Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
garblegarble
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
garblegarble
3y ago
No, I can't think of any case where that would be useful (although I do take umbrage with a local datetime being referred to as a 'naive' type - it serves a very specific purpose as a primitive... if it's naive then so i
92.
▲
by
garblegarble
3y ago
>2. That’s why you WANT time zones in this example. It will automatically stay at 10am instead of bouncing around when DST rules change. Respectfully, no - in my scenario, if I used a timezone then I would show up at the wrong time for m
93.
▲
by
garblegarble
3y ago
Here are two counter examples that sprang to mind: 1. I've set my alarm for 8am local time every day. I don't want my alarm clock to go off at 3am just because I'm in a different country 2. If I have an appointment at 10am ne
94.
▲
by
garblegarble
3y ago
It's going to be obvious to other players that you're wearing a camera, and who would want to stay in a game where it looks like another person is cheating? Making other players leave seems like it would be the fastest way to get
95.
▲
by
garblegarble
3y ago
I was shocked by this... and here I thought I'd seen enough incompetent processing of datetimes that nothing could surprise me.
96.
▲
by
garblegarble
3y ago
I've found rinse aids that include citric acid end up pitting stainless steel cutlery over time, so I avoid them
97.
▲
by
garblegarble
3y ago
Are you sure that's not a half-remembered version of Russian propaganda[1], where they tried to blame NASA Astronaut Serena Auñón-Chancellor with drilling holes into a Russian spacecraft, ostensibly so she could get sent back to earth
98.
▲
by
garblegarble
3y ago
This may only be applicable to certain CPUs - e.g. sha512 is a lot slower on M1 $ openssl speed sha256 sha512 type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes 16384 bytes sha256 146206.63
99.
▲
by
garblegarble
3y ago
A lot of that is essential complexity of the problem space - media is HARD: so many different specs, each one interpreted differently by each vendor. In case you're not aware, though, there has been work on improving the CLI - it was d
100.
▲
by
garblegarble
3y ago
This does the trick quite nicely (unless you want them textured, anyway) - https://github.com/fzwoch/bsp2obj I tried it on the map BSPs from https://github.com/fzwoch/quake_map_source/tree
101.
▲
by
garblegarble
3y ago
bsky-social-lkzsp-7x7ja bsky-social-p4vwr-nrthu bsky-social-bdu6c-6tbv4 bsky-social-fkpgk-oestw
102.
▲
by
garblegarble
3y ago
I notice in your coverage plot, the UK National Grid data appears to end mid-2023... have they stopped providing this data?
103.
▲
by
garblegarble
3y ago
That seems like the Right Way to do it... As annoying as it is to have to update every sudo reference -> doas, it forces you to think about everywhere you're using it, rather than waiting to see what breaks and then trying to fix it
104.
▲
by
garblegarble
3y ago
The text seems to imply that's not a list of features not implemented, it's a list of features not implemented that don't output a clear error
105.
▲
by
garblegarble
3y ago
For sure! But that mistake has already been made, and has been in the wild for years, so removing those features (and proposing yourself as a replacement for the original) is now a breaking change
106.
▲
by
garblegarble
3y ago
I'm assuming this project's aim is to replace sudo, in which case hand-waving away "Leaving out less commonly used features" is a bit worrying. What are these features? How uncommonly are they used? In which way will it
107.
▲
by
garblegarble
3y ago
A major benefit (for me) is that I can build it with "make" and know that it just works. Now perhaps this is a skills issue on my part (because I'm not a Python dev), but I've had endless trouble with Python-based ML pro
108.
▲
by
garblegarble
3y ago
>This may be less pain overall (much like making many small pull requests vs one giant rewrite). Just shows how different people are in their approaches to dependencies... I can't imagine working like that, the idea that everything
109.
▲
by
garblegarble
3y ago
>Yaml also doesn't have any proper ide support, so it's an annoying loop of restarting the app and staring at logs looking for typos. Since YAML is a superset of JSON, I highly recommend converting[1] it to JSON, and then just
110.
▲
by
garblegarble
3y ago
I agree with you about the feeling of helplessness on the child labour in chocolate episode, but John Oliver covers such a wide range of other topics, many of which are much closer to home (and therefore something you have a chance of being
111.
▲
by
garblegarble
3y ago
The (sometimes) unwritten part of all of those demands is a reasonably priced ad-free version. Down the thread somebody stated an annual revenue of $18 per EU user, so they want to make an order of magnitude more revenue on you in order t
112.
▲
by
garblegarble
3y ago
Wouldn't you also need to factor in that the people who will pay to avoid ads are likely already using ad blockers (or hate ads, in which case showing them an ad for your product is likely counter-productive)?
113.
▲
by
garblegarble
3y ago
>Most researchers aren't there to make a commercial product. They're there to publish well-cited research papers, talk at top conferences, and have access to far more compute than anyone else can offer. I think that's core
114.
▲
by
garblegarble
3y ago
I can't understand the mindset of any AI researchers who stay at Google at this point. You're there (presumably) because you want to make an impact. You gave them a massive lead with a novel approach... and Google failed to capita
115.
▲
by
garblegarble
3y ago
>You own what you produce with it too As I understand it, this is incorrect - all of these outputs are the creations of a machine, and as such are not eligible for copyright protection
116.
▲
by
garblegarble
3y ago
Yep, absolutely terrifying! There's always the old chestnut that if your country doesn't develop it, another country (or well-funded faction) with fewer scruples will, and if they think that their tech gives them a significant adv
117.
▲
by
garblegarble
3y ago
I'm no expert, but aren't rocket launchers recoilless? As for why the dog form factor, I'm betting it'll be because a dog form factor makes for a much cheaper swarm than a tank form factor. Less mass to move about so a s
118.
▲
by
garblegarble
3y ago
No worries, confirmation bias comes for us all at some point!
119.
▲
by
garblegarble
3y ago
Do you have more info on that? Their support site[1] claims their data collection is much more constrained than that (although still seems like a lot). 1: https://nvidia.custhelp.com/app/answers/detail/a_id&#x
120.
▲
by
garblegarble
3y ago
I too thought this when I was 26, I'm 36 now. The world won't stop you, but you may also find you just don't care enough to - you have to be spending a lot of time interactively engaging in a culture to be native to the memes
More ›