Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

Personally I would like to see some ImportC "workflows". How are you supposed to use it and what benefits does it give compared to traditional approaches. Also what's some things you see it being used for in the future.

To me ImportC is highly interesting since I believe (among many things) D will be that kind of language that binds things together. Maybe in some remote future there's even some kind of ImportCpp, who knows? I'm not making any comment on if that would be blessing or a curse though 🤓

On 10/9/22 20:04, Walter Bright via Digitalmars-d wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

The reddit thread of the latest blog post had quite a few "why D when there's Rust et al?" questions. How about expanding on that and giving us an outlook on things and where you want the language to be in the months and years to come?

Many moons ago, Andrei joked that Rust skipped leg day. But say, did they leap us while we were bikeshedding in the forums? Are we still able haul the heavy loads like we claim we can? Is it time for us to adjust our gym routine to counter the back pain we are feeling every now and then?

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

Do you want to talk about something that exists or something that might exist?

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

When is ImportC finished?

I mean this less in a technical sense but more in the big picture sense so that someone knows where things are going and how you're planning to get there. If a company has a large base of existing C code, how easily will they be able to compile it with a D compiler and then put some D functions into the codebase? Particularly if they don't want to add a garbage collector, and if they want others to be able to use their D code from C/C++.

I think it would help to talk about finishing an important addition to the language and showing folks how they'll be able to use it, as opposed to a new idea in an early stage.

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

D 3.0 :p

On 10/10/2022 7:04 AM, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

Why shared library support is not sufficient and what can be done about it.

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

Design errors in programming languages. Generally but also past errors in D and those still present in the language.

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

How would you program an airplane flight control in D, if the compiler was certified for that?

On Sunday, 9 October 2022 at 18:04:36 UTC, Walter Bright wrote:
> Instead of guessing what people would find interesting, what would you all like to hear me talk about at the upcoming D conference?

I think the point about recoverable and non-recoverable errors needs to be repeated again and again. Would be nice to explore the edge-cases and difficult categorization of errors, perhaps there is a kind of underlying formalism.