There are, IMO, quite a lot of badly tagged questions and... not very good tags. Some of them were discussed on meta recently; some of these discussions show, IMO, that users who created these tags don't always understand tagging system of Math.SE well enough.

On Meta.SO one needs 500 rep to create a new tag, on SO — 1500 (!) rep. On Math.SE — just 300 rep.

Since its creation Math.SE has grown a lot. Maybe now it's time to (ask to) raise reputation requirements for creating new tag on Math.SE? (Say, to 1000 rep?)

(Idea suggested by Tito Piezas III in this comment.)

  • 1
    $\begingroup$ Somewhat related feature request: Warning for new tags $\endgroup$ – Grigory M Jan 6 '14 at 15:58
  • 1
    $\begingroup$ I didn't even realise the value was variable among different SE sites. I agree this would be a good idea, seeing as it seems to not be hard to change. $\endgroup$ – Dan Rust Jan 6 '14 at 16:23
  • 2
    $\begingroup$ Great feature. I still think that the warning for new tags is a great idea, but this seems like a first step in reducing tag noise. $\endgroup$ – Asaf Karagila Mod Jan 6 '14 at 16:47
  • 3
    $\begingroup$ I would put it even higher. A low limit was ok in the early days of the site, but it is time to raise the bar now. Ideally I might want to require experience on Meta, because exposure to relevant discussion is only available here. $\endgroup$ – Jyrki Lahtonen Jan 6 '14 at 16:56
  • $\begingroup$ Is there a case to be made for a canonical tag destruction thread, similar to the one for reopen votes? I ask because "probabilty" got created a few days ago, I posted about it in the Tagging chat, and so far it seems to have escaped moderator attention. $\endgroup$ – hardmath Jan 6 '14 at 17:18
  • 1
    $\begingroup$ @hardmath Maybe Tag merging and synonyms should be used for this purpose $\endgroup$ – Grigory M Jan 6 '14 at 17:33
  • 3
    $\begingroup$ @hardmath For cases like "probabilty", just retagging the question(s) with the typo'ed tag should be enough. If a tag is unused, it is automatically deleted after - iirc - 24 hours. Of course if a typo surfaces again and again, sterner measures must be taken. $\endgroup$ – Daniel Fischer Mod Jan 7 '14 at 0:40
  • $\begingroup$ @hardmath Sorry, I've misread your question — that thread is for discussion of changes, for obvious typos just retag the question $\endgroup$ – Grigory M Jan 7 '14 at 14:25
  • $\begingroup$ Erm... what now? Do we need to do to something to get the attention of the powers that be? $\endgroup$ – Grigory M Jan 8 '14 at 21:58
  • $\begingroup$ Allegedly they monitor the feature request tags through meta sites. There is a chance that this will take effect soon. $\endgroup$ – Asaf Karagila Mod Jan 9 '14 at 19:04

After a quick check on this, we've increased the threshold to the suggested 1000 reputation. That seems a saner number than 1500 for here. Only a handful of tags are created in the 1000-1500 reputation range, many of which were things like and that are quite frequently used.

  • 4
    $\begingroup$ Wooooooooooooooooo! (And I'm speaking for at least several users, I'm sure!) $\endgroup$ – Asaf Karagila Mod Jan 16 '14 at 0:34

I agree; there is much less need for new tags on an established math Q&A site than on a technology-oriented Q&A site. On a tech Q&A, users are likely to bring up questions about a new gadget or a new version of some software that was just released. Here we don't normally get questions about mathematical areas that were just recently invented. Unless we count kalle numbers. Or , which was invented just an hour ago.

I suggest setting the limit at 1500 as on SO, simply because it seems SE prefers to avoid too many special cases among the sites. But either 1000 or 1500 would be an improvement.

  • 1
    $\begingroup$ Perhaps we are in need of a kalle-numbers tag for such innovations. $\endgroup$ – Alexander Gruber Mod Jan 6 '14 at 17:20
  • $\begingroup$ -1 for bringing up that infernal thread; +2 for the rest of the content. $\endgroup$ – Asaf Karagila Mod Jan 6 '14 at 17:24
  • 13
    $\begingroup$ Retagging someones questions as kalle-numbers would certainly make for a powerful insult. $\endgroup$ – Michael Greinecker Mod Jan 6 '14 at 19:26

You must log in to answer this question.

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