AtticusTheGreat

Currently fleshing out my boggle-like empire. serpentinegame.com. I've got ELO rated rooms, a daily room, and a new infinite room. I've been doing this since 2008 but recently rewrote everything from scratch and seeing how far I can push the online multiplayer boggle niche.

Whenever this kind of disagreement happens with my team, we put it to a vote and then the winning convention is enforced across the codebase from then on. There are tons of things in programming that come down to opinion but the important thing for me is that the team has a set convention and sticks to it.

In the old days the user would submit the form, the validation would happen on the server, then return a result. If the result was a failure, you would render the errors on a new page load. If the result was a success, you could redirect the user to "confirmation" page or reload the same page but with a different output.

I did try Firefox 57 on my Android phone and found the scrolling to be unpleasant with lots of artifacts and stuttering. It's still a much better experience with the mobile version of Chrome.

I only skimmed the article but I agree with the conclusion. I evaluated both HAProxy and Nginx for a high-volume/low-latency load balancer cluster (500k requests/s) and HAProxy simply beat the pants off of Nginx. It has an extensive and well documented set of configuration options, tooling, and reporting, and it performed flawlessly on production (after much toiling). I couldn't ever quite get Nginx to handle the same load without falling apart at the seams.

I've got two NUCs, which I use as HTPCs running Fedora. They're fantastic machines. Really tiny, quiet, and plenty powerful enough for HD movies and light gaming/emulation.

The thing is, all of these details are implemented in the git command line tool, not in a library. So it is pretty difficult to write a tool that works exactly the same way the command line tool does, or if you do manage to do it, it'll be fragile since the command line tool could change behavior at any time.

This is exactly my stance on the matter. Don't litter the code with unnecessary comments that will never get acted upon. Instead, make a story in your bug tracker and act upon it.

Getting 60TB of usable space in the cloud is pretty expensive. Honestly I have no idea why he needs so much space, but if he does need it all, he probably didn't make out too badly. One of the main benefits of the cloud, though, is that you only pay for what you use, and he has to over-provision from the beginning, so let's consider that.

So he's currently using about 10TB of 60TB of usable space. If he uses Amazon S3's standard storage, he would be paying about $230/mo. If he uses infrequent storage that is $125/mo. That goes up as his usage goes, so when he's using 30TB that will be $690/mo and $375/mo respectively. He also has the benefit of high speed ethernet with the home NAS, unless he has fiber 1Gig internet, in which case speed is probably a wash. I'm not sure if there are other significantly cheaper cloud storage solutions at that scale.

So I'd say he hasn't done too badly for himself, though he probably could have saved some on the hardware by getting cheaper parts.

Such as?

As someone who applauds this pivot, I never saw much value in the whole convergence idea. Why would I want my laptop and phone to be the same device? This seems to follow the misguided idea in software that if two ideas are similar, just make a single generic idea that solves both problems. But then you get into leaky abstractions and have to make lots of sacrifices to get it working well for both use cases.

I know you probably meant "gold standard" but I was amused thinking about what a golf standard might mean ;)