Timeline for Hide trivial comments
Current License: CC BY-SA 4.0
49 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 18, 2021 at 12:13 | history | edited | CommunityBot |
replaced http://data.stackexchange.com/ with https://data.stackexchange.com/
|
|
| Jan 18, 2021 at 11:45 | history | edited | CommunityBot |
replaced http://blog.stackoverflow.com with https://blog.stackoverflow.com
|
|
| Mar 26, 2019 at 3:16 | comment | added | Rob | It would certainly help if the comments flagged as irrelevant by the person whom wrote the question or answer were deleted, instead of the flag most often being declined - now I end up leaving them for someone else to respond to, making the message clear, eventually; assuming that they return and check for other responses or any updates to the post. | |
| May 23, 2017 at 12:37 | history | edited | CommunityBot |
replaced http://stackoverflow.com/ with https://stackoverflow.com/
|
|
| Mar 20, 2017 at 10:31 | history | edited | CommunityBot |
replaced http://meta.stackexchange.com/ with https://meta.stackexchange.com/
|
|
| Apr 6, 2016 at 18:58 | history | edited | Jon EricsonStaff |
edited tags
|
|
| Apr 1, 2016 at 18:15 | comment | added | ᔕᖺᘎᕊ | @JonEricson would status-declined be kinda appropriate for this? | |
| Dec 20, 2015 at 11:22 | comment | added | Bruno Lowagie | One thing I'd like high-rep users to be able to do, is to automatically move comments to the question. Often low-rep users don't know they can edit their question (due to their low rep they need someone else to approve the edit) and they clarify their question by posting a comment (sometimes they even paste source code in the comment, which then becomes illegible). I often copy/paste such comments to the question, but it would be great if I could just click on the comment to extend the question. As a result, the question would improve for those people who overlook the comments. | |
| Dec 16, 2015 at 22:17 | comment | added | Dronz | Great idea. Hmm. Ideally, maybe there would be a way to have hidden comment digression threads saved but hidden (unless opened) like this too. | |
| Dec 16, 2015 at 6:41 | comment | added | Ian Miller | Can this formula be tweaked on a community by community basis? For exam math.SE comments do not need to be as long as other sites to be as benefical. | |
| Dec 14, 2015 at 19:28 | comment | added | AGE | @JonEricson would be worthwhile to also add a feature (privilege) for reviewers to vote to/remove altogether trivial comments, can't tell you how many times I wish I could do that | |
| Apr 24, 2014 at 13:52 | history | edited | CommunityBot |
Migration of MSO links to MSE links
|
|
| Dec 5, 2013 at 19:59 | history | edited | Shadow Wizard | CC BY-SA 3.0 |
Move critical information from answer to the question to avoid some confusion, link to newer related discussion
|
| Dec 5, 2013 at 19:47 | comment | added | Shadow Wizard | @Jon any updates? If you want I'll stop pinging but it's something I consider important plus I'm curious. Oops, just noticed Jaydles answer - sorry! :) | |
| Dec 5, 2013 at 19:22 | answer | added | Travis J | timeline score: 5 | |
| Dec 3, 2013 at 23:10 | answer | added | JaydlesStaff | timeline score: 20 | |
| Nov 20, 2013 at 18:39 | comment | added | Shadow Wizard | @Jon just roll it out here on Meta and see what happens. :-) | |
| Nov 20, 2013 at 18:00 | comment | added | Jon Ericson Staff | @ShaWizDowArd: We are actively looking into ways we can test the concept. As David mentioned there just isn't enough information to know how much signal we might lose if we hid all unupvoted comments. I hope to have an update (probably just a strategy for looking into the problem) by the end of the week. | |
| Nov 20, 2013 at 12:45 | comment | added | Shadow Wizard | @Jon any updates, two weeks later? | |
| Nov 7, 2013 at 20:24 | comment | added | mxyzplk | This is a great idea; on RPG.SE we are strong believers in comment deletion of trivial and discussiony comments after a short time, they detract from real answers. If they were simply hidden then we'd get the benefits without the work and hurt feelings of comment deletion. | |
| Nov 7, 2013 at 16:56 | answer | added | Jon EricsonStaff | timeline score: 17 | |
| Nov 7, 2013 at 4:28 | comment | added | Manishearth | @JonEricson twitch twitch | |
| Nov 7, 2013 at 2:28 | comment | added | Jon Ericson Staff | @Adel: Well, to be fair, they are already paying my salary. (I got my stickers and a full week in NYC last month. ;-) | |
| Nov 7, 2013 at 1:40 | comment | added | Caffeinated | The effort here is massive . Admins, at the least please hand the man a crate full of stickers and 2 nights in your town | |
