13

On another discussion I recently suggested that we could try having a recurring event here in meta, similar to what Photo.SE does (they hold a contest). It could even help build community integration.

Since this caught the imagination of a couple others, we though we'd bring it up as a new meta post for discussion.

So...

  • What do you think?
  • Do you find it useful?
  • Do you have suggestions on how to make it work?

The way it currently works on Photography:

  1. The top answer to the poll is set to be the photo of the week
  2. The selected answer is moved to a separate hall-of-fame question
  3. Downvoting the answers is forbidden (only upvotes are considered to determine the score)
  4. Users can submit one photo per week, without repetition
  5. Minor edits are allowed
  6. Photo critiques are supposed to be done in chat
  • I think it's an awesome idea. Would we do this as community wikis? Or attributed answers? Would there be any issues created by having excellent knowledge over here in Meta as opposed to all such knowledge being on the main vi stackexchange site? – Wildcard Feb 15 '16 at 20:23
  • 2
    I like the idea but I suggest that the contest have a theme for each week, but not like a challenge. Just to keep focus of the answers and avoiding that the same tip posted each week. – vappolinario Feb 16 '16 at 13:48
  • @Wildcard Attributed answers might work better, it would even enable us to vote on single tips individually. Plus, reputation in meta is not earned for votes. – Roflo Feb 16 '16 at 17:11
  • @vappolinario, how would that be different than just having a well-written question posted on the main site? Hmmm...perhaps we just need to write more community wiki question/answer pairs.... – Wildcard Feb 16 '16 at 17:20
  • I'm not sure yet what to think of this as such, but I do feel that this site is too small for a once a week contest. Once every fortnight or month would probably be better to start with. – Martin Tournoij Feb 18 '16 at 4:38
  • @Carpetsmoker you're probably right. If we think the community response will be slow, a tip-of-the-month would do. We'd still need to agree on the mechanics (which I assume will be equal regardless of the frequency). – Roflo Feb 18 '16 at 15:18
  • 1
    From a pragmatic point of view, is this something we can actually do? On the photo site it's put in the header, but this is not a standard option or feature that I have available (as a moderator, I can only modify some help files). I can add featured to meta threads to make them show up in the sidebar, and AFAIK that's the only way we can do this... It's a bit of an abuse of the feature, though... – Martin Tournoij Feb 22 '16 at 12:32
  • I have to rescind my earlier support for this idea...I think we just need more Q&A sharing knowledge (written as community wikis) to capture the best tips in "canonical" form that can then be pointed to and linked to by others. – Wildcard Feb 25 '16 at 7:49
  • If we're just going up rehash the same suggestions, then there's no point to this question. – muru Feb 26 '16 at 16:54
  • @muru Ok, maybe we should remove it from featured. Ah, you beat me to it. – Roflo Feb 26 '16 at 17:00
4

Firstly, the main problem is maintenance and quality. And Maintenance is hard. And by maintenance, I mean both of the poll question, and of the past selections. For the poll, we need a way for selected entries to be removed from the poll, so that other posts can have their chance. The way Photography does it is not directly workable for us - move the top entry elsewhere. They have just a photograph, after all. Where do we move an entire post? Creating new polls repeatedly is not workable, IMHO - what do we do with the other entries to a poll, and the votes they gathered? Users could re-post entries with new polls, but would people be willing to vote twice, thrice or more on the same post, if it were reposted? A single, ongoing poll is the best option.

Secondly, there are three ways the general Stack Exchange site layout lets us "pin" something to the main page:

  1. Featured meta posts
  2. Community events
  3. Chat events.

The first two require moderator action. I think the third can be done by room owners, but I have never tried those before.

I have a rough plan for how this could work:

As with Photography, a continuous poll question from which the top answer will be selected each week. The posts should be community wiki, so that most users can improve it. Again, down voting will be forbidden - only upvotes will be considered. The top answer will be linked for a week in the community bulletin (perhaps as an event, since the event link is only required to be on the Stack Exchange network, whereas featured posts are always questions).

