argee

Does this help stable Safari in any way? I have a WebKit bug I’m tracking that’s fixed in STP but not in stable and it seems like there’s no telling whether it’ll ever land.

I've already had to add code to "normalize" addresses with dots and plusses. That should've been my first sign to move to a different design.

In general it's bad practice for a user to expose someone else's email address to an application provider without consent

Thanks, that makes sense as a rule of thumb. I'll move towards invite code (and work domain/SSO) based access. It's also a bit awkward to be able to add existing users to a workspace without any sort of confirmation, so my system needs a bit of rework regardless.

I have a question I wasn’t sure how to ask HN that is tangentially related to this. I’m the developer of an app where people can add other users to their workspaces, via email. The way this currently works is that your email gets added to the workspace and then when you log in with that email, you get access.

Now, this of course fails with “hide my email”. What’s the appropriate/standard way to handle this? Should I be using urls with invite codes instead of associating the email itself with access? I’m a little hesitant to do that since the code would have to persist in session and the business logic would have to occur on the login portal, since the mobile/desktop apps don’t share web cookies. Is deep linking the answer?

I’ve helped someone do this in a pinch for a visa, but the guidelines explicitly forbade using an AI, including to touch up photos. As the rules catch up, you’re going to be doing this kind of thing at your own risk. I can’t imagine governments liking it.

That’s the problem. If all (or zero) models were trained on it, it would be fine as a benchmark.

Is the model "thinking of the last word for the next line already" or is "the word that came last in the second line was already showing high p from considering only the first line"? Shouldn't we care about such a distinction?

This type of person is extremely annoying and comes across as both desperate and incompetent, don’t follow this advice. The phrasing of these "voicemails" reminds me of the time I purchased an expensive piece of equipment specifically through eBay because I didn’t quite trust the buyer, and then was bombarded by such a pity parade after the equipment didn’t quite work (the delivery was another story of astounding incompetence) that I cancelled my eBay return request out of a combination of disgust and the desire to never interact with that person again.

Besides, "I want to be rich, help me" is only going to attract bottom feeders, who will then proceed to feed — on you.

<marquee> is deprecated

For some reason the Indian government LOVES marquee, to the point where it's almost a hallmark. Looks like marquee...finds a way.

[0] https://www.mygov.in/

[1] https://www.mea.gov.in/

[2] https://ociservices.gov.in/onlineOCI/

[3] https://www.passportindia.gov.in/psp

One week. Three senior engineers. $10,000. We commit to a reduction target up front, and you pay in proportion to how much of it we hit.

Commitment ain't what it used to be.

Sorry, I chose to use the same terminology as the comment I replied to. Should I have started with the explanation you’ve posted, would that make my comment better?

You can read about why Ian made the new knot as an improvement over secure knots specifically here. [0] But hey, maybe Ian was unknowingly tying a Granny knot instead. ;)

    Comparison To Other Regular Shoelace Knots
    
    It was much easier to prove that Ian's Secure Shoelace Knot is more secure than any of the regular shoelace knots. Using a pair of shoes with round, slippery laces, I tied one with my Ian Knot and the other with my Ian's Secure Shoelace Knot. Despite tying both to approximately the same average tightness, the Ian Knot came untied two or three times a week whereas the Ian's Secure Shoelace Knot never came untied.

If you have round laces, Ian’s knot reliably comes undone. Ian’s secure shoelace knot is better in that case, though it’s slightly slower and trickier to do. The double Ian knot is interesting too, but it (and several other alternatives) won’t have the “pull just once to untie” mechanic.

We care indirectly through cost. Hydroelectric, solar, or wind power are often among the cheapest electricity sources, for example. Beyond that, no we don't care. That's why if people want change we leverage policy on cost, via subsidies, surcharges, taxes, tariffs, what have you.

To a consumer, an amp remains an amp — so they get the cheap one.

I just saw this video today [0], and I was thinking…this girl must be made of tokens if this is how she works.

[0] https://youtu.be/uGwDuvSqgYI?is=CRrIT31HITipIHXi

Indeed, as it gets more commoditized it feels more like swapping electricity providers. Who cares whether you get your electricity from IBM or the state of Texas? An amp is an amp.

Try it without "reasoning". As you can see in my example (and GP), it meanders to correctness eventually after emphatically being wrong, and most reasoning modes hide that from you.

If LLMs worked the way people want to believe they do, there’d be no reason to start in the wrong place — a computer should have the facts!

This repros nearly 100% of the time on most LLMs, even the most advanced ones: https://share.gemini.google/u9NwYu7lbgxe

instead is like a stupid database of all world knowledge with the appearance of intelligence.

A "stupid" database would be better, based on what I get when I ask whether all of Oregon state is North of New York City. Indian English has a word for it: oversmart.

Putting in at least as much effort writing as your audience will reading

This is the big one.

Ideally not sharing generated content if you can't do the above. If you must, then explicitly disclaiming your use of AI

This is the reason why people get mad about AI generated open-source PRs and repositories. Rather than contributing thoughtfully to the commons, you make it a dumping ground when you do this.

That first Twitter testimonial they’ve clipped and featured (with # likes) is remarkable. They made a positive testimonial of Cursor that of course got retweeted by cursor and then quoted that retweet as if it were a testimonial by Cursor for themselves.

I’m a gamer, #1 is not my dream. Games, as with any other work of art, are also an exercise in curation on the part of the developers. Without that filter, and that common experience with other players, I might as well scroll reels and get an equivalent experience.

I barely knew CakePHP, so every file looked wrong to me. I knew Laravel, and I loved working in it. So on my own time, unasked, I ported the thing piece by piece, swapping in Illuminate packages until the old framework was gone.

Title is somewhat clickbaity, because these actions are not what people are going to think of when they think of a rewrite in a corporate environment.

By the way, I was an early employee at Amazon.com and the website’s rendering engine was rewritten three times, each time unlocking a new level of productivity without which the site (as well as the velocity of teams) would have slowed to a crawl, and the company would have probably died.

I guarantee you've thought more about this than I have, but the first impression I had of the "second chance" pool was that it would essentially be a repost of the top-level post and not the comment threads. I think part of the reason people bring it up is because they see the same post PLUS the same comments with new timestamps and feel disoriented.

Strange, it opens just fine...as long as you aren't logged in (to X).

So, why now? Emojis have existed for a while, haven't they? Whereas Craigslist has been using images up until now (at least for the top nav items).

unnecessary visual distraction

I think I personally see emojis used in this manner as unnecessary visual distraction, because it detracts from whatever self-consistent design system you had going on (when used for high visibility items like front page headings). Emojis don't even render the same on every platform, so its a move that dilutes your design language.

Even if it's a useful visual guide, I would wager nine times out of ten you'd be better off with a self-consistent icon set...depending on what you're going for, of course.

Nice sleuthing! So it's quite a recent change. It stands out to me that the items up top (faves, acct, etc.) which already had icons also got emojified.

Prior to that, they had already introduced emojis to draw attention to their Craigslist charitable fund: that probably made this a much easier decision.

I wouldn't find it remarkable anywhere else, but Craigslist has built a reputation on not doing this kind of thing.

Almost sounds like a bunch of people crowdsourced a better textbook than was otherwise available to them. If your database is truly "secret" how are people going to add to it or benefit from it?