| Nov 6, 2013 at 23:13 | answer | added | Esoteric Screen Name | timeline score: 6 | |
| Nov 6, 2013 at 22:24 | comment | added | Jon Ericson Staff | @Flexo: If you want to, sure. Oh, you mean automatically delete them? Probably not. ;-) | |
| Nov 6, 2013 at 22:00 | comment | added | Flexo - Save the data dump Mod | Can we delete them instead of hide them when they're hidden and not loved for long enough? | |
| Nov 6, 2013 at 20:34 | answer | added | David Fullerton | timeline score: 39 | |
| Nov 6, 2013 at 19:35 | comment | added | Esoteric Screen Name | @JonEricson I don't think it's square one at all. The crux of the suggestion is your improvement to the algorithm determining which comments should be hidden. Here's a query demonstrating your suggestion but only applied if there are > 3 comments. It hides 15,782,861 comments (19.7m using a threshold of 2 instead of 3), still a huge jump over the 3m currently hidden. View count rather than (or in addition to?) comment count is a good idea, though... | |
| Nov 6, 2013 at 19:30 | comment | added | Jon Ericson Staff | @Esoteric Screen Name: That sort of takes us back to square one in my opinion. Now waiting until a question has enough views might be a more promising threshold. For a low traffic question, there just aren't enough potential voters to even see a comment. Hmmmm... | |
| Nov 6, 2013 at 19:19 | comment | added | Esoteric Screen Name | Would there be some minimum threshold number of comments posted before hiding takes effect, even if the comments aren't high enough score? I'm lazy and don't want to click to see the single comment on an older question. This would also help alleviate the low traffic problem that McNab raises. | |
| Nov 6, 2013 at 18:49 | answer | added | voretaq7 | timeline score: 9 | |
| Nov 6, 2013 at 18:34 | comment | added | Shadow Wizard | @Jon fair enough, thanks for considering! Another thing I have in mind: reputation. Usually users with high rep know what they're doing and will post better comments (not talking about Meta, lol!) so for every 10K rep add one weight point, up to some limit of course. What do you think? | |
| Nov 6, 2013 at 18:30 | comment | added | McNab | @JonEricson - I've read over your very well researched bit of work again and I don't think I can. Disregard my comment about the comments disappearing over time. I just worry that hardly anyone really votes on comments much in the tags I monitor and often there are short, crucial comments that have been left to expand on an answer. I'd rather see all the comments, maybe this more suited to a user setting. But +1 for your work on this :) | |
| Nov 6, 2013 at 18:26 | answer | added | Manishearth | timeline score: 2 | |
| Nov 6, 2013 at 18:17 | comment | added | Jon Ericson Staff | @McNab: Could you help me out with an example or two of the sorts of comments that would be improperly hidden? If you have a post in mind, you can test which comments will be shown with the query that simulates the comment weight hiding algorithm. It's a lot easier to think about this with concrete examples. | |
| Nov 6, 2013 at 18:10 | comment | added | McNab | My problem with this is that it's another one of these good ideas that suits the heavily trafficked tags where many experienced users routinely vote on everything. On some of the less popular tags, one or two upvotes would be considered a popular comment and I would expect to see every comment disappear over time quite frankly. I would suggest some kind weighting in the calculation that takes overall voting patterns in the tags into account. | |
| Nov 6, 2013 at 17:56 | comment | added | voretaq7 | @Asad On Server Fault we routinely encourage editing the question / posting an answer so the comments become obsolete, and the comments discussion can then be tossed town the nearest memory hole in accordance with the "comments are ephemeral" logic. This works well because the SF mods are somewhat OCD, but for huge communities like SO the community at large could do the encouraging & then flag comments as "obsolete" so the mods can quickly scrub them. | |
| Nov 6, 2013 at 17:55 | comment | added | Jon Ericson Staff |
@Sha Wiz Dow Ard: Hmmm... I hate to mess with comment flagging, which really is a strength of ours. But it does seem like the obsolete and too chatty flags could be used to reduce comment weight rather than outright deletion. As a moderator, I found these flags very difficult to handle since they are so often judgement calls. But wiping out 2 upvotes seems disproportionate to me. (I'll keep thinking about this; seems like a tweak we could apply after we've seen how the feature works in practice.)
