* Posts by Peter Gathercole

2924 posts • joined 15 Jun 2007

British jobs for British people: UK tech rejects PM May’s nativist hiring agenda

Peter Gathercole Silver badge

Re: Don't forget us oldsters! @AC

Sorry, parts of that comment did concentrate too much on young people. I agree that cross training is required as well, but I think that it was covered to a degree by the "training their own people" statement at the beginning of the comment.

I am in my mid '50s and do face the same problems you do, but I hope that my core skills will remain in demand long enough to see me through to retirement, but I am beginning to wonder.

Peter Gathercole Silver badge


The problem that you point out is true, but probably not because of what you have stated.

Both sides in the referendum campaigned on negative agendas. As a result, the people in the exit camp were campaigning against Europe, rather than for a particular model of Britain's future. It thus became a wide church, covering a range of people who are out-and-out xenophobic, all the way to those who are prepared to live in a global society, just not one controlled by Europe. There really was no one issue other that Europe itself that the voters could agree on.

And for the campaign that wanted to stay, they really did overplay the fear card, claiming consequences once we had left that are unlikely to pan out to the extreme cases thay envisaged, so much so that many people just didn't believe it.

As a result of these broad campaigns, we have a situation where even if you just look at the leavers, whatever deal is chosen eventually will upset significant numbers of people. It's unavoidable. There is no way there can be a solution that will satisfy a majority of the UK population.

The referendum was just not thought out properly, and was framed in a way that, because David Cameron thought he could not lose, did not actually ask what kind of Britain people wanted.

I would like to have seen a third option of "kick it down the road for a few years". I believe that most people would have voted for that rather than stay in or leave, which would have shown Europe that the UK was not happy with the way things were going, and put them on notice that we really could leave.

I myself have no problem with working with people from other races, creeds and colours. I have no problem with them living and working here as long as they are doing something that cannot be provided by the local workforce (but please see below about my views on training).

My view was that I no longer wanted to be shackled to a group of countries that were becoming inward looking and so bureaucratic that it was going to become impossible to achieve any change, while the rest of the world was moving in a different direction. So I voted to leave. I did not want to be part of a United States of Europe, which I believe is the direction it is going.

I understood that in the short term, there would be economic costs, but my belief is that within 10 years, as long as we don't end up with some membership-lite deal, the UK will pull ahead of Europe.

Where I do have a problem is with a predilection for businesses to employ people from abroad in preference to training our own young (or even older) people to do the jobs. This not specifically a European issue, but was complicated by free movement within Europe. The lack of training upsets me greatly, as I want to invest in future generations. As I look at it, it should be a no-brainer to make sure we have a adequately trained and experienced native workforce. We have to do this for the sake of our children and children's children.

Peter Gathercole Silver badge

I'm agahst!

The thing that the article completely misses is that employers are not prepared to invest in training their own people or new workers!

You cannot expect the education, even the further education system to generate the people with knowledge and skills that all companies want. It takes +5 years to get a course up, running, and generating trained graduates. By the time this has happened, the skills required have changed. Windows and .Net a few years ago, Linux and Python now, who-the-hell-knows-what in the future. It's education's job to provide a good fundamental grounding, not fully rounded workers. That comes later as they gain experience.

It has in the past, and will remain in the future, the case that to get the skills needed, you can go out and fight in the market for a limited resource, or you can invest in apprenticeship and cross training of the people available. Find someone who has aptitude, fresh out of education, and mould them into what you need.

And you know what. Training UK people keeps the money in the country, and enriches the available skills, and generates jobs for UK residents. It will also continue the UK's reputation for being somewhere to come from abroad to to learn, something that is being increasingly eroded in recent years.

Businesses have become lazy, inward facing, and too focused on profit. They need to take on some responsibility themselves, instead of wanting to steal doctors, nurses, IT skills and many other things from countries that need to keep them themselves.

SSDs in the enterprise: It's about more than just speed

Peter Gathercole Silver badge

Re: Not all about performance @AC and spinning HDDs

Whilst it it true that there are storage systems that do spin down disks when they are not in use, this is not the norm. The problem with spinning disks is that frequently, the most likely time they fail is when they are spun up after being spun down for a period of time.

I recently was in charge of an estate with ~5000 disks in it. It ran 24x7, and was only powered down if there was work being done on the site UPS. When we powered it back up, we inevitably had problems with some of the HDDs, leading to disk-recovery procedures needing to be invoked.

