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.
Did we ever agree on a standardized way of handling disambiguation entries?
I guess probably we should have disambiguation pages that carry the label
category: disambiguation
I started doing this with
Any opinions? I guess there are many more entry names that would deserve to be disambiguated.
I think we should not create a special category. The word like connection which is generic is a disambiguation page BECAUSE it will be hit by people with various motivations. Who will hit connection (disambiguation) ?? I would never type nor search for such an entry. If the word is suggestive but with multiple meanings than it will be hit and then it may be sometimes wise to create it with list of more specific redirects AND possibly with some terminological discussion. This there is connection on a bundle, Ehresmann connection, affine connection. Thus connection is almost TOC. It will be hit by many or searched by many (unlike connection (disambiguation)). On the other hand, when it is clear what is the preferred notion for nLab then a redirect can be just under literature, or a sentence in idea section. Thus I think I would prefer to have a sentence “For another notion of a monad in projective algebraic geometry see Beilinson monad” within the entry monad than creating a separate disambiguation entry which will be never used. Finally why complicating with more and more conventions, making the newcomers more difficult to enter the esoteric world of nLab. Who wants to remark on disambiguation always can.
Who will hit connection (disambiguation) ??
Nobody directly. But we need to encapsulate the “There are also the following notions by this name…” You don’t want to type that list each time, nor could you. You want to just say “see ’term (disambiguation)’ for more meanings of the term’”.
You don’t want to type that list each time, nor could you.
What is “each time” ?? I can not imagine more than 2-3 times (each of which would anyway require a remark and redirect). There are just two entries for monad which should have a sentence with a link to another version. This is two usages. Any usage elsewhere of word monad will use and point to one or another, no need to confuse people there. “There is also Beilinson monad and monad in nonstandard analysis” is not any shorter phrase than “for more notions see monad (disambiguation)”, on the contrary, the second needs going to the link to get to some content (already the reminder “Beilinson monad” is an information while “see monad (disambiguation)” requires downloading a new page to get to the information that there is an entry Beilinson monad). It is just making it two step process via the intermediate entry. Disambiguation is needed for those who mistakenly get hit the wrong opr two general version of the notion. Hence it should be either a remark or it should be a content of a page with a generic name without glitches and brackets. Like connection is a much better name for a top level group of notions than connection (disambiguation).
If some information can be put in one sentence than I want to see it right there where it is pointed to and not needing to go to another entry for it. We are having more and more links, tables, titles and the content is more and more diluted and sparse distributed with this.
TIm Porter already mentioned adding one more meaning of “monad”. Then it is not inconceivable that somebody feels like adding Leibniz’s orginal notion. And tomorrow somebody invents yet another meaning. In ten years the next one. And so on. I think this should be all kep in one place, otherwise it will get out of control. You’ll be the first not to care about adding cross-links by hand ;-)
5: Even 7-8 monad notions if they become relevant fit into ONE not long sentence which has an advantage of having a real content (and math will anyway go into separate entries), this is less space than a bold large font title (kind of which you promote). Imagine how many times you write big section “Related notions” (which could be written more compactly as a list at the beginining of bibliography). Why you advocate one principle for semi-synonyms or related while another for homonyms ? Related notions can also emerge in 2 and 5 years and via Tim Porter and inventions etc. Why don’t you create a special entry like adjunction (related notions) then ? Don’t you think that the list of related notions as one learns more mathematics also gets out of control ? (I can not imagine that list in average being shorter than the list of mathematical homonyms in fact).
On the other hand we have also “historical notes on…” for deeper comparisons in terminology and history (we have now only two such entries I think). historical note on Grothendieck topology, historical note on characteristic classes.
But let me be clear. I am not against existence of monad (disambiguation). I am just against of setting it as a standard. Most disambiguations are among two notions only (so better list it both times) or are of the form connection which is just OK and whcih can be hit by a newcomer of external search.
For what it’s worth, and without going so far as demanding a standard to be followed, IMO the suggestion made by Urs in #1 makes a lot of sense. I’m certainly accustomed to this convention from Wikipedia.
It does make sense at wikipedia which covers hundreds of area and it is not supposed to be maintained in groups but systematically item by item. Some level of improvization and practicality and consideration of size and information density is more relevant to nLab than absolute systematicity. It is like a difference between a manual and textbook. Textbook is more friendly and manual is impeccable.
1 to 8 of 8