* Posts by Lee D

4259 publicly visible posts • joined 14 Feb 2013

Ex-barrister reckons he has a privacy-preserving solution to Britain's smut ban plans

Lee D Silver badge

Re: Mind of a teenager

If you let a 14 year old have administrative / software installation rights on your home PC (or even personal laptop), it's already game over.

How do you think schools cope? They stop kids being able to install or run stuff. It's all there, built-into the OS.

So now you have to ask Dad for his password to install the DRM-disrepecting browser, or you have to circumvent Windows security. Neither are impossible, but neither are easy either.

Honestly, you're far more at risk of your kid accidentally installing a spyware webcam app that puts them at serious risk, than them downloading a bit of Page 3. Why you'd think that access enough to do the first is somehow necessary for a kid, I can't fathom.

Sure, "most people" won't. And that's on them.

Apple to keep Intel at Arm's length: macOS shifts from x86 to homegrown common CPU arch, will run iOS apps

Lee D Silver badge

Re: RIP Hackintosh

Got nothing to do with people discovering that MacOS is just shiny interface over poor hardware, then?

I ran a MacOS VM inside VMWare for years - allocating it all the resources of the Mac hardware that people were running it on, it looked very slick. But it's all looks. Sure the slidey bottom back was smooth as silk (pre-generated cached bitmaps in a range of sizes to make it look like it was shrinking/growing the icons in real-time distortions). But under the hood it was pathetic in performance.

In reality, the hypervisor running that VM had something like 3-4 times more resources, and laughed at running MacOS in a VMWare box that out-specced anything Apple were selling at the time. And it was running on a laptop. A second-hand laptop. A nice one, no doubt, but not some £3000 monster. And it could virtualise MacOS with equivalent specs while I encoded video and played games in the background. It laughed at what MacOS required and the MacOS VM still worked faster than a real Mac.

MacOS was all show, and it wasn't that long ago that I did it.

I'm not sure that there is a serious hardware person out there who thinks Mac hardware is well-specced, certainly given the price. Going to Intel showed just how far behind they were.

Lee D Silver badge

An Intel Windows image running in a ARM OS virtualisation hypervisor is *not* going to be virtualisation. It's emulation, effectively.

You can only virtualise when the underlying architectures are the same, and even with any amount of clever tricks, it's going to be dog-slow in comparison.

Lee D Silver badge

Re: Cock of the walk one week, feather duster the next

Serious question:

What does he do with it?

Because that's more money than I've ever spent in any single transaction, even financed, unless you count a house purchase (and even that, the deposit was cheaper).

If I spent £30k on a PC, it would literally blow anything Apple out of the water so far they'd be in orbit.

It barely cost £6k for a 768Gb RAM machine from Dell, with serious server processors. I can't justify that amount of RAM in my professional life running entire networks, what the hell justifies that amount of RAM in a single machine for a single user?

I know you'll say video-editing or something, but there are far cheaper ways to get a machine capable of that kind of spec than a Mac.

Lee D Silver badge

Re: Probably not a big deal, if you already have reasonably solid code.

Your 2015 Macbook Pro is more than I've paid for every machine I've ever used professionally in the last ten years, even for a second-hand purchase of a 5-year-old unit.

I never said I wasn't an amateur. I'm quite literally a homebrew / hobbyist coder who has dozens of active projects, on dozens of platforms. And not one of them includes Apple because of their policies. I've been doing that for 20 years, and no other platform have I flat-out refused to code on because of the restrictions.

I work on Intel/Linux, and I want to cross-compile from one machine to all targets, and test on-platform when it comes to things not working on a particular platform. You pretty much cannot do that for only one platform. There's a reason coders were buying Macbooks - because that's the one platform from which you could compile for Apple and bootcamp/cross-compile to literally EVERYTHING else. And now Macs will be ARM, and so Bootcamp will be a stunned sloth.

Sorry, but the idea of having to have a particular computer just because it's the only way to compile software for that target is completely alien to everything I've done in the last 20 years.

I'm not "afraid" to pay it. I refuse to. Because not one other mainstream mobile/desktop platform in the world is asking me to do that, or even hinting that it'll make things easier if I did. I code, compile, test, submit, and I'm done. No money changes hands. And at no point is my development environment or working desktop setup determined by anything except "what I already have / like".

Lee D Silver badge

Re: Probably not a big deal, if you already have reasonably solid code.

Same.

The coding won't really change much, except the compiler or cross-compiler in use.

I spent a lot of time cross-compiling to ARM for a number of handheld consoles (weird Korean things like the GP2X which had two ARM processors and ran off AA batteries). The cross-compiling was the easiest part of porting anything. It "just worked".

Contrast with when I tried to compile a piece of software that also worked on Windows, Linux, x86 and ARM, 32 and 64-bit to any Apple platform - which basically is almost impossible without owning an expensive up-to-date Mac, paying for a developer's licence, and using XCode on that machine as the way to compile it. It literally stopped me even trying to support Apple devices. Bear in mind that that same code ended up on Wii homebrew, PSP and all sorts of other platforms with little more than a recompile and a tweak.

