I mean, they had to somehow procure stock from startups that are long out of business. Items nobody actually wants, so what's the point of having _more_ items nobody actually wants?
(EDIT: oh. whoops. they're miniatures haha)
@gkoberger
From 11/3/2008, 9:01:56 AM till now, @gkoberger has achieved 20060 Karma Points with the contribution count of 1765.
Recent @gkoberger Activity
I'm not a fan of scrum but.. wow this is a horrible argument.
There's so many things that aren't backed up with any facts. It's just a list of random Bad Things™, and somehow scrum gets blamed for it? I have no clue how developer mentorship being nonexistent, for example, has anything to do with scrum.
Hey, it's tough. Engineers need both autonomy and guidance to get things done, and the balance isn't easy to find. But just throwing false accusations like wage suppression doesn't fix the problem... I think the author just needs to find a new job.
From the README: https://6b.eleuther.ai/
I really appreciate putting the effort in on any open source project, however GPT-J and GPT-3 are not even close.
I couldn't figure out any prompt that resulted in GPT-J returning anything that wasn't gibberish, whereas GPT-3 is almost always within the ballpark.
They aren't banning ads for vegetables or gym memberships. They're just banning what I assume are mostly scams and fake supplements.
He almost definitely tortured and killed his neighbor. He never stood trial because he fled the country and has been running since.
He isn't tech's Hunter S. Thompson, he was a lunatic who hurt people. Even if you don't believe he (or his bodyguard) killed the neighbor, there's countless other stories where he hurt people (including his crypto scams).
He was found guilty in civil court in Florida and forced to pay $25M; he never stood trial because he fled.
Also, presumption of innocence applies to the law, not to HN comments.
I feel like a lot of people in this thread are forgetting he likely killed his neighbor in 2012 and has been on the run since.
I mean, sure he built some antivirus software a few decades ago, but I don't think he deserves the eulogies he's getting here. He's not tech's Hunter S. Thompson; he's a murderer who spent the last years of his life screwing over people with crypto scams.
I mean, he very likely killed his neighbor in 2012... so I feel like "colorful" undermines how much of a monster he's been for over a decade.
How to write API errors that keep your users moving forward
2 points • 1 comments
This is great. A lot of my favorite people don't self-promote interviews/articles/podcasts/etc on Twitter, so I love this.
I think it's a grass-is-always-greener thing?
Early in your career, you lean into one or the other. And then years later, after you're confident you're right, you find yourself trying the opposite paradigm and liking things about it.
Both have pros and cons, and if there was a correct answer we'd all just go with that one!
I disagree with the above advice (as an employee, you should never ever state the number first), however I think the assumption behind this advice would be that you the company believes it's competing against other companies for you. So, they wouldn't do the lowest because they want to outbid the competition, and the higher number acts as an anchor.
But again, as a candidate, never say the number first.
From the man page, Gnu Stow is something completely different. It just manages symlinks locally.
Plus, it seems to require a good amount of understanding to use. I can't imagine asking new engineers to set this up on their first day.
I think it would make a lot more sense if you could just do multiples of the base amount, rather than arbitrary breakdowns.
This sounds employee-friendly, but it's total BS.
You'll just be getting less equity at a higher strike price every year, so it's just a sneaky way for these companies to give employees less. They can still say "we're giving you $100k in stock this year", but it's a lot less stock since you're not locked into a strike price.
If you want to leave after 1 year (post-cliff), you can leave under either scheme and get 1 year's worth of equity. This doesn't solve any "golden handcuff" problems, it just hurts employees.
I think it's a much smaller signaling risk, especially given how much time has passed and the number of founders (9 total, including Musk).
The Paypal Mafia tends to structure their companies this way... a large, tiered system of founders and junior founders. This is more like losing an early team member than, for example, a Collison.
(Also, it's ironic you cited OpenAI, since it did lose a founder... Musk)
I don't think that's what it is. It's a bit hard to get an invite (although allegedly there's 10 million users now, so not that hard), but once you're inside, everything is open and free and accessible to anyone.
You could argue that things being realtime cause FOMO, but that's true about conversations. Everything is so "on-demand" now; there's something nice about the ephemeral, realtime nature of it. It encourages spontaneity.
This article makes me wonder if we'll ever see another social network ever again. Our opinions on privacy and security have shifted dramatically (for the better, in my opinion), however imagine an author writing about Facebook a year in under a 2021 lens.
Facebook (and others) grew because a lack of privacy and security was the goal. Images were public and tagged, conversations were viewable by anyone, the minifeed made it easy to see every change people made, etc. It was exclusive (college only) but also very public. The high amount of content and low amount of friction made it spread virally. Now it's much more locked down, but that's okay... it already hit a critical mass.
The current consensus about privacy and security seem to be at odds with what's necessary to make a social network work. The ability to connect with anyone at any time with no friction is what makes social networks thrive, but also what enables harassment and abuse.
I think it's likely we've already seen the last great social network.
If you only talk about the tech behind it, then you're correct. However there is a scarcity of time and attention, which is what Clubhouse aims to capture. People want to be where things are happening. That's why land is abundant, yet I pay $2600 for a one-bedroom apartment in a city.
I've been using Clubhouse for a year, and seen various iterations. Earlier, there was so much human connection. I talked to amazing people almost 1:1, and it was like being part of a small club. As it grew, it became more like a 24/7 conference where you could drop in to hear any topic you want. Less my thing, but still fascinating.
site design / logo © 2021 Box Piper