Discussion on book policy

David Zaslavsky

Jul 16, 2013 16:04

OK seriously though: if we want to have important discussion this week there's always this:

13

The Stack Exchange model does not work well for "big list" questions, and these are banned network-wide for good reason. However, many of the other stack exchange sites allow or even encourage list questions that do not fall under the "big list" category. On the other hand, Physics.SE currently ...

discussion

ManishEarth

Yep

@Manishearth Well my point was not that the rep system is all-important, it was that questions which supposedly should be made CW are actually not good questions (for us). — David Zaslavsky 3 mins ago

Good point, but in this case are we really sure of them being not good?

Colin McFaul

I'm sort of here. I have to go grab lunch.

ManishEarth

@ColinMcFaul Ah. Food.

Physics Meta

0

Q: What affects visibility of a question on google?

Question header more or less sums it up. If I had in mind that I wanted a question to be visible to a google search using particular keywords, would it be sufficient to just include those words in a question body? Is there anything on the SE end that affects this? Or is it purely up to the search...

support

David Zaslavsky

@ManishEarth I think I am.

user54412

Jul 16, 2013 16:07

I'm also here - but it is lunch time, and neutrino cupcakes sound pretty good

ManishEarth

Why though?

That's what I've been trying to hammer out

(Btw, are we talking about books or lists? IMO those two are rather different)

David Zaslavsky

(Lists, although many book recommendations are also list questions)

ManishEarth

I mean {x: x is a list} - {x: x is a book}

user54412

I always thought \setminus was best implemented as \\` rather than -` [Edit: GRRRR - cannot format]

Crazy Buddy

@ManishEarth Sorry... Had to dig somewhere else..! Back now ;-)

David Zaslavsky

Jul 16, 2013 16:09

I'm not saying those questions aren't useful, just that they are not appropriate for the SE model. Stack Exchange is for questions that prompt answers which can be evaluated on their correctness. That's the whole reason we vote on them.

ManishEarth

@DavidZaslavsky Me too, but what if books were handled as in my answer? Now there's an evaluation of how the book is too

That issue is still there, I agree

user54412

I really like @ManishEarth 's idea... but I do have worries

user54412

(about books)

ManishEarth

Me too

user54412

I mean - these can be opinionated

Crazy Buddy

Jul 16, 2013 16:12

Ohh... Books... I love books..! Are we changing FAQ? (so that I don't have to VTC anymore) :P

user54412

e.g. I think most "classic" physics books (Goldstein, or Jackson, or anything written by Griffiths) are awful

ManishEarth

@DavidZaslavsky Btw, you may want to read the discussion in TL starting here. I was curious about the reasons behind disallowing book recs, and it seems like most don't apply here

user54412

do you really want me discoursing on how those books are bad?

ManishEarth

@ChrisWhite maaaybe :P

twistor59

@ChrisWhite yeah, rant away

ManishEarth

Jul 16, 2013 16:13

I personally haven't found much wrong with Griffiths though.

user54412

well, I suppose that would provide amusement...

ManishEarth

But I don't rely much on books in the first place.

twistor59

I learned QFT from Bjorken and Drell. That's why I'm so crap at it.

5

Mostafa

Is this question on-topic? I answered because didn't find an answer in the main list. Also, I flagged it as opinion-based.

ManishEarth

It was my luck to pick up the worst two books to start learning GR and QM. Misner-Thorne-Wheeler and Dirac. Oops.

Whoops

David Zaslavsky

Jul 16, 2013 16:16

@Mostafa It is not, good flag. (Tip: if you think a question should be closed, don't answer. Answering kind of counts as a partial leave-open vote in the mods' eyes.)

ManishEarth

@Mostafa Currently off topic.

twistor59

@Mostafa if someone asks which book to self study fluid mechs, the best answer is to recommend a good therapist!

user54412

LOL

ManishEarth

@twistor59 Also an extra supply of upside down triangles.

David Zaslavsky

@ManishEarth Still though, it just makes the answers more verbose, but it doesn't make the question better.

twistor59

Jul 16, 2013 16:17

@ManishEarth hah yes, you can never have too many of those!

ManishEarth

@DavidZaslavsky Hmm. But ultimately isn't a good question one that, while not confusing answerers, leads to very useful answers?

David Zaslavsky

One of the key points in evaluating the appropriateness of a question for an SE site (notice I'm not saying "quality") is whether you can evaluate any given answer as correct or incorrect.

Crazy Buddy

@twistor59 What? As an aeronautics undergrad, I shouldn't be discouraged like this..!

twistor59

But @ManishEarth did you really think MTW was that bad? I kind of like it.

David Zaslavsky

@ManishEarth Sure, but not all good questions are appropriate for SE.

twistor59

Jul 16, 2013 16:18

@CrazyBuddy oops!

