yifanl

15 ms·

271.
by yifanl 2y ago
You must have an incredible lack of imagination, the story of software is the story of mayflies.
272.
by yifanl 2y ago
It's fiction, but it's a depiction of a society that's amoral of technology to the point of immorality. A world where any technology that might be slightly be useful becomes used up of every bit of profit that can extracted a
273.
by yifanl 2y ago
Seems obvious to me that once the incentive is made that maximizing views at all cost occurs, the platform will (slowly or quickly) eventually cater only to the types of people most eager to provide views. Are they okay or not isn't ev
274.
by yifanl 2y ago
You can rather easily set up semi-hard rate limiting with a proof of work scheme. Will very trivially affect human users, while bot spammers have to eat up the cost of a million hash reversions per hour or whatever.
275.
by yifanl 2y ago
Pascal's Wager is a terminating line of thought, adding a tech coat of paint doesn't make it any more worth discussing.
276.
by yifanl 2y ago
Things would be substantially more productive if boardroom meetings allowed more definitive responses than "We'll circle back to that later". This is the other extreme.
277.
by yifanl 2y ago
What privacy/security features make this meaningfully different than Microsoft's offering?
278.
by yifanl 2y ago
From a practical perspective, its easier to demand a US-based company to break non-US laws.
279.
by yifanl 2y ago
Anything can be anything else tomorrow, which is why we must criticize what's in front of us today.
280.
by yifanl 2y ago
I would likely be just as put off by a picture of Spongebob or Goofy or Goku in a readme as Anya, fwiw.
281.
by yifanl 2y ago
I would be rather embarrassed to spend $100 billion dollars creating a new feature that's not better than the legacy feature. (Notably, this isn't even a rewrite, this is greenfield work!)
282.
by yifanl 2y ago
discussed 20 years ago https://ansuz.sooke.bc.ca/entry/23 > I think Colour is what the designers of Monolith are trying to challenge, although I'm afraid I think their understanding of the issues is superficial
283.
by yifanl 2y ago
Arguably the best game of 2023, Alan Wake 2: https://wccftech.com/alan-wake-ii-recoup-expenses-tencent
284.
by yifanl 2y ago
Curiously, linked discussion has been downranked to Page 19 at time of posting.
285.
by yifanl 2y ago
https://www.nytimes.com/subscription/cooking.html You'll have to scroll down (and visit each individual page) to get to the separate subscription plans, because dark patterns, but it is still available as an optio
286.
by yifanl 2y ago
> Also, and I hate to say this because nobody likes this model, but what if news took on a cable TV model where you pay $20/month and subscribe to a few different pubs, maybe even allowing you to make your own bundles? You're d
287.
by yifanl 2y ago
Past performance is not indicative of future returns.
288.
by yifanl 2y ago
What does a community crisis mean exactly? If every member of the community wants one thing, and the leadership doesn't, then the community wins, every time.
289.
by yifanl 2y ago
Correct. If you're offering me money to link you, then presumably you believe your content is not of high enough quality for me to link you otherwise.
290.
by yifanl 3y ago
Replying to myself, after re-reading from a different perspective, I wanna walk back on "uninteresting". At first I was expecting more out of why Gomes was kicked, but I realized I answered myself. And the choice for Raghavan spec
291.
by yifanl 3y ago
Err, I'll walk it back a little. Corporate decisions are just people's decisions, and though it's probably not just Raghavan, it was _somebody_'s decision to have Google spam our homepages. Maybe we just need to be bette
292.
by yifanl 3y ago
Oh, yes, I agree that we should name and shame, but I suspect the list of names to be a lot longer than one or two people.
293.
by yifanl 3y ago
> unrealistic expectations From my (admittedly limited!) experience, unrealistic expectations are often set only when you want to push a senior off the team, and then as the new exec comes in, they'll "re-evaluate" the tra
294.
by yifanl 3y ago
I disagree, because this ends up with implying that if you just got rid of That One Fucking Guy, then everything with Google Search would be good. Which... is not a claim I'd agree with without extremely convincing evidence.
295.
by yifanl 3y ago
I think the blogpost spend a lot of time focusing on the uninteresting part. If it wasn't Raghavan, it'd just be someone else, Google (the corporation) wanted more search query metrics and Google is large enough to enforce its wil
296.
by yifanl 3y ago
One of the differences is that humans are very good at not doing word associations if we think they don't exist, which makes us able to outperform LLMs even without a hundred billion dollars worth of hardware strapped into our skulls.
297.
by yifanl 3y ago
If you're using it for tadks like generate templates and fill out boilerplate, they're great. As long as you arent trying to offload any decision making, LLMs are extremely efficient tools.
298.
by yifanl 3y ago
I'd argue that none of the output is useful, even if some proportion may be factually correct, if you're using LLMs as a decision-making tool.
299.
by yifanl 3y ago
> I want architecture students to see that the flexible, modular, all-purpose and all-choices box of a room isn’t always what’s called for. It sounds right — surely your client wants a space that could be anything you need it to be — but
300.
by yifanl 3y ago
I guess, the broad definition is just someone who's good at selling their thoughts to other people. Maybe it's because they have a popular podcast, or maybe they were a former head of state. There's no general reason I should

More ›