Posts by rg287

US Navy backs right to repair after $13B carrier crew left half-fed by contractor-locked ovens

Re: Contracts have been this way for decades

The techs have to follow a script just like a foreign CSR even when they can see that a wire got pinched when the last person rammed the module into place. No fixing the wire and no skipping any steps on the forms.

There is kind of a logic there. Say something gets broken on exercise. Maybe even a pre-deployment warm-up training. Do you fix a thing, or do you say "Nah, get a new one. We're not deploying with something we broke".

It's not your money, and you wouldn't want to be responsible for doing a field repair in training on something that breaks - perhaps with operational or fatal consequences - in combat.

But by the same token, do those people then have the skills to perform a field repair on deployment?

Congress: "Our soldiers are immensely smart and capable and should not need to rely on a third party contractor to maintain their equipment."

Procurement: "Is it squaddie-proof? Because soldiers will break it and the Marines will try to eat it if it looks like a crayon."

I jest. A bit of inter-service ribaldry.

But yes, the idea there's anything in a f-ing ship's oven that can't be fixed by the ship's engineers is madness.

Apple goes glass whole as it pours new UI everywhere

Very shiny.

But have they fixed Finder or Spotlight yet?

Sequoia has been bloody awful for basic functionality. The ability to preview .CR3 (Canon Mirrorless RAW) files in Finder is spotty to say the least. Sometimes it works, sometimes you just get a black 6000x4000 rectangle. Not sure why. Used to work, but they borked it.

Spotlight frequently doesn't find installed applications, instead offering to perform a web search (and yes, I've been through every forum thread, rebuilt the index, etc).

Meanwhile Finder is unfathomably slow. Save a file and it can take 30 seconds to appear in the finder window, which is just ridiculous on an M3 with NVMe storage and 24GB of RAM. Are you rebuilding the whole f-ing index?

Worse yet, Finder (or an underlying dependency) is apparently still single-threaded in places because a slow SMB share can lock all your Finder windows. Trying to move a document between two local folders? Sorry - if the thread is hanging on a NAS for an unrelated operation, you'll just have to wait. Which is weird because simultaneous transfers are perfectly possible. But seemingly certain operations like trying to index a remote folder will lock the whole thing up. This never used to be a problem, but is now. I suspect someone has tried to improve Finder and make it more responsive by predicting what you might want to do (on an assumption that storage is so fast you can afford to do some speculative stuff?), which has the opposite effect when you're accessing a HDD-based share.

If I go to "Save Image" or a PDF in a browser, I hear the disks in the NAS spin up before the Save Dialog opens. I'm not actually wanting to save it to the NAS - just my Downloads folder. But at a guess, the underlying file management system is saying "Aha, they have a share mounted. They might want to save the file there so we'll make sure that connection is awake before we let them do anything", rather than presenting the dialog and then pinwheeling if and when I select a network share to interact with. This presumption creates a pause of multiple seconds between clicking "Save" and the save dialog actually appearing, which is a right nuisance.

The saying used to go "What Intel giveth, Microsoft taketh away". But it's true of everyone - despite order-of-magnitude increases in compute and CPU speed, and massive growth in available RAM, the underlying UI responsiveness is seemingly not a priority for developers, despite being the single most important UX factor. If you want your UI to feel sluggish and slow, making users wait as you bring windows into focus or simply hanging when you browse to a different folder is how to do it. We managed this in 2005. How do we not manage now?

I was able to tab through 20MB RAW files quite happily on a 2008 dual-core Macbook (with SSD upgrade). Why do we struggle with doing the same on NVMe storage in 2025?

Blocking stolen phones from the cloud can be done, should be done, won't be done

Re: Nice idea

As a public service, isn't this what they should do.

To a point. But that assumes public opinion represents informed opinion, which - thanks to the efforts of Murdoch, the Barclays and Lord Lebedevedev-dev, is not a sure thing. Some sort of moderating expert input is often an important component.

Of course it's a difficult thing to balance - a gilded tower "Well the plebs are wrong" approach is not a good thing - it encourages echo-chambers within the halls of power. Even if the plebs are just being bigoted because the Daily Zeig Heil told them to.

But conversely, it's also undesirable to just do what the racists want because DEATH and INVASION sells papers and the flames of intolerance and willful-ignorance have been fanned by billionaires who want an unregulated, low-tax business environment and have found that pandering to that corner of society can be quite profitable - if you don't care too much about the wider social consequences.

Some people will ask how that squares with democracy, and that's a tricky one - there's a reason we have a minimum voting age and don't let 10year olds vote - we assume that some competence is required, and that most 18 year olds can form some sort of a meaningful opinion. Where do you balance straight democracy with tyranny of the masses?

Nonetheless, these billionaires have managed to convince the proles that the reason their public services are dreadful is because of a statistically-insignificant number of immigrants, not because they voted for the people those billionaires told to vote for - people who then implemented a deeply regressive tax system and stopped investing in public infrastructure (and indeed sold much of it to the private sector who then asset-strip it for private gain).

Turkeys voting for Christmas.

It's incredible that the biggest topic in that corner of politics is "stop the boats" not "why are taxpayers subsidising commercial banks to the tune of £35Bn/yr?".

Re: Nice idea

Civil forfeiture != proceeds of crime.

Civil forfeiture in the US is an affront to basic common sense and decency.

But across Europe there are plenty of proceeds-of-crime statutes where Police can have assets frozen temporarily (as evidence usually) and then a court can distribute those where appropriate - whether it’s drug dealer’s BMW, the proceeds of embezzlement or fraud, or other nefarious gains.

Your ransomware nightmare just came true – now what?

That seems stupid. Establishing a working relationship should make things easier for them.

This should be the case. It's a bit like lawyers mediating an acriminous lawsuit. The clients can be screaming and shouting, and then the lawyers go off to a room quietly and say "Okay, this is what our respective clients want. How do we get to an acceptable midpoint?"

They can be objective, unemotional about it, let the client do the whinging and then go have a to-the-point conversation.

Of course, not all lawyers are good at their job, and some get invested in their client's hubris (*cough* Giuliani). But court cases and lawsuits tend to go quicker and smoother with some detached professionals in the room rather than two parties screaming at each other. Of course some parties are perfectly capable of carrying themselves as well. But many are better off with representation.

Re: Stop paying. Stop making excuses for piss-poor IT.

While it looks good on paper, legal penalties for ransomware payment will create a perverse incentive to avoid disclosing the attack and give the attacker additional leverage over the victim, thus likely making the ransomware problem even worse.

Good luck with that. Kevin Beaumont calls most major ransomware incidents whilst the victim is still denying they've had an incident. And the leverage ransomware offers is 50% disruption, 50% threat of disclosing sensitive exfiltrated data. So if attackers know that orgs won't pay if it's known they've been popped, then attackers can't threaten to disclose or advertise stolen data on dark markets. They have to keep it secret so that paying the ransom remains an option for victims.

And of course if the infection is bad enough that the incident becomes public knowledge, the org also can't then pay. Realistically, most incidents would then end with "Hey, we'd love to pay but our execs aren't willing to go to prison now The Reg has reported it, sorry.".

Re: Stop paying. Stop making excuses for piss-poor IT.

If you are going to make it illegal, make it an automatic jail sentence.

And C-Suite-only. Unless there's clear and unequivocal evidence of culpable negligence below that level.

The presumption should be that staff were doing the best they could with the resources afforded them by the Board, and that the buck stops with people whose job description starts with "C".

KDE targets Windows 10 'exiles' claiming 'your computer is toast'

Re: Alarmist?

There has never been a technical reason for creatives to prefer Apple. Creatives prefer Apple kit for the same reason arty types use cigarette holders and wear berets.

The early Macintoshes was pretty standout (particularly the 512k). Jobs was obsessed with design and typography, and this showed in the Mac's support for fonts and type. Before Windows was a thing, MS released Excel on Macintosh as the launch platform, and then Photoshop was Mac-only from 1990-1993. So in that era of 1985-1995, there were strong and compelling reasons for visual-creative businesses to run Apple for both business and colouring in (in the audio world of course, some key applications like Sibelius were Acorn-only). Apple had a growing momentum in the design sector, plus credible business alternatives to Lotus1-2-3 (the "killer app" for the IBM PC and compatibles).

There have always been alternatives, but there have indisputibly been periods where Apple was the technical leader in some sectors. The OS-stability helped as well - fewer BSODs. For music, Pro-Tools didn't come to Windows until 1997.

