Joost de Vries

As an ex-Twitter boss, I have a way to grab Elon Musk’s attention. If he keeps stirring unrest, get an arrest warrant | Bruce Daisley theguardian.com/commentisfree/…

“There is a 50-50 chance AI will get more intelligent than humans in the next 20 years. We’ve never had to deal with things more intelligent than us. And we should be very uncertain about what it will look like.” ~ Geoffrey Hinton

@adamwarski @alexelcu @lukasz_bialy Maybe have a look at Lucas Rytz' phd thesis. It delved into the question of checked exception ergonomics. I remember something like: with checked exceptions the default is 'no exceptions'. But yeah: ergonomics will make or break capability based programming a la future scala

Yes, exactly, with same benefits as `using Async` has over `throws InterruptedException`. Checked exceptions are considered broken, so now comes the question, aren't then capabilities broken the same way? The hypothesis would be that no: checked exceptions are broke because they don't play well with lambdas, and require any callers of lambdas to be aware of any checked exceptions that might be thrown. With capabilities, we don't have this problem. However, that's only one theory of why checked exceptions are bad, there are many others as well ;)

@andrestaltz I've seen Ishikawa diagrams used for this en.wikipedia.org/wiki/Ishikawa_…

I'd like to see a project management system based on a chain of "why"s. That is, instead of just stating tasks to be done, inform the reason for a task, plus the reason for that reason, etc. Might shed a lot of light on what matters and what doesn't.

@alexelcu Lucas Rytz analysed this in his PhD thesis infoscience.epfl.ch/record/196412 He quotes this article and proposes a solution that addresses the unfortunate pragmatics of java checked exceptions. news.ycombinator.com/item?id=101317… Introducing an effect (throwing) and removing (catching)

@headinthebox Maybe human hallucinations are mispredictions as well. So then the term would be fitting after all.

Shouldn't we call "hallucinations" simply "mispredictions". The former implies perception which is not there in a model, whereas the latter precisely describes what is going on mechanically.

@_JamesWard @epascarello Have you considered Deno + Typescript instead? I find it very productive. Works out of the box most often. Also has top level await :-)

I think the biggest source of pain comes from wishy-washy and changing foundations. Example: I wanted top level await, much time on StackOverflow with many out-of-date and incomplete or unhelpful answers, converted to mjs, which broke Jest, back into the StackOverflow quagmire, then had to figure out different exports & imports, again into StackOverflow morass, eventually figured it all out. It is really hard to evolve something as legacy as JS because it isn't just the tech, it's the ecosystem that has to move. Java certainly has some of these same challenges but have tackled them with much more care and with more solid foundations.

Been spending time in the JavaScript / Node pain cave lately to remind myself how good I have it in the JVM ecosystem.

“You need to imagine something that is more intelligent than us by the same degree that we are more intelligent than a frog. It’s all very well to say: ‘Well, don’t connect them to the internet,’ but as long as they’re talking to us, they can make us do things.”

@davefarley77 Language workbenches martinfowler.com/articles/langu…

What tech-related product/service did you think was going to be a hit, but instead failed terribly? 😂

Functional programming is an ideal fit for event logs

Interesting discussion on what Loom will mean for async programming in Scala contributors.scala-lang.org/t/pre-sip-susp…

@urs_peter Akka and kotlin experiment gist.github.com/joost-de-vries…

Occurrent 0.14.0 is released with a new "Domain Query DSL" and a #springboot starter project as well as some other good stuff. 🥳 See changelog here: occurrent.org/news/2021/11/0… #spring #java #kotlin #eventsourcing #occurrent

"Duplication is far cheaper than the wrong abstraction" A classic sandimetz.com/blog/2016/1/20…

The economist on the need for real time data economist.com/leaders/2021/1…

Arrow computation expressions eliminate the need for flatMap style programming. Exciting direction.

@CrowdHailer ...and here's the archive of the write up web.archive.org/web/2017060308…

Please stop polluting our imperative languages with pure concepts. another example of #erlang having good ideas 1st youtube.com/watch?v=449j7o…

Just completed the course. coursera.org/share/d53f634e… It was fun fiddling with Scala 3. Tx Julien