Posts by Peter Gathercole

Kernel-memory-leaking Intel processor design flaw forces Linux, Windows redesign

Re: Hmmm... @J.G.Harston

That is true, but for volume data moves was mitigated by DMA from disk directly into memory-mapped buffers in the process address space, using the UNIBUS address mapping registers, which allowed raw DMA transfers to addresses outside of the kernel address space.

Of course, not all PDP11 models had the UNIBUS (or, I presume a similar QBUS) feature, but pretty much everything after an 11/34 would have. I had an unusual 11/34e that also had 22-bit addressing, which made it much more useful.

Re: Hmmm... @AC

That's true. Having a address space change would have to disable speculative execution, because it would also have to try to predict which address space it would be in.

Actually, thinking about it, it still has to, because if the mapped page is protected from view, there still needs to be some mechanism to lift the protection to allow the speculative execution of the branch of code, before the decision is taken. But in theory, the results of the branch-not-taken should be discarded as soon as the decision is made, so that the information gathered could not be used. Maybe there is something in the combination of speculative execution and instruction re-ordering (not mentioned yet) which allows data to be extracted from later in the pipeline.

Maybe this is the problem, and if it is, it's probably a design flaw rather than a bug, Interesting.

@ lsatenstein

I think you'll find that Core, i series and Xeon processors are all installed in sockets.

Atom processors are designed in packages intended to be soldered onto system boards. Everything else is in sockets that allow the processor to be replaced. But the problem here is that Intel keep changing the socket design, so you just can't put new processors into old motherboards.

This means that if you are upgrading a system piecemeal, rather than all at once, you end up having to replace not only the processor, but also the motherboard and probably the memory as well.

I would very much like Intel to be forced to support older sockets for longer, so you could give a system a relatively non-intrusive processor upgrade without having to tear the whole system down.

Re: Hmmm...

I think we need to return to PDP11, where you had an alternative set of memory management registers for program and supervisor (kernel) mode. When you issued the instruction to trigger the syscall, the processor switched the mode, triggering an automatic switch to the priv. mode registers, mapping the kernel to execute the syscall code..

This meant that it was not necessary to have part of the kernel mapped into every process.

IIRC, s370/XA and Motorola 68000 with MMU also had a similar feature. I do not know about the other UNIX reference platforms like VAX (the BSD 3.X & 4.X development platform) or WE320XX (AT&T's processor family used in the 3B family of systems - the primary UNIX development platform for AT&T UNIX for many years), but I would suspect that they had it as well.

I first came across the need to have at least one kernel page mapped into user processes on IBM Power processors back in AIX 3.1, where page 0 was reserved for this purpose. In early releases, it was possible to read the contents of page 0, but sometime around AIX 3.2.5, the page became unreadable (and actually triggered a segmentation violation if you tried to access it).

Brazil says it has bagged Royal Navy flagship HMS Ocean for £84m

Re: Whats in a name

When the RN was at it's zenith, warships were named after all sorts of things. The before mentioned HMS Pansy was almost certainly a Flower class corvette, all of which were named after, um, flowers.

It used to be that capital warships were named after famous people, characters from mythology, or an adjective (like Victorious).

Lesser ships have been named after all sorts of things, like counties, towns, and as you get down to the more numerous ships which followed a letter (destroyers, frigates etc.) like the Amazon class all started with "A", with names from all sorts of word category (e.g. Amazon, Antelope, Ambuscade, Arrow, Active, Alacrity, Ardent, Avenger).

With the smaller number of warships recently, there has been a desire to keep certain names going (for example Victorious, Vanguard, Audacious, and Ajax), although for submarines, they are apparently following letters as well.

IIRC, Ocean was quite unusual, as there had only been one previous HMS Ocean, which was a Colossus class aircraft carrier.

One interesting part of Royal Navy tradition is that battle honors for namesake ships are carried across to the new ship, and I believe that the wardroom silver- and crystal-ware is also moved to the new ship.

If this is the case, you can imagine there having to be significant storage space for the wares from all the ship names that are no longer in use!

AI smarts: IBM pushes out 'faster than X86' POWER9 servers

Re: POWER to the people! @AC

Late to the comment trail, I know, but AS/400 was the hardware platform, and used to have it's own processor types, although they adopted (and some say saved) IBMs PowerPC processor platform, with Rochester picking up 64 bit systems with the Amazon (RS64) processor when Austin dropped the ball with the failed PowerPC 620, which barely saw the light of day outside of IBM.