These days of course it's much of a muchness - Adobe/CO/Serif/BM/Avid. Cross-platform, all acceptably stable. Pick your poison, unless you've wedded yourself to FCP/Motion or Logic Pro. Which is a valid choice - but as you say, not a technical choice as such. You're not getting inherently better output from FCP than cross-platform alternatives.

Datacenters have a public image problem, industry confesses to The Reg

Re: Cry me a river.

You could have a nonprofit datacenter, but you've basically just reinvented the colo. We have private colos already and they work fine. In many cases, there's not a lot that a nonprofit one would do that a private one wouldn't also do.

Well, I've defined a particular flavour of colo that doesn't really exist (Tier 0.5-1 if you will). Back in the wild-west days there were "data centers" operating on a wing and a prayer with non-redundant power supplies and transit (although some would tell you they had - that was perhaps "aspirational"). But if you wanted cheap and were willing to accept that risk, then fine. Now it's actually kinda hard to find colo that isn't Tier 3-4, with a list of certifications as long as your arm. And that's an important part of the ecosystem, but it's also excessive for a lot of people. It adds cost to business and eliminates risk that some users would be happy to accept for non-critical workloads.

The only possibility, cutting costs by eliminating normal utilities like UPS or inter-user security so they could cut prices, could easily be done by private ones but it's probably good that it hasn't.

Security is important of course. N+N/N+1 power, N+1 generator and diverse fibre - even a UPS - actually isn't necessarily a requirement for lots of people. People spend a lot of time on it, and then their office broadband goes down (usually single-homed, single point-of-failure) and the highly available, five-nines cloud service isn't accessible! Of course in that case, you're wanting on-prem anyway.

There are some interesting community projects out there. NYCMesh have done an incredible job building an infrastructure commons across the city. Their primary interest is connectivity, and they encourage self-hosting. What I've described is kind of the colo equivalent - run well, to a good standard, but taking a pragmatic view to risk and the diminishing returns of additional redundancy, etc. Not "bad" but saying "This is what you get. It's less, but it's cheaper and more accesisble". But perhaps also obsolete these days given higher upload speeds from residential connections that allow more self-hosting (unless you want to run SMTP, etc in which case you need a tame ISP who will open those ports).

The TL;DR is that for some people, you don't need a hundred-million pound custom campus with all the utilities. A converted office unit with good physical access controls would do the job, but because the market all feel the need to accredit themselves up the wazoo to be taken seriously, everyone is stuck with higher costs. In fairness, there are now some "edge" datacentres emerging who claim "Tier 3 alignment" rather than certification, which seems to show a move back to that.

Re: give useful/wanted examples

Show a picture of a data center with typical uses. That rack contains your isp email,that rack contains Netflix movies,that rack contains BBC radio/iplayer. Realistically it will be more boring such as company xyz data. Arguably government databases are important ( I couldn't sell that sorry). If people can relate to what it contains they maybe more tolerant.

The problem being that they can't do that - and they know it. The bulk of these new DCs are going up for AI because old DCs can't support the per-rack power/cooling demands. If they were doing something genuinely useful like "here's your email", then yes, people would be tolerant. But they're being asked to watch greenfield sites get paved into GW-scale bitbarns (plus gensets & ancillary infrastructure), and to give up their water supplies for evaporative cooling (and somewhere to dump the toxic brine), all in pursuit of a chatbot that lies sporadically.

And for (the island of) Ireland in particular this is a touchy subject. Loch Neagh is still in the grip of total ecosystem collapse. Turns out the 2023 algal outbreak, collapse of the eel fisheries and subsequent wetland-bird population crisis wasn't just the "bad year" that politicians were praying for (so they didn't have to fix it). It's as bad as it was. Ireland's largest freshwater body is fecked, and likely to remain fecked for the foreseeable future. In fairness, this collapse is not due to datacentres, but has rather focussed attention on the (mis-)allocation of (finite) natural resources and the economic effects of their collapse.

Re: Cry me a river.

These hyperscale bit barns don’t host community projects or serve local needs - they’re energy-guzzling, water-hungry fortresses for Big Tech to hoard compute, automate layoffs, and extract value from everyone else's data.

That's the crux of it. DC's are important. Our modern world is built on digital services. Even our phones are mostly VOIP now. But this gold rush for 150kW racks to power AI workloads is not actually benefitting anyone. Sure, we see reports that tech vacancies are up, with people hiring AI-"specialists". But how much of that is down to actual business and how much is it execs stuffing some money into "AI" because of FOMO. Sure, I'd rather see the cash go out in salaries than as a stock-buyback or dividend. But it's not actually generating value or product. And really, these massive AI datacentres don't benefit locals (or anyone really) the way a telco datacentres does (which is legit utility infrastructure).

But it's all part of the act - muddy the waters that "Datacentres are essential infrastructure". When the truth is that Some DC are essential infrastructure. Some are luxuries, and we should be more discriminating in which luxuries we indulge in.

You know what would be an amazing concept? A community data centre. A room with some mesh-fronted lockers run by a non-profit CIC who has some IPs and a 1-10Gb backhaul. For a network provider like CityFibre, the traffic would be mostly upload and therefore free (since their customer traffic is going to be heavily download). So if you were reasonably on-network geographically, then they should be able to charge a fairly sensible monthly connection fee. Basically a step up from a well-organised hackspace.

You get a power socket and ethernet port in your locker and that's that. You can drop in a couple of RPis, an HP Microserver, whatever. £20/mo, done. All to the benefit of the local community for local workloads.

If the power goes off, you lose connectivity for a while. But does that actually matter? At my first proper job (late 2000s), our public website and customer portal was run on a Dell tower server in the office hanging off an SDSL line. Which part of our infrastructure represented a single point of failure? All of it. The server, the crappy 5-port 100Mb unmanaged switch, the dreadful BT router... I had most faith in the cabling - because I crimped that Cat5 myself.

That wouldn't be tenable today given the growth of that company. But for lots of firms or enthusiasts... it'd probably be fine? How much are we paying for 5x9s hosting when really "available most of the time we're awake" is sufficient. (See also: Low Tech Magazine). There's a fair case to be made that modern DCs have excellent PUE values and are more efficient. It's possibly also the case that distributing our compute to smaller on-prem deployments can be cheaper and offer better experiences to local (LAN) users.

Not everything needs to be K8s in AWS with 99.999% reliability, because at the end of the day - most of those big boys go to sleep occasionally as well when they drop a region from their routing table or fudge their BGP.

For a lot of people, what I've described is enough. They don't need to pay £nnn/mo for a quarter or half rack. But as it stands, if you're not at that point, then you're directed to spin up some VMs with a hyperscaler. I suppose that concept is undercut by the fact most domestic lines these days have at least 20Mb upload, which is fine for most "self-hosting" situations, provided you can get a static IP.

Ukrainians smuggle drones hidden in cabins on trucks to strike Russian airfields

Re: Unintended (?) consequences

Ukraine has been treated exceedingly gently by Russia, despite your deluded ramblings, Putin is a moderate, with a restrained influence. The Russians care about the Ukrainian people, so they want to harm as few of them as possible.

Yes, penning a rambling revisionist essay about how Ukraine has always been part of Russia and then lobbing glide bombs into civilian housing is definitely a gentle and restrained thing to do to the people you care deeply about.

Re: Unintended (?) consequences

And almost certainly intended and why Ukraine has published such detail on how this was achieved.

Along with the comment that "What’s most interesting, and this can now be stated publicly, is that the 'office' of our operation on Russian territory was located directly next to FSB headquarters in one of their regions," which may or may not be true, but has probably caused absolute panic and seen every FSB regional office raiding neighbouring units and businesses to work out if it might have been under their noses.

Aside from being a wildly successful and audacious operation in terms of it's anti-aircraft objectives, demonstrating an ability to run covert ops deep inside Russian territory will have put the frightners on from Kaliningrad to Beringovskii. Oh, you thought Vostochny was too far east and out of Ukraine's reach? Maybe. Maybe not...

Re: Ukraine did

"Those who want the spread to lie that Ukraine is helpless without the US or that they're doomed to lose the war are wrong."

Quite.

Trump: "‘I’ve been watching for years, and I’ve been watching Zelensky negotiate with no cards. ... He has no cards. And you get sick of it."

Zelensky: "You mean this card?" [Plays King of Drones]

Has the US even said thankyou? For eliminating 30% of Russia's nuclear-capable strategic bomber fleet?

American science put on starvation diet

Re: 1 makes 4

The same applies to new road schemes.

