for just answering the question, showing work, and explaining how the answer was obtained. Easy questions aren't bad questions. I find the "basic arithmetics" designation disingenuous here, since it's entirely possible that a person who understands "basic arithmetics" would have no idea how to answer this question.

Raphael

@ArtemKaznatcheev It's quite clear what the asker wants. It's not clear what they tried.

@Patrick87 You mean 21381?

Or 21326?

I think you still mistake the point, though. It's not that the problem is a bad one per se, otherwise it would not be an exercise problem (actually, I do cringe at what some people deem appropriate for higher ed, but that's another discussion entirely).

It's that there is no question.

And that's really independent from the level of the problem.

My guideline is: if that was a student/colleague that came into my office and asked, what would I do?

If the person says: "Look at exercise 3.a). What's the solution?", I'll tell them to try and then come back.

That's just common curtesy; why demand another person's time on something that I was not ready to devote time to myself? (And that's even worse online, because typing an answer takes longer.)

And that's what I expect of askers here.

Case in point: almost no make any effort to improve their question. Very few even react to the closing comment.

Those who do come back and do display serious interest and/or effort, we help. See here, for instance (and note in particular the lengthy chat).

I really don't see why Computer Science should be a problem-solving black box, how that would actually help people (in particular, but not only, students) and why the result would be a site that still attracts people that are qualified to give good answers.

That's my position, anyway.

See also here:

138

A: How much research effort is expected of Stack Overflow users?

A lot. An absurd amount. More than you think you are capable of. In fact, asking a question on Stack Overflow is the absolute last thing you ever want to do. You want to avoid it at all costs. You want to think of it as a horrible shame1 that will forever haunt you and pass down from you to your ...

Now, we can argue whether closing immediately is the best approach. Should we not encourage to improve first, and only close if there is no improvement? Two issues.

1) We have no tool support for that. [I proposed some ideas.](http://meta.stackoverflow.com/questions/217494/how-to-train-your-growing-community-to-self-moderate)

2) People, in particular new users, keep answering dumps, thus rendering our policy void. We can't prevent that without closing (*or* vote-hammering those answering, as used to happen on [cstheory.SE], but I don't like that at all).

Also, @Patrick87, if you browse some questions that have not been closed, you'll note that it's not much we require. Even the most naive yet independent thought qualifies a question to remain open. Those we close (for the "dump" reason) are almost exclusively dumps with nothing else attached to them.

Juho

18:14

@Raphael My two cents for your two points: closing is a much better idea than "trusting on downvoting". I think downvoting is quite rare, and some people simply don't downvote even if they think an answer is bad (either because they find it disrespectful i.e. personal, don't want to lose points themselves, or whatever). If a question is closed, it can still be edited, so no harm done there (and closing a question is no personal attack, either).

So, IMO, rather close aggressively

Raphael

@Juho I'm glad you agree. However, it can't be denied that closing feels rude to some, especially when it happens with a canned comment that may or may not fit the question very well. I think that's unavoidable and we don't need people to stick around that take offense at criticism, anyway, but opinions may vary.

vzn

18:52

dislike aggressive closing.

eg there are many posts eg on tcs.se meta complaining about "too aggressive" closing... also issue in recent elections.

are you guys complaining about "Finding all paths with lengths in a fixed interval in sparse graphs"?

Juho

@vzn Can you link to some (recent maybe?) meta post where aggressive closing is discussed on TCS.SE?

And yes, that was the question

vzn

juho it is not really very recent. ppl probably just gave up long ago....

there are lots of posts, have many bookmarked/in my faves. might do a blog on it someday.

as for so-called "vote hammering answers" to hw questions.... huh?!?

Juho

My guess is that on TCS.SE, questions are usually closed for being "not research level", and not because they are not showing effort

and I'd guess that's what they have been talking about

vzn

there is just too much navel gazing about what constitutes quality questions all over se imho....

juho you were just complaining openly how arbitrary/unpredictable their policy was wrt question on SAT benchmarks....