IBM i was previously called OS/400.

One reason that IBM i persists is because it is a very business friendly system. Before things like Apache and the other open software packages were grafted on top of the POSIX compatibility layer, many organizations did not employ specialist system admin/operations staff. It was sufficiently simple and menu driven that the general running could be given to ordinary admin staff with little training, and all of the hardware type stuff was handled by IBM CEs.

But it is a propriety system, and you have almost complete vendor lock-in, which is why most consultancies will suggest ditching them. But that does not mean that they could still be the best solution for some companies.

Yes, your old iPhone is slowing down: iOS hits brakes on CPUs as batteries wear out

Re: Economy

I think you need to look at bigclivedotcom's channel in YouTube for his battery tear-downs.

There are differences between expensive and cheap rechargeable batteries, but they are probably much less than you might think, and it's the embedded electronics that are often the biggest difference. As long as there is some charging protection and over-current fuses, both of which are now *very* cheap to add to a battery (using Chinese produced single chip solutions), they might fail, but not catastrophically. Things have moved on hugely in the last few years.

Of course, if you buy the cheapest, there are likely to more corners cut, but I've bought replacement batteries for phones and laptops from Chinese sellers for years, and not had any problems.

The only faulty phone battery that I've had was a branded Nokia battery for a 7110 (although it could have been a counterfeit, it was bought from a high street phone accessories shop), which suffered an internal short and overheated, although it did not explode or catch fire.

When it comes to SD cards, buying them from supermarkets is nearly as cheap as on-line, and will very rarely give any trouble at all.

Re: Battery shape?

The chemistry of ordinary car batteries that start a fossil fuel car, and those that run EV's is very different.

Starter batteries, which are single batteries, need to provide a very high current (40-100 amps depending on the type and size of engine) for a matter of a few 10's of seconds, and then get charged over the next 20 minutes or so using relatively unsophisticated, and generally rather poor power.

EV batteries need to provide reasonably constant current draw for a few hours, and are then charged using sophisticated charging hardware from a clean supply normally over a number of hours. There are multiple batteries that each contribute to the overall current, and you can do some clever things with switching them from parallel to series for short bursts of power when accelerating.

This means that the chemistry and physical design of starter and EV batteries is very different, even though they look similar from the outside, and also means that starter batteries tend to age faster.

Merry Xmas, fellow code nerds: Avast open-sources decompiler

Re: This is game-changing stuff

I was going to say something very similar, and add to it by saying that decompilers are not exactly new.

I know it's a bit clumsy, but various debuggers have decompilers built into them to turn lumps of machine code into something more readable.

I mean, dbx and gdb have been around a good long time, and I used adb, cdb and in fact the original db (on UNIX edition 6) 35+ years ago.

Whilst I would not want to decompile a complete software suite using one of these tools, investigating interesting bits of code has always been possible.

IBM reminds staff not to break customers in pre-Xmas fix-this-now rush

Re: Someone not following best practices

The idea of starting the period of restricted change half way through December is to allow time for any cockups that do get in before the freeze to be fixed before the real Christmas shutdown.

In most cases, it is not really a full 'freeze', because changes to fix operational problems may still have to be made, but it is really to hold back on any non-essential service affecting changes that may inadvertently cause a problem. Many organizations still allow changes in their non-customer facing systems.

Blighty flogs Qatar a bunch of missiles and Typhoon fighter jets

@Matt Hawkins

Steam catapults need a steam plant. Not necessarily nuclear.

Britain invented the steam catapult just after WWII. This was way before nuclear propulsion was an option.

But you are in a way, quite right. We don't have any ships with steam turbines any more (probably the last built was the Type 82 destroyer HMS Bristol), so there isn't any serious steam generation in HMS QE or PoE (these are IEP - Integrated Electric Propulsion involving diesel and gas turbines driving generators and electric motors), and there is not enough electrical generation for EMALS, although I think that EMALS actually use a kinetic storage device to charge up and rapidly dump the electrical power that is needed to launch aircraft.

Disk drive fired 'Frisbees of death' across data centre after storage admin crossed his wires

Re: What no-one has mentioned yet... - DRUMS

When I visited the machine room in Claremont Tower at Newcastle University in 1978 or '79, they had a drum acting as the swap space on the IBM System/360 Model 65.

What I remember is that the side was replaced by a perspex panel, and you could see the multiple fixed heads arranged around the spinning drum, so there was no seek component of the access time, merely the rotation time of the drum.

