noodlesUK

I think there’s a big difference in commitment between the level of friction in signing up for a service and spending a couple of min on orienting myself and clicking play and watching the happy path go for 30s.

Respectfully I think this is a mistake. You should not have released the pricing update without making this self-service.

I've evangelised using Tailscale at our company and the 100+(20 * user) limit was very comfortable for us at our level of scale. Cutting that to 50 + "contact sales for pricing" means that our use-case is no longer supported. We use Tailscale as the control plane for remote access to embedded devices we ship around the world - we use tagged devices far more than user devices. I think most tech-focussed companies will also have far more tagged devices than user devices. Anything that discourages us from tagging a server or a random dev machine and putting it on Tailscale will reduce our use of the service and ultimately reduces the "stickiness" that you are probably aiming for.

I also think capping Tailscale SSH to 5 devices on the standard plan kneecaps the value. It's probably in the top 3 most "magical" features of Tailscale. Who has only 5 servers?

I'm happy to pay more if that's what is needed but I resent the "contact sales for pricing". I waste so much of my time jumping through sales hoops with vendors and have a strong preference for vendors that will openly tell me on their website how much my use case will cost.

I think this echoes very true in a lot of places, not just in the US. Here in the UK I'm pretty sure the police/the state more broadly know perfectly well who is doing a lot of the low level quality-of-life crime in most areas, but for structural reasons either can't or won't bother acting in many instances. Investigative work has never been easier: oftentimes there's multiple cctv angles of offences being committed, endless digital records, etc., but unless something can be done with this information in the real world, it's useless and actually takes resources away from other areas of public services.

Increasing the quality of the panopticon has all the downsides we talk about regularly on HN, and if you can't do anything useful for society with the data, it only ends up hurting people.

I think it’s really cool how good the newest video codecs are, but here I am finally able to (only very recently) safely use HEVC and assume that pretty much every end user device has hardware support. It’ll be a while before AV1 reaches that level of adoption and even longer before AV2 does. See you in a few years.

I’m curious: who is operating these bots and to what end? Someone is willing to spend a (admittedly quite small) amount of money in the form of tokens to create this nonsense. Why do any of this?

The difference is between x64 machines and ARM machines. The no-name x64 machines have excellent software support because they all run EFI and have fairly ordinary hardware. The no-name ARM boards have cobbled together bootloaders and require specific U-Boot magic most of the time to even get them online.

I think that for 95+% of companies that write software, it's not particularly sensitive to the ISA of the processor it's running on. Anyone who is writing code in a high level language like python or java doesn't care whether they're on x64, ARM or RISC-V. Compilers essentially already do this. There are specific situations (SIMD extensions, cryptography instructions, etc) where the differences sometimes get exposed, but it's pretty rare that this isn't abstracted away.

This is great!

I’d love a similar document for ARM NEON as well.

Is the exterior of the car not public yet? Why is the only detail about the control cluster?

It's also completely unloved. Even MSFT Azure's own documentation regularly treats it as a second class citizen to GitHub. I have no idea why they don't just deprecate the service and officially feature freeze it.

Honestly that's the case with a lot of Azure services though.

Even just a year or two ago its web interface was way snappier. Now an issue with a non-trivial number of comments, or a PR with a diff of even just a few hundred or thousand lines of changes causes my browser to lock up.

It shows you the level of quality to expect from a Microsoft flagship cloud product...

Can someone in GitHub senior leadership please start paying attention and reprioritise towards actually delivering a product that's at least relatively reliable?

I moved my company over to GH enterprise last year (from AzDO) and I'm considering moving us away to another vendor altogether as a result of the constant partial outages. Things that used to "just work" now are slow in the UI, and GH actions fail to schedule in a reasonable timeframe way more than they ever used to. I enjoy GH copilot as much as the next person, but ultimately I came to GH because I needed a git forge, and I will leave GH if the git forge doesn't work.

Thank you for this. I was mistakenly under the impression that it was not possible to raise these concerns with DVLA anonymously.

Do you know what the process that follows this looks like? Are they just asked to self-certify again? Are they told that someone has reported them (even if they aren't told who it was)?

Older people in the UK already have free bus passes and various other substantial concessions regarding public transport. Cars are dangerous, and if you can't see clearly, you're obviously not fit to drive. It's true that there will be negative impacts on people who will fail the eye tests, and we should be compassionate, but ultimately those people aren't safe behind the wheel, and put other peoples lives at risk, not just their own.

