saair

🐘

[saair @msquad]

@_nicojs @mattpocockuk https://t.co/YkAGzLio3Z

[saair @msquad]

@steveruizok Semantically, the ordering of unions isn't meaningful so I wouldn't encourage relying on it, but in practice they do seem to have predictable ordering so you can use some ugly tricks to pull the last one out from a union:

https://t.co/o8dZBvep6m

[saair @msquad]

@steveruizok type Last<Tuple extends ReadonlyArray<unknown>> = Tuple extends readonly [...infer _, infer Last] ? Last : never;
export type LastFruitType = Last<typeof FRUIT_TYPES>;

[Jason Leopold @JasonLeopold]

NEW #FOIA SCOOP/🧵Took me 6 yrs to pry this out of NSA

A few wks before @Snowden's leaks were published, 2 NSA employees contacted NSA IG to blow the whistle on unauthorized surveillance

IG substantiated it

W/@KatrinaManson @WilliamTurton @rj_gallagher
https://t.co/zh8gh2e6H0

[Elisabeth Bik @MicrobiomDigest]

Science Has a Nasty Photoshopping Problem
* Unfortunately, many scientific journals and academic institutions are slow to respond to evidence of image manipulation — if they take action at all.

My opinion piece @nytimes - #MyGiftArticle
https://t.co/VLC0wb2hor

[saair @msquad]

@alexandrosM @majohmo @mikko https://t.co/RgrtCzTweG

[saair @msquad]

@shadowy_cod3r @markdalgleish @dan_abramov No since child2 won’t be rendered to null this point. It is just a jsx element (a special kind of object) that when rendered will render null. ReactDOM.render at the root of the React app basically initiates the rendering for the tree of elements.

[saair @msquad]

@drosenwasser @typescript Are the bulletpoints actually meant to be conjunctive with AND or disjunctive with OR?

[saair @msquad]

@drosenwasser @typescript Thanks for the reply and all the great work on TS! What’s not really clear to me in the PR is the second bulletpoint. In the cases where B DOES use infer, under what conditions is source still assignable to A extends B ? C : D?

[saair @msquad]

@khyperia Ahh, right! I misinterpreted that earlier... so I guess SpiderMonkey is actually doing exactly what it should per spec (the spec is just not particularly developer-friendly here)

[saair @msquad]

@khyperia Would be very surprised if it has something about the start and end being braces though, since I'm sure they are working off of some kind of Abstract Syntax Tree

[saair @msquad]

@khyperia Yeah, I think there just has to be a bug in both V8 and Webkit. I think SpiderMonkey is the most consistent here. Honestly though I wish all the engines would just throw some kind of syntax error when they encounter "clearly ambiguous" syntax like this.

[saair @msquad]

@McCormickWrites Thank you for this beautifully written & moving piece... The senselessness of it all is exasperating, and yet here we are unable to do anything but try to make some sense of it

[saair @msquad]

@stephos13 @TheQuartering not what happened:
https://t.co/YrF1UA3wOR

[saair @msquad]

@el0rede @Nappingisbomb @bernie_3rd @realDonaldTrump Except that is not what happened... the 100% of votes you saw were not given but rather *taken away* from Biden because the wrong number was added by a typo earlier - the images were intentionally reversed by Republican disinformationists
https://t.co/YrF1UA3wOR

[saair @msquad]

@realDailyWire @realDonaldTrump @MattWalshBlog Maybe you at @realDailyWire and @MattWalshBlog should take a stand and stop spreading lies?
https://t.co/YrF1UA3wOR

[saair @msquad]

@dan_abramov This feels icky in part since providers might or might not actually consume from other provider ancestors. The hierarchy is artificial and interdependencies are unclear. Allowing parallel providers would be a cleaner and less ambiguous representation of the mental model.

[saair @msquad]

@spoike @ChromiumDev @Paul_Kinlan It appears in the current Canary build as well (at least when using template literals in JSX)