And most of the feedback in Aras' post is vendor oriented feedback. The games industry could also spend some money and time working on their own tooling. Clang is open source, you could land papers and features if you actually wanted to.
-
-
We have spent enormous amounts of time and money on tooling. Every studio spends some amount of money reinventing reflection/serialization stuff because we desperately need that metadata. Some systems are cheap macros, some are fully custom builds of Clang.
1 reply 0 retweets 6 likes -
Joshua T. Fisher Retweeted
Right, and people have been working on getting you that metadata for years in the Reflection TS that will do pretty much everything you want. Have you been paying attention to it to give feedback? Christer implies that work isn't software engineering. https://twitter.com/ChristerEricson/status/1154597139856551937 …
Joshua T. Fisher added,
This Tweet is unavailable.4 replies 0 retweets 2 likes -
Replying to @Timewatcher @MikeNicolella and
My implication was that real problems (like modules) have been ignored for 20-30 years (and meanwhile the problem has gotten harder so we get a weak module concept), while 'enthusiasts' run the show, adding irrelevant fluff (e.g. <=>) that adds no practical sweng value.
5 replies 0 retweets 16 likes -
Replying to @ChristerEricson @MikeNicolella and
The context of your tweet did not read that way, to me at least. Still not everyone works on every feature. Do you think piling the people from all these library features onto modules would make it work?
1 reply 0 retweets 0 likes -
Replying to @Timewatcher @ChristerEricson and
The problem is that C++ basically has no direction, and that is probably an intrinsic consequence of (design by committee) + (historical baggage of how C++ evolved). If C++ had reasonable priorities, it would be in a way better shape today.
3 replies 0 retweets 4 likes -
Replying to @hugoamnov @Timewatcher and
I believe if a game dev powerhouse with a strong technical head had absorbed (C++ standard + compiler) as its core tool and evolved it in a tight R&D loop to fit its imperatives, C++ would be much saner and efficient lang than what C++ committee has produced in the past 20 years.
2 replies 0 retweets 1 like -
Replying to @hugoamnov @Timewatcher and
The result would have been a language for developing games. C++ is not that language. It's a language for games, finance, audio, telecommunication, systems programming, embedded, automotive, aerospace, science, machine learning, server infrastructure...
1 reply 0 retweets 1 like -
Replying to @timur_audio @Timewatcher and
There's not a single characteristic of what game developers require from a language (C++) that wouldn't benefit every single one of those industries, and in a scale that we can't possibly even imagine yet because we just aren't there yet.
1 reply 0 retweets 0 likes -
Replying to @hugoamnov @timur_audio and
There often seems to be a tension, in practice, between popularity and language tightness - Bjarne's quote rings true. There's no shortage of C++ replacements around that major studios could have put money and effort behind, but they didn't. Perhaps they're accepting a tradeoff.
4 replies 0 retweets 0 likes
Arguments like this are true until they aren't. You can't extrapolate them to infinity. On my part, I made this tradeoff for a long time, but the cost became too high to continue paying. Expect more people to make that decision over time.
-
-
Replying to @Jonathan_Blow @hugoamnov and
Big fan of your work. As a long time D user I'm completely with you - it'll be fascinating to see which languages get usage, and where.
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.