At Durham, we had a PDP11/34e with RK05 drives in a DEC 19" rack , and when the drives were busy, the whole rack rocked forward and backward quite violently as the voice-coils moved.

Later, I looked after a system with 80MB SMD drives. The worst that we had happen was the platter brakes seizing, making one hell of a racket, and a minor head-crash. We did have one pack that had the bottom guard platter bent making it a little unbalanced, which used to sing, but we only used that to hold an infrequently updated system backup.

VINYL is BACK and you can thank Sonos for that

Re: I never gave vinyl up

Was not a rumble filter, was modified motor mounts, a different profile belt and a pulley to match the belt profile. All hardware, no electronics.

The motor now floats much like a Rega.

Voyager 1 fires thrusters last used in 1980 – and they worked!

Re: V'Ger

Um. Queller drive. Space 1999 Series 1?

Let me look it up.

Ah yes. S1E6 Voyagers Return. Excellent. Have an upvote.

Anybody know of a UNIGRAM.X archive?

Re: Anybody know of a UNIGRAM.X archive?

I think you may have me confused with one of my then colleges, probably Jan-Simon (surname withheld, as I haven't talked with him to check, he was previously at Imperial College, so knew Sun kit really well), or Paul (ex. of ICL and OSF) who were performing the sysadmin at the UK AIX Systems Support Centre, although 1991 would be around the time that I started picking up what Jan-Simon was doing as he was preparing to leave, so it is possible.

Jan-Simon had big shoes to fill, and I was standing on the shoulders of giants when I moved in to try to manage what he was instrumental in setting up. I believe he's with Google now.

I'm flattered that you even remember the team. Few people either inside or outside of IBM do now.

Re: Anybody know of a UNIGRAM.X archive?

Jake,

I did not take it any further, as I've been busy myself (mainly work related - still a wage-slave even if as a 'consultant').

If you have actual contacts who may know, I am still interested. I should have followed up with TPM, but I think I got on his bad-books when he was posting HPC stuff on the Register, as I was a bit picky with corrections to one of his articles about the UK Metoffice, where I was working at the time.

Probably all in my mind, and he probably wouldn't remember anyway, but me agonizing about the past and being too self-critical are a couple of my failings.

Way back, you wrote a post that implied that we had crossed paths some point in the past, but try as I might, the only contact I remember with anybody on the US West coast was when I was working on UTS on Amdahl mainframes while at AT&T. Most of the contact I've had with people in the US has been east coast in AT&T and IBM (and possibly DEC thinking back).

Anyway, glad the fires are less of a problem. I have a traveling friend who was caught up in them with his family a little, but they're OK too.

Re: Anybody know of a UNIGRAM.X archive?

Who downvoted jake? I'll admit that Maureen O'Gara appears to have a less than stellar reputation, especially over the SCO stories of the noughties, but his information added to my quest to find a Unigram.X archive.

Anybody know of a UNIGRAM.X archive?

Back in the 1980s, before the Web existed, I used to read a tech news bulletin called UNIGRAM.X that was distributed via email, particularly UUNET.

It was, I suppose, a bit like the Reg. without the tabloid headlines and comments, distributed several times a week. There was also another link with the Register, in that according to his biog. Timothy Pickett Morgan was the editor at one time.

It was a subscription service, at least for a while, although it was very common for someone in a company to have a subscription, and then distribute it to other people in the company to read.

What I would really like to find is an archive of the news items, but almost unbelievably, Google et. al. have captured almost no information about UNIGRAM.X.

Is this an example of information falling through the cracks, being neither old enough to merit historians re-constructing the history, nor new enough to have been hoovered up by the Internet's web crawlers?

If anybody has any information, or would just like to reminisce, I would be very interested in reading their comments, and especially interested if such an archive exists.

Pro tip: You can log into macOS High Sierra as root with no password

Re: This is a deliberate feature and it's because Apple cares.

Both "sudo su" and "sudo sh" have problems, in that they will not load the root environment, or run the profile.

You really need "sudo su -" to get the full effect as if you had logged on.

The End of Abandondroid? Treble might rescue Google from OTA Hell

Re: What About It Moto? @RJ1

EE don't provide updates to phones that they supplied, so why do you think they will provide OTA updates for one that they didn't?

If your phone uses stock Moto firmware, it should be possible for you to get the code to put it on yourself.

