Hacker Newsnew | past | comments | ask | show | jobs | submit | svl's commentslogin

FWIW, it's reported in Dutch news - https://www.nrc.nl/nieuws/2026/01/26/dertigduizend-doden-sla... - with reference to this time.com article - https://time.com/7357635/more-than-30000-killed-in-iran-say-... - and a lot of caveats about how the figure can't be verified.


duckduckgo: https://noai.duckduckgo.com/?q=how+to+configure+arducopter+g...

Optionally with a custom CSS rule to block the starting video block: [data-layout="videos"]


Regular DDG doesn't seem much better than Google: https://duckduckgo.com/?q=how+to+configure+arducopter+gps

AI summary and videos dominating it.


AI summaries are actually rather good, in the general case. I know they take away traffic from source sites, but they work well for search users.

Videos, on the other hand, are a cancer.


> AI summaries are actually rather good, in the general case.

The google ones for me seem to be talking about something unrelated to what I'm looking for a significant percent of the time because it interprets it as a much more popular concept that looks like my query if you squint very hard. Other times it pulls up seo-spam tier answers that are plainly wrong.


Sadly they explicitly don't support that. :(

> Now that there is a Steam Deck Native build, is Baldur’s Gate 3 supported on Linux?

> Larian does not provide support for the Linux platform. The Steam Deck Native build is only supported on Steam Deck.


Of course they don’t, it would be crazy to say they would support all the different possible distros and configurations that people might run, when the majority of users are in steam deck. But that doesn’t mean it won’t run, just that if you have issues, they don’t promise to fix them. Seems reasonable to me.

BG3 already ran well enough on Linux, so I imagine this will only make it run better, official support or not.


I don't think they have to do that for a Steam Linux release. Steam has Linux native runtimes to provide games with a consistent environment across distros. https://github.com/ValveSoftware/steam-runtime/blob/master/R...


Ah, that’s true. Steam Linux is a smaller target than Linux in general. You’re right.


I'd expect it to work anyway. Under Steam at least. There is nothing special about the Steam Deck/SteamOS that's not available on other distros when running Steam, afaik.


Well, most game companies will only tell you the game only works on something very specific, say Ubuntu 24.04, and everything else is untested/unsupported. That doesn't exclude the game will work perfectly fine on other distros, which is usually the case.


No discussion about roundabouts can be complete without a mention of the (mostly) Dutch "turbo roundabout", where the lane you take going into a multi-lane roundabout depends on where you want to exit, and you can't / are not allowed to switch lanes while on it:

https://wiki.openstreetmap.org/wiki/Tag:roundabout%3Dturbo

https://en.wikipedia.org/wiki/Roundabout#Turbo_roundabouts

https://www.arcadis.com/en-us/knowledge-hub/blog/united-stat...


This is literally just how you're supposed to go round any roundabout. Right lane if you're going straight ahead or right (to any extent), left lane otherwise. Anything else _will_ cause crashes, because vehicles will necessarily have to cut across each other to exit the roundabout.

The "turbo roundabout" might make this explicit, but it's not different.


This is hugely oversimplified and doesn't really correspond to real life. Not all roundabouts are symmetric and not all have four entry-exit pairs. Many roundabouts have two lanes on some entries, but a single lane on others, similarly for exits. In scenarios like this you will inevitably have to switch lanes in some scenarios. It isn't really as big of a problem as you make it sound though, since roundabouts naturally have everyone go slow, crashes are very rare so long as the layout is clear.


> Not all roundabouts are symmetric and not all have four entry-exit pairs.

I didn't say or imply this. The rule works for non-symmetrical roundabouts without issue. To phrase it differently:

If your exit is to the right of a hypothetical line extending across the roundabout in your direction of travel upon entry into the roundabout, go in the right lane. Otherwise, left lane.

> In scenarios like this you will inevitably have to switch lanes in some scenarios.

No roundabout I've ever driven through in the UK has required lane switching, unless I was in the wrong lane to begin with.


A turbo roundabout is directionally biased while "any roundabout" doesn't have to be. A turbo roundabout also does not allow u-turns which becomes quite the limitations for road systems wanting to utilize medians for left turn control.

E.g. a standard 2-lane by 2-lane roundabout intersection may just as well look like this https://i.imgur.com/jqhMxW4.jpeg. Note the entrance markings allowing all lanes to go straight with 1 alternative turn direction per lane choice, the exit markings allowing dual lane exits in all directions, and internal markings allowing u-turns (the roads in this case have medians farther out). It has some of the downsides you mention but also some upsides in exchange for allowing slightly more lane flexibility. Regardless, you're definitely not supposed to follow the turbo's rules in that roundabout.

Now you could "no true Scotsman" it and say all the other roundabout types aren't roundabouts because they are supposed to be like turbo roundabouts to be so... but that still leaves needing the distinction in types, for which everyone calls one a turbo roundabout and other variations different types of roundabout.


It's only the difference between CI enforcing code style vs manual PR reviews that have a checkbox for code style. They accomplish the same, but one is infinitely better.


It seems this would solve the problem with normal roundabouts where you have a lane you should be following but know that a vehicle in an adjacent lane is likely to infringe on yours.



Oh come on the Swindon one is the original and best :)

https://en.wikipedia.org/wiki/Magic_Roundabout_(Swindon)


Hemel's is much more fun to drive around though, it's bigger and has the traditional lack of signage and faded lines as a bit of an extra challenge.


> In 2009, it was voted the fourth-scariest junction in Britain

Only fourth? Pssh!


