Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
gojomo
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
76 ms
·
841.
▲
by
gojomo
6y ago
Some think the Adenovirus-vetor vaccines have more risk of certain side-effects, especially if (by chance or a prior vaccination) the person was previously infected by a related Adenovirus. By contrast, the mRNA vaccines do the same thing i
842.
▲
by
gojomo
6y ago
For which vaxes? How do you know this?
843.
▲
by
gojomo
6y ago
Your physician is wrong, as separate from just the Vitamin-D mechanism, greater lifetime sun exposure is correlated with many net-positive health outcomes, especially with regard to cancer & cardiovascular health, that outweigh the ri
844.
▲
by
gojomo
6y ago
Twitter's CEO Jack Dorsey has repeatedly been summoned before Congress to be dressed-down by legislators, who regularly threaten to "do something" that increases Twitter's liabilities or otherwise encumbers their busines
845.
▲
by
gojomo
6y ago
Reminds a bit of the massive theft of Mercedes from Car2Go in the US last year: https://www.caranddriver.com/news/a27192115/thieves-car2go-s... Criminals will find your weak spots & exploit them until closed.
846.
▲
by
gojomo
6y ago
Even tiny sensitivity would be better than nothing – or something that's so slow & expensive it's impractical. But other antigen tests purport to have 90%+ sensitivity. And the kinds of "PCR positives" they miss are
847.
▲
by
gojomo
6y ago
This article doesn't mention PCR tests, so perhaps we're being smart enough to accept the often-more-relevant-for-infection-control antigen tests for this purpose. (I'd not yet bet on it, given how bureaucracies have made alm
848.
▲
by
gojomo
6y ago
In order to stop seeing "millions of new cases", every source of new cases should be reduced. International travel is an orderly place to apply a filter, just as hospitals test all admits (even those without symptoms). Blocking al
849.
▲
8 years ago: HN front page 100% Aaron Swartz stories
(web.archive.org)
12 points
by
gojomo
6y ago
|
2 comments
850.
▲
by
gojomo
6y ago
You have an interesting specific question, about actions against a particular service (Parler). But I think your poll results may be muddied a bit, because of the the side musings about TBL's data decentralization efforts, internationa
851.
▲
by
gojomo
6y ago
But if architected for openness without central censorship chokepoints, such affinity declarations can’t alone prevent participation. Mastodon’s tech is used by Gab, for example. At best this can: * influence the observable qualities of the
852.
▲
by
gojomo
6y ago
Why would people be discontented? It's just numbers. West coast office work would happen 11a-7p USA Time, while the Midwest/Texas does 9-5 and the East Coast does 8a-4p. Here in California, bars would be open until 4a. Prime resta
853.
▲
by
gojomo
6y ago
A good 1st step, then abolish US time zones. Works for China & India! (While all the normies are groggy, it should then be easy to slip a change to nationwide UTC in as the coup de grace.)
854.
▲
by
gojomo
6y ago
For seeing an overall mortality benefit in normal times, maybe. But it's the D-related-compound levels in your body right now (or seasonally) that have a noticeable effect on fighting current viral infections. It takes about 7 days for
855.
▲
by
gojomo
6y ago
It's the default most places. Especially with larger training corpora or vocabularies, the negative-sampling tends perform better - & I don't recall notable situations where the HS mode is better. (I recall seeing some hints i
856.
▲
by
gojomo
6y ago
Ah, but I've now looked at their code, and it's not the only change! They've also eliminated the `reduced_window` method of weighting-by-distance that's present in `word2vec.c`, Gensim, and FastText. What if that'
857.
▲
by
gojomo
6y ago
While I'm unsure of this paper/implementation's main claims without a closer reading, the Appendix D 'alias method for negative sampling' looks like it might be a nice standalone performance improvement to Gensim (&
858.
▲
by
gojomo
6y ago
While I still need to read this paper in detail, I'm not sure their only change is to this scaling of the update. The `koan` CBOW change has mixed effects on benchmarks, and makes their implementation no longer match the choices of the
859.
▲
by
gojomo
6y ago
Gensim considers the `word2vec.c` code, from the original authors of the Word2Vec paper, as canonical and seeks to match its behavior exactly, even in ways it might deviate from some interpretations of the paper. If there's an actual b
860.
▲
by
gojomo
6y ago
Sure, but it's a beautiful disaster that's helped explore the space, & both train & incentivize people for various offshoots & successors that will resolve all of Dillinger's concerns, and more.
861.
▲
by
gojomo
6y ago
Also: if company has 'standard' vesting documents & has taken outside funding, is the lawyer who set those up still a trusted neutral resource for both founders? Even in their role as "company's lawyer", they co
862.
▲
by
gojomo
6y ago
Agree completely that poster would want the right kind of lawyer – someone who specializes in this kind/scale of business, and that reaching out to acquaintances/etc who've been in relevant situations may be as valuable or mo
863.
▲
by
gojomo
6y ago
Find a lawyer/negotiator with experience in this area you can talk to directly, high-bandwidth. They'll echo many of the themes here, but be able to get (& give) more depth/detail. This is a negotiation that's alread
864.
▲
by
gojomo
6y ago
That's a negotiating position.
865.
▲
by
gojomo
6y ago
Wow, surprising PDF partisan downvotes here!
866.
▲
by
gojomo
6y ago
Some org might have decreed internal fine distinctions, but in common parlance, both terms are used for the same kind of after-event writeups.
867.
▲
by
gojomo
6y ago
In this domain, the two terms are synonyms. And neither "post-mortem" nor "incident report" appear on the log page, making either an equally fair synthesized title.
868.
▲
by
gojomo
6y ago
That Google is choosing to use a PDF (!) as the official incident-reporting media is as confidence-destroying as was the outage.
869.
▲
by
gojomo
6y ago
…which would be plenty of info for the primary endpoint of "saving lives". The government's desire for a legible set of stats should take a backseat to the urgent need to inform people how to protect themselves. Let's he
870.
▲
by
gojomo
6y ago
Not seeing even a single such prompt visiting this site in Brave.
More ›