If you are "TRUSTED BY 4,300+ MARKETERS" and "<200-ish, but changes every refresh> Users Signed Up Today", then you've built a multi-million dollar business in the last 20 days. Congrats...
I’m a customer of his uptime service (https://apex.sh/ping/), and following up framework (https://up.docs.apex.sh/) with interest, but haven’t used it yet. Perhaps he is more focused on career and family, and less on open source? If so, good for him.
>Could this be re-purposed for detecting anomalies/outliers in time series data?
If you define anomaly as something unexpected then yes. In this case, if the reality differs significantly from the forecast (=expectation) then it is an anomaly (according to our definition). In numeric univariate case, there could be positive anomalies where you get more than expected, and negative anomaly where you get less than expected.
My guess would be yes. I'm thinking this could be used to find out how effective a particular marketing campaign was. Just compare the forecast with actuals and the difference would be the number of sales/clicks you got from that campaign.
I work on the project discussed in the article. We're running puma in clustered mode, with multiple workers per dyno and multiple threads per worker. 1GB is plenty in this context - each worker is pretty lightweight. At the time this was written we were running ~10 2x dynos, but we've since switched to 3 performance-m dynos with more puma threads/workers.
> I don't think it's any less crazy than being forced to chase a GC white whale for two weeks on a tiny memory leak to avoid a huge rate hike on your hosting bill.
The main issue we were facing wasn't the hosting bill, but high (for us) traffic, that was leaking memory on every request. Under low traffic it was unnoticeable, but at peak load the leak would cause dyno memory to max out pretty quickly, which would cause timeouts and increase the traffic to other dynos, causing cascading failures. Having more memory available would definitely have made things a lot easier, but we would have run out eventually either way.
The web UI for the Transmission bittorrent client, around 2006-ish. I wrote it in prototype.js and then discovered jQuery, and ended up rewriting the whole thing.
Just a random data point, but one of my apps with a definite non-technical demographic gets 97% of logins via facebook, 2.5% via twitter, and 0.5% via email/password registration. I haven't tested it obviously, but I don't think removing Facebook login would do much for conversions.