> I'm "pulling this from" the settings on my browser. When I look at the check box in preferences it's unchecked and greyed out.

The setting being greyed out indicates that the entire functionality is disabled, and the value of the setting is irrelevant. That's a common pattern in Firefox settings when some other setting invalidates the current one. In this case, it's most likely because you have "Allow Firefox to send technical and interaction data to Mozilla" unchecked.


Direct from the source: https://www.autoriteitpersoonsgegevens.nl/en/current/dutch-d...

The full PDF with the investigation results e.a. as sent to clearview: https://www.autoriteitpersoonsgegevens.nl/en/system/files?fi...



If you care to read deeper, including comments by many of the principal parties involved, there's no better place than https://file770.com/ (also linked to from this news article). It's hard to follow if you come in fresh, as it's a very active blog, with many pages of insightful comments on each post, but worthwhile.

Here's the list of all posts from the last month which are primarily about this. (There are also daily roundup posts, and lots of small and relevant tidbits in there, too.)

  https://file770.com/chengdu-worldcon-releases-2023-hugo-nomination-statistics/
  https://file770.com/2023-hugo-nomination-report-has-unexplained-ineligibility-rulings-also-reveals-who-declined/
  https://file770.com/chengdu-hugo-administrator-dave-mccarty-fields-questions-on-facebook/
  https://file770.com/hugo-controversy-hits-mainstream-news-a-chengdu-vice-chair-comments-in-social-media/
  https://file770.com/zimozi-natsuco-guest-post-the-hugo-awards-evil-fall-is-a-watered-down-affair-and-certain-issues-to-watch-out-for/
  https://file770.com/dave-mccarty-makes-statement-about-his-facebook-responses/
  https://file770.com/worldcon-intellectual-property-announces-censure-of-mccarty-chen-shi-and-yalow-mccarty-resigns-eastlake-is-new-chair/
  https://file770.com/decoding-the-tianwen-project/
  https://file770.com/the-2023-hugo-awards-a-report-on-censorship-and-exclusion/ (the big one)
  https://file770.com/2023-hugo-awards-related-statement-by-kat-jones/
  https://file770.com/diane-laceys-letter-about-the-2023-hugos/
  https://file770.com/glasgow-2024-announces-kat-jones-resignation-as-hugo-administrator/
  https://file770.com/chengdu-worldcon-wont-account-for-sponsorships/
  https://file770.com/cheryl-morgan-dave-mccarty-resign-from-wsfs-hugo-award-marketing-committee/
The original nomination statistics dropped on January 20th (already bizarre; usually these are released as soon as the Hugo awards are awarded). In the month since then, fans have been puzzling together just what happened. Chances are that more will come to light in the coming weeks.


As others have said, between competing bids for who'd host Worldcon in 2023, China got the most votes (from registered fans at an earlier Worldcon).

However, there was also controversy about this, as a very large number of those votes lacked an address (a minimal safety feature to give some indication that votes belong to distinct people). The responsible committee decided not to count them, but was overruled. If it wasn't for that, Chengdu wouldn't have won the bid.


That's an obvious scam that any trained eye spots in 5 seconds, and very typical for more primitive undemocratic societies like Russia and China. I am curious why nobody reacted since at least few must have realized this.

Once you clarify you deal with scam no voting can overrule anything, you can easily go public and expose people if they would keep pushing for it. This is 2024, we have (at least some) tools to deal with such things in such situations.


According to Charles Stross's explanation of how Worldcons work, you don't really need to fake votes - you can just buy them at $50/pop:

> My understanding is that a bunch of Chinese fans who ran a successful regional convention in Chengdu (population 21 million; slightly more than the New York metropolitan area, about 30% more than London and suburbs) heard about the worldcon and thought "wouldn't it be great if we could call ourselves the world science fiction convention?"

> They put together a bid, then got a bunch of their regulars to cough up $50 each to buy a supporting membership in the 2021 worldcon and vote in site selection. It doesn't take that many people to "buy" a worldcon—I seem to recall it's on the order of 500-700 votes—so they bought themselves the right to run the worldcon in 2023. And that's when the fun and games started.

https://www.antipope.org/charlie/blog-static/2024/01/worldco...

No scam involved at all, to be honest - just regular voting-with-money.


Please note that Kevin Standlee is a good egg here, and was not involved with this iteration of the Hugo awards and everything which went wrong with it.

His only fault was saying something impolitic at the wrong time, which people feared would be misunderstood due to his position as Chair. (You can see the difference in censured/reprimanded and in the reasoning behind it, but it's easy to skip over that and just retain the list of names. Kevin does not deserve to be lumped in with the others!)


Thanks for this. Kevin has been a part of the Bay Area SF fandom community for a very long time, and I’ve never known him to be anything other than a good, honorable person. I was saddened to see his name on the list of people associated with this, and would have a hard time believing he’d ever act intentionally to undermine Worldcon or the Hugos.


I think this is important to note, and a good call-out.

I think Kevin’s statements were dangerous and problematic (from a maintaining the Hugo trademark perspective) but were more of an administrative/legal concern, regarding the organization’s ability to effectively police the trademark.

His statements also probably weren’t wrong about what was possible, but just not something that an entity that represents the Trademark holder should say in public.

But, he wasn’t associated with the disaster that was the awards presentation. McCarty and other administrators of the award committee should be banned from being near this process for life. Standlee hasn’t done _anything_ so problematic, and seems to have good intentions about the whole thing.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: