qotta_, qotta and JeanDiscord[m] joined the channel
#@HongPong↩️ its on Laravel and Vue JS which is way closer to what I know in webdev than mastodon, and i have been trying to follow activitypub stuff for quite a while (and webmentions and so on) so, i'm really pulling for it (twitter.com/_/status/1165800934443761665)
[Michael_Beckwit, [tantek], [vendan], ZapierDiscord[m], gRegorLove and [snarfed] joined the channel
#[snarfed]i may have figured out why bridgy instagram broke. someone added a bunch (all?) of their individual IG followees to their reader via granary, so IG fetches spiked
#aaronpkhmm that doesn't track with ownyourgram getting blocked around the same time, curious
#aaronpkI wonder if I suddenly got a bunch more users? That's something I didn't check before
cweiske, [Rose], jeremych_ and [grantcodes] joined the channel
#[grantcodes]Just ran into a issue with the wordpress plugin & experimental micropub queries - and I'm not sure it is particularly a wordpress issue, but perhaps something that hasn't been discussed much yet
#[grantcodes]And most endpoints don't support this, but the wp plugin still returns the list of posts, even though they are not valid for the query
#[grantcodes]So the actual generalized question is should micropub endpoints throw an error when there is a query parameter they don't understand or not?
#[grantcodes]The spec says it should throw a 400 if "The request is missing a required parameter, or there was a problem with a value of one of the parameters" - which doesn't really cover unrecognized parameters
jjuran, [Rose], gRegorLove and vika_nezrimaya joined the channel
#[grantcodes]Yeah, that's fine - I think there are only maybe 2 implementations max (one being me). The issue was I sent that experimental query and got a response with all posts.
#[tantek]are you using a set of rotating / random proxies, or will this just be burn one proxy at a time as they ban IPs?
[KevinMarks] joined the channel
#[tantek]on another topic: I'm working on ICS support for my events through H2VX, and it looks like my hCalendar (backcompat publishing) support is not quite there, so I'm considering how much work to put in backend code vs publishing markup
#[snarfed]IP bans seem to be temporary. at least they were in 2016
#[snarfed]hoping still true, so i can use this just for a week or so and then switch back to direct
#[tantek]and requires the 00 seconds artificial precision
#[tantek]ok I think I'm going to publish floating times for my "local" events (because that's the 99% use-case for me personally), and on the rare occasion I'm posting/hosting an event somewhere else, I'll put the Z time somewhere visible so I can mark it up explicitly and have that work
#[tantek]seriously can't believe we're still dealing with such broken ICS support in 2019. It really is the RSS of calendaring 😛
#[tantek]correction: FeedParser at least has a public test suite for RSS 😆
#aaronpk[tantek]: if i see an "add to calendar" button on your website for an event in san francisco next week, but I'm currently in New York, and I click and add the event, it will add it in my local time, and then it will be in the wrong time next week on my phone
#KartikPrabhuI actually have had that problem with some stuff I added to Google calendar
#LoqiTimezones are names used to highlight regions on the globe that are used to define a uniform time for society's use (legal, finance, social) https://indieweb.org/timezone
#aaronpkpersonal feelings about timezones aren't really relevant to the discussion ;-)
#KartikPrabhuwell I am glad you guys are stuck on Earth and don't have to deal with actual time ;)
#[tantek]aaronpk, sounds like a bug in your calendar reader. it should keep the time floating continuously so when you arrive in the destination time zone it works
#aaronpki guess that's how an alarm on the phone does it
#[tantek]sounds like your calendar reader is asserting an artificial precision of a specific time zone when you add your event, then errantly locking you into that
#[tantek]so it doesn't know to "float it" when you arrive at your destination time zone
#aaronpkthis is actually why i use a paper planner
#Loqi[Aaron Parecki] I haven't found a digital planner that can properly deal with how often I hop timezones. Sometimes you just have to stick to paper. Most planners are so big that I don't want to take them with me, so I'm very happy I found this stick planner in Japan... https://aaronparecki.com/2019/03/22/7/stick-planner.jpg
#aaronpkthat exact issue tho... i want to write "9am-1pm" on my calendar, be able to schedule something else after it at "2pm", and then on that day when i'm in a different timezone, have it still say "9am"
#[tantek]This means there is likely a (small but lucrative) market for a digital calendar/planner that *does* do proper/nice/smart/reliable handling of timezones as a paid service
#aaronpkin any case, afaik all calendaring software will make an assumption about the timezone so you'll have the problem i described if you don't specify
#aaronpkif you want it to actually be useful today
#[tantek]also please correct me if I got anything wrong
#[tantek]aaronpk that's a pretty broad assumption regarding an entire class of user interfaces
#[tantek]unless you've like designed numerous calendaring software UIs yourself, I'd say you lack the experience to make such a grand statement of UX assumption
#[tantek]lots of mediocre examples != impossible to make a good example
#aaronpkdo you want the button to work well for most people? or do you want to try to use it as leverage to get software developers to improve calendaring apps?
#[tantek]it'll work well for most people since most people going to an event are already in that event's timezone
#aaronpkeither is fine :) and I definitely fall in the latter category for certain things myself
#[tantek]and yeah I'm ok with posting public test cases that break calendar software as a way to encourage their devs to fix
#[tantek]is not going to bite at the nerd snipe of designing a new calendaring UI that has a better projection of the 4D space of calendar events into a 2D display.
#[tantek]alright I think I'm going to go with *some* of the latter aaronpk, but not with calendaring apps but rather mf2 parsers. I'm going to be switching my datetimes in my h-events to depend on value class pattern support (which I know many mf2 parsers have still not quite caught up to)
#[tantek]so who knows what will break. will be interesting to hear from anyone consuming h-event dt-start dt-end