I'm not worried about the ARM side. That's probably the most sensible decision Apple has ever made. But the development side, and the ability for Apple to create their own, custom, bespoke chip and throw whatever they want into it... that's going to be pretty scary. They'll make their things incompatible with everything else, and you'll have to use XCode to compile for them, and only the latest XCode, and that will only be updated on the top generation of Macs, and no cross-platform compiler will produce working code. That's what I see happening.

Literally the advice the last time I looked was to use cross-platform development / debugging tools like Eclipse on every platform "except on Mac, where you just have to use XCode, either directly or as the compiler in the background of the Eclipse IDE".

Apple's monoculture is going to hurt developers a lot. And, now, they can't even have a Mac with Bootcamp so they can have all their platforms on one device - Intel->ARM translation is NEVER going to be fast enough to be comfortable.

Lee D Silver badge

So this means that Macs will be pretty much useless for running Windows via Bootcamp, one presumes?

Email innovator Hey extends an olive branch in standoff with Apple, tweaks code to make the iGiant appier

Lee D Silver badge

Re: A subscription fee for email‽

I used to pay for Hotmail, back in the day.

People run their entire businesses on GMail and pay for business inboxes.

Exchange costs a fortune when you take into account the CALs, and more if you use it via the Cloud.

Hell, I pay for a domain name, a dedicated server, etc. for my own personal domains - just so I can collect or refuse email in the way I want (I own the entire domain, so I can just make up anything@mydomain.com and it will work... and then when some company spams the email I gave them, I know who they were, and I can just shut that alias down). Not to mention that, then, nobody knows where that email actually goes (it ends up in a popular webmail), but I can literally change the destination in seconds, or even copy every email to multiple places for redundancy. Oh, and I give out forwarded address to friends and family.

I haven't changed my domain in nearly 20 years, and have always had paid-for services behind that (the domain itself, the servers, etc.), and I have a complete archive of every email I've ever sent and received despite my "provider" / destination mailbox changing dozens of times over the year.

Personal users aren't the target - and aren't profitable at all anyway. Hotmail/GMail don't make any money out of personal users, except as a branding exercise and "consumer data". As you suggest, they can go anywhere and get a free email address. But private users serious about their email and the millions of business users... that's their market.

Can't get your Pi fix online? The Cambridge shop's back open for business, Brits

Lee D Silver badge

Re: An idea for allowing hands on action...

Welcome to the rest of the world where people are wearing masks and gloves in every shop they go into, because they have half a brain.

Lee D Silver badge

Re: An idea for allowing hands on action...

Could you tell me how it's not related to healthcare in the current climate?

Faxing hell: The cops say they would very much like us to stop calling them all the time

Lee D Silver badge

Re: My first modem..

I'm still running an active Hylafax connected to a USB fax modem (hard to find non-winmodems, but boot sales are very useful!).

Granted, we've not had a fax in years, but the system is still there and works. Users just "print" to Winprint Hylafax Client printer (freeware), it gets sent over the network to the Hylafax server, gets queued and sent out as appropriate, and then the user gets an email confirmation with PDF attachment of what they sent.

Incoming faxes do the same in reverse, popping a PDF attachment to an email to a given email address. For the last 2 years, it was nothing but spam-faxes, so nobody even bothers to check any more.

But it's a fun afternoon project, but definitely a dying technology.

Lee D Silver badge

I think the phone number was only a fax system, so when you called it back, all you got was a fax tone and nothing else.

Lee D Silver badge

Once had a fax try repeatedly to connect to my parent's house number. Drove us all mad (I was living at home at the time).

After a day or so of constant calls and no way to trace them, and before I could cobble up a fax modem to do what it was supposed to, we called BT and had them intercept the line. Weirdly, they claimed the only way to trace it was to take over our entire phone line for the whole day and we'd receive no calls whatsoever... kind of a self-imposed DoS.

Anyway, they intercepted the line. And then gave a phone call to a nice chap working in a central bank who was faxing incredibly sensitive documents to what he thought was another central bank, but was in fact our home number. The BT guy took a certain amount of joy in explaining it all to us, not least because said bank needed to then contact us to ask us to please, please, please never reveal what had happened and to destroy anything that we'd received (I was so annoyed that I couldn't get the fax modem working - think of the extortio.... I mean reward!).

We eventually got our phone line back, a grovelling apology, and never did get another fax call.

Good luck using generative adversarial networks in real life – they're difficult to train and finicky to fix

Lee D Silver badge

This basically describes every AI I've ever heard of.

As far as I can tell, any kind of machine learning:

- Starts off useless.

- After computer-years of training gets average-to-good results for basic categorisations or tasks.

- Plateaus just after it becomes useful.

- Cannot be "untrained", and to retrain to include extra parameters, data, etc. requires far, far, far more effort and time than just throwing it out, starting clean, and retraining from scratch (it's like a stubborn old middle-ager who won't learn the new computer system, so you have to replace him and start all over again with new staff). The reason for this is clear - it's entire life, it's been told "this is right" and as soon as you introduce another subtlety or data that's borderline, it has to be trained enough to overwhelm ALL of its lifetime training that it had been specifically selected for. It's like the old guy who learned Novell and now needs to learn something new, except that he literally cannot leave his Novell training behind and his entire career is based solely on how good he was at Novell and not anything else.

