spondyl

Hah! I had watched this just last night. I have a Fisher & Paykel Dishdrawer so this prompted me to check the instruction manual and sure enough, I had been putting Rinse Aid in the pre-wash area. I don't even really know what Rinse Aid is honestly but it's fun having some things be a black box. Turns out the correct spot is turning a knob, pulling it out and pouring it down a hole containing a glowing red light. I had assumed there was just some sort of circuitry down there and doing so would be a horrible idea. Thanks Technology Connections!

Another, not incompatible explanation is that it's also just easier to develop for a handful of known iOS/iPadOS targets compared to Android's unbounded set of screen sizes and device specs.

Commit Mono apparently: https://kau.sh/blog/commit-mono/

The fairly popular Renovate bot provided by Mend is being renamed to just "Mend" shortly. The timeline for this doesn't seem set in stone but some users have raised concern that some workflows may break if you make assumptions about the name of the Github app being used.

Thanks for the reminder. I went to reset my password when the email went out but when following the reset flow, I hit a Cloudflare page (due to the origin presumably having crashed) and got sidetracked

Any plans for the HN API to expose highlights? It's a neat high signal collection that I have some custom bit of code that translates it into an RSS feed but an official endpoint would be neat :)

Not to beat on frontend but I vaguely remember that the UI was rebuilt with React a couple of years back and the other day while browsing a sale, specifically hovering over Final Fantasy XIII would crash an entire sales widget, printing out a React error. I wonder if this is just a case of framework bloat getting out of hand (again)

Can you... talk to a human for support? Perhaps I'm just used to SSO tax billing pages where I expect the right most column to mention that. I was partly expecting it because it'd be ironic to see people complaining that a model hallucinated only for some engineer at Google to shrug and be like "Nothing we can do about it"

> Not true, you can compete with the quality of the deployed service _separate_ from the development of the software.

That is true in a literal sense but (anecdotally) from the point of view of an engineer deploying ie; Redis, there is no real space for that angle of competition when the choice is between having to go through procurement hurdles to sign a contract with Redis Labs versus say; spinning up Redis on AWS which has zero hurdles because there is already an organization wide agreement in place for example.

The competition there isn't a level playing field as the deck is kind of stacked for most businesses where engineers don't have free reign to procurement what might be objectively the best hosting solution?

I've never really thought through this in any depth before now so don't consider this a great fleshed out argument, just an observation from my personal experience.

As a user myself, I'm not sure. Assistant is a separate view so it's easy to never interact with it but yeah, I can see why that might be a work policy issue. You could try requesting that as a feature in https://kagifeedback.org. The team are quite responsive there and historically Vlad the CEO reads every post.

> As a consumer of Atom feeds I find whole content feeds like that quite convenient, provided they don't get into the thousands of entries.

Atom does have an RFC for supporting pagination so it is possible (client support provided) to have both a short feed as well as full content history: https://www.rfc-editor.org/rfc/rfc5005#section-3

For Wordpress sites, `?paged=X` can be used to paginate through RSS feeds, although that's more of a Wordpress standard than an RSS/Atom standard: https://codex.wordpress.org/Pagination

My personal gripe is that clients checking for query params would seem to rule out pagination of static feeds (so you'd end up shipping a single huge response to the client) but now that I look closer, I guess you could achieve it via Atom, you'd just end up having to reshuffle every feed whenever pages get updated.

I went through some of this recently when standing up some home software to generate RSS feeds for sites that don't have them :)

I came here to say this too. I assume the developer only uses dark mode and didn't test light mode.

> I know some trading firms use lispy languages and some game studios (Naughty Dog?) also use them.

Naughty Dog had somewhat famously written Crash Bandicoot and Jak and Daxter using Lisp although they don't appear to use it these days

https://en.wikipedia.org/wiki/Game_Oriented_Assembly_Lisp

Ah, the game I was referring to is https://en.wikipedia.org/wiki/The_Wonderful_End_of_the_World

The poster is Ichiro Lambe of https://en.wikipedia.org/wiki/Dejobaan_Games

Oh damn, even Ichiro lurks here!

To this day, I still replay TWEotW once every year or two :)

> if you were going to slip something unpopular into a foss project policy, wouldn’t you want the public updates to be vague enough to not technically be lying about anything while also being plausibly innocuous?

If a company was being competently "evil"[1] then it probably would look like this!

Personally, I don't know that I consider Mozilla competent enough to reach that bar though given a lot of their previous blunders seemed like, well, blunders and not finely crafted acts of trickery.

> that seems like the kind of thing they’d clear up pretty easily if it were that innocuous

I suppose in this scenario, if it were innocuous and this is just some automated mirroring thing that someone triggered without realising the optics of, I wouldn't then automatically assume that same organisation would have a level of coordination to recognise or put out a blanket statement about the issue?

I mean, you'd think surely some amount of Firefox/Mozilla folks are very online and this would be raised internally but if this is downstream of some process owned by mostly legal and non-internet/chat using folks, it might make sense that they a) take some time to be notified, b) take some time to realise a lot of the internet is saying "What the fuck" and c) take a while to figure out what to do about it (ie; issue a press release to not make it worse? someone higher up acks and is like reverse whatever this mess is?)

