Hello kind friends!
Anna the websquid here, with an update on the site. In short, we are aware of the various issues afflicting the site, and I’m working on them as fast as I can. This is a part-time gig on my part, so the amount of time I have to work on issues is limited. But we know the captchas suck, we know about the mobile formatting issues, we know pages are loading slow, we know about the 502s.
The best way to report site issues or complaints: check our issue tracker, and if you don’t see the issue there, you can send an email to email@example.com. (Alternately, if you’re the sort of person who githubs, feel free to open an issue directly on the issue tracker.)
Some technical notes, for the curious:
I’m currently prioritizing the latency/availability issues. Our current webhost is proving to be a bad fit for our needs in a number of ways, chief amongst them the difficulty in setting up meaningful monitoring metrics. As a result, we haven’t even been able to reliably identify where in the stack the latency issues are occurring, or whether the 502s are related to backend server latency. The 502s are being returned by hosting-service-provided Apache reverse proxy frontends, and our application server backend (running on gunicorn) doesn’t report any errors at all, except for the occasional “worker timeout” that doesn’t correlate with the periods in which the 502s spike. It’s a bad situation, and I’m digging us out of it. This will probably result in some brief downtime sometime this weekend while I migrate the site to a new hosting platform.
Once we have the major operational issues under control, I’ll turn my focus onto the most painful user-visible site bugs. They’re all on my radar, I just have limited cycles to spend.