Well yes, induced demand occurs on all transit. But on roads it's a bad thing (increased maintenance, then people complain about potholes, and complain when you close the road to patch the potholes, and the scheme was usually conceived to "ease congestion", which it fails at - see America, "just one more lane bro"), whereas induced demand on Public Transport is a good thing - you want transit ridership approaching 100%, because that pays back your infrastructure overheads and minimises costs per-fare. And riders who represent modal share from car to transit mean there are fewer cars on the road - the best way to ease congestion is not to add more roads/lanes, but to improve public transport.

And of course, roads don't help apprentices, non-drivers, etc. They entrench a car-first society and make it marginally more convenient for people who already drive to get where they're going marginally quicker (until congestion makes it a wash).

Building roads through Stoke-on-Trent for instance won't improve traffic. No road scheme can save the city when the M6 is closed and all the blocked traffic is clogging the roads. Whereas urban rail or a tram network with dedicated running paths would be unaffected by road chaos.

Roads are important, and we must ensure those needs are met, but we've now built a very brittle society where the roads get clogged and the areas grinds to a halt, whereas in London a bad accident doesn't impact the Underground (or Overground).

Re: 1 makes 4

I don't think that, outside of Ponzi schemes, you can find other schemes that boost economy on that scale.

In the UK and US, public transport infrastructure - but primarily because it's been ignored for decades and is now in a dire state, so every time you open a new station in the UK it blows it's passenger projections out the water.

e.g. "Year 3 ridership projected at 50k pax/yr, but actually hit 120k in year 2. Treasury had delayed it for a decade because they didn't think it'd see 40k/yr or generate a good ROI".

This weird and remarkable phenomena whereby making it easy for people¹ to access employment and education improves economic output, improves tax receipts and lowers expenditure on benefits.

1. Including 16year old students/apprentices, people with heath conditions that preclude them from driving and low-income workers who can't afford a car thanks to wage stagnation. It's so strange how making it hard for people to join the workforce is bad for the economy. I just can't explain why that might be. _/s

Empire of office workers strikes back against RTO mandates

Re: Cuckoo land

I disagree with that argument. It's like saying that since not everybody can ice skate, there should be no ice rinks.

No it isn't. Building out a car-first society where access to an industrial park is exclusively via private car is absolutely discriminatory. Of course people can choose to drive, but the development should also include bus stops (or indeed a train/tram station if such a network exists, or passively provisioned to accept an extension from an existing network). Along with parallel walking/cycling routes, the businesses should have secure bike storage (if you think this imposes an expense, wait till you see how much paving a large car park costs - a bike shelter saves money compared to paving the equivalent number of parking spaces. You can buy less land and do less paving. Fewer potholes to fill as well).

Nowhere have I argued for not having cars. I have simply argued for making them a lower priority than public transport which everyone (including car-owners) can use.

Re: Cuckoo land

My experience of the M6 in the West Midlands and the adjacent bit of the M5 Is that one could easily and safely watch a movie from the drivers seat if one wished - in fact several. Being called upon to drive would be at best intermittent.

Well yes! Usually because you're parked behind someone who got that equation wrong.

Re: Cuckoo land

Perfect - now it’s ‘I never said take cars away, I just want public transport so good people won’t need them!’ That’s adorable.

"Now". No, no, no. That's how it was to start with. If you read my post properly.

It's not adorable. It's calling for better. In any German city >200k (so, places like Stoke-on-Trent, Wigan or Plymouth) would have a comprehensive tram network, light rail - all under the S-Bahn banner.

Meanwhile, back in reality, the system is wrecked, underfunded, unreliable, and decades away from being that utopian alternative you’re fantasising about.

Yes. It is wrecked. The best time to plant a tree is 20 years ago. The next best time is today. So why the throwing up of hands and "ah well, it's screwed, forget it, we'll just lock into a car-first society forever". You say "people are drowning now". But you aren't proposing any solutions other than "Well, we'll just carry on with cars then". Which by your own description is drowning people. They are forced to own a car for survival even if they can ill afford it..

You had a nice Eurostar ride with a glass of wine? Congratulations - but comparing that to the daily filth, chaos, and exhaustion of Britain’s broken local transport is laughable. For countless people, it’s not a choice between a cosy bus ride and a car, it’s a choice between a car or no job, no income.

YES. I F-ING KNOW THAT'S THE CHOICE. THAT'S WHY GOING ON COMPETENT PUBLIC TRANSPORT IS SUCH A WAKE-UP CALL OF HOW MUCH BETTER THINGS COULD BE UNDER COMPETENT (local & national) GOVERNMENT. IT'S WHY I AM RESENTFUL WHEN I AM FORCED TO GET IN THE CAR TO DRIVE TO SOMEWHERE THAT OUGHT TO BE A BRIEF TRAM TRIP.

You talk about how ‘driving is lost time’ - tell that to the shift workers, the night workers, the ones outside metropolitan bubbles who can’t even rely on a bus showing up, let alone on time.

YES. I KNOW. THAT IS WHY I AM RESENTFUL ON THEIR (and my own) BEHALF THAT LOW EARNERS ARE LOSING THOUSANDS OUT THEIR PAY PACKET RUNNING A CAR THEY CAN ILL-AFFORD. WHY I SIGH EVERY TIME MY COLLEGAUE IS LATE TO OUR MORNING CALL BECAUSE THEY'RE DRIVING THEIR SON TO COLLEGE BECAUSE THE BUS DIDN'T F-ING TURN UP.

Pretending the answer is just ‘fix public transport’ ignores that for decades no government has meaningfully done that. People are drowning now. They don’t need wishful thinking. They need solutions that exist - not smug lectures about a system that only works in theory.

In theory? This is not theory. This is practice. It works in practice. We just don't do it. It took decades for Amsterdam to sort itself out. We have to start somewhere.

Look, I Know. This is the de facto situation. And it will take time to fix. That doesn't mean we shouldn't start now. The solution is not to say "Oh well, that's adorable in theory but it won't work". What solutions do you propose? What alternative do you propose to unaffordable car dependency?

And actually, some fixes are quick. We could improve bus services massively in 12-18 months - if government financially backed councils to nationalise or regulate services - not merely giving them the theoretical power to do so. Tram networks can be rolled out in 5 years - not decades. If they can secure funding.

There are short-term fixes. But Reeves won't sign off on them. She's too busy subsidising commercial banks to the tune of £35Bn/yr.

Re: Cuckoo land

Public transport can work well for commuting if there is a single route, fairly direct, between your home and place of work - and it's adequately time-tabled. Many decades ago that existed where I live. It took half an hour to go by bus to the nearest town with four buses an hour, one of which served the next most outlying village. In peak times those were doubled up.

There is, nonetheless, a major call for it though - every time a new rail station gets opened (improving the network effects of that line by adding a node), it tends to blow out it's passenger projections in 18months. The reticence of Treasury to fund more station projects despite the massive ROI is bizarre and counter-productive. And there are places like Penicuik - an Edinburgh dormitory down - which don't have a rail station at all. From Penicuik it's a 30-50minute drive into Edinburgh, or 90minutes on the bus (which visits all the little villages). A train could cover the 10miles in under 20mins. There's some exceptionally low-hanging fruit which would make a major difference to traffic in our cities.

Even so it only supplied a fraction of the transport needed because most people were able to walk to work as the workplaces were close by; this was the England of industrial villages.

Well yes, this is a significant issue. I live on the edge of a small West Mids town (within the ring road). The former light-industrial (garment-making) site at the bottom of our road has long gone and been redeveloped as housing. Other business sites in the centre of town are moving to edge-of-town with no real active-travel options, just a dual-carriageway. They're being replaced by more housing, from which people will presumably have to drive to the out-of-town locations.

The loss of mixed-development is a real social problem, leaving us in this decreasingly tenable position where people get in their car and queue down the bypass to get to an industrial estate, and then do the same in reverse at 5pm. The loss of co-located employment in residential areas is deeply problematic. People didn't need a car-per-adult because people could walk into work. And of course people walking into work don't need somewhere at work to park their car. Win-win for them and the employer. But now the area is purely residential and everyone has to drive out to get to work. If you'd told employers in the 1950s that their office building would need the same footprint of land again for car parking, they'd have laughed you out the room. How much do you think they're paying their staff? And yet, we're now in a position where we have hollow cities - not as bad as the US, but in areas quite problematic (Stoke-on-Trent) - where office blocks are separated by acres of tarmac and otherwise-able non-drivers are barred from employment because they can't get there by other means.

And then the government have the gall to suggest they cut disability benefits and push people back to work. HOW IS SOMEONE WITH MILD-BUT-DISQUALIFYING EPILEPSY SUPPOSED TO DO THAT? THIS IS NOT A RHETORICAL QUESTION.

