It's for you!

https://i.imgur.com/NNgrSyP.png

On Sunday, 12 November 2017 at 08:59:05 UTC, Satoshi wrote:
> It's for you!
>
> https://i.imgur.com/NNgrSyP.png

If you're actually taking bets on that...then put me down for $10_000.00

on the MSFT fanbois that is ;-)


Nice stuff with Trinix. I'll cross you off my list of fanboys...for now.

On Sunday, 12 November 2017 at 04:40:21 UTC, codephantom wrote:
> On Sunday, 12 November 2017 at 01:00:46 UTC, bauss wrote:
>> I'm tired so I will end my post here.
>
> And I'm going to end all my posts here, cause I'm sick of arguing with MSFT fanboys, who want to restrain D's development by tying it into propriatery, closed source, bloatware.
>
> Besides being more productive, it also seems like more fun, than responding to MSFT fanboys (although that's been fun too).
>

I told you once and I'll tell you twice.

I'm definitely not a MSFT fan boy.

If I were one, I wouldn't have written a whole framework to replace my ASP.NET projects with D.

I would probably have .NET languges installed too, which I don't.

The only thing I have in my development environment that's related to MS is their linker from Visual Studio, but I don't have anything else installed from VS.

For the simply fact, I don't write any .NET code privately. I have some projects I still maintain, but I can do that without VS.

All projects I have that I currently work on are written in D or C.

-----

On Sunday, 12 November 2017 at 08:59:05 UTC, Satoshi wrote:
> It's for you!
>
> https://i.imgur.com/NNgrSyP.png

Laughed really hard.

On Sunday, 12 November 2017 at 16:47:02 UTC, bauss wrote:
>
> I told you once and I'll tell you twice.
>
> I'm definitely not a MSFT fan boy.
>

Well, you were pretty quick to jump into the middle of a conversation, just to have a long..drawn out....go at me, because I had some critical comments to say about C#  - a language which I know all to well.

Only a MSFT fanboy would defend C# with that level of passion.

What else am I meant to assume?

On Sunday, 12 November 2017 at 16:47:02 UTC, bauss wrote:
>
> I told you once and I'll tell you twice.
>
> I'm definitely not a MSFT fan boy.
>
> The only thing I have in my development environment that's related to MS is their linker from Visual Studio, but I don't have anything else installed from VS.
>
> All projects I have that I currently work on are written in D or C.

ok..I'll take you off the list too..for now.

I think maybe I need to get a cute gravatar, like you and Satoshi.

Maybe a little kitten or something.

Then maybe people will be less eager to attack me....

On Monday, 6 November 2017 at 13:02:43 UTC, Ola Fosheim Grøstad wrote:
> On Monday, 6 November 2017 at 10:12:11 UTC, Jonathan M Davis wrote:
>> All it does is take the expression
>>
>> x ? x : y
>>
>> and make it
>>
>> x ?: y
>
> Yes, that is an issue because it means that typos no longer are caught. E.g. if you accidentally comment out or delete the second expression.

The commenting out case can be prevented by making ?: an actual operator, so ?/**/: would be an error. (Also I think it's regressive to argue invalid code becoming valid is a good reason to prevent introducing a feature).

>
> Which is why I think ?? or some other choice would offer better usability.

I think ??= reads better than ?:=, which looks like Pascal assignment.

On Saturday, 11 November 2017 at 05:20:39 UTC, codephantom wrote:
> I just saw this about the new 'damnit' operator, for C# 8.
>
> https://github.com/dotnet/csharplang/issues/556

The principle is a good one - by default you cannot dereference something that can be null, you get a compiler error instead. If you are confident it isn't null, you use a special operator to override the compiler check. This is better because:

1. The programmer has to acknowledge that the reference is nullable (except where the compiler may be able to prove it is not null).
2. People reading the code are informed that potentially the reference is null but the programmer thought it wouldn't be, in this particular case, documenting the programmer's understanding.
3. Reviewers are freed from checking r's possible assigned value in all code paths every time r is dereferenced.

This solution is probably less disruptive to existing code than removing null altogether like Rust, which encourages the programmer to always handle the not-null and null case for every dereference. (Although half the battle is having non-null types).

On Monday, 13 November 2017 at 14:44:55 UTC, Nick Treleaven wrote:
> The commenting out case can be prevented by making ?: an actual operator, so ?/**/: would be an error.

Yes, that sounds reasonable.

> (Also I think it's regressive to argue invalid code becoming valid is a good reason to prevent introducing a feature).

In terms of usability you want to catch typos so you want some redundancy, but I don't know whether this is a likely typo.

But in general, it is possible to construct a language where all strings are valid programs. That would also mean that all typos would go un-noticed in that language.

On Tuesday, 7 November 2017 at 13:43:20 UTC, user1234 wrote:
> On Monday, 6 November 2017 at 20:14:17 UTC, Meta wrote:
>> The functionality is probably a good idea, but a library solution is doable today without any acrobatics.
>
> Show me a library solution that works fine with IDE completion (so for the safe navigation operator, not the Elvis one).

Well, the IDE can know when a particular standard library symbol is being used. It can detect this case and provide special autocompletion suggestions for it.

Yes, this is not perfect. But where standard library symbols are commonly used, it is a workable solution.