EE have a bad habit of getting tweaked firmware from the handset manufacturers for the phones they supply, with a slightly different EE specific model number, and a different ROM id. They then do not provide the updates, and the changed ROM id prevents you putting the stock manufacturer ROM on them without significant effort.

I've been caught by this twice with 'phones that came from them. I'm now seriously looking at not upgrading my current phone, but buying an unlocked phone and dropping to a sim only deal.

New UK aircraft carrier to be commissioned on Pearl Harbor anniversary

Re: Coffinships

The only way you can claim that the PoW batleship was obsolescent would be if you said battleships as a whole were obsolescent. The KGV class PoW was a modern ship, having been completed in 1939, commissioned in 1940, and sunk in 1941.

By battleship standards, it was modern, with contemporary propulsion, protection and armament.

The design was hampered by the London Naval Treaty, which put significant limits in the way of a good ship.

British battleships were not designed to fight in close quarters in range of land based aircraft, they were designed to fight other surface ships. That said, the British experience of fighting in the North Atlantic, North Sea and Mediterranean showed that they could still serve a useful purpose in protecting against and deterring enemy warships, even while under air attack.

If the Home Fleet had not existed, German Navel Raiders would have torn the Atlantic convoys to shreds in the area where land based aircraft could not reach.

WWII was the cusp of the change to air dominated warfare, but in that conflict, it was still necessary to have significant surface ships as well as aircraft.

PoW not a lead ship.

The previous Prince of Wales was the second ship of the five ship King George V class of battleships. Other ships, in order of launch were Duke of York, Anson and Howe.

There are some references to Vanguard being a KGV class, but in reality it had more similarities to the cancelled Lion class, which was an evolution of the KGVs.

Intel drags Xeon Phi Knights Hill chips out back... two shots heard

Re: So give them what they really want

It really does depend on exactly what you're doing with an HPC.

If you're doing any type of simulation, then HPC comes down as much to communication and shunting data around between processors/nodes as it is computation.

The flow is generally a computation cycle followed by a communication cycle to prepare for the next computation cycle.

Until you specialize your communications into silicon, moving data around is much better done using a general purpose CPU that an FPU/APU.

A proper HPC system is a balance of multiple different technologies.

Teensy weensy space shuttle flies and lands

Re: How many flights?

Unfortunately, real aircraft are not that strong.

This was the first Thunderbirds episode shown, so dates from around 1965, over fifty years ago.

The effects still stand up now. Good old British brute force, ignorance and an explosives license at it's best.

Interestingly, in the episode "Terror in New York", Thunderbird 2 is crash landed using foam to ease the landing, something that is now done in reality.

Official: Perl the most hated programming language, say devs

Re: Shell scripting tools @MrBanana

But that's the point. Basic System 5 (n)awk, sed, ksh88 and the other tools will probably never change, and that means that it will always work as you expect.

No matter how good the writers of gawk at. al. are, there will always be enough differences to trip you up once in a blue-moon, normally when you can least afford the time to problem solve. Also, the exact version number becomes important as the tools evolve.

I know this is a very backward looking view, but it's served me well over the last 35 or so years (before that, you're talking the original. much more limited awk from UNIX edition 7, and probably Bourne shell or maybe csh - did you know that somewhere in 2BSD circa 1979, there was a shell called vsh which worked uncannily like the later Norton Commander utility).

If you want a real laugh, dig out the Edition 6 shell documentation at TUHS! Two character variable names, with a significant number reserved, no functions and very minimal looping constructs, and a much less usable environment to allow variables to be inherited by child processes.

@EveryTime

I wish I could up-vote you more than once.

- A fellow user of sh, awk and sed, but also cut and paste.

Re: Ahhh Perl @g e

You obviously haven't tried to read someone else's APL. There's no trying to work it out by 'reading' it, because there's no English words contained in the program, except the comments and any text it's trying to output.

In my experience, it's the one language where "It's all Greek to me" is descriptive!

Official Secrets Act alert went off after embassy hired local tech support

Re: Too many stories like that one. @AC

The problem with laptop batteries is that if they are at the stage where they can't even provide power for 30 minutes, at the end of that period, the voltage will take a sudden dive, effectively crashing the laptop.

As the warning is based on either the battery history and/or the voltage delivered by the battery, it often does not give the system enough time to spot and report a battery issue before it's too late!

Car trouble: Keyless and lockless is no match for brainless

Re: Door or boot not shut

The Mini Moke used the instruments from the standard Mini 850 (the donor vehicle that the Moke was made from), so that would be just as materialistic.