A household with at least two working members is very likely to need at least two cars because those members are likely to be working in quite different places.

Yes. This is the problem. It should not be necessary for someone to require two tonnes of car (which has to be parked somewhere both at home and work) to get themselves and their lunch to their place of work. Public transport should be good enough to allow at least one, if not both of them to go by bus/tram/rail and reserve the household car for exceptional use.

Re: Cuckoo land

‘no benefit to driving’ - spoken like someone privileged comfortably inside Zone 1-2, who’s never endured the slow, grinding misery of commuting beyond Zone 3.

No. Sounds like someone who lives in the West Midlands whose commute used to take between 25minutes and 2 hours depending whether some tosser had been watching a movie on the M6 instead of... y'know... driving. Not that my commute involved the M6, but when someone spilled their car/caravan/HGV and closed the motorway, the traffic looking for a parallel route would bring the region to a standstill. Happened about once a month.

Like I say, WFH is preferable, but I wish I had had the option of taking a tram, or a bus with dedicated bus lanes. But like many, I'm in the position where public transport would have taken 90minutes vs. 25.

Sure, you ‘save’ a grand, but you trade it for something far more valuable: your time, your energy, your sanity.

I mean, I've just been to Brussels and back via Avanti and Eurostar. It was much less effort and much saner taking the train than driving up to Manchester airport, arsing around in a crowded departures lounge, being sandwiched into a teeny seat (even basic Eurostar is a step up from airline Premium Economy) and then having to drive home again on the way back (as it happens, I was in Eurostar Plus, which meant I got fed and had a glass of wine, so I really wouldn't want to be driving at the other end).

Sitting on a bus or metro for half an hour reading is a much saner use of my life and my time than braving the roads and all the lunatics who have forgotten how to drive (particularly since 2020). Even if I save 5 minutes on the journey, I've gained half an hour of reading time vs driving.

And this is the point really. People say "public transport is slower". And sometimes it's catestrophically so. But if the choice is 20minutes driving or 35minutes reading... actually I'm good with reading? Driving is "lost" time. Reading isn't.

As for the claim that needing a car is ‘madness’ or ‘discriminatory’ - no. What’s madness is pretending that Britain’s fragmented, underfunded, decaying transport system can actually serve everyone.

Sure, so f- the disabled then. If you can't drive, suck it up and spend your life being a political whipping boy as politicians decide whether or not to allow you benefits, whilst also denying you the infrastructure to actually access education of employment. What sort of vile rhetoric is this? The answe to Britain's public transport infrastructure is to fix it. Not to give up and say "Those who can, drive. Everyone else, tough titties".

For countless people, a car isn’t luxury or preference - it’s survival.

And you don't see any problem with that?

And stripping away that option under the banner of ‘fairness’ only deepens inequality - turning mobility into a privilege, cutting off access, and pushing more people into hardship.

Wait.... you haven't read the f-ing post have you. What a troll you are. I didn't propose stripping anything away. I said public transport should be so good that people would willingly forego a second car because they didn't need it. Reading comprehension.

For reference, this line in particular:

Not out of an authoritarian "we'll tax the bejesus out of second cars", but because public transport should be cheaper and of comparable speed for getting a worker, laptop and their lunch to a place of work.

Re: Cuckoo land

but relying on public transport means paying more, waiting longer,

Depends entirely on where you are. In London - or even quite a few bits of the South East, traffic means there is no particular time benefit to driving, and public transport will be quite a lot cheaper. And you can genuinely relax, read, listen to a podcast, etc (not half-listening whilst you hopefully concentrate on the road!). Rest of the country you're screwed (bar maybe Reading, or living in Glasgow or Newcastle, which have Metros).

People cringe at the idea of a £1500 season ticket, but when you point out that's half the price of a £200/mo car finance plan (£2400/yr before you add petrol, parking or any tolls such as the Dartford Crossing), they realise they'd quite like the £1000+ back in their pocket. One-car households should be the norm. Not out of an authoritarian "we'll tax the bejesus out of second cars", but because public transport should be cheaper and of comparable speed for getting a worker, laptop and their lunch to a place of work. Cars are great for outings, big shops and Ikea runs, but the idea that every worker needs a car because it's the only viable way to access their employment is madness - and discriminatory against those without licenses, disqualifying health conditions, etc.

The idea we're habitually using two-tonne metal boxes to get individuals plus laptop to work (and providing sufficient parking for that metal box to stand idle for 8 hours) is insane use of land, resource - and bad for the economy. Far better to be spending money on British-made buses and UK-tax-paying drivers than on German/French cars.

All of which is inferior to WFH (where possible, Doesn't work great for nurses or surgeons), which is cheaper yet. But we should still have fantastic and universal public transport as the default option, servicing the needs of everyone.

Ex-Meta exec: Copyright consent obligation = end of AI biz

Re: Just for the record. . .

This strikes me as more of a battle of the wealthy vs the super-wealthy.

The fact that the small minority of musicians and artists who have made it big are choosing to speak out on behalf of the industry at large only speaks well of them. The vast majority of musicians and artists do not make seven - or even six - figures. The theft of their IP is every bit as problematic as ingesting the Lennon-McCartney songbook is for it's rightholders. More problematic in fact, because they can't just say "well, I've got £1.2Bn in the bank so it doesn't actually matter if my royalty income collapses. I'm not going to suffer hardship - it's just the principle of the thing". For those jobbing artists... it is their rent and their livelihood.

Of course we're still in this perverse position where the MPAA and RIAA are chasing teenagers for torrenting a couple of albums and claiming that this amounts to losses of £10k/track instead of hefting their entire legal weight against Silicon Valley as they should be. Or organising against the likes of Spotify who pay pennies on the pound to artists, despite providing nothing more than a delivery service - not having taken the risk on funding studio time, production, touring, artist development, etc.

Some signs of AI model collapse begin to reveal themselves

"The model becomes poisoned with its own projection of reality."

Ahhhhh, now I get it. Liz Truss makes a lot more sense when you understand she's a late-stage LLM that's been consuming her own kool-aid output, CJD/BSE-style.

Europe plots escape hatch from enshittification of search

Re: distributed search

Stations used to be very common for this as you'd have your stopping or slow train in the platform (stopped, naturally) and the express would continue through on the through tracks running between the slow lines in the platforms. In some cases stopping trains also used to be shunted into sidings to let express trains go past.

These days we don't do any of that, because the sidings are mostly gone and the express tracks through stations have been ripped up to "economise" on track maintenance. We do still have fast and slow lines though so some services can overtake others.

Well yes, which is why opposition to HS2 was so asinine. The entire point was to provide a segregated route for express trains because quad-tracking the West/East Coast/Midland Mainlines would involve demolishing a lot of houses to widen the corridors (even in towns where the train isn't stopping) and rebuilding stations to accommodate extra lines/platforms. Which would be far more expensive than HS2.

Much of the problem is rising line speeds. A nippy local train can mix with an 80mph express train much easier than a 120mph express train, which is why various stations on the WCML closed in the 2000s when the WCML Route Modernisation pushed speeds up past 100mph - even if the stopping service can peak at 100mph, it spends most of it's time accelerating/decelerating. So the disparity in average speed got larger.

Shoving a slow service in a siding is a dreadful option though. Aside from all your local passengers being sat in a siding, you then have to reverse the train out onto the mainline, and then the driver changes ends again to take it forward. Which only works if you are running a very low timetable. Reversing trains murders your trains-per-hour throughput on a line. Most journeys are local/regional so you want to prioritise your local services (in the UK we do the exact opposite - prioritise long-distance and assume people can drive short trips).

The answer is a dedicated inter-city network (TGV, etc), and then you can dedicated your legacy network to high frequency local services (4+tph), regional trains and freight (which will just match its average speed to mix with other traffic).

Re: distributed search

I don't think the network should ever be private, but I think private trains can work, with adequate regulation and strong enforcement.

The problem is, private operators don't want to run mostly empty trains at midnight - despite them being vital for shift workers and suchlike.

So your regulation has to be so robust (to the point of specifying minimum service requirements - "You can run whatever services you like, but there must be trains running till 0000 and starting at 0500 or whatever) that you end up squeezing out any "creativity" in service patterns. A train operator can only ever give the track owner a wishlist and they'll be assigned train paths based on the line's maximum TPH, what other "competitive" operators have asked for, etc. There's no meaningful competition there.

Re: distributed search

Running a train is no more a natural monopoly than an airliner.

The actual tracks, fair enough, I see that, but the trains not so much.

