Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
bcbrown
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
63 ms
·
571.
▲
How Atigeo Uses Semantics To Make Search Interactive
(gigaom.com)
3 points
by
bcbrown
14y ago
|
0 comments
572.
▲
by
bcbrown
14y ago
A business catastrophe, not a usability or market catastrophe. "I think margins will be destroyed for a bunch of people."
573.
▲
by
bcbrown
14y ago
From the article: "Shares slumped as much as 42 percent in extended trading, dragging Facebook’s stock down as much as 9.7 percent." The title seems pretty accurate to me.
574.
▲
by
bcbrown
14y ago
I agree that there's nothing new here, but I think knowing C is still useful. It's as close as it gets to a lingua franca in our field.
575.
▲
by
bcbrown
14y ago
Because it gets them pageviews.
576.
▲
by
bcbrown
14y ago
Try turning sending referrer information on.
577.
▲
by
bcbrown
14y ago
Readability is just client-side code. Think of it as just a different implementation of a browser that ignores some markup and renders some markup in a non-standard way.
578.
▲
by
bcbrown
14y ago
I don't think that would work. You don't need copyright protection for closed-source code, because it's closed; no one can see it anyway.
579.
▲
by
bcbrown
14y ago
That author sounds like he has an axe to grind. What did Ballmer/MS do to "throw Nokia under the bus"?
580.
▲
by
bcbrown
14y ago
Are you suggesting that research into the top 10 causes of death isn't a lot more than the $9 billion spent on the equipment needed to discover the Higgs? The NIH gets $30B/yr[1]. I don't know where the $9B figure for the LHC came from, but
581.
▲
by
bcbrown
14y ago
I'll state up front that I have no familiarity with Csmith or compilers. You can characterize a random test as a random walk through a many-dimensional space, right? And the problem is that a random walk will naturally spend more time close
582.
▲
by
bcbrown
14y ago
I think one approach would be to use RNGs in a slightly different way. Instead of a 'random walk' of various API calls, define a possible outcome category, and choose a random number to be the quantity of that outcome. So for the bounded qu
583.
▲
by
bcbrown
14y ago
I found this to be very true for me, in a different context. Last spring I quit my job to take a six month road trip visiting as many national parks as I could. It had been a dream of mine for several years, and I loved it. What I found, th
584.
▲
by
bcbrown
14y ago
Microsoft is going to have a very hard time moving to an annual development cycle for Windows, as opposed to the 2-4 years they've been at. I don't know if they can solve that problem.
585.
▲
by
bcbrown
14y ago
"You get to impact the lives of millions and millions of people." That, plus "knowing secrets" was the coolest part of my time at MS. But after several years working on Win7 and Win8, I hated my job, and every week I thank my stars that I'm
586.
▲
by
bcbrown
14y ago
The article just boils down to "people interning at Microsoft have a positive impression of their employer." That's hardly a surprise, especially if they're interviewing people now, when they've barely begun their internship.
587.
▲
by
bcbrown
14y ago
Is it what the populace wants? Well, given that there's been little to no innovation in native Windows applications in the past five years, I think it's fair to say the populace doesn't want the status quo. Microsoft has to do something. I
588.
▲
by
bcbrown
15y ago
Since moving from C# and Visual Studio to Python and Vim, the one thing I've missed the most is VS-level debugging support, where I can step through a script. Is there any standalone, non-browser tool that's equivalent to this?
589.
▲
by
bcbrown
15y ago
I doubt the "free power" would be enough to compensate for the energy cost of getting a dedicated server satellite into space, plus the construction cost of building the satellite, plus the increased power cost of communicating with it.
590.
▲
by
bcbrown
15y ago
https://www.coursera.org/saas/class/index might be worth looking into.