David Zaslavsky

Stack Exchange was never intended to be a repository for all useful information on the internet.

ManishEarth

@twistor59 As a book to read once you learn GR? Sure. Trying to learn tensor calculus and GR from it. Noe.

@DavidZaslavsky Of course

user54412

@twistor59 For learning?!? or for weight lifting?

Crazy Buddy

@DavidZaslavsky: Hah... That very few know

ManishEarth

@ChrisWhite Resnick-Halliday-* is multipurpose :P

twistor59

Jul 16, 2013 16:19

@ChrisWhite lol! yeah you can tell the old relativists by their bicep girth

user54412

@ManishEarth Don't get me started on those people

ManishEarth

@DavidZaslavsky However, as noted in my answer, some of the things that make a question bad for SE don't apply here.

The price and outdatedness for example

David Zaslavsky

I know, I'm not even taking those issues into account.

ManishEarth

You can't easily refute the popularity contest thing, though

user54412

@DavidZaslavsky playing devil's advocate here - we're not a programming stackexchange, and already a number of our great answers can't be tested for correctness by the OP

David Zaslavsky

Jul 16, 2013 16:22

OK, but in principle good questions still do have correct and incorrect answers.

user54412

the validity of books recommendations would be vetted by the votes cast by (presumably knowledgeable) users

ManishEarth

@DavidZaslavsky In this case, while the question doesn't have a generally "correct" answer, it will put the visitor or OP in a position where he knows enough to choose his own

David Zaslavsky

I agree though, I have been somewhat skeptical of how well the whole SE model works for physics precisely because answers are not easy to test. But if you discount accepting of answers, it kind of works.

ManishEarth

@DavidZaslavsky Exactly

David Zaslavsky

@ManishEarth No, you cannot choose a correct answer when there is no correct answer.

ManishEarth

Jul 16, 2013 16:23

@DavidZaslavsky "choose his own for himself"

For example, if I want to learn about uhh, QM, (broadness aside), and I post the question and get answers elaborating/ranting about the book, I can choose which answer works best for me. A different answer will work best for a different visitor.

David Zaslavsky

@ChrisWhite It's not only knowledgeable users who vote on list-like book recommendations or other lists. A lot of people will have only tried one or two of the books in the list and they'll just upvote the answers which mention those.

user54412

what if the phrasing were "What are the ~~strengths/weaknesses~~ features/missing parts of the graduate-level texts in general theories of turbulence?" or something like that

David Zaslavsky

@ManishEarth No, I mean when there is no correct answer to a question, you cannot make one answer correct by saying "this works for me."

twistor59

@ChrisWhite yep, something like that might work. Much better than "which is best"

ManishEarth

@DavidZaslavsky Sure

While accepting doesn't work

For each reader there is a "personally correct" answer

By virtue of the elaborations in the answer

@twistor59 Oh, we're not phrasing them as "which is the best" in the first place

David Zaslavsky

Jul 16, 2013 16:26

@ManishEarth It's highly misleading (IMO) to call that "personally correct." I'd say the proper word is "favorite."

ManishEarth

@DavidZaslavsky maybe

basically, "works for me"

user54412

personally I wouldn't mind if the "accepted" feature were removed from most of the sites on the network

4

ManishEarth

Unlike an SO book rec question (where you get advice, which is the sum of all favorites), these will be questions where you are provided information and can make the choice yourself.

@ChrisWhite AFAICT my policy will probably make the question be edited into something like that

David Zaslavsky

@ChrisWhite Hm, well, that is better. Questions which prompt answers that have a self-contained evaluation of the major books on the topic don't suffer from the problems that plague list questions. Of course there are still some potential issues with such questions but we could work through them perhaps.

@ChrisWhite I could get behind that ;-)

ManishEarth

I definitely don't want "what is best" questions

@ChrisWhite /agrees

David Zaslavsky

Jul 16, 2013 16:30

OK, so here's an interesting case study: the question Mostafa just linked to

0

Q: Book recommendations for fluid dynamics self study

I have my bachelors degree in engineering science: computer science, option electrotechnic. I have a good understanding of basic mechanics and dynamics, calculus (differential equations). The books I'm now having a look at are: Fox and McDonald's Introduction to Fluid Mechanics, 8th Edition Fu...

fluid-dynamics books education aerodynamics

IMO the question is badly phrase. "What books would you recommend?" <- highly opinionated.

But the answer is good

and is the kind of answer that a good book recommendation question should encourage.

ManishEarth

yeah

Well, sort of

Basically, the question shouldn't ask for "best". It can ask for "good", but it needs to ask for elaboration (i.e. the features, the prereqs, and the style)

David Zaslavsky

Actually I would say it should ask for "standard"

ManishEarth

Hmm, that could work

David Zaslavsky

"Good" is fundamentally subjective

user54412

just throwing this idea out there - what if we had a template for asking book questions?