I know that in the archival cold-disk vaults, they end up with a complex RAID/mirroring environment such that data was stored on *several* hard disks, so that when the inevitable failures happen, the data is still safe. You also have to cope with bit-rot.

ZX Spectrum Vega+ will ship on time, developer claims amid doubts

Peter Gathercole Silver badge

Re: Arer we surprised? @steamnut

It's a completely different manufacturing environment now. If you have the design and the board layout, and are not using anything particularly esoteric, then there are companies in China who are queuing up to build these things for you.

Just look at Alibaba, and there are adverts for companies that will build to your design, and at volume that you want. It's a bit late for volume shipments before Christmas (shipping by sea can take 6 weeks), but it may still be possible to get them air-freighted in if you can accept the cost.

Chances are these things are in a container somewhere on the ocean. Lets hope it is is not on a ship owned by Hanjin Shipping Company!

Peter Gathercole Silver badge

If I remember my brief encounter with a ZX81, I bought an assembler called ZASM or something similar where you loaded the assembler program into memory, then added the assembler code as REM statements, and then ran it. It assembled the code into more REMs somewhere else in the program. You then deleted the assembler and the source, and wrote any extra BASIC around the machine code.

I already knew 6502 assembler, but I learned Z80 assembler on this setup. It kept me busy while my BBC micro arrived.

Sinclair fans rejoice: ZX Spectrum Vega+ to launch October 20

Peter Gathercole Silver badge

Re: Hmmmm...

If you've never played Elite on a BEEB with a Bitstik, then you've really missed out!

Excellent control, and the throttle on the twist joystick.

Oh, and a 6502 second processor helped (full screen in mode 1, all galaxies in memory, and smoother as well).

If we can't fix this printer tonight, the bank's core app will stop working

Peter Gathercole Silver badge

Re: Some time ago... @ShortLegs

I love all this retro stuff, so it's no bother to explain. What I am talking about here is the 80 column IBM or Hollerith card that was common in the '50s, '60s and '70s, but was mostly obsolete by the '80s. I learned to program PL/1 on them, and also used them in my first job.

A punched card was exactly as it sounds. It was a card measuring 187mm by 83mm. It had rounded corners, with one or other top corner cut off to allow quick card orientation checks.

They had rectangular holes punched in an 80 column by 12 row matrix, with each column representing one character. Not every manufacturer used the same encoding type on the cards.

Generally speaking each card represented one line of information or a line of source code. Most card punch machines would punch the holes representing the character, and would also print the character at the top of the card, so that you could read the card.

Most languages (but not all) reserved the first or last six columns to hold a card number, so that cards could be ordered. When punching the cards, it was normal to step the cards number by 10, so that it was possible to insert cards into the deck without having to repunch the whole deck. Most compilers would allow you to miss the numbers altogether, but if you ever dropped the card deck.....

Punching the cards was done on a desk-sized machine with a keyboard and a transport mechanism which allowed blank cards to be fed in to a hopper (normally on the right), typed one column at a time, and then moved into a hopper on the left (cleverly turning the cards over to keep the order). The better card punches also allowed you to copy a card, one column at a time until you got to an error, and then punch the rest of the card.

Alternatively, you could get hand punched that allowed you to punch by hand the correct holes for each character, but you had to be desperate or very clever to use one. Some people claim also to have blocked cut holes by carefully placing a 'chad' (the cut out rectangle of card) into a hole to 'edit' a card, but I'm a bit sceptical myself.

Column 7 in most languages was a card-type indicator. Typing a C in this column normally meant that the cards was a comment. This may sound wasteful, but the comments would be included in the listing from the compilation. In some languages, like Cobol and RPG, the card-type indicator was used to identify the section that the line was in (Input, Calculation, Output, Exception).

Cards were read in a reader that pulled one card at a time (at a rate of 2-10 cards a second) across either an mechanical, air-jet or optical reader, which worked out which holes were punched. You kept card decks together with elastic bands, and just for security, most people would use two bands, in case one broke.

Creasing a card, or allowing them to get damp or worn at the edges would often cause the cards to jam in the reader, normally meaning that the whole deck would be rejected. Having a deck rejected meant that you missed the slot and have to put your job to the back of the queue once the problem was rectified. Remember that systems were mostly single tasking (at least for compilation streams), so they processed the queue in sequence, one-at-a-time.

I'm not sad to see the end of those days, but having lived through them, I feel that it imposed a rigour that would benefit modern programmers if they had to experience it.

Peter Gathercole Silver badge

Re: Some time ago... @earl grey

Thanks. I don't necessarily differentiate between chain and band printers. My bad.