- Gets written up / deployed / sold off while it's still useful but before anyone can actually do anything about improving it, so the new owner / the person lumbered with taking it on basically has a read-only system that will never improve.

- Distances itself from all claims of being mere statistics, despite being basically a entirely statistical model.

- Cannot be modified knowingly. The same way you can't just stick a knife into someone's brain to extract a memory, especially without knowing exactly where that particular memory is. You can't correct behaviour, you can only try to train it out (overwhelm it), and you have no idea where that behaviour arose from, what metric it's actually operating on, or why it's doing what it's doing. For all you know, it identified that banana in the image because of the Getty Images copyright that other photos didn't have, or something equally ridiculous (e.g. the average background colour of the central left-third of the image).

- Are far too complex and random to analyse post-training.

- Are not even necessarily reproducible between two identical trainings.#

AI is bunk. It's "sufficiently advanced magic" to fool the casual onlooker (idiot in a hurry) but I have real trouble positing it as something to be trusted in any manner. So it helped you upscale some old movie to 4K. Whoopie. No harm done, time saved. But could probably have been done with just a few filters anyway. But for anything serious... get out of here. It's throwing dice into a canyon until you've thrown enough to make a vague outline of Jesus's head and then selling it as an art-creating computer.

Health Sec Hancock says UK will use Apple-Google API for virus contact-tracing app after all (even though Apple were right rotters)

Lee D Silver badge

It's based on the proximity of your phone anyway.

Leave your phone on the desk and someone walks past when you're not there and you're now linked.

Drive past a cyclist. Linked.

Stand on opposite sides of a wall. Linked.

And that's if they have BT turned on and are using the app.

Lee D Silver badge

It's a lovely expensive waste of time, that's for sure.

I don't have Bluetooth on unless I'm using it, so it won't detect me at all.

If you walk past someone's phone, presumably that counts as "contact", rather than any actual physical contact.

The people who aren't running it but have BT on are just going to be BT MAC addresses, you can't do anything about tracing them really.

Then, if it works, you can tell a lot of people - those who are voluntarily walking around with BT on, the app enabled, etc. - to go into lockdown, and they can just ignore you anyway.

I honestly don't see what we're trying to achieve here. One guy walks through town with the app (showing he's quite diligent anyway, presumably), days later discovers he had the virus, then we have to inform hundreds of people (also likely the ones diligently distancing) that they might have it, and they can't go to work.

Just... continue lockdown.

Customers of Brit ISP Virgin Media have downloaded an extra 325GB since March, though we can't think why

Lee D Silver badge

Not really...

All those Meet/Zoom/Skype etc. conferences would account for that much on its own. And they loosened a lot of restrictions because of that.

HTC breaks with tradition to push out 2 phones someone might actually want to buy

Lee D Silver badge

They're all too late.

I held on as long as I could for a decent phone that had everything I want (which would include 5G purely because I use it as my main Internet connection), but I gave up.

Earlier this year, I bought a Samsung XCover Pro. Removable battery. Headphone port. MicroSD expansion. Waterproof. Latest Android. Dual-SIM. A camera (I don't care beyond that, but apparently it's got a few fairly-snazzy ones). Not stupendously large. A single button shortcut for putting the torch on. All the usual things you'd expect. Fast charge, USB-C, including host (I've already used it to run an inspection camera on a long fibre optic, plus an RTL-SDR to pick up air-traffic and radio, not to mention being able to just plug in a keyboard and mouse in a pinch). And the GPS is amazing - GPS, Galileo, GLONASS and Beidou. It fixes in seconds and it's accurate to 2m even indoors! And 4G.

(I mean it lacks the IR blaster port my old phone had, and it has the stupid front-facing-camera-inside-a-notch junk, and the microSD and SIM are still underneath the damn battery, but it's all a really good compromise for my purposes).

They took too long and my old S5 Mini was dying, and I clung on for 5G as much as I could - and still there just aren't enough phones/packages/coverage to justify it. I'll have to just suffer the 80Mbps that my phone gives (far better than the 4Mbps I was offered as a landline! I told BT where to stick that landline) on 4G.

I don't think I was unreasonable in wanting a removable battery. The headphone port is a nice bonus (and may well form the basis of an IR blaster if I can find the right hardware, but a cheap one off Amazon didn't work). And it outperformed whatever that recent iPhone thing was in almost everything (okay, it doesn't have a 4K display but... I want a small display anyway, resolution at that size barely matters!).

Was debating between the Pro and the Field Pro but couldn't find much of a difference and the thing feels pretty solid. I don't use cases and screen protectors and nonsense - I expect the manufacturer to provide sufficient protection!

But it would have been really nice if they'd put 5G in it... it would have neared perfection even if I couldn't actually use it yet. But, to be honest, based on the phones I'm seeing with 5G, I honestly think I'd rather carry a 5G wifi box (the 4G ones are tiny nowadays, I'm hoping the 5G will be the same, and they can charge from the host USB-C port in a pinch) and the XCover separately if I ever feel desperately left out.

Whose side you on, Nominet? Registry floods .co.uk owners with begging emails to renew unwanted .uk domains

Lee D Silver badge

Re: I used to pay...

