Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
gcp
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
20 ms
·
181.
▲
by
gcp
9y ago
Uneducated people should be protected from ideas that are proven to be bad. Like capitalism? Or free Negroes? Western Culture?
182.
▲
by
gcp
9y ago
The point that's being made is that, indeed, as private entities they have the right to ban this content from their sites. But due to the SIZE of these operations, their actions practically amount to censorship. Saying that the right
183.
▲
by
gcp
9y ago
The situation was called out clearly for what it was. It was his right to do it, and it was understandable that he did it, and he should be able to do it. The problem is that the centralization of these services gives them so much power tha
184.
▲
by
gcp
9y ago
I get some tiles with the most popular websites. So actually the same as the new tab page in Firefox.
185.
▲
by
gcp
9y ago
Ok, so one country in Europe is excluded. This is quite far removed from "remote hiring limited to some particular countries". A large part of Mozilla's workforce is spread out through Europe, so there can't be too many
186.
▲
by
gcp
9y ago
AFAIK it's more correct to say that only some particular countries are excluded, but I'm not HR and I have no idea where your are getting that impression from, so I won't comment further.
187.
▲
by
gcp
9y ago
I think there should also be a switch to make it faster.
188.
▲
by
gcp
9y ago
Differentiated salary does not imply low wage. It means you don't get 150k USD/year for a Junior Software Engineer in rural India.
189.
▲
by
gcp
9y ago
So if you on the technical side of things, you better like maintenance work, hehe. It depends on what you work on. The Servo people feel differently ;-)
190.
▲
by
gcp
9y ago
I'm sure it also depends a lot on your manager. Like in any corporation, really.
191.
▲
by
gcp
9y ago
giving Firefox users a browser that looks like, and feeling like Chrome. Yawn. What did we do this time? Make it faster? Follow platform guidelines or modernize the design style? Adopt an extension model that can be combined with replacin
192.
▲
by
gcp
9y ago
https://careers.mozilla.org/listings/?location=Berlin Quite a bit.
193.
▲
by
gcp
9y ago
I worry it's a sinking ship The ship already sunk once, it just keeps resurfacing every time the water drops because the competition suffers from the CADT Model wrt their attitude towards the internet.
194.
▲
by
gcp
9y ago
Getting hired directly by Mozilla requires them to have a (minimal) legal presence in your country. If that isn't there, you'll be hired through an intermediary, but that is - as far as I know - possible essentially worldwide.
195.
▲
by
gcp
9y ago
These kind of things are why board games are the initial preferred testbed for a lot of this stuff, though :-)
196.
▲
by
gcp
9y ago
One typically keeps pools of trained networks to combat this.
197.
▲
by
gcp
9y ago
It's a highly specific procedure that happens before there is interaction with the opponent, so without "handing over" the understanding that having creeps on your high-ground is good, it's very hard for the learning to
198.
▲
Reverse Engineering x86 Processor Microcode [pdf]
(usenix.org)
73 points
by
gcp
9y ago
|
9 comments
199.
▲
by
gcp
9y ago
fp16 is usable for training and inference, so the Vega would have been good for both. I care about inference performance too, anyway. CPU is only a (much slower) fallback if the OpenCL drivers are broken.
200.
▲
by
gcp
9y ago
Well, cuDNN doesn't run on AMD cards, so if you want your software to have good performance on all consumers' systems (as well as avoiding being vendor-locked), cuDNN isn't an option. Also, you have to get explicit permission
201.
▲
by
gcp
9y ago
I regrettably agree there. I spent time doing OpenCL implementations of some of our stuff, but most recent decisions from AMD's side seem to indicate they'd rather get rid of it as well, in favor of some CUDA compatibility layers.
202.
▲
by
gcp
9y ago
There are two questions there that are really completely distinct, even though you're conflating them: 1) OpenCL vs CUDA In reality, performance of these is identical: i.e. writing the same kernel in OpenCL versus writing the same kern
203.
▲
by
gcp
9y ago
The author tweeted out much more: https://twitter.com/rygorous/status/895422691963936768 https://twitter.com/rygorous/status/895456126849826816 My interest is for "Wilson, Johnston
204.
▲
by
gcp
9y ago
"Surprisingly, native FP16 operations are not currently exposed to OpenCL" Given that these cards have potential to crash the price/perf of current deep learning solutions, this is more than regrettable.
205.
▲
by
gcp
9y ago
SPEC is very useful, but it's harder to get results, much less people who have access to the suite, and there is a strong influence of the compiler (SPEC2017 will tone this down early on, but there's almost no results out). I beli
206.
▲
by
gcp
9y ago
Awwww :( That's really sad. rr is important enough for C++ development that I agree it's dealbreaker.
207.
▲
by
gcp
9y ago
AMD specifically claims they do not suffer from the gcc segfault thing. So I guess it depends on what you mean by "issues?"
208.
▲
by
gcp
9y ago
How does npm5 compare to yarn? I didn't notice anything, but I'm using yarn everywhere, so...
209.
▲
by
gcp
9y ago
This wasn't an issue for Bridge either. As soon as you can build a probability distribution over possible states, you can use Monte Carlo like methods.
210.
▲
by
gcp
9y ago
I wonder if it's come too late. Couldn't it be the opposite? Blizzard was willing to do this release exactly because SC2 is dead?
More ›