Juho

Yes... but how is that now relevant?

vzn

18:59

relevant?!? its a Q/A site

Juho

It is, but at the same time I think it shouldn't be a HW dump site

vzn

agreed the hw policy is controversial & tricky to manage and frankly an issue all over se as the recent meta posts noticed there indicate.

nobody labels their questions as "hw" so is it our job to try to play "hw PI"?

look one of the most important guidelines is ignore/downvote stuff you dont like

as for closing ... save it for stuff that is "not a question"... (ok yes it is out there)

are we here to answer questions or be policemen? alas imho mods get carried away on that sometimes =(

Juho

I think we are here to ask and answer questions, but "bad" questions should not be answered

vzn

as for downvoting helpful answers to questions that "seem hw", that seems really capricious to me

Juho

That's why HW questions with zero effort should just be closed and then edited, if possible

vzn

19:05

so dont personally answer the "bad" questions! but closing them implies value judgement for all other users ie "no other person on this site could possibly find anything redeeming about this Q/A"

you guys keep talking about "hw questions" as if they have little helpful labels on them!

Juho

You on the other hand make it sound like it is absolutely impossible to tell what question is HW with no own effort

vzn

if it "looks/smells" like hw, fine, label it & downvote it! warn others! whatever... imho closing CS-related q's & downvoting helpful answers is too extreme.

Juho

I think homework questions are fine, as long as they include OP's own effort

I remember several cases where a HW question was posted with no such effort, but when it was asked for in a comment, the OP clarified what he/she had tried

vzn

roughly the only way to definitively identify hw is if the poster states it is, which is rare. asking for clarification in comment & holding off on an answer seems ok to me.

Juho

Holding off an answer is fine yes, but too many users (I think) are unaware of our policies. Therefore, closing is much safer

vzn

19:15

heres the deal. imho, overcome the [quite human] tendency to focus attn on bad stuff. focus attn on the good stuff & the bad stuff will take care of itself & float away....

Juho

I definitely disagree. If you just let the bad stuff be and ignore it, it'll invite even more bad stuff.

vzn

haha that seems to be the Great Fear of se mods everywhere

mostly unproven.

Juho

I don't see why it isn't totally valid

vzn

the real issue on the sites is lack of participation...

time is valuable...

R. points out himself that most ppl/newcomers rarely stick around....

Juho

Having a lot of "bad stuff" on the site is not going to attract any more "good" users

2

vzn

19:18

obviously nobody is encouraging bad stuff.

Juho

Answering HW questions with no effort is encouraging bad stuff

vzn

but the idea that by attempting to kill/eradicate all the "bad stuff", all the shiny high-participation/engagement users will show up.... "question" that

Juho

Definitely sounds like a better way of attracting new users than having a lot "bad" stuff on the site to me!

vzn

nobody is encouraging bad stuff

Juho

Yes, we try to fight that

vzn

19:20

stop "fighting" then! put down the "fight"! "fighting" is surely part of the problem... waste of energy

se recently heavily reworked its close policy to put stuff on hold.... get the memo on that?

Juho

Well... I don't think it's a waste of energy to keep the site healthy

But we can have different opinions, and waste energy on different things :-)

What's exactly the difference between closing a question and putting one on hold?

vzn

prefer not to waste energy think you would feel the same :)

right now its not much different. but se mgt thinks its a systemic issue with the site

hence the recent attn to it

Raphael

@Juho None.

vzn

Blog - Stack Exchange: The War of the Closes

Jay Hanlon on June 25, 2013

It pains me when I hear people say that our sites are unfriendly, or that we chase new users away. But it’s a hard problem, because our highest priority has always been the quality of content on our sites. And it still is. We can’t lower our standards. We won’t.