Now, how do we eliminate this entry? We shouldn't lose knowledge, so just deleting it is out of the question.

My suggestion is Github. Stack Exchange's blog system is in limbo, so it doesn't seem to be a good target. We can create a Github organisation dedicated to our Stack Exchange, and use Github pages as a repository for past tips. Since Github pages are also git repositories, if someone wants to suggest a change, it can be done via a fork-and-pr. So, a given post is not totally lost. The syntax is highly similar, so only minimal touch-ups would be needed.

Unfortunately, an external link cannot directly be added to the community bulletin (not without SE intervention, at least). So, when the post is selected, the mod/whoever is in charge will make a Github page for it, add a link to this page in the post, for people wanting a permalink, and then a mod will make it an event lasting a week.

This creates some problems:

  1. The post passes out of your ownership, and out of the site itself. This is very troubling to me. Sure, many of you may have Github posts, and perhaps you could make the commit that adds the post, but I don't know how this issue can be handled gracefully.
  2. Any comments the post gathered will be lost. Comments are supposed to be transient, but they often contain useful information.
  3. Github pages doesn't offer any comment system of itself. We could use Disqus the way Stack Exchange blogs now do.
  4. ...

The problems with Github pages are mostly eliminated with Stack Exchange's own blog system:

  1. Site blog entries automatically get added to the bulletin.
  2. No touch-up will be needed - the syntax should be identical.
  3. The post would be under your authorship.
  4. And for those maintaining the poll, the workflow would be easier.

However:

  1. As I said before, the blog system is in limbo. New blogs are not being created. They might be willing to do this, but I'm not sure how much willing for a beta site.
  2. Comments on the poll entry would be lost.
  3. No easy way for the community to update past entries (as far as I can tell, which is not much).

That last point may not seem that important, but I believe one of the greatest features of Stack Exchange is that a post's age doesn't work against it - a number of grand old answers across the network are still as useful after five years because we have been able to keep them updated.

  • I'm not convinced about using Github (or any other non-SE site). Anyway, I just found out that Fitness.SE (also in beta) has a blog. Are you sure new blogs aren't being created? Of course, if we do start a blog, we'd have to make sure maintenance is good. – Roflo Feb 16 '16 at 23:06
  • 1
    @Roflo I'm going to go out on a limb and say Physical Fitness is part of the reason why new blogs aren't being created: existing ones had pretty much died. See meta.stackexchange.com/q/244467/270345. Note: last post on Fitness' blog was in 2012. – muru Feb 16 '16 at 23:07
  • Ah, thanks for the link. Perhaps you could add that link to your post? Maybe where it says "New blogs are not being created"? – Roflo Feb 16 '16 at 23:13
  • @Roflo Sure. I wrote the entire thing on the app, and the earlier post had blogs as a suggestion, which is why I didn't. – muru Feb 16 '16 at 23:15
  • 1
    @Roflo (and anyone else interested in the blog idea): Also check out related discussions on Magento and World Building metas. – muru Feb 16 '16 at 23:22
  • 1
    I think that one difference with the photo site is that the photos are not really something you can post at all in the "normal" SE Q&A system, while Vim tips are. So while I think it's an interesting idea, I have some reservations about whether it's also a good once, and if it isn't just replicating the SE functionality already. – Martin Tournoij Feb 18 '16 at 4:41
  • @Carpetsmoker well, one can do same thing with photos that one can do with tips: Convert to Q&A. Take a photo, ask how some flaw in it can be corrected, or how to process it to get a particular effect, or ask how a pic was shot or processed, etc. Of course, that's hard, and most people don't want to do that, just like they don't always want to make questions and answers out of something they found interesting - they just want to show it to an audience that they know will find it interesting as well. We have such an audience, and it seems people willing to both show and see such things... – muru Feb 18 '16 at 4:50
  • ... It's just that Q&A isn't necessarily a good or an easy way to go about it. – muru Feb 18 '16 at 4:50
  • 1
    @muru Unless I misunderstood something the photo contest, it seems to be just a "best looking photo"-contest. Nothing to do with effects or flaws or whatnot. I don't see how you can put that in a Q&A format. Whereas the "vim tips" are 100% about practical problem solving and fit very well in the Q&A format. – Martin Tournoij Feb 18 '16 at 4:52
  • @Carpetsmoker sure Vim tips can be posted as Q&As. It's just that people aren't necessarily interested in writing up Qs for them. – muru Feb 18 '16 at 4:54
  • @Carpetsmoker Also, the thing that started all of this is that we don't usually accept list of tricks in our Q&A format. I often find myself browsing through odd Qs and As just to see if there's something I didn't know that could be done. Regardless of whether I need them or not. – Roflo Feb 18 '16 at 17:12
  • @Roflo As I explained in this answer, I don't think we need a "special question" for tricks. If you want to find random "tricks", then just look at all questions ordered by votes or all answers ordered by votes... – Martin Tournoij Feb 18 '16 at 19:46
  • 1
    @Carpetsmoker I get that. But if I want to learn something new (unknown unknowns), I can't ask people to tell me something I don't know that I don't know. Argh, I hate writing sentences like that (lol). – Roflo Feb 18 '16 at 19:59
5

