@Pat TL;DR I have a constitutional right to shoot people and I'm itching for an excuse to do it
the gibberish generator of
the happy c++98er without a supporting library: http://ix.io/3inZ/cpp
the troubled c++17er with a supporting library: http://ix.io/3ivL/cpp
random number generators take seeding object by lvalue reference, brilliant. I get it, the library was written for c++98, but someone fix it already -_-
@valleyforge exactly, kids that are not raised right deserve to die, courts can argue all they want, but we got it figured out. And yes if are charging with a knife that's a gunfight, there is no force other than deadly force.
@valleyforge hell yeah, raise your kids right or we gonna kill em, this is not wild west!
@zpartacoos equally productive? and kdenlive? freakin kdenlive? seriously? It might work for you personally, but you will not stay competitive for long if you plan to run a video production/editing business on it. It still doesn't have support for optical flow ffs. And then you have the audacity(hehe) to tell me I must edit the audio in a separate program? Oh yeah and for optical flow I'll just open up slowmoVideo, such a productivity boost, I love constantly switching bewteen programs that were never meant to work together. At least say cinelerra and would shut up, since I actually know nothing about video editing, and yet still can tell kdenlive sucks.
If you can meaningfully modify it, you shouldn't be modifying it to suit your needs, you should be modifying it to suit the needs of those who can't modify it, otherwise it's not going anywhere. Find a professional video editor, learn their workflow, make kdenlive work for them, make it work for them better than anything else, do it couple of more times, get good at it, then start selling warranty on your custom installs (or more likely encounter an insurmountable obstacle give up and move on).
The only thing free systems have over the propitiatory alternatives today is that if configured properly they are more stable and requires less maintenance. The individual apps might be lacking but the base system is not going to break or slow to a crawl over time for no apparent reason, and its behaviour would usually be predictable and repeatable. It's also auditable. What's any of that good for? Selling warranties on customized installs.
And don't you put VSCodium next to vim I'll cut you.
re: language
@amiloradovsky@functional.cafe @newt From what I've seen the implication is a sort of a narrow-mindedness, considering your expertise and focus the most important of all. It's an easy ad hominem attack, as it's a common knee-jerk reaction. After all if you didn't consider it important why would you become an expert, and if you are focused on one thing, you are likely to disregard others, since... well that's the definition of focus. Both as an argument and an insult it's obviously weak, hence the invention of new words to conceal and confuse.
I also sometimes notice hints of tribalism in certain uses of the vague plural "techbors", I presume because of some economic disparity and common business practises that isolate certain technical experts from general public.
The idea that a general purpose programming language designed for portability and efficiency should map directly to your favourite machine architecture (thus the implementation of it on any other being a virtual machine of said architecture) is both relatable and delusionally egocentric.
what we say to save our skin: a garbage collector is more efficient than smart pointers in real world projects.
what we actually mean: if you want to write spaghetti code, with smart pointers you'll hit a brick wall pretty fast, while a garbage collector will delay this face flattening experience long enough for you to abandon ship.
the absolute state of #programming
also not a fan of these pure code specs in the standard (apply and make_from_tuple), spelling it all out in english they probably would've avoided these weird inconsistencies... reminds me of some video codec specs, where it feels like they just copy pasted the implementation: "there ya go, cryptic c code, that's your spec"...
so whether tuple_cat works with tuple-like types is implementation defined,
apply doesn't work with them period (except standard tuple-likes),
and make_from_tuple apparently is like apply in c++17, but is supposed to work with them in c++20, and yet no implementation complies... I must know who is responsible for this! some pussy ass library vendor no doubt >:(
Consider that everything is hard to navigate with a touch screen.
Consider that, of all text editors, vim is the best text editor for a mobile device.
And finally consider a real keyboard
https://www.ebay.com/itm/Motorola-Photon-Q-4G-LTE-Use-your-own-NANO-Sim-card/174709456691?epid=118366470&hash=item28ad7ebf33:g:x84AAOSwRkNfyD36
@freemo nah, std::search takes care of that, it works with ranges that you specify using begin and end iterators/pointers, a very c++ thing. The split itself works fine and handles all the edge cases, it just has undefined behaviour, which you can't identify in any way other than sheer unbridled pedantry.
old code new question: can you spot the undefined behaviour? omg! and how can I ever recover from this?