On the panel discussion at Dconf day 3 (page 4)

On Sunday, 10 September 2023 at 02:12:17 UTC, Adam Wilson wrote: > For the moment EH is a problem, but it's known problem, and one that was discussed at length at DConf. From what I gathered Walter wants to move in the direction of a Herb Sutter style system where exceptions are returned up the stack silently via a shadow value in the return statement. Would this require explicit propagation or call stack refactoring? That would be bad news.

On Sunday, 10 September 2023 at 17:22:55 UTC, German Diago wrote: > On Sunday, 10 September 2023 at 02:12:17 UTC, Adam Wilson wrote: >> For the moment EH is a problem, but it's known problem, and one that was discussed at length at DConf. From what I gathered Walter wants to move in the direction of a Herb Sutter style system where exceptions are returned up the stack silently via a shadow value in the return statement. > > Would this require explicit propagation or call stack refactoring? > That would be bad news. You can read the C++ proposal for more details: https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p0709r4.pdf

September 11

Re: On the panel discussion at Dconf day 3

Posted by Adam Wilson
in reply to claptrap

Permalink

Adam Wilson

Posted in reply to claptrap

Permalink

On Sunday, 10 September 2023 at 10:32:52 UTC, claptrap wrote:

They could have just spent the money on youtube ads, reached 10x as many people and targeted people who actually might be interested in it.

Possible but unlikely TBH. Ads are a terrible tool for this. The better argument would be "spend the money to get a name like Scott Meyers".

Tbh the presentation seemed lacking in substance to me. "The truth about D" was not an accurate name for the talk and there was no real explanation of how IVY would help D. Thats actually what people wanted to know, how is IVY going to help D.

I think most people are still non the wiser on that point.

IMO, the purpose of the presentation was to set-up the Hackathon day sessions, they made that pretty clear IRL, don't know if it came through on stream.

The titling and all that, it's marketing, the "D" part would be swapped out were it for any other language and the talk kept mostly the same. That's how these talks work. I've sat through many of them at bigger conferences. It's just part of the game. Getting bent out of shape over it is pointless.

When I saw the talk on the DConf schedule I briefly googled Ucora/IVY, read about it for five minutes, figured out what it was, and having correctly deduced what it was, I opened my laptop during the talk and was hammering away on ODBC stuff for the duration. I completely tuned it out until Mr. Colvin's most excellent question. But that is a totally different topic.

And frankly, it's not the only talk I tuned out, some of the talks held no interest to me, others I found really quite engaging. Mike Shah's students for example. Other folks have different interests, that's the nature of one-track conferences, so be prepared with something else to do, and be kind to others by sitting in the back. :)

On 9/10/2023 8:35 PM, Adam Wilson wrote: > And frankly, it's not the only talk I tuned out, some of the talks held no interest to me, others I found really quite engaging. Mike Shah's students for example. Other folks have different interests, that's the nature of one-track conferences, so be prepared with something else to do, and be kind to others by sitting in the back. :) The good stuff happens between the talks. The talks are meant to inspire conversation. Personally, I had a fine time meeting new members of the D community and reconnecting with existing ones. I enjoyed giving my talk, too, and hope people found a few nuggets in it they could apply to their own work.

On 22/03/2024 9:58 PM, surfcrone wrote: > IVY = bureaucracy + religion, recipe for disaster, and we are already seeing the effects: the language is now in a deep freeze state What are you talking about? We've just come out of a year long focus on maintenance. DIP queue is open. Named arguments has just been completed. String interpolation merged. UAX31 character ranges has just been merged (two more PR's to go for full UAX31 identifiers).

On Friday, 22 March 2024 at 08:58:14 UTC, surfcrone wrote: > IVY = bureaucracy + religion, recipe for disaster, and we are already seeing the effects: the language is now in a deep freeze state Hah! I can see why you think that to be sure. I just feel differently. Sometimes you have to slow down to speed up. By stabilizing the language, all the things D is missing will start coming into play like libraries, IDE support and so forth. A vast majority of tools and libraries built with D will not compile today so are unavailable. Imagine if you built a tool and it worked forever, like in Java or C#. Instead of constantly needing to rebuild old tools and libraries, you could work on new ones. That's what problem they are solving. Additionally, the editor issue is getting help with the language server and other things. I think if editions are done right, you will see D in the top 20 tiobe index finally and that alone will give it a bump in popularity. I even think OpenD will be a good thing for D. Mainline D will continue to attract companies and people who want stability. OpenD seems to be moving towards a more modern entry-level approach with a single tool to do everything and including tons of libraries from the get-go. The cross pollination is going to be hard to compete with. Adam Ruppe is one of the best developers I know and the Mainline D language has some of the brightest minds in language development. In short, I have never been more excited about D!

On Saturday, 23 March 2024 at 03:22:49 UTC, harakim wrote: > Sometimes you have to slow down to speed up. By stabilizing the language, all the things D is missing will start coming into play like libraries, IDE support and so forth. +1 > A vast majority of tools and libraries built with D will not compile today so are unavailable. Imagine if you built a tool and it worked forever, like in Java or C#. Instead of constantly needing to rebuild old tools and libraries, you could work on new ones. That's what problem they are solving. +1 > I even think OpenD will be a good thing for D. Mainline D will continue to attract companies and people who want stability. OpenD seems to be moving towards a more modern entry-level approach with a single tool to do everything and including tons of libraries from the get-go. +1 > The cross pollination is going to be hard to compete with. +1

On Sunday, 10 September 2023 at 02:12:17 UTC, Adam Wilson wrote: > On Saturday, 9 September 2023 at 19:06:24 UTC, ryuukk_ wrote: >> [...] > > For the moment EH is a problem, but it's known problem, and one that was discussed at length at DConf. From what I gathered Walter wants to move in the direction of a Herb Sutter style system where exceptions are returned up the stack silently via a shadow value in the return statement. It's actually a pretty elegant solution. It's not here now because of the pause, but I also think that this kind of exception handling would be more efficient. When can we expect to see any actual progress on this feature?

Forums