Nonsense. If Concorde took off immediately after a little twin-prop islander running to Jersey, it was not delayed by that aircraft. They have to get their takeoff slots but once in the air they can go their separate ways. Likewise running National Express/Greyhound coaches on the freeway/motorway.

But an express train cannot go just after a slow local-stopping train. They're on the same track and it'll be held up. The answer there is to set off the express train first. Which is fine if you only want to rn one express train per day. But if you want to run another, then you're going to have to wait a long time for the local service to trundle off and build up a good lead/headway before your next fast train can dispatch. This is why (in the UK) it's common for a route to have maybe three trains per hour, but they're all within 20minutes of each other. Then there's a half-hour wait for the next one. It's because they batch off a bunch of fast trains, then the slow trains and then wait half an hour before the next lot of fast trains can go.

To run a train on a railway you need permission from the track owner. They have to schedule you in and assess your speed/performance profile to ensure you don't conflict with other traffic - either running into slower services ahead or getting rear-ended by an express train.

On any reasonably busy section of the UK network it's a one-in-one-out system. If one operator wants to run a new service they need to boot someone else out. Whereas it's trivial to (say) run an extra bus or coach service down the motorway. Worse yet, if someone wants to run an incompatible service - say a 100mph express service that blows through a load of intermediate stations without stopping - this could involve cancelling multiple local or regional services to clear more headway for the fast train.

Consequently, there's no real competition - train operators can bid for train paths, but the track operator (Network Rail) ultimately decides who gets them, often having to abide by regulatory public-service rules that says every station needs at least n services per hour and you can't just leave half the stations without service because one express operator is willing to outbid the local operator.

There's no meaningful space for competition in railway operations. Competition for rolling stock procurement, sure. Competition for hardware and infrastructure procurement. But in terms of running trains, it's a tightly intermeshed system and at the end of the day, there's one scheduler setting down who can use a section of track at a particular time. Who gets which train paths. What priority will be given to express vs. local services.

Re: distributed search

You evidently don't remember the days of nationalised rail and utilities. I do. Shit customer service, huge under-investment in assets, poor technical and operational performance, and massive inefficiency.

British Rail went through some difficult times. You've identified a core issue of under-investment which is not the fault of BR and did not get any better under RailTrack (it got fatally worse, to the point the Government were forced to renationalise RT as Network Rail).

Until 1980, BR was run on 5 regions, inherited from wartime divisions. In 1980 we got sectorisation, with BR split into six businesses:

* InterCity

* Rail Express Systems (parcel freight)

* Provincial Services

* London & South East

* Trainload Freight

* Railfreight Distribution

Breaking BR into sector-specific business units saw the new orgs responsible for driving their verticals - whether that was local passenger services, container freight or long-distance travel. By 1985 services, reliability and passenger numbers were all growing, driven by a new sense of purpose and clearly targeted improvements - InterCity might assess that a section past a freight yard was bottlenecking their long-distance services, develop a business case for improvements or an additional passing line and get it built. In many cases people had not really been paying attention to the express services because they were busy ensuring the freight was reliable and express services just had to crawl past. But now there was a champion for them saying "let's sort this section out" in the same way road bypasses and motorways were being built to avoid town-centre bottlenecks.

The emphasis on business and profit centres played well under Thatcher. Ridership rose, as did freight volumes.

It's important to note that it's nonsense to imagine rail got better under private operators because it was privately operated. RailTrack cut maintainance and treated engineering as below them. They were - by design - a contract management agency, with a focus on sheholder dividends. The decision makers had no engineering background and didn't understand the reports they were supposed to be shuffling from one subcontractor to another. As a result, people died. Yes, passengers were impressed when the likes of Virgin ordered shiny new trains via the Rolling Stock Companies (ROSCOs - who then lease the trains to Train Operating Companies). They were required to do so under the terms of their franchise, and the government could just as easily have ordered new rolling stock - but they didn't want it on their books, in much the same way George Osborne did austerity which the BoE had to counteract with Quantitative Easing to stop the UK entering a prolonged depression. You're just picking whose books the debt appears on.

This of course is asinine because noone can borrow cheaper than the government and cost-of-(private)-capital will always result in higher fares (or energy prices! See: Hinckley Point)) eventually. Particularly when the ROSCOs do not have profit caps as the TOCS did. Even today, a totally disproportionate chunk of your fare goes to the ROSCOs.

The fact is, rail operations have been all-but nationalised since 2020. The TOCs all went into administration and have basically been run by the DfT since then on a concession model (similar to TfL). CrossCountry for instance do not set their own timetables or services. They are contracted to deliver a timetable by DfT with the government paying a flat fee and receiving the farebox. Despite an indifferent service, every rail service outside of London is well above 2019 (pre-Pandemic) levels. In 2022, long-distance leisure was 125% of 2019 ridership. Of course politicos like to tell us that "ridership hasn't really recovered" on the basis national ridership has only just reached 2019 figures - because London commuting remains on it's knees. But that fact that London was the only city with really good rail service and dominated the statistics masks the fact that every other train in the country is rammed, and politicians seem reticent to build the solution (HS2 - which provides a dedicated long-distance route and lets you (re)open lots of local stations and run local services along the lines where London-Glasgow services are no longer slamming through).

The worst part of this is that privatisation saw lumpy rolling stock procurement. There was a big-bang of companies buying new trains in the late 90s and then nothing. As a result, most of the UK's rolling stock factories were closed down by their owners or consolidated. BR had always maintained a rolling programme of fleet renewal. The consequence is that rolling stock either has to be imported, or is built at "pop up factories" which are spun up for the purpose at relatively high per-unit cost. It also means we don't have enough stock for special events or contingencies. For-profit ROSCOs refuse to buy stock speculatively, or to cover surge. They'll only buy what they have a confirmed lease for. This leaves us in a pickle when a fleet of CAFs have to be withdrawn pending chassis inspections.

There could be a place for private firms in rail - as simple contracted service providers, the same way TfL's buses are run. The private firm just runs buses on a timetable with no initiative to add or withdraw service. The farebox - and risk - lies with TfL, meaning than loss-making public services (e.g. late night buses) are not withdrawn as uneconomic, screwing over shift workers. It can instead by cross-subsidised by profitable rush-hour services. Whereas out in the provinces, the bus operators just don't bother.

No-boom supersonic flights could slide through US skies soon

Re: This is all for billionaires like Musk

You sure they require 1900m?

I was quoting quite a specific metric, because it's readily available and a good rule of thumb:

a balanced field takeoff is a condition where the takeoff distance required (TODR) with one engine inoperative and the accelerate-stop distance are equal for the aircraft weight, engine thrust, aircraft configuration and runway condition. For a given aircraft weight, engine thrust, aircraft configuration, and runway condition, the shortest runway length that complies with safety regulations is the balanced field length.

I had 1900m in my head, and on closer inspection that's right for the G650ER and G700. Other models range between 1500-1700m, and could go shorter in a light configuration/half-fuelled. So from your 1500m airfield, it could be a subset of Gulfstreams, or light-loaded ERs which aren't going far or have to refuel (in the same way BA's London City-USA business route takes off light from London and refuels in Shannon).

So Overture could likely go a bit shorter than 3000m if you short-fuel it for a regional hop and have a light load (which is hardly worth doing - they likely won't get high enough to engage boomless cruise before they get back on their approach descent).

And that last point is why these probably won't be that popular as billionaire playtoys - it's going to be a really expensive way of making regional hops for the sake of going really fast on (relatively occasional) longer runs. Which the Emiratis won't care about, but a lot of businesses will. It won't really replace big Gulfstreams. I suppose some of the charter companies like Netjets/Flexjet/Vistajet might have a couple which would then be assigned for medium/long-haul charters with smaller jets for regional/European hops.

Re: This is all for billionaires like Musk

This is probably possible in the US, enabling the slot competition to be sidestepped, although I wonder how many private airfields will have a sufficiently long runway.

Although it isn't readily findable on their site at the moment, various (slightly dated) sources list a balanced field length of 8,500ft-10,000ft, which is 2500m-3050m in real money. So a significant airport. There won't be many private airfields able to support it.

Boom claim to have identified 600 worldwide routes. Which actually isn't very many when you think about that as a full mesh.

People do keep talking about the Overture as a billionaire plaything and that's sort of fair, but we should remember it's a 60metre aircraft seating 80. It's twice the length of a Gulfstream G600 (which seats ~20 and has a BFL of 1900m).

Some billionaires do have large jets (Trump's 737, various Emirati jets), but Overture is really large enough to be luxury-airline class. Although quite why airlines would want to undercut their own (highly profitable) subsonic First Class business is an open question.