ManishEarth

Jul 16, 2013 16:33

@ChrisWhite Yes, that would work

David Zaslavsky

Could work

ManishEarth

That's what I've been thinking of right now

David Zaslavsky

@ManishEarth The whole reason we might be able to allow some book recommendations is that there are standard books in various subfields. So a question that asks for an overview of the standard resources in a given field would not be a list question, and would not be overly opinion-based.

ManishEarth

@DavidZaslavsky Yep

user54412

> I am interested in learning about <specific topic>. (I am self-studying, and I know ____ already | I am in school and this is to augment my studies at _____ level) What are the standard books I should be considering? In particular, what do they include, what are they missing, how much extra background do I need for them, and how far along can they take me?

ManishEarth

Jul 16, 2013 16:36

@DavidZaslavsky I thought that when "good" is coupled with a request for details as per my meta post and Chris' chatmessage, it wouldn't really be subjective, but you're right, it's better left out

@ChrisWhite Style is important too

i.e. how rigorous they are, etc

David Zaslavsky

@ChrisWhite Here's my issue with that though: that allows an effectively infinite variety of book recommendations depending on what each individual poster's level of preparation is.

ManishEarth

So... we don't mandate the prereqs in the question, we mandate it in the answer

Makes it useful for everyone. Though it will increase A-per-Q as well

David Zaslavsky

One of my problems with existing book recommendations which is not related to them being list questions is that they tend to overlap slightly, enough so that they are kind of duplicates but not quite? It results in a ton of duplication

@ManishEarth That could work.

user54412

@ManishEarth that's an improvement

ManishEarth

@DavidZaslavsky Good point. But afaict we have that much duplication with other questions too

David Zaslavsky

Jul 16, 2013 16:39

@ManishEarth Not that much.

ManishEarth

Not that much less, either :P

David Zaslavsky

Except for a few questions that seem to get asked over and over again, like why there's a c^2 in E=mc^2, but those it's much easier to close as duplicates of each other

ManishEarth

@DavidZaslavsky 'course

Reminds me, there probably are a bajillion of those which haven't been closed yet

user54412

@ManishEarth I agree. Actually, I was considering proposing a CW post (I know, I know - CW is evil) to accumulate the Hubble expansion questions into a list of 5-10 questions that, when gone through in order, tell everything one needs to know about the subject. Then I realized there are > 100 near-duplicates on the subject...

ManishEarth

@ChrisWhite Why CW? Make it a canonical long post

And then amass all the rep and start laughing evilly

twistor59

Jul 16, 2013 16:42

I really should know this, but what actually is Community Wiki?

David Zaslavsky

@ManishEarth yeah, I can't keep up

user54412

@ManishEarth I didn't want to answer the question - I wanted to amass a set of links to where specific answers could be found

user54412

with a brief summary of each link

ManishEarth

@ChrisWhite Ah. CW then

David Zaslavsky

@ChrisWhite OK, see now that's what community wiki should be used for. A question that's really a resource to be collaboratively assembled by the entire community.

ManishEarth

Jul 16, 2013 16:44

@twistor59 It's for posts which are meant to be a community collaboration. It used to be also for questions where the voting will be colored with opinion. For that matter, it still is, but it's only supposed to be used for questions in special cases

Also, on meta <2k users can't edit period, so CW is good for collab

twistor59

@ManishEarth thanks! Where are the links to it?

user54412

@twistor59 Basically it lowers the barrier to editing, and negates any reputation changes (essentially it removes ownership)

ManishEarth

Blog - Stack Exchange: The Future of Community Wiki

Grace Note on August 19, 2011

When you mark a post community wiki on a Stack Exchange site, that means …

this post can be edited by anyone with 100 reputation

this post does not generate any reputation for anyone when upvoted or downvoted

The main advantage of community wiki — more editing — was nerfed when we introduced suggested edits. With suggested edits, anyone, even an anonymous user, can edit anything — so long as another experienced user reviews and approves their edit.

This leaves many wondering — what’s the point of Community Wiki? …

Btw guys, the privilege and badge pages have changed

@twistor59 physics.stackexchange.com/help/privileges/community-wiki

user54412

@ManishEarth ?

ManishEarth

@ChrisWhite physics.stackexchange.com/help/privileges

user54412

Jul 16, 2013 16:46

they changed a while ago?

ManishEarth

@ChrisWhite physics.stackexchange.com/badges

Colin McFaul

@ManishEarth Yes, and they are super awesome.

ManishEarth

@ChrisWhite Yeah, I know, but most haven't noticed :P

@ColinMcFaul :D

David Zaslavsky

Yeah, I can't keep up with all the changes the SE team makes

twistor59

@ManishEarth thanks for the info!

