wekempf

.NET Developer

[wekempf @wekempf]

@aiDidThisToo @unclebobmartin Pure ignorance. In no way was it a “copy paste” of the jvm. Ignoring the Java history that led to the creation of C# is far less egregious than this statement

[wekempf @wekempf]

@LumiereEtoile @polydao No, and easily provable. Learn to debug your conversations.

[wekempf @wekempf]

@Akiessling4 @VigilantFox "They are against genocide."

"From the river to the sea."

Who's intellectually dishonest?

[wekempf @wekempf]

@Krazzier1 @FrankDTank3118 @DOGE Because you don't buy VSCode licenses. Even in the case of the C# DevKit if you're large enough you have to buy a VisualStudio license, not a VSCode license.

[wekempf @wekempf]

@nickchapsas Well... It is free, and you can't buy a license for it. What you're referring to is the license for the C# DevKit and how you have to buy a _VisualStudio_ licence to use that extension in some cases.

[wekempf @wekempf]

@davidfowl PowerShell scripts using InvokeBuild and Azure DevOps YAML pipelines. As much of the logic as we can in the PowerShell scripts so that it runs locally and is shifted as far left as we can.

[wekempf @wekempf]

@davidfowl There's some benefit in "using a single language", but honestly you'd be better off using a language catered to the job. I still want a language solution... config files with embedded language extensions are evil.

[wekempf @wekempf]

@ABlekh @davidfowl Both are shells. Sounds like apples to apples to me. "rm -r -f ." The command and argument name lengths are not an issue.

[wekempf @wekempf]

@mussabur @davidfowl That take is insane.

[wekempf @wekempf]

@nastymorbol @okyrylchuk If so, I don't get the reasoning?

[wekempf @wekempf]

@CJavierSaldana @Dave_DotNet I think you, and some enterprise companies, are over simplifying. There's no added stability in starting with an LTS. Yes, any update has risk. That's ANY update. And these "no STS" mandates include new projects as well, which is crazy given the short difference in time.

[wekempf @wekempf]

@KarenPayneMVP Natural types rather than better your inference? I get it, but that's less useful.

[wekempf @wekempf]

@guyrleech @okyrylchuk Too many people here confusing type inference with dynamic typing. Your code is strongly typed when you say "var foo =10;". Foo is not a dynamic value. It's an int. All the same behavior and performance as if you'd written "int foo = 10;"

[wekempf @wekempf]

@RaiuLyn @okyrylchuk No type safety issues here, so why would you be "casting around"?

[wekempf @wekempf]

@nbevans @okyrylchuk A good language gives you a choice, but makes you feel gross when you're being explicit.

[wekempf @wekempf]

@VinceZalamea @okyrylchuk @generic_sf This is why I dislike dropping the async suffix from the name. But the compiler will complain regardless, and the type isn't helping me to understand the code.

[wekempf @wekempf]

@conradonline @okyrylchuk Because types are for the compiler, not the programmer.

[wekempf @wekempf]

@okyrylchuk @nastymorbol You have to specify the type for returns in C#. One of many cases where type inference doesn't work in the language. F# is superior here, and if argue it's NOT superior to use explicit types for return values there.

[wekempf @wekempf]

@VinceZalamea @okyrylchuk That's why I prefer target typed new in many cases. But... this is explicit typing. Not implicit/inferred typing where the type is simply never specified, which is what the op talked about

[wekempf @wekempf]

@paulhuizer @okyrylchuk Why are you thinking so hard about types when you read the code? Do you think so hard when using a dynamically typed language? Types are for the compiler, not me.