I'm afraid your problem is really one of thinking that your customers are somehow going to go to the wrong domain when I very much doubt they even know what the right domain is anyway. People don't use domain names like that any more, they just hit a Google or other link and find you, then bookmark it.

There are literally so many other domains out there that I can't guess what most places would use at first attempt, even when I know their business or website inside-out. There are companies whose official name is .uk.com, personal users on .co.uk, and huge companies on .net and other TLDs. Hell, the whole .eu debacle should show you what a waste of time it is - companies went mad to snap them up and now if they were British they have to forfeit them anyway, and someone else can have them.

My old workplace did this for about 10 years, trying to buy up every related domain, and they soon realised that the "wrong" domains just sat unused for years (because it would have been incredibly simple to prove they were passing-off on the basis of the actual name, no matter the TLD), and people were literally registered more deliberately and then telling that place, in the expectation that we'd buy their domain too.

It got stupid, and a stop was put to it, and now they have one domain and that's it. They get far more hassle from other random, unrelated companies doing things like putting their opening hours on generic "opening hours dot com" websites, getting them completely wrong (in some cases just making them up), and then when customers google what the opening hours are for your company they somehow think that random-website-on-the-other-side-of-the-world knows better than the actual, main company website. And when you contact those pages to demand removal or correction, they want to charge you to do so.

Buy a domain. Stick with it. To my knowledge, I've never knowingly typed in a .uk (without a "co" or something before it) or a .hotel or a .travel or a .anything - I may have clicked on a link to them or been redirected to them, but I've never typed them in or tried to remember them.

And any website that's passing off as you? They could do it on a billion similarly-named domains, including the unicode tricks to look like normal characters that look almost identical. The problem there is fraud in claiming to be you, not what name they actually host under. They can find a thousand domains that are confusingly similar to yours in an instant.

Lee D Silver badge

Re: Is this Fraud?

It's what I would refer to as "an interesting Sunday afternoon spent sending a letter by recorded delivery".

Go on, Nominet. Try it. Because I will do what others don't have the time to do and report it as an unfair/misleading/fraudulent business practice.

It'll cost you far more than you'd ever make from me to respond to me, and I won't stop just because you reply saying sorry.

Literally, I'm in need of some entertainment at the moment. Do it. Send me the email. See what happens. Just *try* to imply that I'm somehow needed to pay you for that domain, that I have ownership of that domain, or that if I don't "renew" I'll lose something I had before.

This is no different to just sending someone a letter and telling them that if they don't pay £30 they'll subscription will expire, when literally there has never been any such subscription in existence. Just because I own *another* domain name does not mean you can talk to me about one I don't own as if I did.

And, maybe, if this is really your business practice now, you should seriously review why you can't make money from millions of people owning UK domain names and paying you every year, when all you have to do is run a bunch of NS for them. And why you think that tricking your customers into buying something they never wanted on the basis of confusion and fraud is a good way to do business.

DevOps to DevOops: Docker Hub proves so secure that 430 Docker images out of 2,500 have no vulnerabilities

Lee D Silver badge

Age old problem.

Because unless someone remembers to update every Docker image on a regular schedule, it's just going to be out of date.

But if they do, it's going to screw lots of people over if there are changes / bugs beyond the security update, people who would have been working fine without the update.

Docker is really just a bunch of "other people's VM's" in essence, anyway. It's no more secure than anything else, because of that.

You'd think there'd be some kind of automated dependency/security tool by now that realises that a dependency is out of date, updates it and rebuilds everything that was reliant on it (or contains an unannounced copy of it, which is far more likely!). But no.

Docker - like all similar containerisation technologies - just pushes software updates behind another layer of obscurity and complexity, it doesn't actually fix them.

An Internet of Trouble lies ahead as root certificates begin to expire en masse, warns security researcher

Lee D Silver badge

At one point, a certificate inside a Java .JAR file, associated with APC UPS software expired.

Except, we didn't know that. All we knew was that every server suddenly went to 100% CPU, couldn't log in, couldn't do safe mode, couldn't do anything in terms of diagnosis on the machine itself. And when you looked at the data, everything matched recent backups, including the software JAR file that hadn't changed in years. Even if you restored from a known-good backup, same problem as soon as the clock updated (but, again, you wouldn't know that, it would just work for a minute and then go mad). Literally took down a network without any visible signature and no way to diagnose a live system.

It was only by one of those word-of-mouth fixes that it was determined that if you kill off the APC UPS software by deleting the entire folder, and then restart, the machines would operate as normal.

Eventually APC pushed an update, but it literally took out several networks that I was working on, and many colleague's networks. Obviously one of them noticed that the ones not on an APC UPS or using its monitoring agents to shut themselves down were unaffected, and spread the word.

Then the root cause analysis was a certificate inside a JAR expiring, causing the APC software to throw a paddy and bring the servers to their knees (for whatever reason). APC issued an update with the new certificate and a reinstall of it worked, but first you had to take down every affected machine, manually delete the APC software, then boot and reinstall.

I expect much the same problem in dozens, if not hundreds, of pieces of software when old root certificates start to expire, and they'll be the embedded ones as well as the server.

Play stupid games, win stupid prizes: UK man gets 3 years for torching 4G phone mast over 5G fears

Lee D Silver badge

Re: nutters

Used to work in a primary school.

