I so can't wait for Hylo 1.0 to come out. Tremendously excited by the idea of a mutable value semantics "all the way down" language.

Why C++ doesn't have built-in support of JSON in 2024?

Dave Abrahams with a "live Sean Parent" background for his zoom calls

It looks like someone wants to force a core dump in a "portable" way, i.e. both on linux or windows using x32/x64 machines. I can't speculate on microcontrollers and such. I generally feel like its something people who learned C++ in the 90's (and still treat it like "C/C++" rather than two languages) would write, but there could be domains that still use it commonly beyond just being left in ancient library code.

what lib are u using for the ascii animations

I really hope this makes the '26 standard.

based

Could you please provide me a link to the top10 repository? (the one with the top programming challenges) I could not find it.

This is why C++ is the worst language ever. An abomination.

okay, i think that *(char*)0 = 0; means: cast "0" to a character pointer and "unpointer" it ( i think it is called dereference ) and then assign 0 to it. so to the character '0' assign 0

You need the thing that closes each function down all neatly by clicking on the greater than symbol like in dark basic & Java. Pull out the MASM64 debugger with the void object.

The gist of the talk: not gonna tell you how, join IMC and learn, we are hiring. Here, I saved you half an hour of your life.

Your just saying, is this capable of that? So i can say, i am all capable, VOID OBJECT. Then you cannot have all capability, so you must write every single possible operation out by hand. So you cannot do this, now we must have a big bang to fill up every possible state, no stone left unturned. So we cannot do this, but people spend money, close enough for bills. In our next study; *_use every single register at one time to pivot table array the cache registers_*

I'm not seeing the advantage. Why not instead trigger an event when an agent takes damage and update only the damaged agent? No need for iteration of all agents.

How does c++ *still* not have reflection

Actually I don't know if it's about the difficulty of this presentation or just is my adult ADHD that is to blame for the distortion in my mind, Thanx for the effort anyways

Great talk. Anyhow I still dont understant when structure binding discards const? const auto& [a, b, c] = tuple<int&, float&&, int> creates a confusion because it seems only c is const. P.S. I got it. Structure bidings in this case will work exactly like tuple elements works.

Можешь оставить ссылку на код // Can you give github repository

13:53, but why not use binary protocols like protobuf, if we are talking about the footprint of a message?

Literally the best video I've seen on this topic. Thankyou

This is good stuff. Thank you.

You can call it "Being Sold Out" however Linux, GCC, GNU and Unix all agreed to allow the Microsoft Team to come in to bring both compilers up-to-par into a multi platform conglomerate which will eventually eliminate the need for MingGW or MSYST. They made this agreement in 2014 and the trade off was for Microsoft to continue allowing accessibility for Microsoft's Gaming system, their win32 Lib and Sound and Media Codices for everyone. Microsoft officially calls this "Going Open Source" with their own compiler software. I call this "Being Blackmailed". It's already 2024 and Linux hasn't kept their agreed provision yet. And now Microsoft has added the entire Linux Build System, GNU and the GCC Compiler Libs into it's own massive Visual Studio Application. You can actually build an entire Linux distro or Linux applications in VS. And many are doing so. CPython is one of those teams using VS for every platform. Nice video but honestly everyone doing these videos is out on some Maximum Overdrive Mind Trip to out do the other guy that they are unknowingly leaving the normal everyday C++ programmer in the shadows. Mix things up a bit and do some videos on Lexing and Parsing and developing Compilers, that's where everyone is right now. Lol If you guys would post a single Video on only Lexing you would get the most views.

I love that this guy taught me so much about c++ and that I could probably teach him kinda of the same thing on template and compile time c++ xD

TL;DR try this on your modern Linux machine, you will get a SIGSEGV fault with core dump info

I think that code by it self, its ok. someone is zeroing the first byte of the data segment in some segmented arch where in some platform that uses int like pointers (no hidden info in bits) who said that this code is run on common architecture/platform?