The original Mini had no winding windows (they slid horizontally), no inside door handle (you used a wire in the door pocket to open the door), and no ventilation apart from the windows. But even with all this minimalism, Ford could not work out how BMC/Austin-Morris made a profit on the Mini. Apparently, the secret is, they didn't!

The Moke, which was intended to compete with Beach Buggy VW Beetle conversions, was more hair shirt, however, because it did not even have doors, and the roof, if fitted, was more like an awning, with clear heavy duty polythene splash panels (you could not call them doors) to provide some protection from the elements, and nothing as sophisticated as a roll bar! Would not be allowed now.

IIRC, my Grandmothers Morris Minor 1000 had the same instrumentation, so Austin-Morris/BMC/BL got their use of standard parts. Not like today, where things change every year.

Re: Ghost In The Shell

Come on!

This is "Inner Space" the opening title track of GITS:SAC (Stand Aline Complex) 1st Gig, not the anime film (sadly Origa, the main singer, passed away a year or two ago).

The soundtrack for both SAC gigs (and Solid State Society) is absolutely excellent, but was written by Yoko Kanno, and not by Kenji Kawai (the composer of the original anime films - very atmospheric), or Clint Mansell (the live action film - very disappointing).

I have three Original Sound Track albums taken from SAC, and they're good to listen to as music, but they counterpoint the action of the anime perfectly (try watching "Grass Labyrinth – AFFECTION" Gig 2 ep. 11 and listening to "I do" sung by Ilaria Graziano - reprised in "To the Other Side of Paradise – THIS SIDE OF JUSTICE" ep 25, without feeling a little tearful).

IMHO, this is the finest anime TV series to have been dubbed into English ever.

Re: It's a f'in car!

It's not rust anymore. Other reasons are:

"Oh, that part is not available any more, and there's no alternative", and then finding out that it's a common failure once the car is about 5 years old. They'll be none in the breakers yards either. "They're like gold dust. I'll call you if I get one in, but it'll be expensive because there's high demand, and I don't often see one of those with it not broken".

Or special tools necessary for routine jobs on higher mileage vehicles that are too expensive for the small garages to buy, meaning that you have to pay main dealer labour prices or get rid of the vehicle!

I have a ~25 year old MPV that's got to the "parts not available" state (but I'm not complaining about this, it's about time this vehicle was taken off the road, if only for emissions), but I reckon that it will be significantly less than that for any vehicle made today.

Credit insurance tightens for geek shack Maplin Electronics

Re: You cant have it both ways @Neil 44

That could well have been the case, they had to operate from somewhere, but that was rather a moot point for those of us outside of Essex.

Re: Hobbyist @David Nash

But often, the component desk has less than half-a dozen of each component. I went in to get some capacitors to do an emergency rebuild of a TV power supply, and whilst they had most of them, I ended up buying their complete stock of a couple of the values I needed, and they did not have any of one of them. I ended up having to buy two of the 1/2 value capacitors, and wire them in parallel until I could get the correct one.

If you had a complete project, I would doubt that you would get all of the parts in one visit.

Re: You cant have it both ways

One of my retro possessions is a Maplin catalog from something like 1979 (this was while they were still only mail-order).

It's really funny, but if you get a recent one (do they still publish one on paper? - my last copy was from about 2012), many, many of the item listings, pictures etc. are exactly the same in both catalogs.

The one thing you do notice, however, is how much smaller the newer catalog is, even with the new products that did not exist in the older catalog. Whole sections have pretty much disappeared. I used to use the older catalog as a pinout reference for 7400TTL and 4000 CMOS chips, as it had full schematics for almost the complete series. It also used to have a pretty good transistor equivalence section, and pictures of all of the semiconductor packaging types.

I used Maplin because they were more friendly to hobbyists than RS Components or Farnell (although I did use Watford Electronics as well), but also because I read the magazine Electronics Today International (ETI), and Maplin used to make up packs of all of the components, and some printed case inserts for many of the ETI projects. The full modular polyphonic digital synthesizer, which ran over about 2 years, one module per month was a really major project that resulted in a very usable device, but they did multi-channel mixers, guitar pedals, high quality audio and PA equipment, and even a computer develop kit as projects as well, and Maplin sold all of the kits of parts.

IIRC, for several years, the catalog was pretty much the same year-on-year, with the price list published separately, and new products published in addenda with the price list. If you bought regularly, you would get sent the price list when it changed, and I think it was also sometimes attached to ETI.

