ot

13 ms·

151.
by ot 3y ago
Imagine you use systemd to manage daemons in a large distributed system. Crashes could be caused by a failure in a dependency. Once you fix the dependency, you want all your systems to recover as quickly as possible, you don't want to
152.
by ot 3y ago
> The scheduler performance has nothing to do with the used language. This is correct, but I don't think the article is trying to make any claim about the language being relevant for performance. What I believe the author is showcas
153.
by ot 3y ago
> Total: 26 days 5 hrs 40 mins What's the significance of adding up together the rows? Time passes simultaneously for all these companies.
154.
by ot 3y ago
The repo does not appear to have had substantial changes for ~8-9 years https://github.com/yui/yuicompressor Yahoo announced development was stopped in 2014: https://yahooeng.tumblr.com/post/960981
155.
by ot 3y ago
I think that's not what people would normally call "the distribution" :) systemd, coreutils, ssh, shell, ... come from CentOS. (Hi Pádraig!)
156.
by ot 3y ago
This is not accurate, the system packages are from the CentOS binary distribution. You might be thinking of C/C++ runtime libraries, which are distributed as separate packages used by the internal ("fbcode") binaries, but sys
157.
by ot 3y ago
Not sure if you're being facetious, but if not, different Wozniak :)
158.
by ot 3y ago
SSDs are orders of magnitude faster than HDD, enough to justify the cost differential and investments to rethink the software stack. Octane was an incremental improvement in performance when compared to SSD, and in cost when compared to RAM
159.
by ot 3y ago
That's an apples-and-oranges comparisons, 100M is just the size of the frontend/middleware code (Hack), while 2B includes everything , including configuration and generated code.
160.
A suite of AI language translation models that preserve expression(ai.meta.com)
1 points by ot 3y ago|0 comments
161.
by ot 3y ago
Cannot edit anymore but I realized that my comment was wrong :) You just need 2 threads, as long as one is suspended long enough.
162.
by ot 3y ago
Don't you need 65k threads all contending on the same state for that to happen? Even if your process does have 65k threads, you'd need a pretty large critical section for all of them to be preempted in an unlucky point. That said,
163.
by ot 3y ago
Are you thinking of netmap?
164.
by ot 3y ago
The title is as factual as it can be. If you were expecting a different conclusion, it may have been due to prior biases. If the title said "Mac users" instead of "Linux community", would you have expected a different co
165.
by ot 3y ago
But how would that benefit OpenAI? It would be a "fuck you" move that faintly inconveniences Microsoft and destroys OpenAI.
166.
by ot 3y ago
They just need him on speed dial for that, not physically present breathing down their necks.
167.
by ot 3y ago
> Google co-founder Sergey Brin has been spending “four to five days a week” with the developers. I don't see how that can make the project move faster. I wouldn't be surprised it has the opposite effect actually.
168.
Emu Video and Emu Edit, our latest generative AI research milestones(ai.meta.com)
201 points by ot 3y ago|47 comments
169.
by ot 3y ago
Yes, according to those numbers BoA is technically solvent.
170.
by ot 3y ago
Comparing the balance sheet fluctuations of a bank to its market cap is meaningless. Banks are extremely leveraged by their nature, BA has like $3T in assets and $2.8T in liabilities, so minor swings in assets value move the equity by amoun
171.
by ot 3y ago
How much entropy do you need for that?
172.
by ot 3y ago
The article shows the connection but it doesn't present an intuition of the why , which I think is by far more interesting, and it's key to understanding Fourier analysis. You can take a vector and shift all its elements to the r
173.
by ot 3y ago
For the end user, yes. What would be the alternative?
174.
by ot 3y ago
It is doable, `folly::Expected` supports that: https://github.com/facebook/folly/blob/main/folly/test/Expec...
175.
by ot 3y ago
I've seen early proposals of this and never been a big fan of the syntax, I'm disappointed it made it in to C++23. Monads can be useful if they enable composition across different concepts, but this is syntactic sugar that IMO can
176.
by ot 3y ago
It's 1000x compared to a very naive baseline, exhaustive traversal of the matching subtree. Nobody in practice would do that, storing the maximum score for each subtree so you can prune is a very common technique, definitely not novel
177.
by ot 3y ago
From TFA: > For everyone else, the RDA should probably be boosted roughly 25% up to 1 gram per kilogram of body weight per day.
178.
by ot 3y ago
> Have something substantial to add please The comment was indeed about a matter of substance
179.
by ot 3y ago
I used to do a lot of bit twiddling (for compression) and the fact that shift values of 64 (for 64-bit integers) always have to be special-cased, instead of just zeroing out the value, is so annoying and inefficient. I really wish it just w
180.
by ot 3y ago
Even if you don't use password authentication you may still type sensitive information in a SSH session. For example, password when using sudo.

More ›