[05:34:55] Why does yours work at Z18501 but mine doesn't at Z18513? (re @Al: There’s a performance hit but it’s not bad. I’m wondering whether/when we should have specific untyping (“de-Typing”) as in test...) [07:38:50] Oh, and my hopes that I could reproduce the success of Z12668 at Z18602 were dashed. [07:39:42] I meant: Z17708 and Z18602 (re @Toby: Oh, and my hopes that I could reproduce the success of Z12668 at Z18602 were dashed.) [07:50:27] See Z18568. There’s some issue with the result validation when using Z17414, probably that the Python implementation is preferred over JavaScript. (re @Toby: Why does yours work at Z18501 but mine doesn't at Z18513?) [08:21:24] I think that’s fixable, I’m just not sure how 😏 (re @Toby: I meant: Z17708 and Z18602) [08:21:59] I can't see anything wrong with the Python implementation though... so it seems like there's a bug here? (re @Al: See Z18568. There’s some issue with the result validation when using Z17414, probably that the Python implementation is preferre...) [08:23:04] Live from Wikimania: there is a lot of mentions of Lexemes and a bit of Wikifunctions during the indigenous languages round table [08:24:57] In that it’s timing out silently, I guess… (re @Toby: I can't see anything wrong with the Python implementation though... so it seems like there's a bug here?) [08:25:24] You're right though, when I disabled python and put in a composition of same Reference on Z17414, this now works: Z18513 (re @Toby: I can't see anything wrong with the Python implementation though... so it seems like there's a bug here?) [08:28:11] Yes I suppose it could time out if it had to run against all 7 days on the list. Same Reference is a good solution - which should work on all enumerated types?? (re @Al: In that it’s timing out silently, I guess…) [08:31:26] That was the plan. I think I implemented on quite a few already. (re @Toby: Yes I suppose it could time out if it had to run against all 7 days on the list. Same Reference is a good solution - which shoul...) [11:08:56] Because of Wikimania, some new people are joining and testing things on Wikifunctions [11:08:57] Don't hesitate to reach out and help them [14:39:30] Z18618 I’m not sure that concatenate Typed lists is worth while. It is a little counter-intuitive that you have to de-Type two Typed lists before concatenation, but the result from concatenate lists will be a properly Typed list when all the elements have the same Type (unless the recursive composition is preferred, which is unlikely now we have a JavaScript [14:39:30] implementation). (r [14:39:30] e @Al: I think that’s fixable, I’m just not sure how 😏) [15:21:13] Even less likely now the composition fails for Z18622. To succeed with lists having the same Type, I guess de-Typing is required at each recursion. I have a more efficient solution in mind, but I’d rather leave the test failing with that implementation (indefinitely). (re @Al: Z18618 I’m not sure that concatenate Typed lists is worth while. It is a little [15:21:13] counter-intuitive th [15:21:13] at you have to de-Type two T...)