We could make a contest of screenshots or small animated GIFs of VIM sessions. We could also change the motto for each week, like

  • Craziest window layout
  • Coolest plugin
  • VIM integrated into other Apps
  • Best color scheme
  • VIM on craziest platforms (like smart watches)
  • ...

Also I would think about extending the one week period to two weeks or so.

  • 6
    Interesting. Thought I'd be against purely image-only submissions. A transcript of animated GIFs or an explanation should be present. – muru Feb 17 '16 at 18:39
3

How would this add to the current Stack Exchange format we already have?

The contest on the photography site is not a very good analogy here. The contest on the photo side is intended to:

showcase your best quality work, demonstrating at least moderate skill with a camera and a general understanding of the artistic aspects of photography.

This is useful, because these are photographs you can't really share on the main photography site using the standard method of sharing information: asking and answering your own question. Most questions that would feature those photographs as answers would either be opinion-based or so artificial and contrived that they would be nigh-useless and likely to be closed.

We already tried something like this with Please list your single best .vimrc trick, and as I outlined in my answer to How do we share general Vim experience? Or can we? the results weren't exactly great:

  • There was a lot of duplicate information.
  • There were unclear and unhelpful answers.
  • There is no way to hold a discussion about a particular "tip". Comments are far too limited.

In addition, we already have a way to find "good tips":

I don't see how a single "tip thread" would add to that. It would most likely duplicate information from pre-existing answers, and if the tip is dodgy or outright crap then the only thing I can do about it is point that out in the comments, which are extremely limited and not designed for these sort of discussions (remember, according to the above rules I can't even downvote).


The only way I can see this working is by having a thread which links to the "best answers of the {week,fortnight,month}". This would avoid the problems associated with stepping outside of the "Stack Exchange model" and doing things the platform isn't really designed for, but you can still give people the opportunity to highlight things they find particularly clever or useful.

Personally, I don't really see the need in this, but am not against it either if other people would like to have it.