But we have been working hard to make our sites more welcoming, reminding users that feedback can be clear and nice, and helping new users learn the ropes before they get frustrated. And, as of today, we’ve completely overhauled closing. …

Juho

@Raphael But when I think of it, I think I've seen a question "on hold", and a question that has been "closed". Why would there be two meanings for the same thing?

vzn

19:24

completely overhauled wow! =o

Raphael

They changed the UI: for a few days, questions will be shown as "on hold" instead of "closed". The change was made with the intent to better communicate "not like this, but if you improve your post we can talk" instead of "what the hell gtfo".

Juho

Oh yeah, I see that now, going through that link...

I suppose "on hold" better conveys the meaning of what's going on

Raphael

@Juho Exactly. :)

vzn

it is true users rarely return to fix on-hold questions...

but others may do it....

re:

1 hour ago, by Raphael

Those who do come back and do display serious interest and/or effort, we help. See here, for instance (and note in particular the lengthy chat).

why is that so rare? that is what the site really needs. engagement. stickiness. return users. how is that achieved? it is not obvious.

why can we not help on their 1st visit? why is it reqd that they be dedicated in some way to help them? agreed that is a good metric of how much energy to put into it... but it is rare for users to engage much...

Juho

Maybe because the 1st visit is the last visit, and it's just "damn I only have an hour before I have to turn my HW in, maybe these guys can help me if I just copy+paste here"

vzn

19:34

(lol) agreed there is some of that. dont seem to run across it much myself....

re closing. look at this comment by a Phd who hangs out on tcs.se

? & curious why posting here, it seems clearly a tcs.se level question... also better to cite the tcs.se Q — vzn 2 hours ago

Because of TCS's tough standards ;) — Mohammad Al-Turkistany 2 hours ago

just see so much griping by mods about the users... re tone...

seems, sometimes, no wonder they dont come back....

Juho

I still think the atmosphere on CS.SE and TCS.SE is different

but that's just my "feeling"

vzn

its like a shady/dark teachers lounge....

"let me tell you about my @#%& students!"

the sites are much different agreed.

Raphael

Oh the obsession with Theoretical Computer Science...

vzn

hah! was just waiting for R. to remind me of that again

r. in contrast is obssessed with Computer Science haha

Raphael

They would have a better climate over there if non-researchers did not clamour to be heard all the time, I guess.

vzn

19:43

"non-researchers"... jeez has anyone heard of a "purity test"?

its a purity test for academics

Juho

@Raphael I understand what you mean, and that relates to what we talked about some days ago about the atmosphere on TeX.SE and certain other sites

vzn

Purity test

A purity test is a self-graded survey that assesses the participants' supposed degree of innocence in worldly matters (sex, drugs, deceit, and other activities assumed to be vices), generally on a percentage scale with 100% being the most and 0% being the least "pure". Online purity tests were among the earliest of Internet memes, popular on Usenet beginning in the early 1980s. However, similar types of tests circulated under various names long before the existence of the Internet. Historical examples In 1936, The Indian Express reported that students at Toronto University were "under-g...

r. again you "blame the users"

Raphael

Jup. The more alike the users are and the more common their goals, the better the climate is. I think that's a universal truth, valid in any group/society.

vzn

nobody whatsoever is "clamoring" on tcs.se

the site is like a dusty library with rare flashes of anything....

ssssssssh!!!

seems like weeks go by where no posts get more than 5 votes....

which reminds me of this post

things every tcser should have done once / suresh

things a complexity theorist should do at least once / lance

hey look at that last item! =)

is MAT above a "non-researcher"?!? haha whatever

Juho

19:58

I like Suresh's list more (because I get way more points using it)

vzn

=)

Raphael

20:51

Someone seems to be on a comment-cleanup spree. I wonder who that is...

Juho

Haha, just a little something while my pasta cooks :)

Raphael

:D

Enjoy!

Juho

The pasta or going through comments?

Raphael

Definitely the former, but the latter... I think you should schedule with a shrink if you actually enjoyed that.

