Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
kromem
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
68 ms
·
811.
▲
by
kromem
9y ago
Go & Postgresql on the server. Angular 2+ on the frontend. Why? Go is fast and bare bones, and works nicely as a backbone to an API (especially with a GraphQL intermediary). Postgresql plays very nicely with Go, especially if writing qu
812.
▲
by
kromem
9y ago
Real identities doesn't work when FB threads aren't indexed by Google. If googling myself found comments I'd make on "platform X," you can be sure I'd carefully consider my comments on that platform. Having ano
813.
▲
by
kromem
9y ago
It's not particularly harmful in terms of heredity (roughly the same "danger" as a woman having a kid at 40 in societies where it is not common). Interesting factoid: Freud was married to his first cousin. It's partly be
814.
▲
by
kromem
9y ago
I never understand why we always assume that it's extra-terrestrial. For example, let's say time travel will be possible at any given point in the future. Well, if it were possible in the future, we might expect to see it happenin
815.
▲
by
kromem
9y ago
You let the comments themselves have upvotes. Look at any of the technical subreddits - typically domain experts end up being near the top comments naturally, because they lay out substantive comments. Reddit "flair" is a solution
816.
▲
by
kromem
9y ago
Give it time. Just as a frame of reference, the original iPad sold less units in its first year than the PlayStation VR. Virtual reality is still in its infancy, and honestly AR is a technology that's going to have to build on top of a
817.
▲
by
kromem
9y ago
Because of the study design, there's no control group that would account for placebo. The placebo effect in several meta-analyses (Beecher, Haas) showed an effect in studies of around 30%. Coincidentally, people participating in this s
818.
▲
by
kromem
9y ago
Kyle, I applaud the concept of sharing ideas for entrepreneurs to tackle. However, this list of ideas is dangerously bad, with nearly all of them suffering from either monetization challenges or huge execution issues. Anyone that sets out t
819.
▲
by
kromem
9y ago
The problem is not with egos, the problem is that they have a preconceived notion that they are equal with their coworkers, and then suddenly find out they are not. Had the designation been clear from day one and each individual pay scale e
820.
▲
by
kromem
9y ago
Except that it'll become the same as any other public indicator of performance if public - you don't make X because you perform, you clearly must perform because you make X. Transparent compensation inherently ranks performance, w
821.
▲
by
kromem
9y ago
Even with an 0-day, why was consumer data not encrypted at rest with segmentation for access? If you have the most sensitive information for millions of people, maybe don't only worry about perimeter defense?
822.
▲
by
kromem
9y ago
I'm a big fan of Rand and truly believe that pure altruism is a fundimentally unsustainable and fruitless endeavor (considering the definition is putting in more than you get out, that's kind of obvious). Luckily, most of the thin
823.
▲
by
kromem
9y ago
Not just that, but it's incredibly shortsighted as well. Human life has positive ROI, as evidenced by the fact that our species has done better in the world over time. The largest returns are highly unpredictable too - the majority of
824.
▲
by
kromem
9y ago
Really? Go ahead and avoid anything hosted on or using AWS. Might be a bit difficult, and you may reconsider Amazon's scale relative to Google.
825.
▲
by
kromem
9y ago
I'll answer, though not OP. Medicare is a pretty good example. As best I'm aware, that program's negotiated rate is less than the average private insurer for any given procedure, which is pretty impressive. Public broadcastin
826.
▲
by
kromem
9y ago
I never understand how people point to existing systems that are working for a country, and then make the (valid) point that the success of those policies can't necessarily be extrapolated to another country, but then (invalidly) belie
827.
▲
by
kromem
9y ago
Same for me, though a huge factor in my "re-entry" threshold is resolution. I really do enjoy VR when I get around to playing it, but all I have is presence, but in a world that not particularly beautiful. My favorite VR title is
828.
▲
by
kromem
9y ago
Can anyone cite the parts of the document where he claimed that the biological differences mapped to job performance? I read it, but definitely don't recall that association being made (he did talk a lot about preference and job satisf
829.
▲
by
kromem
9y ago
An interesting anecdote regarding gender bias and tech. In the very early stage of my company, we wanted to outsource some UX work. After an exhausting review of applicants on one of the freelance sites out there, we finally settled on a Pa
830.
▲
by
kromem
9y ago
There are groups that have been more marginalized than "white males" on average . But that was the whole point of the original document in question. White/black/male/female/etc is a really bad proxy for a pe
831.
▲
by
kromem
9y ago
You may want to research the early days of McCarthyism and the blacklist. Is this the first firing that was perhaps an overly sensitive reaction concerned with appeasing a very touchy ideological base? Because I can think of a number of oth
832.
▲
by
kromem
9y ago
To further your point #1 - vet is the lowest paid medical profession, and the highest (anesthesiologists) is male-skewed. But in general, he's right that we're in a scary age with the aggressive and dogmatic ideology that favors p
833.
▲
by
kromem
9y ago
You're spot on about how scary the need for an echo chamber is. I was recently in Russia and the young, educated, and violently ideological foundations for the revolution are scarily parallel to some of the militant social "warrio
834.
▲
by
kromem
9y ago
Are male vets a "freak"? Vets are 80% female. Is the industry closed to men that might want to be a vet? Do we need more outreach programs for men to be vets? Or maybe is it that men are less likely to want to medically treat anim
835.
▲
by
kromem
9y ago
If we're so concerned with the advantages of wealth and education inequality, why are we still using race (a bad proxy, given the spread and distribution around the averages - the whole point of the original document) as the determinin
836.
▲
by
kromem
9y ago
Except his whole point was that only looking at averages and not considering the spread and distribution is insane. Should a son of a several-year unemployed coal miner be discriminated against because of his gender and race, just because o
837.
▲
by
kromem
9y ago
Facebook also has a serious problem in that its news feed is a content recommendation engine with only positive reinforcement but no negative reinforcement. So you end up with a ton of false positives even when actively interacting with the
838.
▲
by
kromem
9y ago
For shallow vs deep, what I decided for my code base was that I'd pick the appropriate depth as a general rule for an object and not worry about I/O micromanagement. If it is cheap to pull the additional data and it makes sense to
839.
▲
by
kromem
9y ago
Basically "Uncle Bob" took the onion architecture, changed it slightly, and took ownership of it. But yeah, regardless of the lame appropriation, a solid concept.
840.
▲
by
kromem
9y ago
I've been using a very similar architecture on an application for about 2 years now, and it's been incredibly smooth to work with, especially in Go (passively satisfied interfaces and enfirced lack of inheritance), though I do fee
More ›