Re: Mould breakers
Ah, yes, Pi Day.
Nobody really knows what the politicians will do, and so how Brexit will turn out. I am not sure that any tech company can launch anything in the UK before midsummer. Brexit adds too much risk.
2113 posts • joined 14 Sep 2007
It's hinted at at the end of the article, but we're missing the difference between population coverage and area coverage. Round here the are many 1km blocks with zero residents, and zero roads. You can check this with the OS coverage available through Bing Maps, a good example is the area north of Brigg. It likely still gets coverage, it has people working there, but nobody lives there. There are similarly empty grid squares on the Lincolnshire Wolds, which might not get good coverage because of the hills.
And internet coverage is not the same as voice coverage.
What a map such as this could be good for is suggesting areas with more people that are worth more checking. It depends what they call a main road for this survey, but the article suggests that they would only be the A-roads on a British map (and not all of them). I can see on my local maps that there are villages of over 1000 people, mostly in one grid square, which are 5 miles from any A-road. I don't know where the phone towers are, but five miles, line of sight. would be a possibility for one.
The population distribution is lumpy on a 1km grid. A map such as this one is only a first step. But it's people, not grid squares, that vote and use the internet.
It suddenly doesn't seem so crazy to suggest that HTML5 support can be included in this process so that we can have some confidence that an update to a web browser, or other program using video, has a consistent, mostly working, place to look. As it is, there are a whole bunch of different media codecs, all supposed to be doing the same thing, except when they don't.
I am wondering how anyone could have spotting a drone flying around an airport at about 3am in December. It could be flying using GPS waypoints, so the operator doesn't need it to be showing a light, but just how are they detecting it? Radar? They're not stealthed but they're not a big target, and a lot of the aviation business relies on transponders for routine use. There will be some radar which uses the echo, has to be in case something goes wrong, and that sort of radar has been spoofed since the 1940s.
It seems possible that there is no drone, and something else is being detected, but, even if the plane doesn't crash, sucking something into a jet engine is expensive.
Opera runs on the Chromium engine now. So their website says, anyway, but some sources seem to use different names for the same thing.
Opera has some nice features, but I am very careful about backing up before I do a version upgrade, there have been some odd glitches.
Chrome/Chromium is appearing in a lot of places. One program I use manages to open three copies of a library called "Dullahan", just when you start it. For sundry other reasons, I am a bit wary. Security is as much a state of mind as a set of tools, and the best locks in the world are useless if you don't lock them.
I can see what you're getting at. The certificate system has a different purpose for this situation. It isn't about somebody such as me, downloading software from a myriad of possible suppliers, possibly via intermediaries, where the certificate is about blocking access to possible malware, now with such things as HTTPS. Secure delivery still needs attention, but once a genuine copy of the software is delivered and authorised for use, the supplier's action (or inaction) shouldn't be able to stop it working.
Yeah, I suppose contracts can set up something like software rental, and that's nothing new. But if you shut down your customer I am sure the lawyers would be interested in the procedures you followed.
I am a little bit unsure about just what the first stage and second stage are on the Soyuz. The basic design goes all the way back to the first Sputnik launch, which five units all firing together at launch: the four boosters and the core. When the boosters have burnt all their fuel they detach, leaving the core still burning.
The pictures show one of the four boosters having a bad separation.
If one control module failed, out of four, it all makes a lot on sense, if what I am thinking of as booster separation is what is been called first stage to second stage.
OK, so I was able to check through the link you provided, which says "up to and including 239", but I had just installed a systemd update and when you said there was already a fix written, working it's way through the distro update systems, all I had to do was check my log.
Linux Mint makes it easy.
But why didn't you say something such as "reported to affect systemd versions up to and including 239" and then give the link to the CVE? That failure looks like rather careless journalism.
One of the reasons I like Mint is that they are willing to experiment with the UI, but they are able to support a cluster of different shells and keep them available in parallel. They might not all be released on the same day, and I doubt that's a good idea anyway, but it is one of the things that makes it more than an Ubuntu clone.
