Technically, I'm completely ignorant about these topics.. And yet I have an opinion: I think web people tend to overestimate their scale and use big sledge hammers to drive small nails.
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
This Tweet is unavailable.
-
-
Like yeah, if you put your data in a database with logic in Ruby, you may need elastic compute to handle load.. But a hash table in C++ probably starts off with two orders of magnitude higher perf, so maybe go with that first.
1 reply 0 retweets 3 likes -
You can buy machines with lots of memory, and CPUs are really fast. There's a huge up front cost when you start using all these fancy scalable/distributed technologies. Very few people need "google scale" solutions.
2 replies 0 retweets 4 likes -
The web is probably the software domain that is most obsessed with orthodoxy in development.
1 reply 0 retweets 3 likes -
I wish I could find this blog post. Basically, there was some new paper about a fancy distributed graph processing library, and someone just implemented dijkstra's on a laptop and beat their numbers by a lot.
3 replies 0 retweets 4 likes -
I think it was
@cmuratori who wrote some webapp from scratch rather than with the preferred stack of JS libs and it ran so much faster than your typical web app that it felt native.1 reply 0 retweets 1 like -
Replying to @fahickman @ssylvan and
That's our mailing list sign-up server. Go to, for example, http://mollyrocket.com and put your e-mail address into the subscribe box in the footer. On fast connections it typically completes so fast, you don't notice it's done...
1 reply 0 retweets 0 likes
By contast, the mailing list provider we had been using took _many seconds_ to subscribe someone :( The low quality of web engineering never ceases to amaze me, nor do the bandied-about excuses.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.