Installed a wifi access point in reception so that parents could jump on to check the school website or whatever.

Got yelled at by a parent about "Did I know that I'm frying young children's brains?!?!?!?".

Then watched her get in and put her kid in her 4x4, not do up her own seatbelt, pin a smartphone to her ear and drive off one-handed past all the other kids while jabbering away on it.

Nervous, Adobe? It took 16 years, but open-source vector graphics editor Inkscape now works properly on macOS

Lee D Silver badge

I keep trying to use Inkscape but my needs are limited and yet it's rarely capable of performing.

One of the things I want to do is import and break open PDF format maps (not complex, simplified maps that you'd find in a brochure or advert), delete unwanted elements, save the result as a SVG or similar. Or grab a scalable logo out of a PDF.

It's a job that I used to use Serif for. I'd never touch Adobe, far too expensive. But the affinity software just whinges after install and never runs, it's been like that since their beta period for me, and they have no solution (three different machines, three different versions of Windows).

It's perfectly capable but it's a "go away and have lunch" operation - it imports, you wait. It draws, you select (and it comes in as a huge group). You ungroup, go have a cup of tea. You return, select again, ungroup again, another cup of tea. Then it runs like a stunned sloth while you try to edit, delete and save.

I'm going to literally download it now and try:

- Windows, 64-bit, .EXE.

- Install (DON'T ADD TO MY SYSTEM PATH, what is this, 1980?)

- Run.

- Ungroup

It's slightly faster but it's still clunky.

Nine million logs of Brits' road journeys spill onto the internet from password-less number-plate camera dashboard

Lee D Silver badge

Not true under the old DPA, let alone GDPR.

Many, many, many organisations have been sued where no actual access of data could ever be proven, but where only the potential for such existed (e.g. posted a disc of data and losing it in the post).

Lee D Silver badge

DPA/GDPR lawsuits don't care about the actual consequences, they care about the potential consequences.

Hospitals were sued for millions successfully because they posted a disc which was lost in the post and couldn't PROVE that it had been encrypted and/or that the data was not available to people it shouldn't be.

Schools too. People are very lax on it, and yet the law is quite clear - the leak of actual data doesn't matter, it's the potential. Potentially even allowing you *theoretical* access (i.e. you could have got in without a password, but nobody ever did) is prosecutable under the same laws. This is one of my big-wig reasons for not giving local administrator access to any user. They can complain all they like, the letter of the law says I can't give it to them if it might reveal any data that they don't explicitly need to perform their job.

I imagine government departments get the usual light slap, but I know if my reg was on there, I'd be filing a claim via the Information Commissioner's Office for a GDPR violation of personal data (and, no, it doesn't have to have my name on there - it just needs to be data that can be linked to a particular person or persons).

Happy birthday, ARM1. It is 35 years since Britain's Acorn RISC Machine chip sipped power for the first time

Lee D Silver badge

Re: Cheers...

Spotted the engineer.

Internet root keymasters must think they're cursed: First, a dodgy safe. Now, coronavirus upends IANA ceremony

Lee D Silver badge

So they're not going to know if the keys are in the parcels and work and that they have them until the day of the ceremony?

Anyone else spot a problem here?

And surely, if this stuff was ANYWHERE NEAR secure, those parties could all have an HSM of their own with which they could verifiably sign a key with another that only they could possibly be in possession of (the HSM and it's associated authentication) and then those keys - if they are in any way secure - can just be transmitted over the Internet (I would add the caveat of "avoiding DNS use" but that much should be obvious).

Safe and locks and stupid procedures opening envelopes in front of webcams is just ludicrous, I'm afraid. Unless someone can compromise 12 - or however many - independent people worldwide simultaneously, grab their HSM, torture them all for their signing info and private keys and passcodes, and sign off something fake without ANYONE noticing... even if they have to do that part of it one-by-one on a web video link...

They've had one near miss. They've set themselves up for another here. It's not going to be long before they totally screw it up because of some other instance they hadn't considered and it'll be game over for DNSSEC.

Education tech supplier RM smacked by UK schools closure

Lee D Silver badge

Re: If you cant sell virtual learning environments

Let me enlighten you:

A VLE is an intranet with a content management system so the school can put files on there, assign them to kids, the kids can login from home, view the files, submit answers or new documents, etc.

That's not the kind of thing you slap in at the last moment if you don't already have one (instead, you jump on something like Google Classroom, which is free to all schools), it's not the kind of thing you just run setup.exe and it works (server provision, port forwarding, security, LDAP integration, etc.), and it's not the kind of thing you want to sign up for a year for just because you didn't have one and think you need one to cover corona.

Lee D Silver badge

Re: Garbage

They stopped making gear years ago.

Their software is mainly just software-as-a-service type nowadays, all web-based.

As a mathematician, computer scientist and school IT manager let me just say:

IT: Good riddance, generally speaking.

Mathematician: RM Maths was actually educationally very good.

IT: RM Maths sucks (install, maintain, users, expense, etc. etc. etc.)

BT's Wi-Fi Disc ads banned because there's no evidence the things work

Lee D Silver badge

The ad watchdog -- the people in charge of verifying advertisements

Puts the telco -- the telecommunications company

On the naughty step -- in trouble

Over (their) --- because of

range-extender promotion. -- their promotion of a range-extender.