|
|
| Nov 6, 2013 at 17:14 | comment | added | Servy | @Asad That happens already if, in a back and forth, only one side of the conversation is getting upvotes (or one is getting more) and the comment thread goes on long enough. So in that case, while it may not be desirable, it's still just staying the same, not getting any worse. In fact, given the proposed change, the less voted on comments would have enough "weight" to be shown for several days without votes, unlike now, where they'd be hidden as soon as there are enough comments for some to be hidden. It's a net wash or win in every way, even if it's not perfect. | |
| Nov 6, 2013 at 17:04 | history | edited | Jon EricsonStaff | CC BY-SA 3.0 |
Clarify what "hidden" means.
|
| Nov 6, 2013 at 9:16 | comment | added | Shadow Wizard | Third class citizens finally getting some love, nice! One thing I would add is that each flag on a comment will reduce its weight. I think 2 points is sensible. Many times old comments get one or two flags that are not enough to auto-delete and get buried too deep in the queue so moderator can take some days until he handles this. | |
| Nov 6, 2013 at 5:41 | comment | added | Jon Ericson Staff | @Asad: That's a very interesting idea. A counter argument is that the OP should probably edit their question in addition to/instead of having a long comment conversation. There would also have to be some way to avoid letting the OP get their comments written in stone by making them long enough. I still think a comment needs to be validated via at least one upvote. | |
| Nov 6, 2013 at 5:41 | comment | added | user200500 | Additionally, for a specific user, their own comments and responses should have a higher weight in determining which comments are hidden for them. This might be too expensive, depending on how pages are generated and whether the weighting and hiding logic goes on the server or client side. | |
| Nov 6, 2013 at 5:34 | comment | added | user200500 | I like this idea. Problem I foresee: in comment conversations between the OP of a question and someone who is helping them, it is very common to see upvotes on the "helper's" comments, and none on the "helpee's" comments. I feel like this might lead to unneccessary annoyance where every conversation in which an asker has sought help from an answerer will be partially obscured by a link. Perhaps the question owner should have extra weight on their comments? | |
| Nov 6, 2013 at 5:34 | history | edited | Jon EricsonStaff | CC BY-SA 3.0 |
Bring up the issue of the cap.
|
| Nov 6, 2013 at 5:19 | comment | added | Jon Ericson Staff | @BoltClock's a Unicorn: Nope. If a comment is 150 characters or longer and has one upvote, it will be shown forever or until deleted, whichever comes first. Shorter comments need more votes, but even the shortest gets displayed permanently with 10 or more votes. | |
| Nov 6, 2013 at 5:16 | comment | added | BoltClock's a Unicorn Mod | Will every comment in existence eventually become hidden given enough time? | |
| Nov 6, 2013 at 4:43 | history | asked | Jon EricsonStaff | CC BY-SA 3.0 |