On my first look I thought that it's not C++, but C. Based on that, my serious guess about why anyone ever would write this code would be that it's for some very specific compiler/device where it would do... something. Maybe cause a very specific error, in order to see that the device is actually loading this main. Maybe the device has some hardware trouble. Otherwise I'd say it's a badly made disgruntled employee bomb. #ifndef _DEBUG #define if(x) if(rand()<0.0001 && (x)) #endif Would be way more fun.

This video is a gem!!!

"Segmentation fault" if someone asked from personal experience

It would be nice if he wasn't constantly dragging the talk to a halt to awkwardly insult imaginary people.

We had an opensource project where people kept asking the same question and wouldn't read the simple "getting started" guide and look through the config file. We put 2 entries in the config file crash1=true and crash2=true at the bottom. If those were set we would essentially run code kind of like this. Just crash. The next time someone came in to our IRC (ye.. this was a long time ago..) with "its crashing on startup!" we could just tell them to RTFM.

It was so dense. You should read a system programming book and take some operating system course to understand what he said.

source code?

If I got this question my reply would be "John Carmack's hellow world?"

SIGSEGV Signal... segmentation?... ... V????

Thank you very much! Really helpful.

another great example why not to use singletons

Mason... is that Mason Murner of M++Meekly fame?

beautiful slides!

I feel like I only understood maybe half of this, damn

I don’t know if this is addressed but there is a glaring problem (at least in some domains) with the cursor concept as implemented here: In the example, what if cursor<bool> binded to the gas stove knob …? The main gripe is that cursors can’t be that generic or we don’t get compile time type safety. You have to limit that interface ( with concepts, for example ). I mean it’s obvious here because of the simplicty of the example, but in a ‘real codebase’ with a lot of domain types and operations on them , type safety makes the cursors unwieldly unless concepts are used (and who has a concepts capable compiler in production …)

I'm so looking forward to std simd.

I'm not that much into low-level stuff, but the talk was very insightful and even entertaining.

I can't speak to the mind of the programmer, but I _can_ tell you what it actually does: *movb $0, 0* [writes zero to memory address zero.] There are plenty of embedded systems where this is a perfectly valid thing to do. (on any modern, high level system (i.e. PC), it. is. not.)

Can someone explain how this was anything other than a waste of 50 minutes? Could he not have made that same point _equally_ effectively in, say, 4 minutes? I am feeling frustrated after having watched the whole thing, so maybe this is misdirected angst, but it seems to me that this whole talk was him explaining that this makes a good interview question for someone in this very specific field of computer science because the interviewee's answer demonstrates how well they know what they're talking about. If I had been attending this conference, I can't help but think that I would have been _very_ disappointed if I attended this in person.

31:00 the fact that clang-format implements some of the rules of clang-tidy just without the need for the full AST, I believe by definition makes it justifiable to be consider as a code analysis tool (of sorts), combine that with its ability to add/remove braces, rearrange includes and imports, rearrange const east/west, remove extraneous semi colons, gives it the capabilities to ensure and enforce code follow a user specific style guide. Ironically becuase of the speed due to the lack of a need for a real parser, most people run clang-format all the time, on commits, and on even on save, many of these other tools listed here are run once in a while often requiring hours of processing and post analysis for the debunking of false positives. When it comes to code conformance I believe clang-format has done more than most tool to ensure consistency.

29:20 you say that clang-format is breaking the compatibility between the settings, this is incorrect, the clang-format team does NOT break the configuration between releases, most options switch from being a "bool" -> "enumeration" -> "structure" during their lifecycle as the complexity of what is being introduced increases (people asking for more knobs and whistles), BUT we DO NOT break compatibility, we continue to provide mappings from the old to the new settings allowing you to stick with the old (if you must), it doesn't break the compatibility, we also don't require people to use the latest version but they cannot have new configuration but use an old binary, for that they must upgrade (and why wouldn't they? if its in their configuration). Some people have expressed that from version to version they can get small changes in Formatting, This often is not so much changes but corrections often caused by improvements to the clang-format parser which more correctly identifies the token real use. It would be nice if you give this presentation again that you correct this mistake.

I like how, after 50 minutes, he goes, "That's the basics; let's talk about some details." Anyway, it was an excellent presentation!

What a useless talk.

put the byte 0 in the address 0

This is disgusting. What the hell are the standard devs doing?