I've seen this. A pop-up asks for a response, you type in text, hit return, and for some crazy reason the program responds as if you clicked on a different button that is hidden behind the pop-up. It's the sort of thing that makes you wonder if programmers are human, or some monster which will be revealed in the next episode of Doctor Who.
The UI, on any operating system, is something people are reluctant to change. We still have, with slight differences elements that have been here since Windows on MS-DOS. Look at how minimise/maximise/close has and hasn't changed.
I have seen ideas for UI changes which might be improvements, but the struggle to overcome all those decades of habit made them more like failures.
Some things Kerbal Space Program does very well, but it has simplifications that build up errors. You learn the basics of changing orbit and rendezvous, the stuff that Buzz Aldrin wrote the book on, but I'd still rather have him at the controls.
Some of this may be things that used to work, which fail badly with new methods being applied since the Cold War. Consider things such as biometric data on modern passports, which is hard to fake. And we have fingerprint sensors on some of our phones. Some Russian Spies, in the old days, managed to use more than one identity, and not every one can have been identified. We're reaching a point where the document isn't use-once, it's the human.
One thing we know is that here in the UK, we were very good at catching enemy agents and persuading them to work for us. And that depended on being very careful about revealing what we knew. It's possibly why some things were not reported to the politicians. So, through a lot of hard work, we identified two Russian agents. We stood a good chance of being able to spot them crossing a border, whatever documents they used, and that could have given away something else.
What is going on?
I can see why the project exists. The existing system is horribly old. And transferring the old data to a new system is certainly an opportunity to deal with some of the retention problems. Though I have to wonder if there was ever the information in the database to identify the records that should be deleted.
But, really, does anyone expect either side in this argument to be saying anything different to what they are doing?
The difference is pretty small, and I am not sure that, in the last few years, GP Home Visits are so good an indicator. My brother is currently in hospital after the GP surgery sent an Ambulance, and that felt like one of a range of options they had, from "come to see us" upwards. And can there be a difference between the almost routine and the urgent cases? (I'm thinking of the elderly with limited mobility.)
Did your report over-simplify?
This whole story is riddled with misconceptions, and where it isn't, it;s all rather obvious anyway. It's essentially automating traceroute and ping and saying that when the RTT and packet loss jumps, the problem is between the last good node and the first bad one.
I was doing that over dial-up internet through Demon in the last century.
This isn't rocket science. And Kerbal Space Program feels more realistic than this article.
The actual trains aren't working all that well either. And the numerous websites telling you about delays don't seem to be working at all. I was watching arrivals at my local station, and the system doesn't seem to know whether a train is late until it leaves the previous station.
Which is odd because the signalling system has to know where the trains are, and has to know which train is which or a train will go down the wrong line.
Buses, you don't even get that sort of detail. Monday, the bus which eventually turned up had a sheet of paper with the service number taped up in the windscreen. And, for one dreadful moment in the middle of nowhere it seemed as though the gearbox had failed.
We still have a local bus company which isn't Stagecoach, and their bus was making odd noises too.
That is one of the critical distinctions.
Recent experience of GDPR-rated consents and settings suggests that internet companies are each allowing hundreds of advertising companies to see my data, and I see nothing to distinguish Google on this. Nobody seems to anonymise the data.
It's not like old-time advertising on TV, when viewing figures were obtained by recording a sample audience, and you had some idea of what sort of audience watched a particular programme, but nothing specific. It seemed to work. The commercial TV companies made good profits. And, if you're old enough, a phrase such as "Ridley Scott's Hovis ad" still conjures up an image.
The stuff bad enough to remember was for the local companies, the static card with the voice-over for one of the local department stores that vanished into BHS or House of Fraser. Or perhaps, in the cinema, the Pearl & Dean advert for the restaurant so good that the chef ate there himself. And we seem to be getting that level of advertising over the internet, without even getting as good a localisation as Pearl & Dean gave you. The restaurant where the chef ate was at least in the same town as the cinema.