I mean, it's really not that hard. And telco isn't really even a British word, we don't use that much, it's come from the US. The rest is just literal English words used to form slightly silly headline.

SpaceX beats an engine failure to loft another 60 Starlink satellites

Lee D Silver badge

Re: "Shows value of having 9 engines"

Unfortunately, SpaceX are not the only entity at risk if they have failures in launch, insertion or orbit (or de-orbit).

Lee D Silver badge

Re: "Shows value of having 9 engines"

"four more recoveries than anyone else has..." risked, because they knew it wouldn't be viable to do too many and there were too many variables and too much money at stake to just let the things blow up because someone didn't service them properly between.

Capita hops on UK's years-late, billions-over-budget Emergency Services Network to keep legacy system alive

Lee D Silver badge

Re: EE Data

I think it's stupid in this day and age to deal with only one vendor.

They're the government. They could literally pass a legislation: As a condition of your 4G/5G operating licence you MUST provide... whatever... on all cell towers, no matter the network.

The extra cost would be reflected in a lower price at the spectrum auction, but I bet it's not £3bn lower.

Then nobody gets a network that's "picked on" in an emergency, the emergency workers get guaranteed access wherever it's theoretically available, and the telecoms operators know exactly what's happening and there are no favourites taking backhanders.

Of course... the latter reason is exactly why things aren't operated in that fashion.

Corporate VPN huffing and puffing while everyone works from home over COVID-19? You're not alone, admins

Lee D Silver badge

Re: Split-tunnelling? Security madness, surely?

What if you're a school or a secure area who, say, requires their web filter to log and filter all inappropriate accesses?

Lee D Silver badge

Re: Not just VPN accounts

Isn't that what Terminal Services was made for?

Then they can literally run their stuff on anything, and the internal corporate network is still secure from whatever junk they've bought/used/borrowed.

How does Monzo keep 1,600 microservices spinning? Go, clean code, and a strong team

Lee D Silver badge

Re: Wow

You'd rather have one big bubble that any little spike inside it can pop the entire thing?

Lee D Silver badge

Re: Banking isn't really a highly computational process

"To put this into context, every fixed line call in Germany has to go through a complete lookup of the portability database. That's a database listing every number that has ever been ported. That's millions of datasets. The lookup works with a simple barely optimized program which rarely takes more than a millisecond to look up a dataset, even on a very modest computer."

I should damn well hope so. Sorry, but it's 2020, and you're doing a lookup from a list of, say, millions of numbers to retrieve a small set of data associated with it?

There's no way it's searching one-by-one... it's hashing prefixes and following trees. If it touches 13, 14 entries for comparision, I'll be amazed. And at 3GHz, even, that's literally taking fractions of a millisecond, even if it takes hundreds of thousands of instructions on an in-memory lookup (a million data rows is NOTHING to keep in main memory).

All the IT ladies (all the IT ladies), all the IT ladies (all the IT ladies), now put your hands up! Oh, still not many here

Lee D Silver badge

When I went to university 20 years ago, there was precisely one woman in the lectures for the CS department, so less than 1%. The maths side, which I was also studying, had something approaching a 40% ratio.

After 20 years in IT working for schools, I have worked under precisely one woman (in a technical sense, not in a "they were the headmistress of the whole company"), who was an outlier and had been in banking IT for decades before and was nearing retirement. All the applicants for her replacement were male. Every time I've put out job ads (and HR are scrupulous about being equal-opportunity), from apprentices up, every single applicant was male. Every IT department I've visited has entirely male staff. We have employed women briefly, but in the "anyone can carry a computer" tier of jobs... not through want of trying, but we just don't get the applications from female applicants. Schools I've worked in have been majority-female staff in general. It just doesn't feed down to IT.

I refuse to let my department become toxic masculinity personified, so we are often the haven of staff, male and female alike, when they want some sanity. But I don't get any female applicants responding to widely-published, heavily-advertised, neutral descriptions of a job that involves nothing gender-specific, even in a female-heavy workplace. Yet we have female staff in finance doing high-level Excel and SQL, and I've worked with female teachers more than capable of teaching coding (some of them ex-COBOL programmers), and there's literally nothing in the job that's female-offputting.

There is obviously some disconnect somewhere - at some level women are discouraged from a career in IT or CS. And short of saying "female applicants welcome" or something (which is going to be construed as sexism or "looking for totty only"), as someone who hires IT staff, I can't do much more.

Helpdesk roles are perfectly well gender-neutral... anyone can man a helpdesk. Anyone can follow a procedural sheet. You don't even need to be in an office, helpdesks often operate remotely. I do speak to a number of helpdesks throughout the working day, but generally it's male. I hate to say it, but the female staff tend to be in the minority, not stretch past first-level support at all, or are literally "secretarial" staff who were answering the phone and just recording details for the IT people (almost exclusively male) to follow up on later, just so that the phone isn't ringing for too long.

