Posts by Lee D

I wouldn't bother.

In this instance, stamp it "Not at this address" and put it back in a postbox.

The longer they are receiving mail (without this happening) at your address, the greater the chance they're affecting your credit record because the banks will think they still live there along with you.

Any kind of contact is going to reinforce that.

Just mark it not at this address, post it back and keep doing it until they stop.

Whenever I move (and I bought a house last year), I do this with all mail addressed to previous occupants. Otherwise you're just making trouble for yourself. As it is, the guy somehow appeared on my electoral-roll registration form this year and I corrected it (and then received a letter from the council addressed to him, which I returned... that's their "test" to see if he still lives there).

You are opening yourself up to card fraud (like the other post below yours here), credit record merging, etc. if you're not careful - I know, because I've had that at previous addresses, even to the point of debt collection people turning up at the door looking for the previous residents.

If it's not addressed to you, you're not supposed to open it, either.

When companies - including councils, banks, lawyers, etc. - get mail marked "not at this address", they start cancelling the accounts automatically because they are being told it's not an address you live at, and if they have no other address they will shut down the accounts and wait for the owner to contact them. I know because I've seen that happen too! Just a few letters marked "not at this address" is enough to cancel someone's credit card, for instance, and debt collectors actually tend to respect the same (but usually after a more persistent contact campaign at first).

Indeed, I have a bluetooth, battery, portable one in the car for my own purposes.

But these ones were hardwired USB, cheap, simple, fast, plain-text, standard, and can print out hundreds of receipts before they need paper changing.

And not long after I built the system, I bought a box of receipt rolls off eBay so they have enough receipts to last them 10 years (since I made the system), and another 10 years on top.

Infrastructure should be nationalised.

It should be paid for by selling its services to consumers via privatised companies.

This way, we get the blanket, funded expansion of standardised services without companies rolling their own proprietary nonsense, interfering with each other, or one "gaining" frequency rights that the others cannot.

And you and I deal with any company we want to, with all of them having the same national coverage, but competing on price, service, support and deals.

Same for all industries - water network should be nationalised, water supply privatised (it would be nice to be able to CHOOSE WHAT WATER COMPANY I WANT TO USE, which currently isn't possible!). National Rail, private rail trains and services. National Grid, private electrical firms. "British Gas", and private energy firms. "British Telecom" and private firms overlaying their individual offerings over the national network.

All-nationalised is terrible.

All-privatised is terrible.

And switching the roles (privatising the network / infrastructure) is terrible.

Nationalise the infrastructure, privatise the resellers and services laid over it to offer to the public.

Re: Million to one chances ....

But what if it's not EXACTLY a million to one?

*cough* SMOOTHWALL *cough*.

Re: re: Stenberg admitted that the flaws found in curl...

As it says later on - if you want to volunteer, get on it.

The use-cases and options in something like curl mean you could easily spend several years trying to get there, and not be able to keep up with curl changes in the meantime.

Also, Rust, etc. are not panaceas and CANNOT act on untrusted data safely either. That's why you have unsafe(), etc. which is literally as bad as C but also can affect guarantees of "safe" Rust code near it in memory if they should fail or miscalculate.

When something is literally designed to act on untrusted data from the Internet, the number of sanity checks you would need are ... well... insane. And Rust etc. can't always fix those kinds of problems, or even deal with them at all.

It's why Rust is no good for device drivers, low-level kernel, memory management code, etc. - at some point in those tasks they have to act on untrusted data without a "type" and then form a well-specified type out of it in an error-free manner in order to be able to use it (e.g. DMA just gives you a memory address where the hardware is putting stuff). Those kinds of things require unsafe() code blocks in order to do, and unsafe is just as dangerous as C but with a false sense of security about surrounding Rust code built-in to it.

Sure, there's a lot that you can make safe in myriad ways, but you can also make it safe by using a safer version of C/C++ with a compiler that throws a fit at everything. And that's a far more viable outcome for a volunteer project than throwing out 30 years of development and starting again in a bug-free manner.

