[02:36:36] Let's take an example of a really simple function: Z16136. What would the expected return value be if I ran Z16136("Thai lunar calendar",8)? [02:37:37] If you make the function properly it should return Month(Calendar(Thai), 9) (re @Toby: Let's take an example of a really simple function: Z16136. What would the expected return value be if I ran Z16136("Thai lunar c...) [02:38:52] Which month is that? [02:39:43] กันยายน (re @Toby: Which month is that?) [02:41:55] Alternatively could be displayed as กย [02:42:56] The only part of this proposal I really disagree with is the creation of 150 persistent objects [02:43:14] But that would happen even if we implemented them all separately [02:49:31] Also Toby can you comment on my [[Wikifunctions:Requests for user groups]] before I close it per the Snowball clause? [02:51:22] It clearly won't pass but I'd like your feedback [04:26:33] That's from the Thai solar calendar. I'm after the Thai *lunar* calendar. (re @Feeglgeef: กันยายน) [04:29:17] กันยายน is a month in the lunar calendar???? [04:30:43] https://en.wikipedia.org/wiki/Thai_lunar_calendar [04:31:15] yes [04:31:27] what i sent lit. means duean 9 [04:38:16] So which index does "month 8/8 athikamat" get? It is also sometimes the "following month" to month 8. My point is, in the Thai lunar calendar, you need to know the year to know what is the actual following month. Unless you just mean the full sequence (which I guess would include athikamat), but then the indexes from duan 9-12 become 10-13, which would be an additional [04:38:16] layer of c [04:38:16] onfusion. So all up, if we are to support this calendar well, it would do better with functions specific to it. [04:39:45] That's a problem for the following month function (re @Toby: So which index does "month 8/8 athikamat" get? It is also sometimes the "following month" to month 8. My point is, in the Thai l...) [04:40:29] Also please do this :) (re @Feeglgeef: Also Toby can you comment on my [[Wikifunctions:Requests for user groups]] before I close it per the Snowball clause?) [04:40:31] Ok, I can't tell that from the Wiki pages, nor did Google translate think so. But I can't read Thai, and that wasn't my point, I just need to know how you propose that your proposal would handle this. (re @Feeglgeef: what i sent lit. means duean 9) [04:40:56] Leave it up to function authors (re @Toby: Ok, I can't tell that from the Wiki pages, nor did Google translate think so. But I can't read Thai, and that wasn't my point, I...) [04:41:19] Maybe we should have a different version for each calendar [04:41:25] And that's fine [04:50:39] When we build the Type you build the indexing. Which indexing do you choose? But no matter what your answer, as a function author of functions very similar to this, it would be much more work to have to deal with all these calendars that I don't understand. And if I'm ultimately writing a function specific to the Gregorian calendar, then why can't I have a type that [04:50:39] will have th [04:50:40] e nice 12-month dropdown, rather than having to choose from every calendar system's months? [04:51:38] So select the Gregorian calendar in the function definition? (re @Toby: When we build the Type you build the indexing. Which indexing do you choose? But no matter what your answer, as a function auth...) [04:51:43] This is not complex shit [04:54:29] It literally took centuries to come up with the calendar rules in the calendar I use. I have no reason to think the others less complex. (re @Feeglgeef: This is not complex shit) [04:55:25] But they all follow the same format (re @Toby: It literally took centuries to come up with the calendar rules in the calendar I use. I have no reason to think the others less ...) [04:55:32] We should take advantage of that [04:55:51] See [[WF:BROAD]] [05:34:00] Done. I hope that helps. (re @Feeglgeef: Also please do this :))