I've met more female IT trainers than any other sub-profession. The techy staff, and especially the most-techy staff, tend to be male. And I say that not from a position of ignorance - I worked under a female IT Manager who taught me more than anyone else I've ever worked under, and I've worked with female teachers who were ex-COBOL programmers and who could happily geek out with me for ages. But the fact is that they are really in the minority. Whether that's manning the phones at a print service company, or programming up apps, or building servers in datacentres. Technically, I know more transgender women in the profession than I know women (I'm sure someone will complain about my wording there somehow implying that they aren't women, etc., and I'm sure my transgender friends will tell them to shut up because "he's okay" and they know what I meant).

How we fix it, I don't know. Going into schools and telling them that IT is a career for women just reminds me of Sheldon from Big Bang on his school visit where he tried to encourage the class of girls to get into science. I can't see it having much of an impact.

Researchers trick Tesla into massively breaking the speed limit by sticking a 2-inch piece of electrical tape on a sign

Lee D Silver badge

Re: Sigh.

You can literally fail your driving test for "failing to make adequate progress".

Guess how I know.

30mph on a stretch of road with multiple *hairpin bends* that were national speed limit... I protested most strongly.

Don't use natwest.co.uk for online banking, Natwest bank tells baffled customer

Lee D Silver badge

I gave up on Natwest many years ago when they told me that I couldn't use any other browser as I had to use Internet Explorer (4?) as it was "more secure". When in reality their online banking consisted of an ActiveX plugin putting a fake padlock icon into a frame that was really just an insecure site.

I mean, it wasn't quite the dark ages of the Internet, but even they should have known that that was a really bad way of doing things, and I knew enough to complain.

I moved my accounts as soon as I realised they were serious and wouldn't be changing any time soon.

I can't imagine their IT has come on any better since then, to be honest.

I've slowly worked my way through all the major highstreet banks for similar things - everything from literally laughing in my face when I applied to a mortgage (so I went to the place next door and got one basically the same day, for exactly what I was asking to), deliberately holding onto cheques for the maximum clearance period despite 10 years of paying them in (because on that ONE occasion delaying it would take me overdrawn for a fraction of a second before the next payment cleared) and don't even get me started on the 2FA device that I "had to" change to a smartphone app, but couldn't without first receiving... a 2FA device in the post that I literally used once to put the code into the app and then threw away.

I've ended up on Monzo, but I'm sure that won't be the last move. At least they do seem to have some semblance of understanding of a secure interface, however.

Ofcom: Rule change to force UK comms providers to tell you when your contract expires

Lee D Silver badge

I'd like to point out:

I don't *WANT* the very cheapest contract every damn time. The whole concept of cheapness being good is something I can't fathom. USwitch will recommend me a bunch of companies that I've blacklisted because their service is just so shite (e.g. TalkTalk Broadband).

The constant race to the bottom, with government-backed initiatives, is something that I don't understand the utility of. Do we only ever buy the cheapest car? Cheapest flat? Cheapest carpet? Cheapest fridge? How many people *ONLY* ever want the cheapest thing?

What they are supposed to be combating is long-term customers getting a worse deal than new customers, really. You don't do that by encouraging companies to constantly push the fact that they're not the cheapest in the customer's face.

I don't *want* to switch utility companies every month, or have you waste advertising and other money on services for me to do so. I want the damn company I've chosen to not raise prices artificially or "after the contract ends" any more than it would cost for someone to get a new contract.

"You could get a better deal by switching to an awful company that's a penny cheaper" isn't good economic sense for anyone.

Let the market operate itself, but put in legislation to enforce that *newer* customers should get no better a deal than is already advertised and available to all *existing* customers of those companies, and which either can switch to at any time.

The one service where I really don't give a damn how cheap and rubbish it is so long as I have it is car insurance. If I've got a properly-underwritten certificate that makes me road-legal, that's it, that's all I care about. And for some reason, every single year, it's cheaper to switch to arbitrary random companies ranging from the co-op to Halifax to the RAC - who all use the same underwriters, all have the same details about me, all utilise the same centralised self-service portals from the same company, and yet my insurance price will double once I've been with them for a year for no reason, and everyone else will be even cheaper than I got last year. It makes no sense, even if you assume they're gouging for customers who then are too lazy to ever move again. Why would the RAC, for example, put their name to something that next year they know the majority of their customers will flee and likely never return because they forced their prices up artificially?

Regulate the damn market, not force companies to try to make you move to the cheapest deal ever year, spamming you each time. If I want to move, I'll move. If I'm unhappy with the service or price rises, I'll move. If the price stays the same but doesn't follow the lowest on the market on a stupid race to the bottom... then unless I particularly care about that, I really don't see why they or I should care, when a 5 -minute check will tell me if I'm being conned. Or I might just decide "Hey, their service is really good, I'll stick with them".

What do we want? A proper review of IR35! When do we want it? Last year! Bunch of IT contractors protest outside UK Parliament

Lee D Silver badge

Re: Offshoring IR35

Because that doesn't sound like an arrangement that the tax office would like to investigate and / or stop being possible at all...

B-but it doesn't get viruses! Not so, Apple fanbois: Mac malware is growing faster than nasties going for Windows

Lee D Silver badge

T'was always a rubbish statement. Nothing is virus-free for as long as there is ever a single, solitary security hole in any component of it.

I hear the same now about Chromebooks and that's just as laughable.

Microsoft ups the ante with fix-fixing patch that leaves some Windows Server 2008 machines unable to boot

Lee D Silver badge

Reason #12497438 to not have Windows Automatic Updates turned on to automatically apply on whatever schedule they like.