My only real basis for all this is I've occasionally run into some compliance/legal types in tech and they can have extremely bad mental models of the company and product they work for so I can feasibly believe this all being accidental but in saying that, I dunno who works for Mozilla and this is very much a stereotype I'm applying.

Anyway, as above, I'm not saying this isn't malicious, just that personally I think the door is still open that this could all be a complete mess that has no real intent behind it

I don't have any special insight on these events but I've seen that commit linked a lot.

From a pure optics point of view, it looks extremely bad but reading some of the comments, it sounds like that repo is basically a mirror of some upstream project where terms and conditions/faqs etc are stored as pseudo-structured data and that they're migrating parts of that repo to some other project?

I feel like a lot of context is missing publicly but if I squint, that seems to be what this comment is trying to express: https://github.com/mozilla/bedrock/commit/d459addab846d8144b...

Don't get me wrong, I love bagging on Mozilla as much as the next person but as extremely bad as it looks, I'm not convinced that it's literally what it looks like?

I don't think it's the tool that's the issue, I think it's the book itself?

If you search the ISBN on the web, you'll get "Zero to One Hundred" with the cover of "Built to Grow" and vice versa.

There's also "Experiment, Build, Scale" which is the book that the visualisation shows, also with the same ISBN attributed to the previous two.

Experiment, Build, Scale seems to be the only book of Stephanie's that is in Google Books while Worldcat has "Zero to One Hundred" with the cover art for "Built to Grow".

Most of the online bookstore pages have this mess so I wouldn't blame the tool for what seems like an upstream data quality issue.

I've been pretty unloving to my various Kobo's and Boox devices without any major issues so I guess I'm just not used to it

I really enjoy my DC-1 but after a quick skim, I didn't notice any mention of the screen's ability to scratch.

I took mine in a backpack up to my parents place and apparently something lightly caused a scratch on the screen so now I just have a permanent little gouge.

Thankfully I've learned to ignore it over time but yeah, don't assume it's as indestructible as a lot of screens.

The DC-1 does/did ship with a padded cover which makes me think it doubled as the engineering fix for when/if they realised that might be an issue.

To answer your question, starting with Pokemon Legends Arceus[1], battles take place in the same overworld that you explore in.

You can see an example of this in action at 3:12 in https://www.youtube.com/watch?v=DR7ZooH9Zos

[1]: This also carries over to Pokemon Scarlet/Violet, which is the latest mainline entry and will presumably remain the new standard going forward? Arceus was seen (and received) as more of a paid beta test for the system since the overworld was a bit empty at times

Previous submissions on the topic, although none were picked up:

https://news.ycombinator.com/item?id=42854927 https://news.ycombinator.com/item?id=43024328 https://news.ycombinator.com/item?id=42996300

This is probably the type of ad you ran into: https://www.404media.co/deepfake-youtube-ads-of-celebrities-...

There is a compilation of games (and a sequel) that are loosely based on those mobile ads available on Steam: https://store.steampowered.com/app/2744230/YEAH_YOU_WANT_THO...

It's called "YEAH! YOU WANT "THOSE GAMES," RIGHT? SO HERE YOU GO! NOW, LET'S SEE YOU CLEAR THEM!"

> House Republicans say they have reason to believe the IRS and Department of Justice are monitoring millions of Americans' private transactions, bank accounts and financial information without legal process using an AI-powered system.

It's worth reading the actual letter that is linked. While the IRS themselves have announced in press releases that they are using AI (and/or machine learning), the citation for the concerns seems to be a Project Veritas sting video featuring one guy verbally claiming these things, presumably without knowing he is on the record.

I dunno that it's a particularly high bar as far as sourcing going but anything is possible of course.

Anthony Bourdain supposedly found the best restaurants by using this trick

> Get online and write something along the lines of, “hey guys, I just had the absolute best chicken rice at [restaurant x] in Singapore, no questions asked, hands down, everything else pales in comparison,” then sit back and enjoy the show as the internet foodie elite each jump into the fray to defend their own picks to the death. You’ll get a much bigger response, more passionate praise, and it’ll probably end up being a little fun to boot.

https://au.lifehacker.com/news/58784/find-the-best-food-in-a...

I know it's a marketing site so the speed is for the sake of the demo but the various animations showing a cursor scrolling past reams of text quicker than any human can scan, and just blindly hitting Accept makes for weird optics.

The idea is great and is essentially the same premise as Beeminder (no relation, just a user)

Instead of pledging an amount up front, you instead provide card details and get charged if you miss your goal.

Once you miss the goal, the pledge multiplies exponentially until you're sufficiently incentivised to do whatever the goal is.

They sound wacky at first but motivation devices are pretty effective, assuming you trust yourself to not game them of course.

Interestingly, The Journal (a podcast from the Wall Street Journal) ran an episode with Anthropic's AI safety team just yesterday.

I had wondered if it was perhaps a PR push from Anthropic to make their safety people available to the press but it was probably just an adaption of an earlier WSJ written piece I wasn't aware of.

https://www.wsj.com/tech/ai/ai-safety-testing-red-team-anthr...

As a bit of anecdata, I used an LLM to refactor a core part of one of my site projects about 6 months or so ago.

I don't really remember how it works, probably because I didn't have to think hard about the design, I just sort of iterated it in conversation which probably meant it didn't stick.

While the iteration part was mostly positive, I don't know that it was worth it in the end to be left with something that I can understand but feels a bit foreign at the same time.