Manding languages
How to get it recognized? It is so on fr.wiktionary.org and on en.wikipedia.org. Reason being it is a recognized subgroup of Mande (and indeed the reason for the existence of N'Ko). These languages (or dialects, or perhaps more usefully, varieties) are to varying but significant degreess mutually intelligible.. This means that from a practical point of view on WT, it is useful to be able to access their respective categories from a subcategory of Category:Mande languages.
English Wikipedia says its language code is man, but that appears to refer to a single language on Wiktionary. If you want this changed, you'll have to request it on WT:GP.
You're confusing Manding and Mandingo, it seems. Rua, you don't need to revert someone's good-faith work just because you don't know what's going on. We can just add it as a family.
I'm not going to make any statement regarding the correctness of the family, but the fact that we don't have it yet suggests it needs someone with better knowledge/discussion to look at it. That's why I deleted it. It shouldn't have a category until it has a valid code, obviously.
You're missing the point. You could have started a discussion instead of reverting and deleting something that was correct, just not implemented in the way it needed to be. The fact we didn't have it yet only suggests that not enough people are working on African languages.
That's why I suggested taking the matter to WT:GP, where it could be fixed. You're already here spewing at me before anyone even had a chance to do that.
The code man is a macrolanguage for a subgroup of Manding (a sub-sub-group of Mande, if you will, including several varieties of Mandinka & Maninka). There is no ISO 639 code for the entire group of Manding languages (covering man + others like bm, dyu, and kao'). The lack of a relevant ISO 639 code shouldn't IMO prevent addition of this category (even if that might not fit conveniently with the cat boiler system (as I understand it).
Metaknowledge, thanks. Will follow up on discussion page there.
Rua, wrt valid code, I think there are some instances where departing from that convention would be useful, tho obviously not without sufficient justification for such exceptions. One of those, I would contend, is "Manding" (of which Mandingo "man" is a subset), and this assertion is based not only on linguistics but also the way this set of languages (or varieties of a language) are perceived by at least a large number of its speakers. A topic for further discussion elsewhere, of course, but please undderstand that my initial edits in this regard were not lightly considered.
Everything on Wiktionary relies on thing having a code, whether it's for a language or a family. So there is really no way to depart from that convention without breaking everything, and having a code to use is a requirement for being able to do anything regarding a family or language. For example, the categories for language families are created with {{famcatboiler}}, which requires a code as the first parameter. So without the code, there can't be a category either.
That said, Wiktionary creates its own codes when there isn't one available in ISO 639. So if there is no code available for something, a new one can be created and added to our database. The database can only be edited by administrators, so you can request one at WT:GP and someone will do it for you. The reason I didn't do it myself is, as I mentioned, that I didn't know anything about these languages and I was confused by the fact that Wikipedia said man was the code for this family, while we already use that code for a specific language. Metaknowledge knows more about these languages, so he was able to judge it correct and added a code for you (along with some unnecessarily snide remarks towards me, just for trying to be cautious). In any case, for more information, you may find WT:Languages and WT:Families useful.