Span hits F#, LinkedIn gets mumbly, and UWP (yes, it's still clinging on) furnished with new toys

Span...

Another workaround attempting to give managed languages the performance of native ones... in a more complex way.

Does UWP have a purpose, yet?

The last time I tried to make a UWP app was long before UWP was UWP. It was back on Windows 8 and, after a brief play, I scrapped the sample and went on with my developer life. The whole experiment can be summed up in one exclamation: "You can't do wot !?!"

So... it's now 2018 and I believe that the sandbox is a little more diverse and that UWP hands developers a little more power but UWP apps are still not true desktop applications and can't do everything. If I was to start a UWP based project, I would have to do two things. Firstly, I'd have to accept that I'd be locked in to UWP and the only way out would be a total rewrite. Secondly, I'd have to completely plan the entire foreseeable life of the project and be absolutely certain that UWP allowed me to implement all the features I would ultimately need.

Sure, I could do these things. Why would I?

Both of those sound like a lot of hard work and a lot of risk and I still don't see the benefits that UWP offers in exchange. It just doesn't offer anything that isn't already available elsewhere -- most of which I have used or experimented with in the past, a lot of which is cross-platform.

Microsoft are asking me to leave a known world of infinite possibility in favour of an unknown sandbox governed by arbitrary decisions. That's like asking someone to give up their somewhat aged but perfectly reliable Skoda in favour of a shiny new toy car. However awesome the toy car may be, it is still a toy!

The only reason I would ever use UWP would be if an employer dictated that I do so and even that seems vanishingly unlikely because UWP is not really suitable for business applications -- most of which now run in the browser, anyway, unless they need very specialised access to local hardware which UWP probably doesn't permit.

Oh... and on the LinkedIn thing: you can now send people a one-line voice clip saying: "You've got mail!". That's totally a use-case!

Anti-F#ism?

a syntax aimed at solving complex analytical problems

I know that these days "solving complex analytical problems" means actually managing to get a meaningful SQL statement together....

Isn't F# just a bit like Haskell, with Microsoftian flavor, i.e. the main feature of it is being strongly typed functional (an idea with us since the 70s, rememberember?)

From Let’s be mainstream! User focused design in Elm (Transcription of a talk by Evan Czaplicki)

The problem with that is that in 1973, ML was introduced. This is one year after C. C was 1972. For some reason, that didn’t work out. Something mysterious happened. So we had another chance in 1990. Standard ML came out. This was actually five years before Java. Java is 1995. We could have gone that route, but for some reason it didn’t happen. In 1996, we had OCaml. This is one year after Java. It has objects. It’s got it all. What’s the problem?

Fluent

Am I the only one who thinks efflouent when I read this ?

"Universal Windows Platform (UWP) apps have a noble goal."

Oh, that's what UWP stands for. Thought it was "Ugly Windows Phone app".

Voice messages on Linked in? Oh dear lord does that mean I'm going to get audio ramblings from recruiters now as well as message spam? This might finally make me delete my LinkedIn account!