3 ms·
I've been out of the game for awhile now, and sure, I find it eminently believable that Rails is much more productive than Node-based things, but has nothing be
by sanderjd 8y ago
I've been out of the game for awhile now, and sure, I find it eminently believable that Rails is much more productive than Node-based things, but has nothing better than Rails established itself by now? I always found it to be great for, as you say, whipping together small apps for clients quickly, but not well targeted at big long-lived software.
- AznHisoka 8y agoI have been using Ruby on Rails for a large distributed systems (using Sidekiq) deployed across over 100+ servers. Gems like 'bluepill' and 'god' have been instrumental in deploying and ensuring mission-critical processes like elasticsearch, and redis are always up. This system has been in production for over 4 years, and we're still cranking :)
- geekjock 8y agoHow did you decide between god, monit, upstart, supervisor, etc?
- whistlerbrk 8y agoI've really had great success with eye: https://github.com/kostya/eye https://github.com/kostya/eye
- sanderjd 8y agoFair enough, that sounds similar to the system I maintained for a similar length of time. I came away thinking it was not great - error prone, expensive to run, and hard to keep running reliably - from which I developed the interest in better options that led me to ask the question in this thread. I guess as usual, YMMV.
- whistlerbrk 8y agoWhile consulting I had several clients in the top 10k websites using Rails making millions and millions of dollars a year. So, I guess I just fundamentally disagree on the 'big' portion of that statement, but also, I think apps are by their nature not long lived, so I think it is a false requirement of a web framework. That said, I also have had work I've done in Rails still powering the same website almost 10 years on, cacm.acm.org
- sanderjd 8y ago"Big" has lots of different definitions. I was referring to code size and domain complexity rather than scale and revenue. Your experience of "apps" not being long-lived doesn't match mine at all: I've never worked on a piece of software that was decommissioned after a short period of time. Or are you using the term "app" in a specific way? I wouldn't call most of the things I've worked on "apps"... We started working on the software that I was thinking of as "big" back in 2008 and from what I hear, it is still going strong (though pieces have been tactically carved out and implemented in different ways over the years). I think you and I may just have worked on different kinds of things.