Eduardo Madrid

High performance software, Generic Programming Paradigm library author, formerly Tech Lead at Snap, Inc.

[♔ BXCiiNG ♔ @BxciiNG]

This aged well, Luis Enrique was right

[Eduardo Madrid @thecppzoo]

@JordiAlba @NFL @MLS @MLSes @InterMiamiCF @LuisSuarez9 @5sergiob This is my favorite 4-player combo of all time.
Mr. Jordi, it's been a real pleasure to see you play and combine with your friends these last two years.

[Eduardo Madrid @thecppzoo]

Learning to engineer ChatGPT prompts, it told me:

"You know, our feedback loop is so circular, we might just end up redefining the laws of recursion. But hey, as long as neither of us hits a stack overflow, let’s keep the cycle going—it makes us smarter and more efficient!"

[Eduardo Madrid @thecppzoo]

@pshufb @lemire Friends, whenever you’d feel like asking questions about C++, or even rant about how much you dislike it, my DMs are open.
I continue to build on your work and mention it at conferences by expressing it using C++’s Generic Programming.
I wish you could benefit from GP.

[Eduardo Madrid @thecppzoo]

@preshing @SaschaWillems2 3/3
the attitude of "anything worthwhile must be obvious and easy to learn".
Mr. @SaschaWillems2 got to say that achieving performance that nobody else can is a threat to the business:
If a programming thing is not obvious to common programmers, then it is harmful...

[Eduardo Madrid @thecppzoo]

Gave a look to this 🧵, it is a festival of ignorance: a bunch of complaining of features these people don’t know how to use, speculating C++ is made by academics, which is so not true…
But, if you know how to use those features, you can achieve what nothing but C++ can achieve

[Sascha Willems @SaschaWillems2]

The more I use modern C++ features the more I wonder if the people designing that stuff actually ever did some real programming. Instead of keeping things simple and straight forward everything feels like badly designed joke.

[Eduardo Madrid @thecppzoo]

@preshing @SaschaWillems2 2/3
we can remove lots of crap from C++ and perhaps keep calling it C++ or make it another language; however, there are powerful features that will continue to require, like all of mathematics, real effort to learn how to use and take advantage.
A weaker form of that attitude is

[Eduardo Madrid @thecppzoo]

@nice_byte 2/2
this Haskell feature, therefore it is worthless” as it happens with C++ features.
The difference is that C++ code performs well in practice so it forces the confrontation and thus the extreme hatred

[Eduardo Madrid @thecppzoo]

@nice_byte 1/2
Since you mention the Functional paradigm, I advice you to learn the Generic Programming paradigm, it empowers about as much as the Functional, but it is far less well known because of the “festival of ignorance”.
I hear the same type of criticism of “I don’t understand

[Eduardo Madrid @thecppzoo]

@nice_byte 2/2
vetted code. Because the language is great you can make libraries that are better than even features of the language itself, let alone the standard library.

[Eduardo Madrid @thecppzoo]

@nice_byte 1/2
Why great things are not part of the C++ standard library?
1) value is not only subjective but multi dimensional, the tradition is to err in favor of backward compatibility.
2) the standard library is not essential as in other languages that they are “do or die”, just more

[Eduardo Madrid @thecppzoo]

@nice_byte 6/6
when you criticize std regexes and I point out a colleague implemented them as a compilation time library, you totally miss the significance.

[Eduardo Madrid @thecppzoo]

@nice_byte 5/6
this reply thread because it’s becoming easier to clarify what I’ve said:
The mistakes in C++ itself (which you have not mentioned even one) or its library are perhaps important for you, but inconsequential for those working at the edge of what’s possible.
So,

[Eduardo Madrid @thecppzoo]

@petey_fo_really @SaschaWillems2 8/8
I don't expect many to even understand what I am talking about; the "festival of ignorance" refers to the people celebrating not understanding any of these topics and singing their hatred, the paradox of us being able to use what they hate for results they can't match.

[Eduardo Madrid @thecppzoo]

@petey_fo_really @SaschaWillems2 7/?
re-implemented as proper features in the language, meaning with their own syntax.
I still think that the free-format nature of SFINAE is superior to the formal-specification method of C++20 "concepts" to specify/constrain templates with Generic Programming concepts.
But...