As soon as passenger rail travel becomes popular enough it too becomes a target for people with bad intentions and you can bet your top dollar the TSA is going to want to install the same level of security theater in rail terminals before letting you onboard.

And yet... You can travel internationally across Europe and you might, might get spot-checked on the train itself by border police when you're passing into (say) Switzerland.

Of course, I appreciate that the US is stuck with the Mall Ninja/Paul Blart characters in the TSA, as well as the Air Marshals who have been bitter ever since they were made redundant barely a year after they were created when metal cockpit doors and a simple bolt largely eliminated any risk of hijack.

The only serious security checks on European rail is on the Eurostar, and that's because the British government is perpetually fearful of dirty furriners. From a security point of view, anyone wanting to (say) try and bomb the Chunnel would be better off trying to get a car bomb onto Le Shuttle than dragging a much smaller device onto Eurostar.

Let me just check how much was a Concorde ticket back then. Ooops!

To be fair:

1. Concorde required a flight engineer - which was one of the reasons BA/AF wanted it gone.

2. Concorde consumed as much fuel taxiing from stand to runway at Heathrow as regional airliners consumed flying to Berlin. I always wondered why they didn't just tug it out to a runway-adjacent hard standing for engine start.

3. When BA first launched Concorde services they did some market research on what people were willing to pay, and it turned out to be double what they'd actually expected to charge. The underlying operating cost was not as high as one might imagine (albeit they never had to amortise any capital cost since the British/French governments sold them for a quid).

4. Concorde never actually turned a loss. It was always profitable - but following the post-9/11 downturn in aviation, BA realised that subsonic first-class was more profitable per seat, and so they were better off dumping Concorde and depriving the market of that option in order to force people into the higher-margin upstairs seats on 747/A380.

So there was a market for this, and I have no doubt there will continue to be a market for this. And using modern engine tech, the fuel bill will likely be more reasonable than Concorde's (which still allowed them to turn a profit!).

All of which still doesn't make this a better option than building out high speed rail (bringing Texas triangle or Midwest cities within an hour of each other), and interlinking those regional networks to allow coast-to-coast sleeper services. Boarding a train at 6pm in NYC or DC, having dinner and a good night's sleep before you pull into LA at 9am is always going to be more comfortable than getting up at 5am to be at the airport for 7am so you can get touched up for your 9am flight and then sit like sardines for n hours. Faster flights don't make up for the fact that airports are a painful experience and - by their nature - usually out of town, well away from the place you're actually travelling to (Los Angeles being somewhat the exception to the rule).

I mena, I have no doubt that Boom will try to get Overture passengers the Concorde experience - a private check-in desk and security suite so they can basically arrive and walk onto the aircraft. But at JFK they'll still have 40minutes of taxiing, and at LAX they'll need a private helicopter or else spend the next hour or two in traffic.

How sticky notes saved 'the single biggest digital program in the world'

Re: Yes but

Let us not submit to the vile doctrine of the nineteenth century that every enterprise must justify itself in pounds, shillings and pence of cash income … Why should we not add in every substantial city the dignity of an ancient university or a European capital … an ample theater, a concert hall, a dance hall, a gallery, cafes, and so forth. Assuredly we can afford this and so much more. Anything we can actually do, we can afford. … We are immeasurably richer than our predecessors. Is it not evident that some sophistry, some fallacy, governs our collective action if we are forced to be so much meaner than they in the embellishments of life? …

Yet these must be only the trimmings on the more solid, urgent and necessary outgoings on housing the people, on reconstructing industry and transport and on replanning the environment of our daily life. Not only shall we come to possess these excellent things. With a big programme carried out at a regulated pace we can hope to keep employment good for many years to come. We shall, in fact, have built our New Jerusalem out of the labour which in our former vain folly we were keeping unused and unhappy in enforced idleness. John Maynard Keynes, 1942

And yet... our neighbouring council are proposing to close their town theatre for lack of funds, whilst the Chancellor insists we are not doing another round of austerity.

It should be perfectly possible for a single earner to keep a family in some modest comfort - yet many couples are working flat out to keep a roof over their heads.

That 10% of the population have zero savings is not through profligate spending (albeit some people won't be helped) but through simple wage stagnation and the insane cost of housing.

Re: Yes but

His design for UC was to be more generous - but of course this was fucked up by the Treasury - who seem to be the most malign force in British government. In that they're very good at cutting spending, but seem to also excel in doing it in ways that actually end up costing the government more money in the end.

I raise a Home Office for pure malice.

But yes, Treasury are responsible for stagnating so much of British government. Inane requirements to flatten spending curves so that infrastructure costs more overall (in actual cash terms) because you're asking contractors to work slower than they could. Ships that we could build in 10 years get stretched out over 15.

This also incurs opportunity costs to the economy because infrastructure doesn't open as soon as it could and start paying for itself. We could spend £1Bn on a new rail line and have it open to paying passengers in 2028, but Treasury would prefer it cost £1.3Bn if they can delay completion till 2032. Nevermind that the entire point of having your own currency and central bank is you can borrow big chunks in one go for those sorts of projects (and nobody can borrow cheaper than the Government), and then pay it back over a time horizon (10-20-50-100years) that the private sector isn't interested in, but which makes total sense when you consider the working life of the asset you built.

India and China have demonstrated the folly of the smooth-spending approach. India electrified more miles of railway last year than there are miles of rail in the UK. China's economy is ascendent. A colleague was blown away by a recent holiday to the Middle Kingdom. Travelling to several large cities, everything was clean, worked, and he described the technology as like looking into the future. I suspect he didn't look too closely at the outflow of the Yangtze, but he was amazed at the level of development. Of course, China does have the flip side of the authoritarian government, slavery of the Uyghurs and everyone earning their social credit on state-controlled social media. But the benefit of robust corporate taxation funding massive public infrastructure is there for all to see, and the authoritarian bits are not a pre-requisite to building infrastructure. Capitalist USA enjoyed enormous growth and managed to build a lot of infrastructure in the 1950s and 60s until corporate lobbyists convinced politicians that it'd be better not to tax their billionaire clients.

It has been proposed that (in common with many European nations) the Treasury needs splitting out along the three roles it currently performs (budgetary - controlling spend; financial - public debt & tax; economics - promoting growth).

At the moment we just get a lot of hawkish "can't have debt" and they just keep us treading water (albeit no small feat in itself).

One benefit would be that the Chancellor and PM would have three senior "finance" civil servants pulling in different directions, at least one of whom would be actively advocating for infrastructure spend - you wouldn't end up in a situation where potentially your top Treasury bod is just telling everyone to slow down and do less. There would be diversity of opinion, and departments trying to make a case for investment would have an ally to help them take the fight to Treasury.

Re: "assumptions don't turn out to be what humans look like when you hit them"

I've had the misfortune to spend time signing-on in two different parts of the UK around 2008/09 when the world was terrible.

The experience between the two was incedibly uniform (in how bad it was, and the staff were as much use as a chocolate tea pot). Happily I was employed by the time the Tories made Sanctions-per-week their KPI.

First time signing on I naively took my CV and what I thought was relevant paperwork. "Don't need that" they said, as they then proceeded to ask about my education and work history...

No advice was offered at any point. They looked at your jobs card (stuff I had applied to) and ticked the box for "yup, they're looking for work". It was quite the realisation to find out that the Job Centre exists purely to administer benefits and has no interest (or competence) in actually helping you find work. It turned out there was a training scheme I'd have been eligible for (and up my street - the field I was applying into) that they were could have referred me to. When I learnt of the scheme I went round on my own initiative and knocked on the door of the centre that delivered it. They had a look and apologised profusely as my eligibility had expired a week previous (based on how long I'd been signing on). They were quite annoyed as well, that the JC weren't referring people who were actively looking for work in the sector they were training for FFS.

Re: "assumptions don't turn out to be what humans look like when you hit them"

Or in other words - there IS a trickle effect. Money trickles UPWARDS

The sort of people who promote UBI schemes tend to propose them within a framework that includes a working tax system and rabid (anti-)monopoly regulators.

70-knot winds so far blamed for yacht disaster that killed Brit tech tycoon Mike Lynch

whether the keel was up or down at the time.

The interim report (linked in the article!) indicates it was up.

They note that the Stability Information Book had data for lowering the centreboard when under sail but not in a motoring condition, which seems like an oversight by the builder.

