Twitter | Search | |
Rob Dodson
Developer Advocate @ Google.
13,569
Tweets
1,350
Following
20,013
Followers
Tweets
Rob Dodson 6h
Replying to @diervo @KevinJHill
yeah there were mixed feelings on stars. I'm going to see if I can figure out a scorecard using . e.g.
Reply Retweet Like
Rob Dodson 9h
Replying to @morewry @_matt
yeah I'd like to do that. I'm not quite sure how to actually do it though. I could also write a specification for each test and just trust folks to implement them the right way using the vanilla custom elements I provide.
Reply Retweet Like
Rob Dodson 9h
Replying to @morewry
yeah it feels like it needs to be a combo of things. Maybe an aggregate score on something like as suggested by other folks.
Reply Retweet Like
Rob Dodson 10h
Replying to @saltnburnem
that's awesome dude, thanks for spreading the word!
Reply Retweet Like
Rob Dodson 10h
Replying to @0xcda7a @justinfagnani
[assigns all new tests to Chris and Justin]
Reply Retweet Like
Rob Dodson 10h
Replying to @saltnburnem
the Chris DeMars bump
Reply Retweet Like
Rob Dodson 10h
Replying to @KevinJHill
Is there an example of a framework that is huge in the enterprise but has <500 GitHub stars? Genuinely curious
Reply Retweet Like
Rob Dodson retweeted
Paul Kinlan 13h
Hello Delhi.
Reply Retweet Like
Rob Dodson 12h
Replying to @agubler_
I should have mentioned that any libraries already on the site would remain. I definitely don't want to throw away work that folks have done. This policy would be directed at adding any new libraries.
Reply Retweet Like
Rob Dodson 12h
Replying to @_matt
Maybe the site could hold tests for X number of popular libraries and then provide a way for other folks to maintain a suite themselves and send the scores back. I'm still not sure how to do that, but I want to believe it's possible :D
Reply Retweet Like
Rob Dodson 12h
Replying to @_matt
yeah I think something along those lines could be possible. For instance, Ember has such a unique build that I can't add it to the site the way it's currently designed so we were already talking about giving them a way to . run the tests and phone back the results.
Reply Retweet Like
Rob Dodson 12h
Replying to @OH_NOES
a part of me hopes it's Facebook A/B testing it or something 🙏
Reply Retweet Like
Rob Dodson 12h
Replying to @_matt
There was some discussion about this a while back But I'm not sure how to do it in practice.
Reply Retweet Like
Rob Dodson 13h
ok which one of you downloaded the focus-visible polyfill 300,000+ times a day for a week.
Reply Retweet Like
Rob Dodson 13h
Replying to @edwinmdev
It may be possible to infer from npm stats:
Reply Retweet Like
Rob Dodson 13h
Replying to @agubler_
There's another aspect which is that when I want to add new tests to the site, I have to do so for every library. As that number grows unbounded it makes it harder to do. I think I'd need to draw a line somewhere to manage the scope.
Reply Retweet Like
Rob Dodson 13h
Unfortunately I've seen 1.0 libraries break. It depends how closely the owner is following semver and what tests they already have in place.
Reply Retweet Like
Rob Dodson 13h
Replying to @justinfagnani @0xcda7a
Updates are all automated by renovate and a bot I built. But I also want to add new tests which gets harder as the number of libraries grows. And is it worth it if a library is not actually used much in the real world?
Reply Retweet Like
Rob Dodson 14h
Replying to @nonswearyphil
I'm mostly looking for a combination of API stability and real world usage. Trying to see if maybe can give me enough ranking factors.
Reply Retweet Like
Rob Dodson 14h
Replying to @edwinmdev
The site doesn't test custom elements but instead tests that frameworks are able to communicate with custom elements. But I get your point, having a bar means that only popular libraries are displayed. That's why I'm curious about finding what feels like an attainable bar.
Reply Retweet Like