The practical details of implementing this are important - is the eye test done at an ordinary optician/optometrist's shop? How are the results going to be submitted to DVLA, etc.? What protections will be in place to prevent people from shopping around for a dodgy optician (as people often do with cars and MOTs)?

I think this is a reasonable and practical step in the right direction. I accept that given the shortage of driving examiners it would be impossible to require re-testing of existing drivers in the foreseeable future, but as the article says, people already get eye tests frequently and often for free, so this is something that can be done without too much additional infrastructure.

A personal anecdote: my grandfather is in his 90s and is not at all fit to drive due to cataracts and various other issues, but he still does "short journeys" because it's convenient and he feels that it's necessary. The UK has plenty of public transport options and places where people can live with amenities close by (though this is not at all universal). Most British towns and cities are very different from their US counterparts in this respect. My grandfather moved house relatively recently --in full knowledge that the house he chose would benefit from car ownership, and in full knowledge about his age. The only thing that will stop him and others like him from putting people in danger is taking away his licence. He has been told by doctors, opticians and family members that he's not safe to drive, but in the absence of any enforcement he persists. I hope that this policy comes in before he or someone else gets hurt.

There are very few which are suitable for integration into other products - I currently build a scientific instrument that needs a fairly powerful SBC to run. Intel NUCs were well supported and documented: all of their firmware was updatable on linux without any issues, they had data sheets with power specs so you could run them off DC predictably, and you could buy boards without a case. There are plenty of small NUC shaped mini-PCs but few that are suitable for integration (at the price point intel was at).

Cheap Chinese mini PCs just aren't well documented and don't have predictable supply.

I agree completely - the NUC segment has a gaping hole post 2023, and faster raspberry pis can probably fill a lot of it especially for small scale commercial stuff.

It really is pretty crazy that some of the more esoteric forms of citizenship have never been rolled into "British Citizen". Almost all BOTCs were given the opportunity to become British Citizens, but not all, and they kept the original status around. BNOs are similarly a somewhat silly situation especially now that it's possible to move to the UK on the special BNO visa (which gives them different/better family reunion rights than normal British citizens). British Subjects essentially don't exist in practice, but they also haven't just rolled that into British Citizen status either (British Subject is the residual status of certain Irish born people who chose to retain the status - they have the right to live in the UK on the basis of their Irish nationality, not on their British nationality which is insane). There's just a perpetual allergy to just rationalising the whole setup.

It would be perfectly politically acceptable to just do away with the statuses that have fewer than say 5000 people and grant them all full-fat citizenship. Generally people who live in the UK are shocked when they find out that holding a British passport does not entitle you to the right to live in the UK.

It’s possible to have been born with multiple forms of British nationality such as being BOTC (e.g. bermudians) and a British citizen at the same time.

It’s also possible for e.g. a BNO to register for British citizenship after a period of residence in the UK. This does not extinguish the original nationality. Most hong kongers with British citizenship are in this bucket.

It sounds like they've dropped the digital ID part being mandatory, but not the digital right-to-work checks being mandatory. I suspect that the UK will end up building something like the US's E-Verify programme, which allows a number of documents to be checked against authoritative sources. It really wouldn't be that hard to build a service that in the first instance allowed you to generate a share code with a GBR passport much the same way people can generate share codes with their drivers licenses or UKVI accounts.

What I have a problem with is just how fragmented and broken the UK immigration system is when you have the misfortune of coming into contact with it. It's (like many such large systems worldwide) a set of policies and rules that have accumulated over time into something that is pathologically poorly thought out. I'm going through the process of renewing my spouse's visa (I'm British), and it's fractally awful -- we've just had a snarky email from our landlord who is worried that the right-to-rent permission is expiring, but it's not possible to apply for a renewal for the visa prior to 28 days before expiry of her current visa. I meet all the criteria to sponsor my spouse for renewal, but the evidentiary burden is insane (I've collected 400+ pages of documents so far). Nobody wants this. It is very expensive and difficult (probably >£10k per person until permanent residency in fees, not including legal expenses) to be compliant even if you meet the criteria, which just leads people falling out of status (to borrow an American term). The government (of all stripes) tries to be "tough" but the only lever it knows how to pull is to make the rules stricter, not making them better enforced or align with some meaningful policy agenda.