Google doesn't seem able to manage that, at times they can't even get the right country.
Butlins has changed since the Hi-Di-Hi era, much smaller than it was and includes hotels on the sites. But just what happened? I'd distinguish between phishing and malware. 34,000 sets of booking details sounds way too big to be the result of a phishing attack pretending to be the local council. A fake email from a local council could be a vector for malware, but how plausible was the email? The scale looks like one site, so it hangs together, but I wonder how robust the system is.
Local councils could plausibly mail out regular information, such as event lists, which somebody might almost automatically open, but why would such stuff get close to the bookings database? Maybe something was sent to customers, but what?
All the switches? What about the display? It'd be a good exercise for your students to work out something like this. Some connectors would be on a Raspberry Pi board anyway, but there would be a lot of extra bits.
And the rule of thumb is still that you buy/make for £x and sell for £2x.
There are a lot of TV channels now showing adverts, and that is pushing the old model to its limits. Perhaps it's why we are never likely to see the quality of Ridley Scott's Hovis adverts ever again. But in those far-off halcyon days the only thing the advertiser knew about his targets was that they were watching a particular TV programme.
Nowadays, as GDPR has revealed, they can't place an advert without knowing your inside-leg measurement.
One has to wonder if the computer has made people smarter, or just biased success in life towards sociopathic semi-literate jackasses.
As for the clothing, Levi Strauss have their label on a pair of jeans, and I reckon that's OK. The logo-laden shirt of your favourite football club is tolerable. Nike have that swirl element in their footwear design, and if that's "in your face" you have other problems and are likely to wake up in hospital. Though the French Connection UK branding seems like a funny-once joke.
But I doubt I would buy these expensive brands anyway. So there's no point in saying I'll boycott them.
I suspect the key point in this is not the detail of the cables affected, but the timing. "You thought you there was no rush..."
And sometimes, "redundant" routing isn't. I remember, back in the Nineties, two US cable links, different operators, apparently totally independent, crossed a river on the same bridge, which fell down.
It's not a wonderful speed, but I have had a decent ADSL connection since the service first became available here, fifteen years ago. It's way below what politicians say should be the minimum, and the difference would matter for a family. And I don't watch very much streaming media anyway.
One of the internet services I use is notorious for problems with router/modem hardware, it is supposed to have a traffic pattern that somehow slows the connection. Whatever the cause, a hardware reboot after about a moth shows a speed increase. I reckon there are more than a few tall stories to explain problems as "not our fault". Something happens, but I doubt some of the supposed reasons.
I would trust a notebook, kept in a secure place, as my back-up to any of the fancy, computerised, alternatives. It's not as convenient for daily use, but it can work as part of a system. Some of the risks for me are different from those of a busy office. Different risks mean different answers.
Recent experience makes me wary of password managers. They're software. Software goes wrong. What then?
When did you last test a back-up?
I recall hearing of a US accounting software company, producing software to handle the annual US income tax returns, which actively supports Wine.
I know of other companies that make a point of treating Wine as another Windows version in testing and development.
And yet there is well-known software, with Linux versions available, which suffers from what seems to be woefully inadequate testing. Some of them depend on specialised *Nix software on their servers, yet struggle to maintain Linux-compatible software for their customers to use.
I see far more variation that your reporter does.
Our dog was a failed lurcher, and one of those rescue animals. She knew to lead visitors to where we were about the farmyard. It was one of those awkward incidents, a local with severe mental disability, who was going around trying locks.
He got a nasty suck.
It was enough.
Security has to match the need.
I have seen some pretty awkward Linux installs, some on the lines of huge archive files that you have to manually open, put on the right place for your system, and link to the right executable for your desktop. No checks for dependencies, nothing.
It's not just Microsoft.