Britain's cyber agents and industry clash over how to tackle shoddy software

Re: BINGO

So these NCSC apparatchiks seem to know nothing about the real world. Their "code of conduct" thing is also close to useless. Hot air without any computer science substance such as "proper scanners, proper parsers as first line of defence" or "fuzz testing".

Ollie Whitehouse came to NCSC after 27 years in private industry - CTO at NCC, research at Blackberry EMEA and Advanced Threat Research at Symantec.

He has literally some industry knowledge - he didn't join the Civil Service straight out of uni and spend his entire career in policy.

At the end of the day, he knows that he's fighting indifference in government and has to play a longer game. If we wanted secure government we'd move Government systems away from US cloud providers. But telling UKGov to ditch Microsoft is not going to get them very far in the short term. They're also having to deal with an environment where - Councils get told not to admit getting ransomware'd by the Home Office when they get caught out. Which is the exact opposite of what you want - post mortem that stuff. So the NCSC is fighting a sometimes-indifferent Civil Service who are sometimes more interested in saving face than anything else.

Once upon a time there were no standards for highly assured software in aerospace, etc. Now there are. There will always be a question over whether vendors apply them (*cough* Boeing) or whether the notional regulator actually checks vendors or lets them mark their own homework. But the existence of those standards is important. It's why Boeing customers have said "We want our own reps on your factory floors or you can take your shonky planes back" and Boeing has had to say "yeah, okay".

Imagine if a victim/"customer" of Cloudstrike said "we want our own QA rep embedded in your dev group to assess dev and deployment processes". They'd be laughed out the room and told "no, that's secure and proprietary". Yet Crowdstrike's epic fail last year was no less damaging than Wannacry, and whislt it's not as dramatic as a door plug falling out in-flight, massive system failures can have lethal consequences. A global outage, hundreds of thousands of boxes bricked and requiring manual intervention. Massive disruption to travel, banking and retailers. There was a national ground stop in the US. And this from a company who had previously tried to block NSS Labs from publishing a report on Cloudstrike's end-point protection in 2017, claiming that NSS had illegally accessed it's software to perform "improper security testing".

I put it that if the NHS or an affected airline said "right, we want an embedded QA rep monitoring your deploys" they have as much right to that as having reps on Boeing's factory floor.

Developing frameworks and good practice is a pre-requisite to holding organisations accountable for not following best practice. And in the process, nudge the "Overton Window" equivalent for what is expected of vendors in the right direction.

Obligatory XKCD.

Re: You should only enforce honesty, security and quality on the tech sector...

It's possible that some will walk from the EU

Well I hope they don't let the door hit them on the way out.

We have plenty of EU/EEA/European alternatives.

Two major OS houses in Canonical and SUSE, no shortage of productivity options between LibreOffice/Collabora Office and OnlyOffice. JetBrains, NextCloud, Proxmox, Acronis, etc. Jolla can even do us a mobile OS in SailFish.

DC/Cloud? I think we can get by with OVH, Scaleway, Hetzner, Exoscale, UpCloud, et al.

Encryption? Ah yes, Norwegian Buypass have their own root and offer an ACME SSL service.

But how do we network? Well, Nokia, Ericsson and Microtik have a lot covered off. There may be things that Cisco/Juniper/Extreme do better. But nothing irreplaceable. Andrews & Arnold dogfood their own Firebrick routers on their network (all software & hardware design, development and manufacture is in UK. A genuinely "sovereign" product).

Actual compute hardware is harder, leaning heavily on x64 IP. But let's face it, we're not going to lose access to US compute any more than the US could stand being cut off from ARM (or ASML!). Relying on PRC and Taiwan (Lenovo/Acer/Asus) in the absence of Dell/HPE/Supermicro would be a bit iffy. There's Fujitsu and Vaio of course, and Medion in Germany (owned by Lenovo, but that's a matter of finance - the HQ & engineering is in Essen). And system builders like Framework or Stone Group.

Seriously. We're fine.

More importantly, Windermere town isn't really on the shore of the Lake. if you drive into Windermere and park up expecting to get an ice cream on the shoreline, you'll have a half hour walk to get there. Those looking to rent a boat on (Lake) Windermere want to go to Bowness (or Waterford), not Windermere town. Windermere railway station is at the top of the town and quite a step from the lake itself.

If Google is forced to give up Chrome, what happens next?

Re: This is madness

Obviously one remedy would be to make Google sell Chrome. But if people really are talking about $50bn to buy it, then those people are going to be exactly as much of a threat to the market as Google are. So a different remedy should be looked at, in preference. Otherwise we risk making things worse.

Perhaps the better thing to make Google divest would be Android. That does more damage to their monopoly, and that actually makes profits, so rather than selling it, they could just spin it off as a private company.

Yeah, the $50Bn valuation is insane. Obviouly you're buying the userbase - I know everyone says that it's impossible to build a full-featured browser engine these days (and in fairness, MS dropped EdgeHTML as too much effort), but I reckon - and I'm spitballing here - that a billion quid would do it. Two at a push.

Spin off Android into a company that needs to make a profit and can't derive cross-subsidy from the Alphabet family, nor benefit from app bundling. Make them license themselves out to manufacturers on the same basis as Jolla/Sailfish OS.

Re: This is madness

Stopping them from getting that monopoly made sense, but that ship has sailed - and regulators did fuck-all. Even when Google were distributing it like malware and using their search monopoly.

This is fundamentally the issue. Breaking up Google post-facto is still arguably the right thing to do (although arguably they should be selling AdSense as the priority), but the issue we have is that anti-trust bodies are so weak and under-resourced that their findings are always years after the fact, and their attempts to patch the damage aren't going to bring back the startups that have been squashed in the process.

Such has been the case for Microsoft's multiple infringements, and even for non-tech cases like the UK Dairy product price fixing scandal (in which supermarkets and processors colluded to fix prices of milk and some dairy products. Case opened in 2004 alleging consumers had overpaid by £270m and finally closed in 2011 levying fines of... £49m. So seven years and a paltry "remedy" that's basically a cost of doing business - after all, the offenders retained £220m of their ill-gotten gains, which some might call "proceeds of crime").

UK Ministry of Defence is spending less with US biz, and more with Europeans

And a big dependency on the US Military's F35 fleet is the 15% workshare the UK has.

Whilst not irreplaceable, a churlish USA would suddenly be tendering rather rapidly for a variety of components before their fleets were grounded for lack of spare parts and consumables.

Whether it's the Rolls Royce lift fan in the F35B, or the Martin-Baker ejection seats used across their fighter fleet - F-35, F-18, T-45 and some F16s. For sure, they can turn to Collins Aerospace, but how quickly can they retrofit thousands of seats, or reverse-engineer compatible components for MB units?

If the US wants to mutually ground both US and British defence assets then they can. And yes, it will probably hurt the UK more than the US (albeit, in such a situation it seems like the UK would have an easier time accessing European vendors than the US would).

But they would be cutting off their nose to spite their face (not that that's stopped them before).

The thing Trump et al need to remember when they whinge about self-sufficiency is the SR-71 Blackbird was built from... Soviet titanium. Laundered through a variety of nations because the USSR was the only plentiful source of rutile ore. Even during WW2, some US military equipment used German-origin machine components laundered through Switzerland (and Germany also relied on certain components coming from Allied as well as neutral nations), because quite simply, they were the best out there. German and British aircraft ended up with Swedish ball bearings because domestic ones kept cracking and the Swedes had got a really good process. They also had an abundance of iron ore.

Via Switzerland, Eastman Kodak openly imported goods to the US that were manufactured in Nazi Germany - this was criticized by American diplomats at the time, but they received no penalties for such collaboration with the enemy. All this globalisation and free-trade they complain about is actually what allowed the US to become the power it is.

So yes, there are many dependencies. And we probably have more dependencies on the US than they have on the UK or Europe. But they have enough reciprocal dependencies that realistically all our hands are tied - unless they want to commit significant self-harm (which apparently they do).

As far as the Russian nuclear deterrent goes... I'm unconvinced that many missiles would make it out the silos given the appalling state Roscosmos has been left in and the rising fault-rate in Soyuz rockets and capsules. Or the condition of their sole aircraft carrier (or their navy in general) and the fact that their entire campaign in Ukraine has largely collapsed to lobbing small missiles, glide bombs and drones over the border. If you're stuck for manpower, trucks, tanks and aircraft then ramp your factory capacity to churn out a small range of low-tech fire-and-forget munitions in large volumes.