Well, at least he found a use-case for an iPhone.

Re: Hope springs eternal

You need to come up with a catchy name first, some sort of "local cloud" pun.

"Run your own Drizzle server" or something like that.

Re: Meanwhile, here in the UK

As someone who reported several sophisticated attempts to take £100,000's from previous employers, including some of the best build-up to accustom people to their interaction, several live phone calls, very detailed supplier impersonation, etc. and each time stopping and recording it officially as a cyber-crime with reams of evidence...

Not once did we ever have any follow-up whatsoever. Not even "no conviction" or "radio silence so they could investigate", but literally not even someone bother to contact us regarding more information, gathering evidence, etc. etc.

I think the national cybercrime reporting sites just end up in a black hole and the data only ever used for statistical reporting.

2 didn't have it.

15 doesn't have it except in legacy support.

You have to have a Speak-enabled ISO, and nobody said that was the default.

Re: Glass cutting

How many replacements before you just put a simple shield or flap over the front of the floppy drive?

"They" never had a clue.

Linux was never really suited for the desktop, mostly because it just doesn't run popular Windows software. This was always the case. People at home don't want to have to break out emulation tools and suffer incompatibilities just to open a Word doc. What killed it though is that people at home would associate "desktop" with "gaming machine", even back then (Warcraft, etc.). If you can't run games, is it still a home desktop?

The situation has evolved over the 25+ years I've been watching it. In that time, I've run Linux as my primary desktop for over a decade at a time at points. Even while managing Windows networks for a living. Of course it's viable. Always was. Just so long as you didn't expect to run the latest game or faddy videoconferencing program on it.

But the fact is that every market where there wasn't that expectation - of computer-doughnuts being able to do the most difficult of things to run the simplest of programs that really weren't suitable for the OS anyway - Linux walked in and owned the joint. Datacentre. Server. Embedded. Smartphone.

Now we're seeing Microsoft people MOVE AWAY from Microsoft-only things to the web... and Linux is creeping straight in to fill the gap. Schools are deploying millions of Chromebooks. Because if all you need is Chrome now to "run Office", then they're far cheaper, easier to manage and easier to lockdown (I know, I work in schools). Increasingly that "at home" laptop which granny just uses for browsing eBay for knitting patterns is becoming a Chromebook or even an Android tablet. Hell, I know of a school where they have ChromeOS Flex on the desktops and Chromebooks for the staff / kids. They don't need anything else any more. For work desktops, it's there and being used.

"They've" been saying the Linux desktop since the 90's because... I've been running a Linux desktop since the 90's just like them.

And now? Now I have a Steam Deck. Literally just Linux, running AAA Windows games on Steam on day of release at full whack, and sometimes better than Windows can run them natively. It's only possible because the emulation layers have improved enormously and standards like Vulkan have given a common base on which to implement things like DirectX -> Vulkan conversion layers. Valve invested HEAVILY in terms of time, money, people and effort to make that happen, along with the Proton people basically performing miracles (and, most importantly, they aren't done yet and still keep working on it).

Increasingly the x86-Windows-only thing is dying. Even Microsoft's own offerings are moving away from that. They want you to have Office on your iPad, so they can sell you Office. It took 30 years but common sense is starting to prevail again.

And now you can have a Linux desktop any time you like, for most people. A Linux *gaming* machine, okay you have to be careful but there are several viable options out there. But a desktop? For work and hobby and general use? Yeah, it's there. Use it.

I can plug my Samsung phone into any HDMI cable (or adaptor with HDMI), plug or connect in a mouse and keyboard (I have bluetooth models of both), and I basically get a "Chromebook" with all my Android apps to boot. Samsung DeX. I've used it to do all sorts of stuff. I literally have a Linux desktop in my phone. And - apart from gaming - I could do my job on that, and all my home browsing and other needs. And with the Steam Deck, gaming isn't an issue either. 1/3rd of my games are "officially" supported, including some new releases out of the box, and I can coax 80% of the rest into working with only minor tweaks.

