vitus

77 ms·

811.
by vitus 8y ago
Without any sort of backing specification, the terminology of 3G vs 4G vs 5G is dumb and meaningless, as we see with AT&T calling their network 5G (E), even though it's no better than what their competitors call 4G. Also, it's
812.
by vitus 8y ago
The requirements for the proposed 4G spec are actually even higher than that -- 100Mbps was for high-mobility clients, whereas low-mobility clients were promised 1Gbps speeds. https://www.itu.int/dms_pubrec/itu-r/r
813.
by vitus 8y ago
That's not inherent to the provided functionality, but rather an implementation detail of existing boxes. There are objectively bad NATs, yes, but extending the lifetime of IPv4 by 20 years and providing isolation between internal and
814.
by vitus 8y ago
> generally which middle boxes are "bad" and the answer is all of them, almost by definition. NATs and firewalls are the first two classes of middleboxes that come to mind, and I wouldn't consider either of them inherently
815.
by vitus 8y ago
FiveThirtyEight actually has a pretty good demo of p-hacking that demonstrates how one underlying dataset can be used to derive any desired conclusion(s) by deciding which factors to include / exclude. https://projects.fivet
816.
by vitus 8y ago
If the extension field is anything like IP or TCP options, some middleboxes will also tamper the hell out of that field and strip unknown extensions, or just break connections. Often-referenced paper in that field: http://confere
817.
by vitus 9y ago
With the caveat that Jenkins is not the main form of CI at Google -- that spot goes to TAP.
818.
by vitus 9y ago
I was under the impression that most rigorous network measurement tests used TCP (e.g. traceroute -T) because it was well-known that routers and other network devices along the path often had a separate fastpath for ICMP traffic.
819.
by vitus 9y ago
https://www.reddit.com/r/boston/comments/5qdq03/mbta_commute... links to https://twitter.com/jimhaddadin/status/824708694441132034 links to the underlying data source, http:&#
820.
by vitus 9y ago
From my brief look into ledbat, it seems like it serves a similar role to BBR, which is in the Linux kernel now. Either way, it looks like the congestion control is just New Reno. https://github.com/google/netstack&#x2F
821.
by vitus 9y ago
Is this the same as the SIGCOMM '10 DCTCP paper [0]? I see that the author lists of both strongly draw from Microsoft, but there's no overlap otherwise. If so, what's changed in the past 7 years? Why publish the RFC now, espe
822.
by vitus 9y ago
Right. While this is an open process, and browser vendors are well aware of these when the spec is announced, the only way to enforce this in a meaningful way would be for all the vendors to coordinate... which I don't exactly see happ
823.
by vitus 9y ago
Worth noting: article is from 2014. It's unfair to ask it to cite trends that started after the date of publication (and likely after the finalization of the research, which probably dates closer to 2012, the cited end date in the arti
824.
by vitus 9y ago
I never said anything about internships being a source of cheap labor. Please don't put words in my mouth. While yes, an intern typically makes less than a full-time engineer, you have to realize that the intern reduces the productivit
825.
by vitus 9y ago
> asking the candidate if they'd be comfortable doing an actual task on the actual codebase, having them sign an NDA, and then giving them a bugfix to perform. They're compensated for their time, of course. At Google, we call t
826.
by vitus 9y ago
Even if that were true, why would that be relevant for deciding between two cities in the same state?
827.
by vitus 9y ago
Off the top of my head, two main points for why Austin is better-suited than Houston for a tech HQ: - UT (@Austin) means a large pool of fresh grads from a top-10 CS program they can recruit. - Existing presence of both Amazon itself and ot
828.
by vitus 9y ago
That's not quite an accurate rebuttal. That modeling study projects a large increase in severe hurricanes, not in hurricanes overall. My understanding is that the models suggest that hurricanes will get more severe, in a fashion that
829.
by vitus 9y ago
Yes, Shannon explicitly stated in Mathematical Theory of Communication that the general form was the same as that in statistical mechanics: "The form of H will be recognized as that of entropy as defined in certain formulations of stat
830.
by vitus 9y ago
Of course it's still hackable. There's nothing stopping you from advertising specific prefixes and hijacking all of Youtube traffic globally. https://dyn.com/blog/pakistan-hijacks-youtube-1/ Oh wait, tha
831.
by vitus 9y ago
Border Gateway Protocol, which is the standard for routing on a super-organizational scale. There are two (main) flavors, but here eBGP (external) is the relevant one. Instead of explaining how it'll route traffic through itself, Googl
832.
by vitus 9y ago
Well, I don't hate fixing performance problems, but I do see the value in deprioritizing them (... indefinitely, sometimes) if they don't render the program unusable by the end users, for precisely the reason you mentioned (&quo
833.
by vitus 9y ago
That sounds rather difficult, given that Google manages that TLD. It's no more plausible than Google registering http://google.bing . And no, they are not alone as far as brand-based TLDs go: https://en.wikipedia.
834.
by vitus 9y ago
> threading is shared memory and multiprocessing is message passing In most cases, yes. You can use shared memory buffers for IPC, though, although it typically requires more effort.
835.
by vitus 9y ago
> software can also influence hardware This happens more often than you'd think. Intel (and later, AMD) added AES primitives to their instruction set to speed up encryption. VT-x (and the AMD equivalent) were both designed to improv
836.
by vitus 9y ago
For about 8 months (Dec 2015 to the end of July 2016) I got daily emails from Amazon with free articles from the Washington Post. Information on this is sparse, so I don't know why it started or stopped. Prime members also get a free s
837.
by vitus 9y ago
We saw this combination historically in European royalty, which was plagued by inbreeding. One particularly well-known instance: https://en.wikipedia.org/wiki/Haemophilia_in_European_royalt... As for more modern times,
838.
by vitus 9y ago
Only sort of. You still have some HOL blocking problems with HTTP2 because you're still running on top of TCP. I recall this explicitly being one of the selling points of QUIC. First resource I found on this matter from a quick search:
839.
by vitus 9y ago
I mean, I see this as a culmination of the response to the OAuth phishing attack that happened a few months ago [0]. Red flags should go up if you were accessing a site yesterday via OAuth, and all of a sudden today on the same site you get
840.
by vitus 9y ago
Yes, but those aren't Byzantine failures. Byzantine failures present with incorrect values, not simply the absence of a value (as seen by a total hardware failure). See the abstract in Lamport's original paper introducing the Byza

More ›