Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
3JPLW
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
35 ms
·
301.
▲
by
3JPLW
9y ago
Yes, there have been lots of discussions about this — typically referred to as "formal interfaces." This would be a 2.0 (or later) feature addition, and it's not yet certain how it would be approached. https://gith
302.
▲
by
3JPLW
9y ago
Just yesterday the Illinois state senate overrode the governor's veto in order to mandate that all schools teach cursive before fifth grade. Major arguments I heard included "reading notes from grandma" and "preventing
303.
▲
by
3JPLW
9y ago
This is easily testable: julia> using BenchmarkTools using OffsetArrays A = rand(1000) O = OffsetArray(A, 0:999) function naive_sum(A) s = zero(eltype(A)) @u
304.
▲
by
3JPLW
9y ago
Yes, these benchmarks do tend to be a lightning-rod for controversy. You have two complaints: 1. Yes, some languages do not make it easy to install and link against a well-tuned BLAS. The benchmarks were very recently updated, and you can
305.
▲
by
3JPLW
9y ago
Yup, it's really amazing what millions of hours of development time will do! That benchmark in particular is a little deceiving — and we take those benchmarks very seriously! See this issue for more details: https://github
306.
▲
by
3JPLW
9y ago
Honest question: did you read the post at all? Even the quick summary addresses this. Chris' whole point is that the biggest benefits of switching to Julia will be felt by the folks that are developing the packages and libraries for o
307.
▲
by
3JPLW
9y ago
Yes, I think most folks do. He's making a distinction between programmers that are consumers of APIs vs. developers that create and maintain APIs.
308.
▲
I Like Julia Because It Scales and Is Productive
(stochasticlifestyle.com)
7 points
by
3JPLW
9y ago
|
0 comments
309.
▲
by
3JPLW
9y ago
Site is down; archive/cache/mirror/ctrl-f: http://archive.is/NHBkq
310.
▲
by
3JPLW
9y ago
The fact that it's not licensed for redistribution or even any regular use is a pretty good reason to not futz with it. https://software.intel.com/en-us/articles/what-if-pre-releas... > Intel grants to y
311.
▲
by
3JPLW
9y ago
Chris — the author of this blog post and lead developer of the DifferentialEquations.jl ecosystem — had a really nice workshop at this year's JuliaCon. You can watch it online here: https://www.youtube.com/watch?v=75SC
312.
▲
by
3JPLW
9y ago
Unlike bicycles, cars have significant blind spots with the A-pillar and such. Also unlike bicycles, cars are significantly more able to seriously injure other road users.
313.
▲
by
3JPLW
9y ago
Interesting; I've also been highly discouraged from attending his class from a recent attendee. She felt he wasn't invested in actually presenting the material and it just felt like an advertising racket to buy his books. Things
314.
▲
by
3JPLW
9y ago
No, I don't believe it is. The "display" on the demo site is simply reading out the state of the RAM. While you can "see" the on bits within the RAM grid, I think it's kinda hard to discern the values from the
315.
▲
by
3JPLW
9y ago
Note the new edit on the blog post: They silently corrected the sample data for Bill Gates, changing his state from WI to WA. Major red flag; if it was wrong in the original data just keep it that way and release a few more samples for ver
316.
▲
by
3JPLW
9y ago
How does that conclusion follow your premise? Premise: US AV companies are deeply connected to the US intelligence communities. Conclusion: Deep suspicion of criticism that a Russian AV company may be connected with Russian intelligence com
317.
▲
by
3JPLW
9y ago
It's not obvious to me why the two (spin-stabilization and hibernation) would have to be coupled. Could it be due to antennas that are required for communication during active mission control? That said, there's a whole lot of no
318.
▲
by
3JPLW
9y ago
My hunch is that it's just an artifact from the shift in imaging technologies from visible wavelengths to infrared (or somesuch).
319.
▲
by
3JPLW
9y ago
Actual paper is buried two links deep and currently seems to be unavailable. Here's the archived copy: http://archive.is/0sxuT https://www.pubpub.org/pub/direct-radio-introspection
320.
▲
by
3JPLW
9y ago
Cache/Mirror/Ctrl-F: http://archive.is/LFRxq
321.
▲
by
3JPLW
9y ago
USPS is a good contender for identity validation, too.
322.
▲
by
3JPLW
9y ago
Looks like they're predicting larger than usual aurora effects as a result: http://www.swpc.noaa.gov/products/aurora-30-minute-forecast Still not getting too far into the lower 48 states, but looks like there shou
323.
▲
by
3JPLW
9y ago
Yes, the TOS does require arbitration (including for actions that occurred before signing the TOS), but it's not clear if it applies to just the child company that is providing the credit monitoring service or if it applies to the ac
324.
▲
by
3JPLW
9y ago
Even worse: In a number of states, the DL number is deterministic based upon name and DOB. http://www.highprogrammer.com/alan/numbers/dl_us_shared.html
325.
▲
by
3JPLW
9y ago
Why must the consumer enroll in this? Why is it not automatic? They already pull all the strings with regard to keeping the consumer from taking out and keeping track of legitimate loans. Why must we go out of our way to prevent them fro
326.
▲
by
3JPLW
9y ago
Then in that case, I suppose it all depends on who is doing the justification. Nearly every police shooting in America is "justified" — be it by a police review board, a prosecutor, a grand jury, or a trial jury.
327.
▲
by
3JPLW
9y ago
You don't need a statically-typed or compiled language to have this sort of infrastructure! Julia does something similar for important changes — and it actually runs all the unit tests of all the registered packages.
328.
▲
by
3JPLW
9y ago
7->T is definitely the most oblique of the four numbers. I'd say that 0 as O is completely unobjectionable, and 1 as I mostly so. It's a lot easier for me to read #c0ffee than #7a57e5.
329.
▲
by
3JPLW
9y ago
The text is the same, but I find the official PDF better typeset and much more readable: https://www.supremecourt.gov/opinions/16pdf/15-1189_ebfj.pdf
330.
▲
by
3JPLW
9y ago
Which points read that way to you? I see one feature that's explicitly tagged as experimental. Everything else reads very matter-of-fact to me.
More ›