Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
fizx
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
73 ms
·
631.
▲
by
fizx
14y ago
I'd donate $10k in a heartbeat to a bi-partisan PAC (or equivalent) whose goal was to fund the campaigns of people who do this. Someone should get working on this.
632.
▲
by
fizx
14y ago
So we agree unlimited vacation really means "vacation at the mercy of your manager." The problem with "quit and find a company that doesn't lie to you about its culture" is that you can't tell if they're lying until you've made the enormou
633.
▲
by
fizx
14y ago
It's a general-purpose web developer conference put on by Heroku. Last year was the most ridiculous conference I've ever been to (it had a sake opening ceremony, catered sushi, Rob Pike, and a rock concert) http://www.youtube.com/watch?v=
634.
▲
by
fizx
14y ago
Have you been paying attention? http://mashable.com/2010/09/24/doj-anticompetitive-employee-...
635.
▲
by
fizx
14y ago
s/young/any/
636.
▲
by
fizx
14y ago
What language are you in? pygments seems to be the best regarded.
637.
▲
by
fizx
14y ago
> demonstrably perform the task required for the role and then beat them up with trivia questions, with puzzles, with nerd dominance quizzes, and with subjective stuff In this world, what happens when a person wants an internal transfer
638.
▲
by
fizx
14y ago
If you like to travel and you're a good interviewer, you might get some free flights and a week of additional vacation as you help with foreign interviews.
639.
▲
by
fizx
14y ago
At large company, it's too much trouble to end the interview early. I ended an technical phone screen after five minutes, when I found a senior rails candidate had fudged his employment history and couldn't tell me anything useful about ra
640.
▲
by
fizx
14y ago
If that's still true, it won't be for very long.
641.
▲
by
fizx
14y ago
Solr will support any combination of analyzers that lucene supports.
642.
▲
Startups Adopt Their Hangouts
(online.wsj.com)
1 points
by
fizx
14y ago
|
0 comments
643.
▲
by
fizx
14y ago
CloudFlare has unlimited bandwidth. I'm not affiliated, and it doesn't exactly answer your question, but perhaps this is a useful thought.
644.
▲
Twitter adds filters -- commentary by an early Twitter engineer
(magicscalingsprinkles.wordpress.com)
4 points
by
fizx
14y ago
|
0 comments
645.
▲
by
fizx
14y ago
One of the reasons I worked at twitter is that they want you to open-source virtually everything. They weren't likely to put up a fuss, and everyone was really proud of bootstrap.
646.
▲
by
fizx
14y ago
In my experience, most boutique development shops and agencies would love to put about 80% of their work through employees, and 20% through trusted contractors. Typically you need to be a product engineer--web, ios, etc. Expect to get pai
647.
▲
Flow: actor-based concurrency with C++
(foundationdb.com)
5 points
by
fizx
14y ago
|
0 comments
648.
▲
by
fizx
14y ago
Two things are important: 1. Your OS will use free memory as a filesystem cache. Ideally, your index is being served entirely from cache. Usually, your working dataset is maybe 1/3 of the size of the index, so if your index is 6GB on disk
649.
▲
by
fizx
14y ago
Good advice in general, but 50% of total isn't a good rule of thumb for the heap. On boxes with more than 16GB of RAM, this will likely be counterproductive. The exception is if you're using cached filters (facets, filter queries) heavily
650.
▲
by
fizx
14y ago
I was under the impression that to be staff+ you had to be a leader (technical direction) but not a manager (performance reviews, people stuff). How accurate is that?
651.
▲
by
fizx
14y ago
Also, if you go by the google ratings, which are similar, you get: Intern Software Engineer Senior Software Engineer Staff Engineer Senior Staff Engineer Principal Engineer Distinguished Engineer Fellow
652.
▲
by
fizx
14y ago
In general, I fix the newgen size mostly to avoid the optimizer choosing something braindead in a pathological case. 50/50 is safe, but not optimal. In general, you should have enough unallocated memory on the box to cover your working dat
653.
▲
by
fizx
14y ago
EBS shouldn't really matter, because with a reasonable heap, he should have 40-50G of available filesystem cache, and 48G of data.
654.
▲
by
fizx
14y ago
That huge heap is extremely counterproductive, because large heaps have terrible GC performance, and you're actually stealing memory from the natively memory-mapped files that make up your index. Try it again with sane GC parameters, e.g.:
655.
▲
Gold digger
(github.com)
1 points
by
fizx
14y ago
|
0 comments
656.
▲
by
fizx
14y ago
It seems to me that he is arguing that languages do not have a total ordering. As such, novices will argue that a language is strictly better than another, when a broader perspective will reveal that this is only true for some use cases. T
657.
▲
by
fizx
14y ago
You really need a "how it works" section. Few people are willing to read the source code to see if you have interesting or useful ideas.
658.
▲
by
fizx
14y ago
> You guys say right on your home page that you're "infinitely scalable." There's a big difference between being infinitely scalable, and being infinitely foolproof. If your PaaS exposes enough power to the user, they'll find a way to
659.
▲
by
fizx
14y ago
I've noticed the difference between modulo and bitshift in java code written in the last five years. If it's getting 99.9+% cache hits, being used in an innermost loop, and jdk7 hasn't gotten around to JITing special cases in modulo, this
660.
▲
by
fizx
14y ago
Anyone know where the datacenters are going to be?
More ›