ManishEarth

Jul 16, 2013 16:47

@DavidZaslavsky They're doing a lot of things to make new user experience more smooth

David Zaslavsky

Indeed. Having full-time developers is pretty great sometimes :-)

ManishEarth

@DavidZaslavsky Yep.

They also have a new UX guy (and Laura, wh was already there though iirc she was in the pitchfork crew)

ManishEarth

@DavidZaslavsky @ChrisWhite Instead of a specific template, why not a boilerplate banner that we add below the question that says "Answers must note prerequisites, style," (etc etc)

user54412

@ManishEarth like in the question itself?

ManishEarth

The question just needs to be of the form "What standard textbooks are there for X, and how are they?"

@ChrisWhite yeah. Though maybe we could get a post notice

On the topic of post notices, something that would be useful for Physics

Or something like that

So we basically edit out the subjectiveness, and add boilerplate text (with a link to meta policy?) that explains how you should answer a book question

David Zaslavsky

Jul 16, 2013 16:54

Sure, I guess that'd help

Crazy Buddy

4 hours ago, by Crazy Buddy

Because, SE engine has something with Safari ;-)

David Zaslavsky

A practical consideration is that this is a lot of work to be doing to keep on top of book questions.

Crazy Buddy

Someone care to address that?

Safari doesn't give anything good..!

user54412

@CrazyBuddy Use a better browser. Consider your issue addressed :P

ManishEarth

@DavidZaslavsky Yeah, for that we need the community's help

Crazy Buddy

Jul 16, 2013 16:57

@ChrisWhite Actually, I thought of changing my browser from Firefox to something nice...

ManishEarth

Though tagging works too (in this case, not a meta tag ^.^)

@CrazyBuddy Chrome!

Crazy Buddy

Opera, IE & Chrome are YUCK..!

ManishEarth

(-ium?)

@CrazyBuddy O_0

Crazy Buddy

@ManishEarth Chrome doesn't support a few JS scripts.>!

ManishEarth

IE10 is pretty nice. Prefer Chrome though

user54412

Jul 16, 2013 16:58

Firefox! It has fire! and a fox!

ManishEarth

@CrazyBuddy Userscripts? Yes it does. Its just a wee bit harder to install.

Crazy Buddy

@ChrisWhite Yeah..! But, I used it for years..!

(along with Chrome)

David Zaslavsky

@ManishEarth I think if we can get a few-click procedure in place to attach the guidelines to any new book question, e.g. a custom post notice that links to a meta question which we will create, that would be good. I'd be fine with book questions in that circumstance.

Crazy Buddy

that's why I set out in a mission to experience a few other browsers..!

ManishEarth

@DavidZaslavsky Yep. If we can't get a custom post notice, what about a banner that we edit in?

David Zaslavsky

Jul 16, 2013 17:00

@ManishEarth meh, I guess that's acceptable, but far from ideal. There are more mental obstacles to going to a meta post, looking up a banner text, and editing it into the question.

Although then again anyone can do it, not just moderators.

Hmm....

ManishEarth

@DavidZaslavsky right

but

Crazy Buddy

@ManishEarth Hey... I've seen it..!

ManishEarth

yeah what you said

:P

David Zaslavsky

(this also being the reason the pro-forma comments script is so much more useful than just a meta post with boilerplate comment text)

Crazy Buddy

Isn't this issue going on somewhere,.?

Started by Chris..!

ManishEarth

Jul 16, 2013 17:01

@DavidZaslavsky yep

David Zaslavsky

Maybe if you create a pro-forma edits script? :-P

Crazy Buddy

4 mins ago, by Chris White

@CrazyBuddy Use a better browser. Consider your issue addressed :P

Grr... @ChrisWhite... ^^

@ManishEarth: Did Safari users had any problems in experiencing SE?

(I like it, coz it's fast, fine, etc...)

ManishEarth

@CrazyBuddy Sometimes. I use SE from an iPad.

But usually, no

Chat has issues in the non-mobile mode

David Zaslavsky

Well anyway, it's past 1 PM which means our scheduled time for the chat session is over! I have to run but feel free to continue discussing.

Crazy Buddy

@ManishEarth: Well, I asked you because you're the meta-specialist..! Just to get a nice link..! :P

ManishEarth

Jul 16, 2013 17:04

@DavidZaslavsky All right. I need to get going too. :)

user54412

If we could come to a boilerplate, I wouldn't mind helping bring book questions up to our standards

ManishEarth

@CrazyBuddy Not sure, meta search

user54412

lunch!

ManishEarth

@CrazyBuddy meta.stackoverflow.com/questions/tagged/…

Cya guys

I may incorporate part of this into the meta post

David Zaslavsky

@ChrisWhite I think we can, it may take a little while

@ManishEarth Note that I think book questions should be treated as a separate issue from list questions, i.e. good book questions should not be list questions.

The h Bar

Participants