But then the program doesn't work, and you ask "support", and they ask if you have the same problem with the Windows version...
My experience has been that the people selling such rubbish are severely clue-deficient, and take the labelling on trust, which as often as not never mentions SMB version support. SMB is SMB is SMB.
So it's a combination of piss-poor documentation from the manufacturer, and low-paid sales staff.
For most of this century the well-informed salesman has been a dying breed, but at least I can download the manuals. But does that help?
Last week I was working on an old Dell workstation, it is good kit and I got a good deal. But the manual (and Dell support) are inadequate on how to fit anything in the front-of-case drive bays. Problem sorted, but it doesn't impress.
I've seen several examples recently with a pattern of US lawyers with limited experience of a field of law collecting large fees for rather feeble cases. Most recently, it was a personal-injury specialist from Texas taking on a Federal Trademark case, and trying to dodge the whole Trademark Registration procedure with a court case. The laughter from IP lawyers was muted, but unmistakable. The style is very different.
Are ICANN that stupid? You would think they would at least have involved a competent German lawyer. Some of the labels and concepts are different, but this is part of the point of having Barristers in the UK. The boundaries have blurred, some solicitors can now do jobs that only used to be open to Barristers, but this does look like what you get if you ignore competent and relevant legal advice.
Though there might have been some time pressure. Things do, generally, look a bit too last-minute on GDPR, and not just because the UK government is running around like a headless chicken on anything to do with Europe. But how much of that is wilful American-led blindness?
I have my doubts about the BBC on a lot of things, these days, but I fear it is a growing awareness of the crapitude of news media, rather than any change at the BBC.
And when it is the frothing anti-EU loonies running the country, I find it hard to blame the BBC for being a bit circumspect.
(The other angle is that, on technical issues, it only takes one journalist to skew things; no names, but there are people writing for The Register who have an obvious political bias on some issues.)
That's well-established, not just GDPR. It's explicit in GDPR, and lawyers like "explicit", but if you didn't have something like that it would be a breach for somebody to put your name and address on a letter they post to you. I've had GDPR opt-in emails warning me that I won't get any notifications of dispatch if I don't opt-in. Which means they're saying they can't fulfil a contract. without an opt-in to everything.
How dodgy is that? A US service gave me a web page with default-on permissions for over 300 companies they share my data with. I tried to count them, as I clicked to "off", but lost track at over 270. As the Good Book says:
"Three shall be the number thou shalt count, and the number of the counting shall be three. Four shalt thou not count, neither count thou two, excepting that thou then proceed to three. Five is right out."
It depends on your set-up, particularly your ISP, but one side effect can be to change your IP address. I doubt it will hide whether or not you are in the EU, but I am not sure it's a good idea to be a fixed target.
I don't think the code in any router/modem is all that reliable, long-term. I don't think it's strictly a memory leak, but something accumulates on mine until performance slumps. And a reboot fixes it. We're talking several weeks of uptime, and there is a downside to frequent reboots, but my system reports over a week of uptime, yet the line has only been up for an hour. Something must have glitched at the ISP.
So I think you might be a little bit optimistic.
There seem to be a lot of businesses lingering on directory sites. Several list the Scunthorpe HMV store which closed in 2013. So you search for a business, Google connects you to a directory site, and you are targeted by several adverts. It doesn't matter to any of them that they're handling false data.
When they're so obviously getting data wrong, I can't really expect them to stay within the law on personal data.
Oh, you used to hear about a "Chattels Auctioneers", and it looked like a defunct business, with part of the sign remaining. Problem is, "goods and chattels" is a term of art in the auctioneering trade, the sort of general auctioneering business associated with house clearance. I found an older picture showing the complete sign, with a business name and that phrase.
The GDPR isn't going to do anything to stop that sort of bad data.
I think I prefer the EU attitude to personal data to that exhibited by the USA.