Linux kernel community tries to castrate GPL copyright troll

Re: Non-GPL feature @FIA

I've actually just read some x86 Linux syscall documentation, and it's pretty much like the PDP11 implementation but without some of the more convenient MMU features, but I realized that I may have used the term "context switch" in a different way than most people would expect.

When I said that some systems did not require a context switch, what I was alluding to was that there is no change in the user mode address mapping registers, and the kernel still tracks time against the process that make the system call (it is still in the same process context). It also may 'borrow' the current process stack.

In a more traditional use of "context switch", you would say that this was actually still a change in context, but because on the PDP11 and s370, the switch to "privileged, supervisor, or system" mode switches to a second set of memory mapping registers (in actual fact a complete duplicate set of all registers IIRC) to get access to kernel code and data spaces, without changing the user memory mapping registers, it leaves the process context intact during the syscall.

In other architectures, for example Power (and x86), there was no duplicate set of registers (not until register renaming became common), so a system call had to do much more saving of the user process context to be able to restore it when the system call completed.

Many UNIX implementations also used the return from a system call as a convenient time to perform a scheduling check, to see whether the current process was still the most eligible to get the processor.

Re: Non-GPL feature @FIA

I've not looked in to the details of Linux in particular (I should really), but on real UNIX systems, it is quite possible for the system call to run under the process context of the calling process without a context switch (in fact for the PDP11 running AT&T/Bell UNIX, it was essential).

I know of at least four different methods that the system call mechanism itself operates (PDP11, s370, SPARC and Power). The main problems are the way that the user and kernel address spaces work, the way that the system call arguments are passed, and whether the system needs to take a context switch as part of the call.

With ancient UNIXes on uniprocessors (PDP11 and s370), the system call had to be non-interruptable, which meant that it was possible to just switch the address mapping within the same process context. This originally made it a cinch to work out the system time that the process used. Less ancient UNIXes with pre-emptable system calls, multi-thredded processes and thread safe system calls had to include a lot of code to handle context saves and saving the timer values as part of the syscall interface.

Re: Non-GPL feature

It is a little more complicated than even that.

IIRC, this has been a source of discussion in the Linux community. Generally, libraries that are compiled in or dynamically linked are often published under the LGPL. This allows for linking in a product without that product having to be published under any GPL license, as long as the library is not altered in any way.

When it comes to the kernel, the problem is not the system call itself which is well known and covered by the interface in libc and other libraries published under the LGPL, but kernel threads or modules that need to access kernel data structures. This requires the modules to know the address of the structures, which in turn requires it to read the kernel symbol table. It is the use of this data that RMS (in particular) believed requires all modules added to the running system to be published under the GPL (the system will mark the kernel TAINTED if there are non-GPL modules loaded).

This was one of the things that prevented OpenZFS from being included in Debian because of a kernel module that it required, IIRC. Not sure how it was resolved in the end.

Wowee. Look at this server. Definitely keep critical data in there. Yup

@Mr Dogshit

Split horizon DNS or other name resolution service.

Pre-order your early-bird pre-sale product today! (Oh did we mention the shipping date has slipped AGAIN?)

Re: Excellent HHGTG reference

I was lucky enough to hear the original first episode on the repeat of the first airing of the radio show, and every subsequent episode on first broadcast.

Somewhere or other, I have some tapes with the whole of the first airing of the link episode and the second season. If I could find them, I think that they would be like gold dust, because the first broadcast of the second season went out before it was really finished, and had some different sound effects and music on the later repeats.

I also had the link episode (the one with the Frogstar fighters) on tape. This was not heard for many years, as it was broadcast twice in the same week, and then neither repeated nor put on the original commercial tapes. It only became available again when the second season was put on CD.

Frogstar Fighter, class C - "That makes me really angry. I think I'll take out this floor"

Zap... Rumble, Crash

Frogstar Fighter, class C - "AAAAHHHHHHhhhh........"

Crunch!

Marvin - "What a depressingly stupid robot"

No, the FCC can't shut down TV stations just because Donald Trump is mad at the news

@terra

The problem is that from the other side of the Atlantic from the US, it is not always possible to directly experience what is going on there. We have to get our information by proxy, and the most immediate ones are the news outlets, particularly TV and social media.

From my perspective, social media is by it's very nature an unreliable source of information. It's very immediacy means that posts are almost always coloured by the posters own beliefs. There is no fact checking, and it is so easy to post partial or incorrect information, and have it go viral. Once information is in the social networks, it's almost impossible to counter. That's not to say it's all wrong, but you cannot use it as a trusted source.

