Time to resume getting my blog on.
Partly, this should serve to remind me to keep my journal (private posts on this site) up-to-date.
And will let me do some longer-form writing without filling people’s email inboxes or producing multipage Stack Overflow answers.
So, what’m I doing that’s brought me back out?
C++ Standardisation
As part of my efforts to broaden my interests outside work and sleep, I started attending conferences such as the excellent CppCon in the USA (hosted by Standard C++ Foundation) and this year, ACCU in the UK (hosted confusingly by ACCU). (I also attended GDC twice, and am hoping to attend 4C: St Petersburg this year, but they’re off-topic for this post.)
I realise that this isn’t a long way outside “work and sleep”… but it’s certainly different from “play video games all weekend”.
It seems to me that volunteering to help with things is an excellent way to get involved with the community. So as well as volunteering as a paper-review for the CppCon Program Committee, I went along to the SG14 meeting.
SG14 — formally “Study Group 14: Game Dev & Low Latency” — is part of WG21 — formally “ISO/IEC JTC1 (Joint Technical Committee 1) / SC22 (Subcommittee 22) / WG21 (Working Group 21)” — is the part of the ISO C++ committee focused on Low-latency and other topics for industries which are not well-served by current standardisation efforts. SG14 started from informal discussions at the first CppCon, and has now grown to encompass:
- Game Development
- Financial and (particularly) High-Frequency Trading
- Simulation
- Embedded Devices
Most of the discussion on the mailing list and in the occasional meetings is about various contains achieving low-latency, no-allocation, no-branching, cache-friendliness (or any combination of those); that’s an area I’m not strong on.
Occasionally a topic I am keen on shows up, around language change or adaption. I’ve volunteered to put some time into one such proposal, “Comparing Virtual Functions” which I will talk more about later.
Since ACCU, where I gave a spur-of-the-moment lightning talk about void{} I’ve also started preparing a small paper proposing a language change for C++ outside SG14: “Umambiguous prvalue void”. That’ll also be a separate post or a series, depending on how things go. I put it out into the wild for the first time today.
So, that’s where I am today. There’s a lot more going on, I intend to talk about it here as I go, so if nothing else, I can get the various ideas I have floating around down on “paper”.
