Видео

13:00 The black art of code generation by Walter Bright 1:16:50 Redub: Configuring, Building, and Iterating by Marcelo Mancini

On next gen phobos, the sys module could have a similar structure as the Phobos one such that it is built on a core that the rest doesn’t import. For instance, mir has mir-core that mir-algorithm uses. mir-core includes a lot of utility stuff that can be used separately.

To skip the lunch break: 4:39:49

4:44:30 to skip lunch break

This is insanely educative

4:42:00 to skip the lunch break!

This is an interesting talk, thank you. I'd like to have KPIs for my performance measured in IVY in my company... Do what I do at my current workplace align with my broader life vision or not? Right now it's hard to tell, when I'm tasked with delivering software and support, but the goals in my KPIs are my managers KPIs not really connected to my daily work? Edit: The "I don't care" discussion was really surprising, but to the speaker did it justice and answered/guided it in a grown up, respectful and thoughtful manner.

Interesting interview, thank you guys

can someone tell me what DIP means?

Love the guidance to reframe in terms of positives. Not only in code but more importantly, with people. Sage advice!

Unfortunately Andrei won't be present on the conf this autumn

Hey Mike, I've been through all three of these books with various languages (C++ using Qt, Java, Kotlin, C#, and Swift), with interesting (and often surprising) differences in runtimes. There's a great C++ channel (The Cherno) that includes several code reviews for ray tracers, and eventually the channel author created his own series for ray tracing in real time out of frustration (because the algorithm described in the books is quite slow). I've been thinking of converting the algorithm he uses within the framework of the books to see if I can speed it up.

Aw shucks, not a single comment to refute the wild rumour going around the D community, that DConf is moving to New Zealand permanently! Dunno how that rumour got started. Anyway, I do love the idea of the inline assembler being done with CTFE rather than the compiler itself. But that means language hooks to do string manipulation and I know Walter hates that proposal. It's an old idea of mine from like ~10 years ago.

I saw some streams from DConf 2006 (?) and the Don talks propelled the idea of arrays ops which are quite a huge draw for some workload in D! The idea that you didn't need expression templates was novel to me.

I can supply a few more on that first conference. There was indeed lunch, though somewhat meager as I funded it and did all the leg work myself. We had a deli spread on thursday and pizza on friday (or maybe the other way around). A few people opted to hit the cafeteria instead, which was in the same building we were in. Thursday night we walked over to a restaurant for a group dinner. I can't recall the name of the restaurant right now. I recall there being 50ish people (though I wouldn't be shocked if it was a total of 70 registrants), which matches what Andrei and Walter talked about. All in all, for a first try and without any staff support except for the videos recording (which was a bit of a nightmare behind the scenes), it went better than I expected. I really miss that building and those conference rooms. It's up on a hill and has an amazing view to the north of all of downtown Seattle and the Puget Sound and on clear days, the mountains to the east and west. Mt. Rainier is visible from other parts of the building off to the south. I tried to pull together a second year in a row for it, but there just wasn't sufficient interest from the community. Not enough people willing to commit to attend or give talks. Several months later the guys managing the Tango project put together a conference in Europe, but I don't recall any of the details and didn't attend. I didn't try again the year after that. For the England change of scenery / location, wasn't it Adam that asked his employer at the time?

Andrei's back! Great talk folks keep it up. That first dconf sounded awesome.

Andrei is back, that's great!

Good to see Andrei!

I wish they listened seriously to Scott Meyers and what he said...

Thabks dinnis sdjn

excellent talk!

New GC? I'm all for it. I'm using D for my presonal projects and "shell" scripts in Linux already. It's great language and it's sad that more people are not aware of it.

Steven Schveighoffer! One of the greatest helping hands of D community. Glad to hear this Conversation.

But 24:47 auto x4 { 0 }; is int as of C++17. Slide at 28:45 was never shown.

Does DLF mean the D Language Foundation?

Why is micro written in Go and not in D?

If the transition from Bugzilla to GitHub issues relies on an internal undocumented GitHub API, there is a risk that they might change that before you have completed the transition... I know a few things about transitioning a large project from one tech to another, and I very much appreciate the work that you do. So I hope you find the time to complete it soonish 🙂

At 10:24 Mike says he made a software for managing the stock in the shop he worked. It's interesting how the product ended up. I have a similar experience. I started learning programming in 2003 with VB6, but I couldn't pursue it because I didn't have a computer then. Later on 2009 I bought my first PC but at that time I have a crush on Electronics. Later on 2013 I restarted learning programing and this time language was VB .Net. One of my friends worked on an organization and they need a software to manage their registration process. I made a program with Access database and gave my friend. Like Mike's manager, he and his boss told me like, "Oh you are awesome man!". But I am damn sure that they didn't even installed it because, there was no further enquiries like "how to use that, how to use this, how to open the report" etc.