Which is not to say that I really want to take that risk, because it only takes one. And maybe in the absence of conventional forces they have actually quietly poured resource into the nuclear arsenal. But what are the odds the Russians have actually been funding the substantial inspection and refurbishment programme necessary to keep nuclear weapons and ICBMs in operational order? The engineering apparently being so complicated the US forgot how to manufacture key components and had to reverse engineer them... The new RS-28 Sarmat missile doesn't really work, which was supposed to replace the R-36M2 Voevoda, which was designed, built and maintained by... a Ukrainian firm.

Microsoft moved the goalposts once. Will Windows 12 bring another shift?

Re: Reasons not to upgrade to Windows 12

But you can still buy a copy of office to install on your laptop to use for X years...

I stand corrected. Although I would state that my comment was not idly made - I went specifically looking for it yesterday to check that I was correct... And I didn't find it. It's just taken me 5 minutes trying to work out how to organically browse to the page you've linked and my god is it a case of "on display in the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard.".

I'd bet folding money there are millions of people on a 365 subscription who wanted a one-time license but simply didn't find it. And that is entirely intentional on Microsoft's part.

Re: Reasons not to upgrade to Windows 12

To some extent it's the stock market's fault. They can't be satisfied with a nice little earner, they demand growth.

This is basically it. And it's the one thing that everyone seems to be right about - the current assumptions underlying our economic model are no longer true. We just diverge on whether we go down the tech-bro corporatocracy and sell the world to the billionaires in some sort of anarcho-libertarian act of onanism, or go the other way and recognise that a profitable business is a profitable business. You don't need to go out of your way to pad your figures to please the speculators at the stock market if you're making money and happily paying your staff and taxes.

In the consumer tech space, most products are way past "good enough", and consequently many industries are on a replacement cycle, not a growth cycle (Reg passim: "The channel stands corrected: Hardware is a refresh cycle business now). My mum is still rocking her Macbook from 2012 (with RAM/SSD updates). It's fine for normal browsing/productivity.

My Dad adheres to the business module from his 1970s engineering course where "you must have growth". And back then it was somewhat true - in a rapidly growing marketplace, a failure to grow meant falling behind. But "growth" then meant improving tooling, improving your assembly lines. Achieving more output per worker hour. Investing (what a word!) in robotic welders and CNC toolstations. But in a static market where population growth has tailed off and the speed of welding robots is limited by the rate of heat flux (i.e. you're up against physics of metallurgy) rather than better controllers or more precise stepper motors. What businesses call "growth" now, economists call "scaling" - opening a second line with additional robots.

And in the software world... "growth" means growth in profits, not economic output.

Take Adobe. Adobe got to the point where basically everyone who wanted Creative Suite had it. People were skipping versions, going from CS3/4 to CS6 or whatnot. As their ideas for new features tailed, their MBAs scratched their heads and moved the whole thing to a subscription model, marketed as a lower barrier to entry and you "got the latest features". Which frankly were pretty limited until the last couple of years when inferencing/AI tools like content-aware fill and selections have been legitimate improvements.

Adobe have seen great growth in profits but not in product development or productivity. Indeed, many of their products have become less stable - to the point of alienating their own ambassadors.

The same could be said of Microsoft. My Dad couldn't believe that he couldn't just buy a copy of Office any more. Why sell you a copy of Office for £150 to install on your laptop and use for 5 years when they can extract £500 in subscriptions over the same period? Of course there's a cost in building the online version and providing you with some storage that you don't want... but not that much.

The fact is, Microsoft Office did basically everything you needed it to do in about 2008. Most "improvements" since then have been fudging about with UI for no particular benefit. Some people prefer it, but it doesn't actually work any better than it did (I'm sure there are some specialist functions in Excel that a handful of users value, but for the 99th percentile we're using it for the same thing we did in 2010 on a dual-core i3 with 4GB of RAM. I say "we". I use Libreoffice).

The world is paying more for less, and people are now waking up to the fact that this smokescreen of GDP growth doesn't equate to improvements in employment or business productivity (and that GDP is an increasingly meaningless measure of a nation's economic output). Whereas in China, GDP growth does equate to them actually making more stuff and developing manufacturing/tooling/development expertise.

90-second Newark blackout exposes parlous state of US air traffic control

The thing about air space and planes is: planes are actually relatively small compared to the atmosphere above a county (city? only for a few seconds!) It's probably much less likely that two planes, flying on their current path, would impact each other than getting struck by lightning

In normal airspace yes. Not in crowded airspace around NYC where aircraft are departing every minute and being sequenced for approach/landing. Newark's sector is busy on it's own. But it's adjacent to JFK, LaGuardia and Teterboro. Not to mention all the rotary (helicopter) traffic around NYC (see recent DC crash).

I have to suspect that anyone reduced to tears or suffering heart palpitations as a result of a 90 second glitch probably isn't cut out for the job anyway.

ATC is a stressful job. People wouldn't make it through training if they couldn't handle stress (albeit the burnout rate is still high). I went through part of the process at NATS (in the UK) and there was a Q&A with a recently qualified controller who was completely candid about the fact that at various points in your training you'd go and have a cry in the loos or your car. They wrung it out of you on a "train-hard, fight-easy" basis.

But what happened at Newark is basically never-never "you're a passenger in a car going over a cliff" territory.

Lose radar? That's bad, but you have headings and you can talk to the aircraft, let them know the situation, keep people parked in their stacks and potentially talk to other control centres to have them take over (at least for aircraft declaring low fuel, etc).

Lose comms? Well at least you can call other control centres and tell them "Hey, you need to move this aircraft and that aircraft".

ATC train for loss of tech. They can fall back to paper (albeit at a loss of capacity).

To lose all radar and comms is basically a matter of "go to the break room, turn on CNN and watch the aircraft crash that you were responsible for". You've got airliners with hundreds of pax doing 400kts and they're basically onto visual flight rules - except you haven't told them that. Because you have no comms. And even if they were all neatly on diverging paths in your sector, at 400kts some of them will be leaving your sector soon (and other joining) and if those controllers are sat around you looking at their similarly dead consoles... well bugger. Your diverging paths weren't necessarily clear through the next sector or the one after that.

It's a high responsibility job and you have been left with nothing. And momentarily that's bad, but recoverable. About 45seconds into staring at a blank screen with no radio and having called everyone you have a landline to saying "we got nothing" you'd be getting very stressed about where the hell the backup is. At 75 seconds? Yeah, you kick back and either laugh or cry because what else is there to do? And I'd be worried about the ones who were laughing.

I don't know what US shift patterns look like of ocurse. Back when I looked at it here in the UK, the general imtimation was that you couldn't spend more than 45 minutes on console between breaks, and usually they had enough staffing that you'd be relieved before making it to 40. In an 8-hour shift you'd be unlikely to log more than 4 hours on console.

Top sci-fi convention gets an earful from authors after using AI to screen panelists

Re: The problem is NOT the AI...

The problem is the QUERY ITSELF, which SMACKS of CANCEL CULTURE:

Due diligence is cancel culture?

No. Due diligence ("including but not limited to") is background-checking your panellists and then letting a human assess whether any findings are relevant or disqualifying (not that I would willingly put AI into that loop). And of course, disqualifying does not need to mean a judicial outcome/conviction. A potential panellist might be coming from a nation where homosexuality is illegal and their homophobic opinions are state-endorsed.

But clearly at a western Con, it may be unlikely that many attendees will want to hear them speak. In which case it would be a poor use of Convention funds to pay their expenses (and potentially an appearance fee) for them to come and speak to an empty room.

The customer is always right. If noone wants to hear a panellist speak, or if that panellist is a reputational risk to the event, then it would be bad business to spend money putting them on an empty stage in an empty room! Also, most conventions have some form of Code of Conduct. It would be asking for trouble to invite a panellist whose has behaved in a non-compliant way recently or in a relevant setting. That's just setting up for having to deal with a complaint at the Con and potentially ejecting your own speakers.

Linux kernel to drop 486 and early 586 support

Re: Hubble Telescope.......

Because nobody is using Linux on those processors. Not modern Linux anyway. Maybe an old kernel that doesn't include lots of "bloatware", or modern affectations like ARM or x86-64 support! For best results, download a version of Damn Small Linux from 2008.

In 2018 these nutters got modern (Gentoo) Linux running on a 486. It took 11 minutes to boot to CLI.

And in fairness, they did get it playing music and serving a webpage via nginx. It also cloned a git repo at a magnificant 50KiB/s, took ~4 seconds to return "python3 --version" and 15seconds to run a "Hello World" script.

If you're actually doing real workloads on a 486 you're probably not using Linux - you're using a RTOS in some embedded application. Booting it into a complete modern Linux environment is a woeful use of a 486's useful but limited resources.