In the UK, I believe we have been lucky to have the BBC, which I would say has been more trustworthy than most. The way I look at it, if the Left are complaining that the BBC is biased to the right, and the Right are claiming that it is biased to the left, then it's probably about in the right place. But even the BBC is prone to sensationalist headlines, and the quality (fact checking, grammar etc.) has declined over time. And they are increasingly relying on other news sources without having the resource to do their own checking.

We do get a view of some of the American TV news outlets here, but Fox News (which I neither liked nor trusted) has dropped off of satellite TV, and I don't really watch either CBS news or CNN. Of other news outlets, Al Jazerra can make interesting watching, but I would not base my world-view on what they say, and frankly, RT is worth watching just to see how bad news coverage can be. The published newspapers appear more commentary rather than news nowadays.

Although we can't really throw stones (being inside the glass house ourselves), US politics appears broken. It appears to be able to be 'bought' by deep pockets, and both the Democrats and the Republicans care too little for the people they represent.

The reason I personally don't trust President Trump is, quite frankly, he is using social media to push his own view of the US, without apparently realizing that many people don't trust the delivery channel. Many of his policies appear misguided or unachievable, and there is a distrust of someone who has clearly come from a radical capitalist background, who suddenly claims to be working for the people.

Some of his policies, like bringing manufacturing back to the US (supposedly to bring quality jobs back), will either increase prices, only deliver low-paid jobs, or just deliver higher automation and no additional jobs, or a combination of all of them. And the policy on The Wall just looked nonsensical. He is trying to turn the clock back on healthcare such that the poorest people will effectively have to rely on charity again. How is that working for 'all the people'. It will help lower-middle class people and above, but not the most needy.

I admit his policies are radical, and certainly don't match previous political thinking, but are they credible? Also, he may be trying to 'drain the swamp' but from here, it looks like he is trying to fill it again with his own Kool Aid cronies.

I guess time will tell whether he is good or bad for the US. One way or another, he will certainly go down in the history books. It's just a shame that the current perceptions of what is happening appear so conflicted at a fundamental level.

Huawei reckons it can strong ARM its way into AI world with new chips

Re: Who?

Tommy, go to the mirror boy, and you'll be Free.

Huge power imbalance between firms and users whose info they grab

"their" data and "your" data.

The idea of "ownership" of data is as complicated as with any other easily reproduced information, and this language is not helping.

The whole concept of possession implies something physical. For example, if I have a unique watch, nobody else can have it at the same time as me.

<pedant>As such, when it comes to personal information, it's really not "your data", it is "data about you".</pedant> Many people may have copies it without denying you. You don't ask an organization to return data about you, you ask to have it deleted. The fact that you exist means that information about you exists. (I'm not going to go all Descartes here, I promise.)

In reality, what people should be talking about is whether an organization has a right to keep particular types of information about you, not whether they own it. They might own a dataset - a collection of data that has an existence of it's own, but basic information is as Jefferson said about ideas, "it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of every one, and the receiver cannot dispossess himself of it". As long as information is known to others, you cannot really claim to be able to control access to it.

I want to be clear. I'm not suggesting unrestricted data retention by organizations, merely that the language used about it should be changed.

This is a philosophical argument, I admit, and I know I will probably be downvoted over it, but, hey, the way it is being presented in the media annoys me.

'We think autonomous coding is a very real thing' – GitHub CEO imagines a future without programmers

Re: Has another five years gone by so soon?

That immediately sprung to my mind as well.

Brit military wants a small-drone-killer system for £20m

@Credas

... but the British frigates and fishery protection vessels were.

Basically, when you get a sharp edge meeting thin sheet steel at any speed, you will end up with a gash in the sheet. Remember, post WW2 ships have no armor.

Sob!

Unfortunately, Lester (RIP) pretty much was the SPB.

Still a loss.

Wonder where LOHAN is now. Probably stuck in a US storage locker, to be shown on Storage Hunters after the rent is due.

You forgot that you hired me and now you're saying it's my fault?

Re: Ah, memories.

I did a stint teaching Computing for a year at a UK Polytechnic, filling in for a lecturer who was on sabbatical (there's a story about how I got the gig. Another day.)

My first lecture was to a group of brickies (brick layers - really, teaching computing to brickies who wanted to move into site management was a thing in the '80s) in a part of the poly I'd never been in before.

