#dev 2023-08-28

2023-08-28 UTC
Loqi___ and [tantek] joined the channel
#
vikanezrimaya
in relation to keybase, I remember my reddit proof breaking as soon as the snoosite shut down the API
#
vikanezrimaya
keybase seems to be almost unmaintained and I wouldn't consider depending on it in prod
bterry, gxt and superkuh joined the channel
#
sknebel
The key case people got aquihired by Zoom
#
sknebel
*keybase, stupid autocarrot
superkuh, btrem, jeremy, AramZS and sp1ff joined the channel
#
Loqi
[preview] [steve-bate] activitypub-testsuite: Server-independent, full-automated test suite primary focused on ActivityPub server compliance testing.
[snarfed] joined the channel
#
[snarfed]
is this different than the revived activitypub.rocks one he made?
#
[jacky]
I _believe_ so
#
[snarfed]
yeah the revived one wasn't literally served on the same hosts
#
[snarfed]
and yes this one does look new
#
[snarfed]
ah there's more discussion in #fediverse-devs:http://matrix.org
#
[jacky]
so many apps
gRegor, gxt and btrem joined the channel
#
[tantek]
so many channels
[dave] joined the channel
#
[tantek]
[snarfed] the interesting distinction between silos/POSSE and mastodon/fedi options is that there's a lot more confidence in depending on open standards for the various masto/fedi options IMO
#
[snarfed]
elaborate on "confidence" ?
#
[tantek]
whereas all the silo POSSE approaches IIRC required proprietary APIs based on getting dev/app permissions etc.
#
[snarfed]
a minority, yes
#
[tantek]
which could change at the whim of the silo at any moment
#
[snarfed]
on the other hand, interop is way harder and less predictable in the fediverse than in silos
#
[tantek]
hence more confidence that depending on an open standard would "keep working"
#
[snarfed]
different confidences for different problems
#
[tantek]
confidence that what you do get working, keeps working
#
[snarfed]
yup, that's one kind, of a number that apply
#
[tantek]
which I'd assert is the more important one, reducing future uncertainty
#
[snarfed]
eh. subjective. but understood
#
[tantek]
whereas short term challenges in figuring out how to make something work "at all" is more of an immediate / known "problem" rather than a lack of "confidence"
#
Loqi
[preview] [Ryan Barrett] Having spent a fair amount of time developing for all of them, my experience has been that most proprietary social network (“silo”) APIs are on the easy/low effort end of the spectrum, followed closely by the Bluesky API, webmention + microformat...
#
[tantek]
yes, confidence (or lack thereof) is a different measure IMO than "ease of implementing" (or lack thereof)
#
[snarfed]
I'd argue they're sometimes long term, ongoing, and still lack confidence that they'll keep working
#
[snarfed]
every time a new implementation pops up, it's interop testing all over again. and moving goalposts with HTTP sigs versions, etc
#
[snarfed]
both sides have problems. my experience has definitely been that building for the fediverse and similar protocols is very different than silos, but so far _not_ that it's easier. just different
#
[tantek]
that's a good point, even with masto/fedi, since there's plenty un/underspecified, every new implementation is going to be a bit of a testing/guessing game
#
[snarfed]
(both building and maintaining over time)
#
[snarfed]
oversimplification would be that silo APIs are significantly easier to get working, fediverse and other protocols have the _potential_ to be easier to maintain. I haven't seen that yet, but maybe it'll turn out to be true over time
#
[snarfed]
(webmention + mf2 are way easier to get working. not always as easy as the average silo API, but much closer)
sp1ff`, barnaby and bterry joined the channel