Oh... but with Windows 10 you don't get a choice... because... feck-you-why?

Maker of Linux patch batch grsecurity can't duck $260,000 legal bills, says Cali appeals court in anti-SLAPP case

Lee D Silver badge

Re: No actual damage

You know that the case has NOTHING to do with the GPL, right?

One guy said "I'm gonna do this".

Another (world expert) said "I don't think you can do that".

The first guy then threatened to sue.

The expert then said "Please don't. I will fight it, it won't go your way. It's an opinion. I'm happy to just drop this now."

The first guy still sued.

Expert won the case. Filed counter-claim for, basically, being a frivolous lawsuit.

Guy appealed.

Expert won the appeal.

Guy now on the line for $250k for his own stupidity.

That the opinion was about the GPL is literally nothing to do with the case - the courts have literally said that the merits of that argument are nothing to do with the case at all, it's whether an expert was expressing his opinion or not. And they ruled it was just an opinion.

However, the GPL is quite literally the most popular open-source licence for a reason. If you want to benefit from code under the GPL, it says that have to give that benefit back to everyone else who uses that code. If you don't like it, don't contribute to GPL code. Hell, you can still use it, it doesn't change the way you use it - only the way you *distribute* it or the way you *contribute* it.

It's quite clear and obvious.

There's a reason that Linux is supported by the world's largest IT companies despite that "payback" clause, and not FreeBSD, etc. They don't like paying back any more than anyone else. But they use it for a reason. My IBM Bladecenter server officially supports Linux, on an equal par with Windows Server. It doesn't have *any* official support for any BSD whatsoever.

Don't like it? Don't use it. Or use it but don't distribute or contribute to it. Strangely, even with that restriction, THOUSANDS of times more people choose to contribute to Linux than to the "open" BSD.

Lee D Silver badge

Couldn't happen to a nicer guy. grsecurity is, basically, just one guy.

From conversations online, he comes across as the biggest twat since Joerg Schilling and his cdrtools "why can't ordinary people just specify every device by it's full SCSI path, no we'll never accept a patch to take a normal device name".

Last time I looked into it, he had to declare how large that organisation is for a Navy software contract... and it's basically one guy without even the money to pay these kinds of legal bills. I'd like to know where he's getting the money from.

And though it doesn't establish Bruce's assertions to be fact, it does prove that they aren't *categorically wrong*. They are just an untested opinion. And, as far as anyone I know is concerned, Bruce is right. You can't impose additional conditions on GPLv2 contracts. And he can't offer the code under *any* other contract as it's a straight derivative of the kernel code. What he tried to do was make a HUGE patch to the kernel to "secure" every single avenue, which is highly tied into the kernel code. A patch which he has thus far refused to break down properly and submit to the usual Linux kernel approval paths. He just expects everyone to take his mega-patch and put it in the kernel outright in one lump. But they won't. So people started breaking it down for him, and taking bits to put into Linux (which is perfectly viable - it has to be GPLv2). He took exception, threatened to cut people off from his code if they did that, including removing their access to it. Then prohibited people distributing his (GPLv2) code.

When someone called him on it and offered a legal opinion, he tried to sue them for defamation.

The guy's a moron who just wants everything his way and must always be right. $260k is a small price to pay for such action, when he could have just said "I disagree".

There. Sue me for that.

This AI is full of holes: Brit council fixes thousands of road cracks spotted by algorithm using sat snaps

Lee D Silver badge

So are you suggesting that satellite-images (of active roads with vehicles) pumped into AI is somehow better than a guy poodling down every road at 10-20mph, actively looking, or even walking the streets, segwaying, skateboarding, whatever?

How the hell are you going to see any more from a satellite than a guy actually driving down the street?

And guess what - if a driver looking for them going down a road looking for potholes, and able to stop and check any time he likes (hi vis optional), doesn't see them... either he's not doing his job, or they don't affect the roadway, or the satellite ain't gonna see them either (e.g. under a parking space).

Lee D Silver badge

Seriously, how much does it cost to employ a guy 40 hours a week to drive round in a van, taking in EVERY road over the course of... hell... a month, let alone a year... and press a GPS tracker button whenever he finds one.

If this is "money-saving", then it's still a ridiculous waste of money.

I'll do it for £50k + vehicle expenses + £30 for the tracker + £20 a month for always-on data connection.

I make that about a £350,000 saving... AT LEAST over the AI, let alone whatever bureaucracy ritual they were doing before.

Very little helps: Tesco flashes ancient Windows desktop on Scan-As-You-Shop device

Lee D Silver badge

Looks like one of these:

https://scanco.zendesk.com/hc/en-us/article_attachments/115026446087/Datawedge_Basic_Configuration_Guide.pdf

Looks like people have been playing with them for years (this mentions 2009):

http://www.barcodedatalink.com/pages/datawedge.php

Looks like we still haven't learned that obsolete general purpose operating systems "just running" an app you want them to, incidentally to all the other background stuff that's left on there, is going to be the way that cyber-apocalypse will compromise us. Not advanced hacking and breaking encryption, but finding Wordpad running on a scanner connected to a corporate network improperly.