I found the room, unpacked my carefully prepared OHP slides, introduced myself and turned on the OHP.

BANG!!!

The bulb blew (very loudly).

Good bunch of guys, really. They took it well, even if I didn't. Was not a confidence building experience, and did not really bode well for a generally miserable year of teaching, which included bearing the brunt of the HNC/D students ribbing, because I did not look any older than them.

Still, I've never been nervous presenting since that year of purgatory.

Bill Gates says he'd do CTRL-ALT-DEL with one key if given the chance to go back through time

Reset buttons on the system unit appeared on quite a few clones of the original PC (along with Turbo buttons), and really became standard with the ATX motherboard standard. That was a real hardware reset, originally in TTL that simulated a power on by getting the processor to do a power-on initialize (there's a RESET pin on an 8088/6 chip) without actually having to turn the power off. It was more friendly to the power supply than actually hitting the power button.

The power button on most PCs is now a software power button interpreted by a small bit of logic on the motherboard (or in it's support chipset). One press generated an ATX reset. Holding it for a few seconds tells the power supply to stop supplying power to all but the standby power rail.

Re: Bad Idea

Remember that the original PC Model F keyboard (Not the AT keyboard we use today) had fewer keys, and there was no dedicated del key.

Missed patch caused Equifax data breach

Re: Typical problem of many large organizations

It is easy to say in hindsight that this patch should have been applied.

But just look at the volume of vulnerabilities, across all software platforms that a company has to watch and plan patches for.

Even in the most proactive organizations I've come across, planning and testing a patch deployment, and arranging for the necessary reduction in service as patches are rolled out can take weeks or months.

What most people don't take into account is that it is quite frequent that a patch changes a behavior or breaks something. One of the past mantras in changing systems used to be "only change one thing at a time", so that you could isolate which component breaks the system. But with the 24x7 nature of many systems nowadays, service outages are hard to arrange, so patches are bundled into releases. Because you are changing multiple components, it becomes important to test a release before deploying it, otherwise you get panned by customers and the press for not testing before release.

So you're stuck between a rock and a hard place. If you spend time testing, you're open to the vulnerabilities while you are planning and testing. If you shortcut the testing process, then you're open to breaking the services you offer.

In my view, and I think it is a very common view, there are two significant things that have to be done.

One is engineer the systems such that you can deploy patches to subsets of the environment while leaving the service running (for example a leg1-leg2 split), so that you don't have to have as many service outages.

The second is that you split your application up into discrete security zones, with the internet facing systems that are most likely to be hacked only having access to data on a transaction-by-transaction basis, with the data being provided under the control of the next zone in. Although this will not prevent data theft, it will prevent mass data extraction, so long as you have decent monitoring of transaction rates, and intrusion monitoring.

The systems holding the bulk of the data, for example the database servers, are in your most secure zones, and you make sure that even if someone gets into these systems, it is difficult to bulk export data out to the internet.

The more zones you have, the more difficult it becomes to hack in and export, especially if you use different technologies for each zone. Hopefully, with enough zones, one of two things will happen. Either the hacker trips some intrusion monitor before getting too far into the system, or they decide that it is just not worth the effort to get any data.

There are many other steps that need to be taken, but these two will mitigate software flaws, limiting the damage. Unfortunately, they have to be designed in from the beginning, and are difficult or impossible to retro-fit. This means that a small quick-and-dirty proof of concept or pilot often needs to be completely re-designed to make it production ready.

But too often, manglement see a working PoC, and decide that it can just be scaled up, rather than the necessary (and expensive) redesign. To them, it's all extra cost that they can't justify. And because many of the people implementing the PoC, especially if they are using newer technologies, are often younger and less experienced, they're not prepared to push back.

The result? Systems that are easy to get to the data through exploitation of only one or a small number of vulnerabilities, and easy to export the data across the Internet, together with a difficult patching process. A recipe for disaster.

Cassini probe's death dive to send data at just 27 kilobits per second

Distance... @BahBoh

Whilst I totally agree with your comment that the distance has not (drastically) changed (in the short-term) (my bracketed additions), the fact that Saturn and Earth are in different orbits, with different orbital periods, means that the distance between Earth and Saturn is constantly changing.

It reaches a maximum when Earth and Saturn are on opposite sides of the sun, and a minimum when they are on the same side.

Currently, I think that the Earth is drawing ahead of Saturn, so the distance will be increasing for the near future.

But you know this...