Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
selcuka
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
30 ms
·
631.
▲
by
selcuka
1y ago
> There needs to be opposing forces among the elite or else we're sure to walk down the path of totalitarianism. This is false dichotomy. Our only options are not either a single consensus, or a balance between good and evil. We ca
632.
▲
by
selcuka
2y ago
Even the commercial product PKLite by PKWARE (of PKZIP fame) was "inspired" by LZEXE [1]: > If you look at the source code of the decompression engine of PKLITE, you'll notice that it looks like the one of LZEXE. [1] http
633.
▲
by
selcuka
2y ago
> "A picture of a schema"? What does this mean? A hand-drawn picture of a database schema. Something like this: https://www.dreamstime.com/stock-images-database-schema-imag... > And by SQL do you mean DDL? D
634.
▲
by
selcuka
2y ago
There are existing comparison services that keep track of prices and locations of grocery items. MCP is just the glue code.
635.
▲
by
selcuka
2y ago
Or you can simply do it using this follow-up prompt. No external tools are needed. Worked for me: Remove all non-visible whitespace characters such as 0x202F from the text. Alternatively, if you have access to the original prompt,
636.
▲
by
selcuka
2y ago
I stand corrected. Claude is enabled on my GitHub.com settings but it's not yet available to PyCharm, so I'm using GPT-4o as well. It's a known issue: https://github.com/orgs/community/discussions&#x
637.
▲
by
selcuka
2y ago
> he has all the luck which no one else has Well, the tagline of the book is "Inheritance. Fraud. Deceit.", so "Lucky" might not be the right word.
638.
▲
by
selcuka
2y ago
Interesting. I'm on Jetbrains+Copilot and I find that the autocomplete is useful 90% of the time. I'm using the Claude backend instead of the default (OpenAI). I wonder if that makes the difference.
639.
▲
by
selcuka
2y ago
This sounds like the perfect application of image processing and machine learning. A computer with a camera attached could monitor the whole operation and warn the surgeon if it detects an object going in but not out. At least it would be m
640.
▲
by
selcuka
2y ago
Fair point, but if it can cope with millions on "dirt cheap hardware" it logically follows that it can do 25M on more expensive hardware.
641.
▲
by
selcuka
2y ago
> Bugsink's also quite scalable[0], but I wouldn't recommend it a 25M/day. Well, your homepage disagrees with this statement: > Bugsink can deal with millions of events per day on dirt cheap hardware
642.
▲
by
selcuka
2y ago
> Since when did we accept, as a society, guilty until proven innocent? At least since the Digital Millennium Copyright Act.
643.
▲
by
selcuka
2y ago
Renewal date is April 24, so unlikely. Even if it expired this year they would still have a week to renew.
644.
▲
by
selcuka
2y ago
Maybe we like comments written closer to our style more, and that motivates us to respond to them.
645.
▲
by
selcuka
2y ago
The matching score is probably the same, or very close in both ways, but this fact does not necessarily help in a three-way scenario: A <-> B: 80% A <-> C: 90% B <-> C: 70% When you search for A the best m
646.
▲
by
selcuka
2y ago
I'm all for threatining them with a socialist revolution if possible. However, I'm afraid they are better prepared this time. In today's world a (metaphorical) guerilla war that targets one small win at a time might be more p
647.
▲
by
selcuka
2y ago
I'm not worrying about billionaires. I'm discussing about hypothetical ways we can tax them. They own the government, and obviously your idea of potentially making them homeless will be immediately rejected and we will be in this
648.
▲
by
selcuka
2y ago
He was a smart man: > After the second day, Jagger kept a clear head and decided that enough was enough, and left Monte Carlo to head home to Yorkshire with his winnings. > When he got home, he retired from the Mill and invested some
649.
▲
by
selcuka
2y ago
> He gets taxed on the value of his shares in year 1 and he gets taxed on the value of the shares in year 2. That doesn't make any sense. If I have $8B worth of shares and I have $2B in cash, and if the wealth tax is 20% I will have
650.
▲
by
selcuka
2y ago
Not sure about the reason for the personal attack, but my comment wasn't a reply to you. Michelin star restaurant was not even your analogy. I guess you are right, though. AI is useful for some things. At least it reads the prompt befo
651.
▲
by
selcuka
2y ago
> If they lie, straight to jail. How do we know whether they lie without a solid definition of net worth? I'm not defending billionaires and I believe they should be heavily taxed, and huge inheritances should be outlawed, but what&
652.
▲
by
selcuka
2y ago
The point is, for most projects there is nothing wrong with writing McDonald's level code. It's not as healthy or tasty as a homemade meal, but it will perform its main function (i.e. prevent you from starving).
653.
▲
by
selcuka
2y ago
> You're basically polling the consensus answer to the question you're asking. There is nothing wrong with going with the consensus answer. Sure, you can't invent an O(N) sorting algorithm with the consensus, but LLMs can
654.
▲
by
selcuka
2y ago
Butterflies are just a means for caterpillars to migrate.
655.
▲
by
selcuka
2y ago
They are doing it for the stable release as well: > The plan is to roll out to 5% of users on the Firefox 138 stable release, ramp up to 50% of users, then ship on all platforms in Firefox 140.
656.
▲
by
selcuka
2y ago
For the curious: https://screenrant.com/garfield-his-9-lives-primal-self-horr...
657.
▲
by
selcuka
2y ago
> simply because people use older versions, use different browsers, etc. Yes, but in this case asking the user to update to the latest version won't fix the problem either. I agree it's a terrible idea.
658.
▲
by
selcuka
2y ago
Wow. This is literally the definition of doublespeak. > The administration also accuses the [60] universities of racism through their diversity, equity, and inclusion programs [...]
659.
▲
by
selcuka
2y ago
Also from the same web site: https://newslttrs.com/yes-in-the-1980s-we-downloaded-games-f...
660.
▲
by
selcuka
2y ago
> within a week How do we know that Gemini 2.5 wasn't specifically trained or fine-tuned with the new questions? I don't buy that a new model could suddenly score 5 times better than the previous state-of-the-art models.
More ›