This farcical situation extends into the UK's broken citizenship model where there are 6 different types of nationality, none of which give any rights you can't build through a hodgepodge of other different statuses. As far as I know the UK is the only country in the world that permits dual nationality with itself!

A government online account which can generate verifiable credentials would probably be helpful in a broad sense but it wouldn't cure bad policy which is rampant in the UK immigration sector. I'd much rather have some kind of digital ID that's clear and authoritative rather than just hoping that Experian has my details right with no recourse if they're wrong.

I think this is a huge mistake at least as far as the office software goes. One of the key advantages to Pages.app and friends is that they are pre-installed and free on MacOS. This will just drive people to M365 and Google Docs.

Whilst I seriously doubt that this will be regularly or universally enforced, I think that gas powered gardening and construction equipment is incredibly loud and annoying, and that especially in built-up areas it's not really acceptable where good alternatives exist. Noise pollution is usually very avoidable, and it can be quite disruptive and sometimes damaging to health.

Further, where I live (in the UK), there needs to be a serious reckoning about modified cars and road noise. At least 5-10x daily people drive through the city centre near my home with cars modified with unbelievably loud pop and bang ECU tunes and exhaust noise (close to as loud as gunfire). This kind of deliberate noise pollution is exceptionally antisocial behaviour and should not be tolerated by society.

I agree that the author lists in various metadata sources and databases are often a bit wrong (weird formatting of names for instance is very common), but many of the cases in the OP article are pretty egregious and far beyond just data entry issues.

Presumably the citation scanner they're using is relying on similar data sources as Zotero in any case to detect these sorts of issues.

Regardless, my comment still stands, it seems like the submission is relying on the actual text of the bibliography being correct, rather than requiring a machine readable citation metadata file of some sort, which would at least allow much of the quality control checks to be automated (and certainly would preclude complete hallucinations of nonexistent papers getting through).

It astonishes me that there would be so many cases of things like wrong authors. I began using a citation manager that extracted metadata automatically (zotero in my case) more than 15 years ago, and can’t imagine writing an academic paper without it or a similar tool.

How are the authors even submitting citations? Surely they could be required to send a .bib or similar file? It’s so easy to then quality control at least to verify that citations exist by looking up DOIs or similar.

I know it wouldn’t solve the human problem of relying on LLMs but I’m shocked we don’t even have this level of scrutiny.

They're also doing something very scammy with ads - if you are scrolling on mobile, they've changed the behaviour on iOS so that if you touch the ad at all, it considers it a click and opens it, whereas it's much less sensitive on ordinary posts and behaves like any other app. This is clearly to increase click-through artificially.

Nope. I was born abroad to a U.S. citizen who didn’t meet the physical presence criteria to pass on citizenship. I came to the U.S. as a child on an IR-2 green card, then when the CCA became law I automatically became a citizen. My parents applied for a passport for me, and in the process the department of state presumably shredded my green card. I don’t have a certificate of citizenship and I’m not eligible to apply for one, as I no longer live in the U.S.

Unfortunately USCIS doesn’t know anything about this (as it was all handled by the department of state), and presumably thinks I’m an alien who abandoned their status.

This argument rings especially true in the U.S. where there is already a primary key in use every day. The SSN serves as a universal enumerator but without canonical data.

If the U.S. wanted to have a national ID system with rules, a defined scope, and redress procedures when things went wrong, and established it in the open, following a democratic process, I would be much happier.

The system we are getting instead has all the downsides of centralisation, with none of the upsides.

I think it's an important thing to remind yourself of sometimes that there's a pretty significant difference between a startup and a small business. In spaces outside HN, small businesses are the default form of business, and there's plenty of merit to running one even in tech - not everything is scaleable, nor does it need to be. If you can identify a niche, generate profits, and provide business value and support the livelihood of your workers, that can be enough. Startups can be very impactful, and the tech sector has thrived as a result of the model, but the growth obsession and leverage is not mandatory.

I know it's a bit worn out but if you want to learn about practical orbital mechanics and have fun doing it, Kerbal Space Program is the best tutor there is. It's a simplified physics model so stuff like Lagrange points don't work, but there's no better way of getting an intuitive understanding of how bodies move in space.

U.S. passports have a unique shielding which prevents them from being read whilst closed. It’s a low tech solution but I think it’s an amazing example of clever mitigations.