Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
Here’s something that confuses me about the “tags” on the nForum. Can a tag contain spaces? The comment “(comma separated)” on the tags input box suggests that spaces in a tag would be okay (in particular, wouldn’t break it into two tags), but I don’t think I see any tags containing spaces in the tag cloud.
Considering there are tags like ’of” and ’kan’, my guess is that spaces act as commas.
I’ve just looked at the code and it would be easy to modify it so that only commas are used as separators. The current mechanism is a little … strange. It “explodes” the tag string first by comma and then by space-and-comma and looks to see if it got more entries by the second method than the first.
Anyway, it would be easy enough to remove that. It wouldn’t back-date, though.
I’d be happy either way, as long as the instructions are consistent with the behavior of the code. The label “Tags (Comma Separated)” definitely suggests to my mind that only commas separate tags.
The ability to use commas within tags would also occasionally be useful, as the extant tags “(infinity” and “1)-category” attest. (-: Obviously we have to use something as a separator; but one might argue that a space is more intuitive than a comma.
Just an observation: I haven’t been using tags and don’t quite see myself starting to use them in the future. It takes yet another few of those seconds that add.
I had always thought – before Zoran told me otherwise – that the “tag cloud” on the left of the nForum is automatically generated by from an automatic search for keywords. That would appear to be useful to me. But I can see that it’s muh more work to implement, unless this functionality already exists somwehere somehow.
Whatever separator is chosen, it would be nice to be able to override it, say by putting quotation marks around a tag not to be broken (as one would in a Google search). So long as nobody wants to put a quotation mark in a tag ….
I find user/tags much more useful than automatically generated. They refer to central topics from the point of view of an intelligent being and not the words which happen to be many times repeated or accidentaly, though are not central. For example, word morphism appears many times in many entries, but is a relevant tag ONLY when we have an entry about some properties/classes of morphisms and alike. for general search one has google. Sites which have an intelligent tagging system have additional functionality. It is often that the topic is hard to ind by used words and the topic works better. Thus many forums have it. I find that many nForum conversations which I remember are hard (or practically impossible) to find as they are not properly classified.
In MathOverfow, one of the features which I use is that all the questions which have tag either algebraic geometry or category theory are automatically highlighted in yellow (in the list), the feature made possible because of classification.
I am sorry that Urs think he would be loosing seconds by putting a tag. In fact at least I have wasted his time more than once by asking him a question which I knew existed in his earlier nForum posts which I could not find, and I am sure others were in that situation as well. So the repetition of explanations because of the organizational issue. If we have tags also in future we can put links from the Lab by keyword (as well as we should have backlinks from Lab to the entries mentioning some Lab entry of its alias, but the latter should be generated by automatic search).
Well, we all know what kind of crap one gets if putting in google word like “category” which have nothing to do with math.
Finally there are people who do not look for a particular known discussion but are simply curious to read the blog by topics they like. I do that when coming to some other math blog. Such reasons are also why most collective blog sites have tag clouds.
As Urs complains about seconds used to create tags, let me mention that these seconds are lost just by the creator of the thread (of comment #1 of the thread) and the majority of other comments in the thread just benefit from being in a labelled discussion. So it is relatively rare that one is the default creator of tags.
There is a tag called “cohomolohy” linking to four nForum threads. This can be merged with “cohomology” (unless I missed the latest developments in homolohical algebra ;)
I don't know how to edit these. Maybe the original poster can?
Amusingly, ‘cohomology’ is когомологии in Russian.
Zoran, though the plural form is mostly used, e.g. этальные когомологии where in English one would say étale cohomology.
Well, this is fortunately a question that can be resolved decisively by looking at any Russian text about cohomology, for example the very nice
re #10: if you dig out the links to these threads, then I’ll fix it
Urs, click on the ’cohomolohy’ tag in the list: or click here.
1 to 17 of 17