Juho

21:18

@Raphael I'm glad the system also acts as a shrink (I have limited flags, and I think I only have 2 or 3 left right now...)

Raphael

@Juho this one still applies, doesn't it?

@Juho Haha, that's just to protect me from the shrink! :D

Juho

Oh right, maybe I misread it actually

Now it's obsolete :)

Raphael

Ruthless!

Juho

Yeah and now, finally, I'm out of those flags :D

Raphael

And bless the system for that. ;)

Honestly though, thanks! It's rare that anybody bothers to clean up.

saadtaame

21:30

How does one become a moderator?

Juho

@Raphael No problem, it was weirdly somehow refreshing

Raphael

@saadtaame On a new beta site, or on a graduated site?

Gilles

@saadtaame run the next time there's an election

Raphael

@Juho O.o

Gilles

which for this site will happen for the first time soon after we graduate

we don't know when that will happen

saadtaame

21:33

CSSE, which is beta at the moment.

Gilles

Moderators were chosen by SE staff a few weeks into the beta

Juho

@Raphael I don't know, for so many nights I've been just editing text, writing/polishing proofs and drawing graphs so it was a nice change I guess :-)

Gilles

unless we can't handle the workload or one of us quits, in which case a slot will open

saadtaame

Who are the mods now?

Raphael

@saadtaame As Gilles said. It probably helps if you engage in community moderation until then. Not only to get the hang of the job, but also to show your interest and aptitude in serving the community.

Gilles

21:34

@saadtaame cs.stackexchange.com/users?tab=moderators

@saadtaame Raphael, patrick87 and I

Raphael

@Juho I see. When that happens, I can't look at screens anymore and have to do something more drastically different.

saadtaame

What are the +'s of being a mod? @Raphael @Gilles

Juho

@Raphael That works too, I usually go for a run and do other stuff before I begin though

Gilles

@saadtaame you don't need somebody's approval to edit, close, delete, etc.

Raphael

@saadtaame Sleepless nights, less time to write answers.

@saadtaame Sometimes a warm feeling of accomplishment.

@Juho Sports?! Woa, woho...

(I'm usually to couch-ish to just go for a run, but sports is definitely good. :))

Juho

21:39

It actually is! If you are not too active it might feel like it wouldn't do much good "brain-wise", but I think it helps

Raphael

@Juho Definitely. I find running to be boring, though. Maybe when I'm older.

Juho

I know, it could be that especially if you run on a treadmill

I know plenty of people who say they don't run because it's boring

I don't really know why I don't find it that boring, even when I've been running on a treadmill lately

Raphael

I guess if you get in a trance-like state, it does not matter.

If have to say I never actually tried running just to run, but whenever it came up during training sessions, I hated it.

Juho

To be honest, I would probably find it really boring too if I didn't have all kinds of data I could get from it

running with a heartrate monitor is so much more fun

and logging the runs you do

Raphael

:D

Nerd-sniped yourself there, nice.

Juho

21:46

haha, I know! :D

Raphael

Do you GPS-log?

Juho

nope

but I get the distance, average/max heartrate, and use a website to log that all

it actually motivates you too when you see how you improve (and makes you realize you are nerd too, of course)

Raphael

I think I get that. Doesn't make me want to go out in case of rain, pollen or Monday. Which probably covers most days.

Juho

Yep. I can't keep myself motivated unless I go really hard

if I'm just like "oh I'll go for a run every now and then", I just won't

I need to have a plan, and I need to monitor how I do. That's how I stay motivated

Raphael

22:01

Yea. My inner jackass is more devious: I won't make such a plan. :D

Juho

Maybe it'll change once you get fed up enough with your inner jackass :)

(Happened to me)

Raphael

We get along far too well.

Juho

For now...

:P

Raphael

Right. So, potato chips now?

;D

Juho

Sure :p

Raphael

22:31

Have a nice day, NAs; good night, CETers!

Juho

@Raphael Nites!

Transcript for