  • I find your arguments about the limitedness of comments somewhat dodgy. Why can't you point out crap in comments? And since a mod has to involved in the poll anyway, it's not that hard to delete entries that aren't up to standard. Also, post enough comments on an answer and you can open a chatroom for it. – muru Feb 23 '16 at 0:16
  • @muru No newlines? No way to say "this is a better way to do it" since posting code beyond a single line is impossible, very limited size, no way to edit/improve comments after 5 mins. And I am not comfortable with mods deciding what is "up to standard". This is not the role of moderators. Chat is a good way to hold a in-the-moment discussion, but the results are volatile and not really something that's easily looked up 6 months later. – Martin Tournoij Feb 23 '16 at 0:29
  • there are plenty of pastebins. Chatrooms are frozen, not deleted - the link remains underneath the post. Why shouldn't we delete poll entries that are crap/duplicates? This already happens on main, albeit through review. Here on a meta, for a single poll, I don't think we'll need the same manpower. – muru Feb 23 '16 at 0:33
  • @muru Seriously, pastebins? I don't want to go off to some crappy external site that may or may not work in a year time just to post three lines of code. I also don't want to open said crappy external sites to read your three lines of code. I don't understand why you're so hellbent on doing this? What advantage is gained? What can we do that we can't do on the main site? – Martin Tournoij Feb 23 '16 at 0:41
  • You can't really post three lines of code in a comment? Just use separate backticks. If not pastebins, Github Gists (or alternatives thereof). Seriously, what is your problem with looking for solutions instead of immediately pronouncing doom at the slightest hint of a hitch? I am not hell-bent on anything - I see a group of people interested in doing something that doesn't necessarily fit in with Q&A and I'm willing to at least try to help them out. – muru Feb 23 '16 at 0:45
  • Come on, it looks butt-ugly. And it may work for three lines in some cases (if it fits in the character limit) but doesn't scale very well beyond three lines. And the fact that some people "are interested" doesn't mean it's a good idea. People are trying to solve a problem ("How can I find new things I never knew about"). This is a very reasonable problem, but the proposed solution is simply not a very good one that fits with the way the platform works. It was attempted in one question, and we had answer that were either duplicates of existing ones, a few crappy ones, – Martin Tournoij Feb 23 '16 at 0:52
  • (cont'd, because comments are not suitable for long discussions) and others that could have been much better served in a question as there are more ways to fix the problem it solves. The entire proposal is like driving in a screw with a hammer. Yeah, it works, but it's a kludge and there are better ways to go about it. Again, as I said in my answer, we could have a "best answer of the ..."-thread that avoids most of these problems and doesn't require work-arounds with pastebin/GitHub/Disqus/whatever. I don't see the need in it, and suspect it will go "dead" after two or three runs, but I'm – Martin Tournoij Feb 23 '16 at 0:52
  • (cont'd again, because it's really shit) not against trying it. I just want to maintain a high signal/noise ratio. Having quality answers is really what the platform is good at and was designed to do, and side-stepping that doesn't really help anyone. – Martin Tournoij Feb 23 '16 at 0:55
  • Also, and I think this is probably the most important part here, you say "doing something that doesn't necessarily fit in with Q&A". What answers do you expect that could not be posted on the main site in a Q&A format? Because I can't think of a single answer that would not also be a good answer on a "normal" question. – Martin Tournoij Feb 23 '16 at 0:58
  • What I said before: People aren't necessarily interested in writing up Qs. Maybe you are, but not everyone is. – muru Feb 23 '16 at 0:59
  • 1
    @muru If you have an answer, then writing a question should not be too much trouble. And yeah, perhaps we should be thinking about ways to encourage people to do that; but this is a question which could have lots more answers than just "have a generic tip thread"... – Martin Tournoij Feb 23 '16 at 1:02
  • 2
    Why would it have a lot more answers? On the one hand, you say once-a-week is unsustainable, on the other hand you say we will get too many answers. You can't have your cake and eat it too. :P Besides if people really do fail to abide by poll rules too often and just post crap, we can shut it down. It's not like we're binding our souls for all eternity to oversee the poll. The point of the experiment is it to try it in a structured way, take precautions to avoid past disasters and see if it works. If it works, OK. If it doesn't, no (or only small) harm done. – muru Feb 23 '16 at 1:08

Not the answer you're looking for? Browse other questions tagged .