I was around for the original green-card lawyers, and now I get spam emails begging me to let them send me spam. They have spent the years since the previous generation of EU law, implemented by the UK Parliament as the Data Protection Acts, finding new victims and new loopholes. And now they're going to have to do that all over again.
'Bliss it was in that dawn to be alive
But to be young was very heaven.'
Back when the system started, somebody in DEFRA specified a higher precision of area measurement than practical surveying allowed, and the Ordnance Survey figures assumed a flat landscape. Not even Norfolk is that flat. Just the ordinary variations in cultivation, year on year, could lead to bigger variations in the cultivated area.
This isn't rocket science. (DEFRA are the sort of people who want to use satellites in a retrograde geostationary orbit.)
The RPA and DEFRA are the latest version of a continuing pattern of failure to adequately handle the EU's direct payments to farmers. Most of Europe has some sort of central record of land ownership and occupation, often for tax purposes, and when the EU started payments for land rather than produce, they already had the basic records needed.
We had to start from scratch.
30 years ago...
It was a big change, you could have expected this sort of mess back then, but things should have improved.
Brexit will be a bigger change. How long will it take for us to sort out the changes from that?
The program was first released in 2003, and had several choices of colour scheme; text, window backgrounds. and the like.
In 2010 v.2 came out. The colour scheme was brown on brown., no choices available, not even an option to switch from light mud on dark mud to dark mud on light mud. A huge effort has been made to increase the loading on the graphics hardware, but the interface colours persist.
Fortunately, there is an alternative that uses human-compatible colour schemes. It is also something of a memory hog.
Since 2010 there have been changes to the major version number. So everything is OK.
Readers should know this, but the Linux Kernel version numbers don't look right.
I checked the Ubuntu link, and the version numbers they use are different. I'm currently running Ubuntu kernel version 4.13.0-39-generic and the patch is in version 4.13.0-41-generic, which has just come up as an update. I don't know why they don't use a format such as 4.13.41 but they have lists, they have versions for different processors, and they all have that extra zero in the version string. So do other Linux suppliers.
The difference between you and the rest of the world looks so consistent that I am wondering just how reliable your reporting is.
This isn't just a Windows thing, it goes back through DOS, but the alternatives are older.
Back in the day, it might have been problematic to handle the different styles, code size was an issue and RAM could be very limited. Once we got Win95 it was time to fix this. Win98 looks to be the big missed chance.
33 years... It should have been fixed 30 years ago
Having a paper notebook in a safe place is a good situation for using a written record.
But what's a safe place?
At one extreme is the sticky note on your office computer's monitor. That's the total insecurity that prompts "Don't write your password down" rules.
Stupid users, it seems, prompt stupid rules. I think, with my personal situation, I'd be more worried about the other end of the chain. The Twitter example resembles other cock-ups I know of, and it could be an instance of poor management of programmers. Specifications and documentation are critical weaknesses.
That's essentially the level of VR tech in the Charles Stross novel "Halting State", more focused on overlays on the real world, though the bank robbery by a band of orcs with a dragon is wholly in VR. I enjoyed the book, and it's about the people, more than the technology.
That sort of activity has been around a long time. It's different when all you have is text - one-handed typing is a problem - but two people interacting doesn't depend on VR.
The one-handed typing problem might be why they only give you one controller.
They had pairs of hand controllers working with the PS3, and the PS4 has a headset as well, hardly surprising since the Oculus Rift project started out in 2012. Oculus might do some things better but it's the apps it has which will matter more.
I think I shall stick to text.
It looks as though Microsoft are splitting hairs over fixing it, saying that because it needs "social engineering" it isn't a software security problem.
A flawed filesystem on a USB stick shouldn't cause a blue-screen-of-death, however it gets attached.
If Department A at MS say it isn't a problem they deal with, and say they have passed the report on to Department B, who do handle those problems, that's OK. Telling you to submit it to Department B might not be the best answer, but it isn't bad.
Biting the hand that feeds IT © 1998–2019