Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
hobs
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
45 ms
·
931.
▲
by
hobs
3y ago
The only thing that's sketchy is Spotify directing everyone to the same content, making stuff that appeals to the audience isn't against the rules.
932.
▲
by
hobs
3y ago
Then punish them instead of punishing everyone, communal bad rules so you can manage something isn't a good enough reason.
933.
▲
by
hobs
3y ago
And this is where if I was a parent I would be ripping the stupid school up for such an asinine rule. As I told my mother when I was in highschool, the only thing they care about is preparing us for life in a factory. The only places I had
934.
▲
by
hobs
3y ago
At the end of the day the recommendation algo is why tiktok grew so fast and is so sticky, users want the companies to remember stuff, just not to use that memory against them (ya right)
935.
▲
by
hobs
3y ago
Absolutely and for this case a different plan mattered.
936.
▲
by
hobs
3y ago
That's exactly what I have done with diffing SQL in lazy mode - just use a server and diff the AST/plan.
937.
▲
by
hobs
3y ago
/shrug that's fair, with only like four things its not that much more readable.
938.
▲
by
hobs
3y ago
It's metadata, so comparing how good a count(*) is is pretty pointless.
939.
▲
by
hobs
3y ago
Makes sense. In my opinion I would just generate fake flamegraphs then - they are much more readable and the format is dummy easy to fake.
940.
▲
by
hobs
3y ago
Good article, just throwing out there that flamegraphs would be exactly what you need for visualizing this stuff.
941.
▲
by
hobs
3y ago
A lot of what you said what right, but fwiw they do have shotguns they go for during the raptor scene "clever girl" and all that.
942.
▲
by
hobs
3y ago
I am really glad that worked! I am pretty paranoid about the downstream breakage of things but I probably just need to let go a little more.
943.
▲
by
hobs
3y ago
How do you deal with the validation problems that crop up, the data entry problems that crop up, and the impedance mismatch between a traditional datastore and the sheet?
944.
▲
by
hobs
3y ago
Chyrp was my first "microblog" and I loved it, it showed me a lot of what you could do with conventions and simple content. It definitely inspired me to look more into web development and coding in general.
945.
▲
by
hobs
3y ago
I don't think we do, its just that just like I don't want to work for big oil, defense, or tobacco companies, I have no interest in working on cryptocurrencies when their biggest contribution I see is rug pulls and proof-of-waste
946.
▲
by
hobs
3y ago
Since nobody answered the question, the reason is its terribly absolutely insanely slow. It's possible, just requiring hundreds of thousands or millions of times as much work as say, a normal lookup in a database.
947.
▲
by
hobs
3y ago
Except for all the people that are born rich, they don't have to do anything.
948.
▲
by
hobs
3y ago
Powershell lets you do this too.
949.
▲
by
hobs
3y ago
I didn't underestimate anything really, I am supposedly an expert in the field. I have used 100GB+ workloads on sqlite, its great, its useful, and its still extremely limited compared to the featuresets of other mutli-user relational d
950.
▲
by
hobs
3y ago
Being able to continue your life as normal is just a calculation of how insulated you are from the real world. The department of homeland security and basically everything snowden brought us. The war in Iraq. The war in Afhangistan. Milli
951.
▲
by
hobs
3y ago
The difference is you are choosing between "learning postgres and have that tool your entire life" and "learning some vendors proprietary extensions so you can stop using it when they shut down" The only way the personal
952.
▲
by
hobs
3y ago
The point is that choosing what appears to be wrong tool for the job and putting a bunch of engineering effort into it makes no sense unless you can demonstrate some special something that makes it all worth it. Sqlite is very nice and fran
953.
▲
by
hobs
3y ago
Nah, there's no reason to accept terrible behavior from people just because they also do some good, the more we push back and frankly punish large companies in real terms the more likely we are to get a peaceable arrangement. Google an
954.
▲
by
hobs
3y ago
Agreed - my single biggest win for recruiters to show up in my inbox is to google search linkedin for the top result for the job role I am targeting, and then change my profile (in an honest way) to ape that as much as possible, pass rate i
955.
▲
by
hobs
3y ago
It in no way turns facts upside down, it simply responds to the word majority and the current events, it's the facts to say that the supreme court did the opposite in the 1970s.
956.
▲
by
hobs
3y ago
To be clear, a majority of people in the USA support access to abortion, its the fact that a minority court was packed with gerrymandered districts and a political party dead set on ending democracy that we got here.
957.
▲
by
hobs
3y ago
Its a common thread with the expectations of "Well I have gotten everything for free up to this point, why don't I get all your time and energy for free too?" I love contributing to OSS but I don't even contribute to MIT
958.
▲
by
hobs
3y ago
A fine on one company for screwing over another company is not "making money from big tech", really low effort comment here.
959.
▲
by
hobs
3y ago
Absolutely, and sometimes labeling does push or pressure corporations, but I just see that the main problem is not labeling the poison we're eating, its making it illegal in the first place.
960.
▲
by
hobs
3y ago
Competition would be nice, getting a label doesn't do shit if you have no choice to change brands.
More ›