Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
blinkbat
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
61.
▲
by
blinkbat
3mo ago
eh, the part about "the only thing between prod and an off-script model is a human reading a diff" -- you realize that humans are also very often off-script, and reading the diff is what we've done for many years as oversight
62.
▲
by
blinkbat
3mo ago
I don't think token usage is a good metric to gamify people into burning through.
63.
▲
by
blinkbat
3mo ago
it's really funny when people advise the fix for widespread sociological problems as "just become an entrepreneur and disrupt an industry bro"
64.
▲
by
blinkbat
3mo ago
my best advice to you is to ditch python at this juncture and focus on a nice, fast, static language.
65.
▲
by
blinkbat
3mo ago
we're so near and literally beyond that threshold
66.
▲
by
blinkbat
3mo ago
it's funny when a site posted on a tech board goes down from traffic. it's funnier when the site is making some kind of principled stance about the power of handmade tools, which presumably made the site, which went down.
67.
▲
by
blinkbat
3mo ago
it was handing me 502s for a minute.
68.
▲
by
blinkbat
3mo ago
I hope ai generated images in articles never becomes the norm
69.
▲
by
blinkbat
3mo ago
what am i to infer from a higher or lower "zaira score"?
70.
▲
by
blinkbat
3mo ago
this isn't even true. you know this isn't true, right? I find it really odd that people would consider current text-gen to be even above average writing. it's not...
71.
▲
by
blinkbat
3mo ago
Cost.
72.
▲
by
blinkbat
3mo ago
Nothing... keep writing, who cares? It's your art
73.
▲
by
blinkbat
3mo ago
Big companies reward being nice over being useful.
74.
▲
by
blinkbat
3mo ago
No
75.
▲
by
blinkbat
3mo ago
Code yes. Art no.
76.
▲
by
blinkbat
3mo ago
what, exactly, were the failures? coding assistants are not visual-first, they are code-first. so it makes sense that they excel more at coding.
77.
▲
by
blinkbat
3mo ago
I don't find that the models are so capable that the workflow can be "much less about prompts". you still need to be in the loop, prompting and validating, to get good output. people will say otherwise -- I say those people a
78.
▲
by
blinkbat
3mo ago
the "in-app purchases" label is reeeeally not worth the in-game tip jar, IMO. move the tip jar out of the game and remove the label. the label is enough to turn many people off (like me).
79.
▲
by
blinkbat
3mo ago
put a big red banner up top on the blog -- "This post was written by AI!" and also tag your HN posts as such.
80.
▲
Ask HN: Why submit your AI-written article?
8 points
by
blinkbat
3mo ago
|
9 comments
81.
▲
by
blinkbat
3mo ago
no desire to read a post written by AI
82.
▲
by
blinkbat
3mo ago
I use it for work, is that productive?
83.
▲
by
blinkbat
3mo ago
"The software should have fits-in-head complexity. Stop with the 500 NPM packages and random dependencies for everything. Its terrible for your supply chain security, so stop it!" program complexity is not only measured in package
84.
▲
by
blinkbat
3mo ago
First thought was this sounds like a hyperlocal sf gripe
85.
▲
by
blinkbat
3mo ago
serving the engineer _does_ serve the business, ultimately.
86.
▲
by
blinkbat
3mo ago
"I listen to podcasts at 3.7x." tell me you're crazy without telling me...
87.
▲
by
blinkbat
4mo ago
nice family. as for the product, I'd never use it - I don't find reading ingredients that difficult.
88.
▲
by
blinkbat
4mo ago
HAHAHAHAHHAA
89.
▲
by
blinkbat
4mo ago
wouldn't these small groups just like... send an email? group chat?
90.
▲
by
blinkbat
4mo ago
you agree?
More ›