Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dbecker
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
39 ms
·
421.
▲
by
dbecker
14y ago
I'm reminded that Google+ hasn't caught every time I log on. Between visits, I think I forget it even exists.
422.
▲
by
dbecker
14y ago
In general, when I see a greyed out comments, I can understand why they were downvoted. But I occasionally see downvoted comments that seem totally legitimate. If someone feels negatively about a comment (for reasons other than it being no
423.
▲
by
dbecker
14y ago
I hear people say good things about Hadoop... though the stigma of it's development at Yahoo has always kept me from investing the time to learn more.
424.
▲
by
dbecker
14y ago
I could be wrong, but I don't think it was closed for off-topic (non-programming). I would guess it was locked for "not having an answer." They say you should be soliciting answers rather than opinions... and, for better or for worse, that
425.
▲
by
dbecker
14y ago
StackOverflow seems to think that these types of questions cause the community to sink to a lower level of discourse. I'd be inclined to disagree, except that the quality of responses their remains the highest I've seen on the internet. S
426.
▲
by
dbecker
14y ago
I tried switching from vim to PyCharm about a year ago. I found it incredibly slow. With vim or ST2, I do my work and don't think about the editor keeping up. With PyCharm, my typing could briefly outrun the editor. The editor would catc
427.
▲
by
dbecker
14y ago
Maybe YCombinator judges applicants based on productivity rather than suffering. I don't mean that as a criticism. He may be a great guy... he may be productive too. But sleeping in a car (or a mansion) is unlikely to be the key to success
428.
▲
by
dbecker
14y ago
I've never heard anyone say this about C... If I've heard it said about C++, that was a long, long time ago.
429.
▲
by
dbecker
14y ago
The R magic is going to be a real win. Great job.
430.
▲
by
dbecker
14y ago
Wes: Thanks! Anyone reading this who wants to get started with Pandas: The early release of "Python for Data Analysis" ( http://shop.oreilly.com/product/0636920023784.do ) is already very helpful.
431.
▲
by
dbecker
14y ago
Both the value (and the experience) of getting a PhD is very field-specific... Out of curiosity, what field? Are you going into academia? I enjoyed my PhD (in economics)... but I think that's pretty rare. It was also less useful to me th
432.
▲
by
dbecker
14y ago
Based on the great reviews of ST2, I switched from vim to ST2 (with vintage mode) a few months ago. It seems good... but it hasn't lived up to the reviews for me. I'm guessing I'm missing some of the functionality. Is there a tutorial any
433.
▲
by
dbecker
14y ago
CEO's say how busy they are, but it seems like most execution is handled by people below the CEO. So I'm left wondering what are specific tasks that are the right level to keep the CEO so busy? The author had a great chance to address this
434.
▲
Why doing a PhD is often waste of time
(economist.com)
5 points
by
dbecker
14y ago
|
5 comments
435.
▲
by
dbecker
14y ago
I didn't expect to learn much when I started reading this, but I was wrong. Excellent post. Thanks.
436.
▲
by
dbecker
14y ago
I'm always surprised when I see internet reviews where people talk about no longer being able to see individual pixels. I can't see individual pixels even if I look closely at my screen... and everyone I've talked to in person has said the
437.
▲
Benchmarks for New Macbook Airs and Macbook Pros
(primatelabs.com)
1 points
by
dbecker
14y ago
|
0 comments
438.
▲
by
dbecker
14y ago
This graph tells me whether a guy prefers to post up on the right side or the left side... high post or low post... shoot mid-range jumpers or drop back to the three point line. A colored dot at the free throw line would tell me where he sh
439.
▲
by
dbecker
14y ago
Wholeheartedly disagree. The Lakers may be up there with the Heat for most hated team, but I think OKC is very well liked (outside of Seattle).
440.
▲
by
dbecker
14y ago
I'm trying to determine whether the "75% less glare" claim means this is usable outside. Anyone seen feedback on this?
441.
▲
by
dbecker
14y ago
The Coursera courses are excellent, but Coursera, Udacity, CMU, etc. are offering a different set of courses than the UW progam. For instance, I don't think any of the current players are offering Hadoop courses... In general, it looks like
442.
▲
by
dbecker
14y ago
Seeing him be so smug makes me want to close my LinkedIn account.
443.
▲
by
dbecker
14y ago
Those salaries seem neither especially high nor especially low... and I don't see how this relates to their lost passwords. If there is a point you are making, can you elaborate?
444.
▲
by
dbecker
14y ago
The author is trying to shoehorn this post into the "everyone should learn to code" debate, but that is a totally unrelated issue. This post talks about switching from (VB to Python). That only happening after they've learned to code.
445.
▲
by
dbecker
14y ago
I don't doubt that ResolverOne is a great experience... but so many businesses are wedded to Excel (and VBA). I expect IronSpread will see much more use than ResolverOne because it will integrate into these offices more easily.
446.
▲
by
dbecker
14y ago
This solves a real business problem for a lot of people. I'm doubly impressed that it was done by a recent grad and someone still in school. You guys have a bright future in front of you.
447.
▲
by
dbecker
14y ago
But it allows the passwords to be stored by someone who specializes in security. It'd be nice if everyone who runs a non-trivial website stayed was a security expert, but we're also busy with other aspects of the site.
448.
▲
by
dbecker
14y ago
I think your employer was probably an exception, and most employers are more similar to the NYTimes author.
449.
▲
by
dbecker
14y ago
The author appears to misunderstand statistical significance when he makes claims such as "the existence of a static type system has neither a positive nor a negative impact on ... development time." The mean development time for dynamicall
450.
▲
by
dbecker
14y ago
Not at all. Even from a Bayesian perspective, having a small sample that is too small to reach strong conclusions does not shift your posterior towards 0.
More ›