Fact is that 95% of everything I do on my Windows laptop (bought for VR gaming), I could do on Linux already. Video editing I use Shotcut. Media I have Plex. Almost everything else is open-source or in the browser. It was only gaming that made me go Windows, and VR gaming at that (a particular weakness because even SteamVR struggles to match features on Linux and Windows, but it's fast catching up). If I was to leave that laptop only for Windows gaming, and had funds or felt the need for a different machine for daily life (I've left it on Windows 10, for example)... Linux would be my go-to.

And I can count the number of people I know with a full VR setup on my elbows. Me. That's it. Everyone else, if they have VR at all, has the ones where they have a tablet inside the headset (and that's almost universally running some kind of Linux).

Next time granny breaks her machine or only has an old clunker and wants to upgrade... stick ChromeOS Flex on it for her. It's free. She can do basically everything she needs on that, and it takes away so much legacy computing that it will make her life so much simpler. And that computer that was slowly falling over will have a new lease of life. If it doesn't work? So what. She had to buy a new machine anyway. But whenever I've done that for people, they find themselves using it for years more - whether alongside a new PC as a convenient Chromebook-like machine, or not even bothering to upgrade at all and just staying with it.

"They" were only ever wrong about public take-up. That's what we really mean. Are the public going to run a Linux desktop? And increasingly they are and don't even realise, while their satnav, mobile phone, and all the things that power their web-based services are Linux already anyway. They don't even know.

But you've been able to use (not just "run", but actually use) a Linux desktop for the best part of 2 decades, minimum. My university in 1997 had dual-boot Linux and NT desktops, and you could do everything you needed to in either. I was a maths student and I used Linux more than anyone else even back then.

I honestly judge IT guys who apply to work with me, or those I deal with professionally, who are still parroting things from the 90's that weren't even true back then, and they haven't bothered to update their information since. Telling me that you haven't even TRIED a Linux desktop in 20+ years of working in IT is like the other classic - implying that open-source software is somehow insecure because "everyone can hack it". Both those things show me a level of ignorance outside of everything MCSA that worries me about your ability to adapt and learn and understand other things.

Last guy I hired as an entirely untrained IT apprentice, I selected the young kid who'd run Linux at home and had a bunch of valid, up-to-date rationale for why he didn't like to use it as a desktop. He'd tried, learned, identified, and rationalised. We worked together through his whole career. He's an IT manager now, and puts all his peers to shame, not to mention MSPs and other outside entities who he has to deal with. And he has several Linux desktops.

Boot up ChromeOS Flex. Buy a Steam Deck. Or, better yet, just install Linux on your machine. It's literally part of Windows now, you don't have to risk your bootloader (those days are gone), run it in a VM. You can install an Ubuntu VM in about 4 clicks from modern Windows. Or, shock horror, wipe a machine and install Linux on it and try it.

You can tell me that it's "not for you" (I *HATE* MacOS, and I can't stand many of Linux desktop environments, and I would gladly surgically remove SystemD entirely), but I question anyone who says "you can't run Linux desktop" or similar nowadays. It means they haven't really tried it or think that everything non-Linux should work perfectly on it.

Re: Ai Dillema

Trolley problem is easier to solve than that.

Do something. There is no win situation at all. Just choose one rather than dither and possibly end up doing both (e.g. hitting the pedestrian, and smash into the truck cutting into your lane, etc.).

People thinking that the trolley problem has a "good" solution are the problem. It doesn't. Just act.

And in 99.99% of situations, slamming on the brake and letting physics determine it for you is about the best you can ever hope for.

Re: Connections..

Or Starlink. Or 4G with a myriad of providers.

For 4 members of staff, you really don't need that much, but also you shouldn't notice the cost of a Starlink.

(And, personally, I hate Starlink and can't wait for alternatives, but you have to consider it).

I lived as an IT Manager for 5 years with having 4G only at home - including for VPN, CCTV, smart devices, TV streaming, etc. etc. etc.

