[14:16:59] <inductiveload>	 what's the best way to get the content model ID for a given content model in a maintenance script?
[14:17:14] <inductiveload>	 look it up from the table, or is there a handy function?
[14:29:24] <Reedy>	 Probably depends what you have currently
[16:19:03] <inductiveload>	 Reedy, I figured it out: NameStoreFactory
[20:22:16] <Krinkle>	 inductiveload: we generally don't proacticely reject visitors without a technologically beneficial reason. We transitioned in ~ 2013 from a UA-based sniff to a capability based feature-test for the optional Grade A JS payload.
[20:22:44] <Krinkle>	 so support generally only moves when there is a new major capability we want to use /and/ require globally.
[20:24:04] <Krinkle>	 having said that, at least in writing ([[mw:Compatibility]]) we could move support up for what is "officialy" tested/prioritised but keep the implemetnation the same, thus silently moving some visitors from Grade A to Grade X despite no apparent difference.
[20:25:29] <Krinkle>	 but in order to stop testing on Safari 9 we'd first have to start testing on it, so there's that :)
[20:33:51] <inductiveload>	 Krinkle, basically, I was wondering about using CSS Grid in something
[20:35:14] <inductiveload>	 to test it, presumably you'd have to find some i-device that hasn't had it's updates and be veeeeery careful not to allow it to update :-D
[20:38:54] <Krinkle>	 inductiveload: individual skins and extension can specify their own support matric. so, depending on what its for you can do as you wish
[20:39:13] <inductiveload>	 sure, I was curious rather than anything else
[20:39:15] <Krinkle>	 If it's for essential page layout in production, then likely no.
[20:39:34] <inductiveload>	 (it's ProofreadPage, which basically doesn't work on mobile anyway)
[20:39:36] <Krinkle>	 anywhere else, or for something that can degrade in a reasonable way, probably fine.
[20:40:39] <inductiveload>	 though I think for the moment it's going to a flexbox rather than grid
[20:41:02] <Krinkle>	 browserstack has macOS desktop vms with Safari 9
[20:41:05] <Krinkle>	 as does saucelabs 
[20:41:22] <Krinkle>	 fully in-browser interactiable and with free limited access
[20:41:33] <Krinkle>	 you can even tunnel your localhost
[20:41:42] <Krinkle>	 although I don't know if that's in the free tier.
[20:41:53] <Krinkle>	 the foundation has an accout with a number of login slots for developers to use.
[20:42:02] <inductiveload>	 (as long as you can get it done within 1 minute ^_^)
[20:42:27] <inductiveload>	 i'm not in a position to actually need to test anything ATM anyway
[20:42:29] <Krinkle>	 saucelabs has unlimited for opensource users.
[20:42:42] <Krinkle>	 via browserstack, oss is approved manually so a bit harder to enter
[20:43:43] <Krinkle>	 saucelabs also offers a higher concurrently for oss, but to get that you have to ask via their support and put hteir logo in your README
[20:43:52] <Krinkle>	 but without that you still get 1 concurrency which is enough for manual testing
[20:44:02] <Krinkle>	 it might time out after 5min inacticity or something like that
[20:44:05] <inductiveload>	 huh good to know
[20:44:50] <Krinkle>	 disclaimer - I use an OSS account for https://github.com/js-reporters/js-reporters and manage it for @jquery and @qunitjs as well
[20:45:00] <Krinkle>	 the latter two use browserstack
[20:46:17] <inductiveload>	 rargh, saucelabs makes you use google sign on if you enter a gmail account -_-
[20:46:38] <Krinkle>	 it's raining personal data
[20:47:33] <inductiveload>	 anyhoo, I have to bounce but thanks for the replies :-)