Around 47 minutes, Scott gets to the point. In C++ you are focused on the tool itself, rather than what you can do with the tool. It is a great graphic he uses. Programmers should be focused on their problem, not on the tool. Then he debunks his own work. "182 Guidelines!". I have posted about this needing to write officious 'style' documents for each project telling programmers what to use and not use in C++. With languages designed from the ground up, these things are just in the language without thinking. And, I agree, that other languages should not require this army of consultants going around to support the language. That is another problem with C++ - it is a language that demands the programmer to support C++. It should be the other way around - a programming language should support programming and programmers. And that is not what is meant in another of those trite C/C++ platitude that such languages are 'just training wheels for beginners'. That has always been wrong, but too many programmers, especially young ones have fallen for it. From Scott's statement to the young programmers, they will spend the next 10 years trying to understand this stuff. Ten years to move beyond being a beginner, really?

"Epicycle on epicycle on epicycle" in C++. What I say is it is fixes that fix fixes to fix fixes. It is never ending. Nice work if you can get it, but really this is costing projects and programming a lot of money, but it is all cost that can be hidden.

It is hard to work out whether Scott is being critical of C++ or making excuses for it. Perhaps a telling quote is when he asks for who is a young programmer and says "well, you can spend the next ten years of your life learning this stuff". Why are schools inflicting this on new programmers who need to learn to program, and without having to learn all this weirdness in C++? The talk of zero-runtime overhead to justify unknown values (whether unititialized or somehow zero) is where C/C++ cheat. Other languages will initialise, and most often a value of zero will be correct and what we want. A C/C++ programmer must then include explicit code to say x := 0. Forget it, and you probably have some obscure bug. But the explicit inclusion at source level is probably more wasteful of processor cycles. So the speed advantage is more than lost. So much for the myth of 'zero cost'. It is a nice platitude and people fall for it, but like most platitudes in C and C++ they are trite and wrong and you lose more than you gain. There was much laughter in the room about how these things are obscure and confusing. But in reality there is nothing that funny about it. The excuses for C++ really need to be more realistically recognised as criticisms.

The comment about PowerPC running too hot and doesn't scale well compared to Intel's chips is quite funny. History repeats itself

mathias lang? is it a new programming language?

Was considering using D, but I guess this talk shows that it doesn't seem to have sane defaults. Sad

The Best!

Gtkd is really a handy library. Interesting to see lots of community members using it now or in the past. Could use some gtk4 upgrade.

Getting into it to make a game and then never actually making the game strikes me as so true. I guess I made my little tetris and number sliding and stuff... but I haven't done any serious games since 2007 and still never finished the one i started back in the beginning :(

I love that Dlang gives me at least the same amount of options as C++. 1. Do I want GC? (C++ don't) 2. Do I want "this" part of code to be managed by GC? (C++ don't) 3. Do I want "this" part of code to be implemented in: strucural. object oriented or functional paradigm? 4. Do I want to write fast scripts for Linux (thank you rdmd)? (C++ can't do it) I can choose at every step, what is best fit for my case. I don't have to give up anything, but I can have something or everything I choose to.

Great talks!

Here are the time codes for each segment/talk. I hope this is useful. 11:00 | Intro to DConf Online '24 15:00 | Compiling code is boring and I don’t want to do it by Átila Neves 1:16:20 | Analysis of the Design Space of a Container Library for D (Academic rigor or pedanticism? You decide.) by Robert Schadek 2:16:02 | Tuples in D- by Timon Gehr 3:16:05 | The Case for Graphics Programming in Dlang by Mike Shah 4:16:05 | Drinking the Tears of D's Competitors by Walter Bright (Hopefully this actually works, RUclips is deleting my comments, I don’t know why)

That intro was absoutely amazing. One of the funniest things I've ever seen.

Really enjoyed the conversation, hope to see more in the future!

Loving these series. Mike sounds like those sports commentators. Great voice 😅

Excellent conversation! #dlang

allocators seem to be done right in newer languages like odin and zig. is it really that hard of a problem to solve for the D language?

I was trying to find out if future is bright for d language. I guess it is.

Great stuff guys. Good interview. Quite comprehensive and understandable. Appreciate all the work Martin! :)

Thanks for your hard work on LDC!

45:49 - Although some people believe that the lack of contributors is entirely due to the D community. That's actually not true! Compiler developers, especially around LLVM, are very hard to find (regardless of technical skills, the biggest problem is lack of interest in such complexity). Today's testimonial showed the total dedication and interest of a contributor who has become one of the main maintainers of the ldc project. Thanks, kinke. 👍

you are a legend kinke, thanks for the good work that you do