I moved house and did get DSL at a decent speed (finally) but I also brought my 4G router AND was on the cusp of reluctantly parting with money to get Starlink if I couldn't get decent speeds.

As it turned out, the broadband is meh but usable and cheap. And because I still have the 4G in it, and that still works well, I don't really care if it falls over until it gets to the point that I feel like asking for a refund.

If Mr Bezos ever gets his network off the ground, I would be sorely tempted to do that instead, if it would provide greater speed.

Also... I've run entire networks with 100s of staff off nothing but dual-DSL lines before now. And even one month where all we had was 4G after our DSL contract was terminated through idiocy on the suppliers part (and then we bought a leased line).

Re: “Shit, what I have done in my life? Now, my career is gone!”

I don't think that anybody even remotely concerned about an environmental disaster would be piloting a ship full of oil and then get distracted by the need for a phone signal, and then ignore charts to order several days of travel to come closer to coral reefs in order to talk to their girlfriend (or whoever).

I once trawled home from their shop on Tottenham Court Road on the Underground carrying several black bin bags full of AT and PS/2 keyboards that I'd snapped up for £10 the lot.

Re: Whose images?

Well, all it needs is another competitor selling images and giving 41p to the creators, and unless Getty are doing something illegal they should start raking it in, right?

It's the same argument about Kindle, etc. My father-in-law is a published author, dozens of works to his name, huge sales figures over his lifetime.

His new books, his agents can't sell and only give him a pittance when they do. He puts the same book on a private website, a dozen other services, he gets a pittance of sales and even though through his own site he gets nearly 100% of the cost, he has to discount heavily to get anyone actually buying them - including the books that were previous best-sellers in physical form and translated into a dozen languages.

He puts them on Kindle, himself, no other middle-men, and he makes money straight away. And though the percentage is nowhere near 100%, he makes money because consumers go to Kindle FIRST almost every time. That's not monopolistic unless Amazon actively abuse that. At one point his books were removed from Amazon, and he devoted his full-time career as an author to getting them up anywhere else, on every other service. He moved immediately back to Kindle the second that the dispute was resolved (which took months) because it had taken him from "making a nice amount on top of his pension" to "you might get a coffee a year if you're lucky" and far, far, far more effort (in terms of marketing, conversion, uploading, etc.) than it was ever worth elsewhere.

In the same way, people looking for stock images are going to Getty first despite there being a thousand other companies doing the same thing. So they can throw the creator only a gnawed bone, and the people getting their images are perfectly happy.

I don't work for any associated companies in this space (unless you count hosting my father-in-law's website for him), by the way. And I don't necessarily think it's fair or right. But your sport photographer friend will know - if he sells his images to ANY other company in that space, he won't make as much as Getty give him, in the long-run. Doesn't mean they aren't conning him, but also doesn't mean they're doing anything "wrong" if there are no serious competitors able to provide the same.

Re: Fundamental issues unaddressed

"Wifi is 20 times slower than a cable".

It's always been true, it's still true now and it will likely always be true.

However the bandwidth is now coming out of the "for casual use, it'll work okay" into "it's so damn fast it doesn't really matter" areas.

The big problem is legacy kit, taking up far more frequency and generating far more noise than required, but with WPA3 that will start to solve itself too and anything not WPA3 capable will be consigned to a bin.

I hard-cable servers, etc. obviously, and at home I'm cabled to my main laptop (mostly for gaming ping!), but I have probably 50+ wifi devices, and my neighbours have a few, and it all "just works", so long as you're not expecting perfection and zero latency and perfect response. And that's on an 802.11n (Wifi 4) network.

To be honest, we're now at the point where something like Wifi 6 or 7 "just works" for almost any application, but we'll all keep servers etc. on multi-gigabit guaranteed connections for a few generations more, I imagine. By that point, everything will be cloud and then all we need is Wifi and core networking on the back-end.

It's come along a lot, and it achieves marvellous things, but it's now at the point where brute-force and ignorance of the problems pretty much just makes everything work together. A bit like when we all started using sensible amounts of CPU and RAM and nobody had to hand-optimise stuff any more.

