Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
vicaya
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
15 ms
·
91.
▲
by
vicaya
15y ago
This is probably true for many people. But some of the best work in history was produced exclusively by people in solitude. In fact, without thinking in solitude first, most people cannot have meaningful in-depth conversation with people in
92.
▲
by
vicaya
15y ago
Focus on interface rather than object. Embrace explicit state with state machines and events.
93.
▲
by
vicaya
15y ago
People should only have strong opinions because they have convincing evidence. The point of the statement is that despite the fact you have convincing evidence from your perspective, you need to consider the possibility that your perspectiv
94.
▲
by
vicaya
15y ago
> entries = new ArrayList<Integer>() {{ add(3); add(4); }} Nah, entries = Arrays.asList(3, 4); Scala is pretty sweet with a lot of syntactic sugars, but it still it doesn't fundamentally make writing correct code any faster, esp
95.
▲
by
vicaya
15y ago
I don't think so. At least 4 companies' codebase have their own assert/contractual macros specifically for this reasons. Such code will never pass a cursory review at these companies. This is a C-noob bug, IMO.
96.
▲
by
vicaya
15y ago
When I was just starting to play with Scala 5 years ago, I found a lot of stuff that's "cool/neat/tight/convenient", I've since watched the language grow into a monster that's trying too hard to be Java++. The xml embedding syntax that look
97.
▲
by
vicaya
15y ago
Well, Chrome must be broken then, as it's a validated HTML 5 page. It works with FF with noscript (with the domain blocked). Frankly, I have a hunch that people who care about javascript security don't use Chrome. No chrome until noscript.
98.
▲
by
vicaya
15y ago
The article is from 2008. Since then, the so called parallel DB goes no where, and Hadoop takes over the world. The main problem of traditional (OLAP) DBMS in the era of big data is that ETL (Extract/Transform/Load) becomes the main bottle
99.
▲
by
vicaya
16y ago
"The only freely-available tool that attempts to provide both of these is zookeeper, but zookeeper is specialized, focused on locking and server management. We need a general-purpose tool that provides the same guarantees in a clean, well-d
100.
▲
by
vicaya
16y ago
For a true lightweight modern web framework in Java that's based on Guice, checkout my little embedded web framework for the next generation of Hadoop. https://issues.apache.org/jira/browse/MAPREDUCE-2399 No XML, no YAML, no config/proper
101.
▲
by
vicaya
16y ago
hsqldb is more than adequate for a diaspora node. diaspora in a jar is completely doable and outperforms any rails/django setup.
102.
▲
by
vicaya
16y ago
http://en.wikipedia.org/wiki/Amdahl%27s_law
103.
▲
by
vicaya
16y ago
Congrats! Diego.
104.
▲
by
vicaya
16y ago
No. The existing Filesystem abstractions conflate namespace APIs and storage APIs and handle both poorly. It's time to break'em apart! What we need is storage atom/fabric APIs that provide individual namespace (tagging, hierarchical or what
105.
▲
by
vicaya
16y ago
Annoying marketing piece devoid of any real technical content.
106.
▲
by
vicaya
16y ago
More reasons to type on glass :)
107.
▲
by
vicaya
16y ago
It doesn't impact me as I use https://encrypted.google.com/ (automagically via the wonderful HTTPS Everywhere firefox plugin.) which doesn't have the instant feature turned on by default. OTOH, it's a very smart move for G from technical
108.
▲
by
vicaya
16y ago
uh, the -g compiler flag to enable debugging will create much more disturbance than a few printf ever could, debuggers are mostly useless for tracking down race conditions as they change the timing completely. OTOH, I'd use a low overhead l
109.
▲
by
vicaya
16y ago
You know, it'd be more convincing if you can come up with reproducible benchmarks to backup your claims. According to This guy's data: http://www.codexon.com/posts/debunking-the-erlang-and-haskel... Haskell is even slower than python (whi
110.
▲
by
vicaya
16y ago
Informative, but should really be called "The Evolution of Writing in Italic and Germanic Scripts" (only one obligatory reference to an ancient Greek example and no mention of many other Indo-European scripts (say Slavic, which includes Rus
111.
▲
by
vicaya
16y ago
Sorry, but real pros don't use Chrome. No chrome until noscript, period.
112.
▲
by
vicaya
16y ago
The strength of Bayesian filters is that they tailor to individuals, i.e., everyone can have a different model. An open source "well-trained" one is trivial to subvert.
113.
▲
by
vicaya
16y ago
Ever heard of EB1? Any PhDs (including candidates) worth their salt (have a few journal papers and good recommendations from their professors) can do it easily with some help from decent immigration lawyers. EB1 has no quota and waiting tim
114.
▲
by
vicaya
16y ago
It really depends on the data. For highly redundant data like web pages with lots of boilerplate header/footer, it can compress better because the bm_pack (first pass of bmz) looks for large common patterns over all the input. For typical t
115.
▲
by
vicaya
16y ago
+1 firefox. I've been resisting Chrome because they don't care about script blocking API. No chrome until noscript!
116.
▲
by
vicaya
16y ago
That's why it's important to know about 83b election to cap the tax rate at 15%. Any start-up lawyer/law firm worth its salt would do this for you as part of the pile of forms to sign when you incorporate.
117.
▲
by
vicaya
16y ago
I'd assume all founders would know about 83b election (us founders anyway.) to cap the tax rate at 15% (long term capital gain)
118.
▲
by
vicaya
16y ago
He should really be using dstat (or something like that) to show number of context switches per second. His benchmark is mostly a measurement of Linux kernel thread context switching, which is a lot slower across cores/cpus. I suspect that
119.
▲
by
vicaya
17y ago
I'm interested to see their interesting algorithms to determine interesting repositories. It's obviously not based on the numbers of watches and forks.
120.
▲
New machine learning challenges/prizes
(hunch.net)
1 points
by
vicaya
17y ago
|
0 comments
More ›