Not saying that's great from an engineering point of view, but for domestic and small business, nobody is going to care about the difference and they'll just Wifi 7 everything and it'll work - maybe with a QoS tweak or two, but it'll work. The use cases of cabled hardware for domestic / small business are getting less and less every day. Even access control, CCTV, etc. aren't going to care about a short blip any more.

Re: How long?

As someone who has multi-dozen-terabyte storage in the cloud for hundreds of users.... I can tell you that a complete backup of every single Office 365 - including all Sharepoints, Teams, Exchange, Onedrive etc. - for every user doesn't take anywhere near 10.8 hours to complete, for a cloud-to-cloud backup to another provider.

And that's as one well-known user, I bet that if you spread it over lots of people downloading different portions of it from computers across the world, Azure would cope just fine.

(P.S. I'm absolutely no Azure nor cloud fan at all).

Re: Were there no signs indicating that the Bridge was out?

Google might drive every road, but they won't know if a road is officially "closed" or not - that's taken from mapping data supplied by local authorities, the same as things like speed limits, etc.

Google can't just guess at those, because getting it wrong would result in things like this being their fault. But if the local authority never "officially" closed the road, that's on the authority.

Re: Similarity to "BMW's pay-as-you-toast subscription failure" article in The Register?

Here, I would agree.

Sadly, it turns out that the ones who do this the most, are the ones who deploy the strongest protections and who get the shirtiest if you bypass it.

It's 2023 and I still have a machine that has to have a USB key plugged in in order for some software to work.

Re: Confused!

No, the concept of Diffie Hellman is to form a secure channel by which you both agree upon a temporary secret - which you then use to share your public keys with each other to initiate an channel elsewhere under ordinary public key encryption - and then discard everything DH because it's no longer needed.

It's literally called a key exchange protocol.

Now, there may be some confusion of terminology and scope here, but ultimately any public key being able to be used to determine the private key is death to all public-key encryption, whether you used DH to transmit the keys or not.

What you're hinting at is that if you use DH in a PFS scheme, you send temporal keys to each other constantly, that's where they are "never shared or even known to the users" and "never stored or published". Hell, you can even mis-use DH directly as an encryption scheme (there are several like that), but it's uncommon.

But the majority use of DH is to exchange keys and go home, and then to leave you to do then something else, somewhere else. PFS uses DH in the way you describe, but not all public-key-encryption bothers with PFS.

And the fact is that a quantum computer of sufficient size would break DH, PFS and most common public key cryptography quite easily, as none of them are quantum-computing-safe.

We do have quantum-safe equivalents of the above, but they are vastly different, and they are also being weakened all the time (and many are not even in active use yet!).

(To simplify the QC-safe thing: At the moment we "scramble" two numbers mathematically to make it almost impossible to determine what the original two numbers were - but there's only really one correct answer, it just takes countless billion-billion-billion attempts to find it by brute-force. With QC, we basically do the opposite - we try to make as many answers as possible appear valid, so that each one is "returned" by a quantum computer, but you still can't work out which one was actually the original message without knowing what the original message was (kind of like a known-plain-text attack). So even though you know all the possible starting numbers, you end up with a billion-billion-billion equally likely and viable number-pairs, and no clue which one is the right one until, basically, you already know what the message contained in the first place to compare it with)

Re: 3G

Have you seen licence-free channels and the mess people make of them?

It'd be swamped to death in seconds and largely unusable, plus you'd have to severely limit the power allowed (especially given the range current given to 3G, you'd want to dial that down if you don't want the guy in the next town interfering with your signal).

Re: What is unclear with these "shops" in cars

Tesla hit that enormously with their "self-driving" (NOT SELF-DRIVING) modes and features.

People, for the most part, just sucked it up.

And Tesla then sued people who tried to re-enable it on cars that had been purchased with it, and then sold onwards.

They'll keep trying to pull this nonsense, until they've found the boundary at which people will or won't tolerate it.