It was probably a band printer, but the basic operating principals were largely similar.

Peter Gathercole Silver badge

Some time ago...

I know some of us do, but I wonder exactly how many of the readers really know what a "chain printer" actually is.

Back in the day, these were capable of 600+ lines a minute. The speed that the chain which carried the letters moved was such that if the chain broke, it could seriously damage the heavy metal acoustic cover. There was a reason why there were cover switches to prevent the printer operating when the cover was open, and it was not just the noise.

In my first job, the single chain printer that they had printed all of the council's rates demands and payment books, the car parking fines, the council house rent and backlog reports, the payment cheques etc, well pretty much anything that the council sent out in the post bar what the typing pool typed.

The printer was on a special I/O channel, and the particular model of Sperry Univac 90/30 could only have one of these printers attached. The printer itself was the size of a large wardrobe, and could not easily be swapped out.

They were a significant part of the cost of the computer system, and it really was a severity one call if it did not work for more than a day. Oh, and when it was down, none of us programmers could work either, because we needed it for the printed output from the batch (decks of punched card) compilation runs for the programs we were working on.

BSODs of the week: From GRUB to nagware

Peter Gathercole Silver badge

Re: Linux Kernel Panic @fajensen

<sarcasm>Gosh, systemd is more magical than I knew! It can be loaded from a missing filesystem, where init can't and then mystically re-populate a newly formatted root filesystem to make the system work! I must quickly switch all my systems to systemd immediately</sarcasm>

I'm a SysV (and earlier Bell Labs UNIX) init diehard. I've lived with it for 30+ years. I can (reluctantly) cope with Upstart, because it still does the init directory thing, but I'm really thinking of trying to find a Linux that does not include systemd because It's too complicated and non-deterministic.

Failing this, one of the *BSDs beckon.

Peter Gathercole Silver badge

Re: Linux Kernel Panic @2 ACs after my previous post

The last real software BSOD I saw was caused by the driver for a graphics adapter in Windows XP, although I have seen the equivalent on Windows 7 caused by not having the correct drivers installed after swapping the motherboard on a system.

I admit that in neither case was it the primary OSs fault, but device drivers. But the driver model on Windows NT 4 onward, where the graphics driver can bring down the whole OS, compared to Linux, where most of the graphics code runs in user mode so the screen crashes, but the rest of the OS functions so you can either re-start the graphics subsystem, or gracefully bring down the OS is IMHO preferable.

Ah, wait a minute. An update caused my middle son's Windows 7 system to fail to boot last weekend. Was fixed, I understand (he fixed it himself), by re-loading the Nvidea driver for his 980.

BTW. I'm old enough that I have worked at the source level on UNIX kernels kernel, and seen (and caused!) real kernel panics caused by code faults.

I've also seen panics (well, 888-102 and 103) errors on AIX (you see this sort of thing in a support centrre)

There used to be a standard X11 screensaver that showed the crash screen of several types of system, including SunOS/Solaris, Macintosh OS (OS9 or earlier) and Windows, amongst several others. Used to really surprise people when they saw it unexpectedly. Was also a challenge trying to identify them all as they cycled round.

IIRC, the Windows mock crash screen had NCC 1701 encoded as one of the fault codes!

Peter Gathercole Silver badge

Re: Linux Kernel Panic

Whilst I understand that it would be interesting to see a real kernel panic as a result of a code fault, this isn't one.

It looks as if the ReiserFS filesystem on device md(9,1) - if I read that correctly, is corrupt, and cannot be mounted, and then cannot umount the RAM filesystem that was loaded during the bootstrap. This looks like it is the root filesystem, and as a result when the kernel that Grub has already loaded tries to start init, it can't.

From this point there's not a lot the system can do, and it takes the very sensible decision to panic with an appropriate message.

There's no fault in Linux, so it's not what I would call a real panic.

Greybeards beware: Hair dye for blokes outfit Just For Men served trojan

Peter Gathercole Silver badge

Fortunately, I don't need it

I am going slowly grey, but will welcome it, as it will finally make me look closer to my real age.

<smug>Still have a full head of hair that is naturally mostly it's original colour with the odd grey one mixed in, at closer to 60 than 50</smug>.

My daughter is getting married shortly, and I've been told I look younger than her fiancee, even though I'm 27 years older! The wedding photos are going to look strange.

Ubuntu tees up OpenStack on IBM's iron

Peter Gathercole Silver badge

Re: Linux with IBM ?

Well, a couple of decades plus a bit.

AIX on POWER (original RIOS processors) was released in 1990 (26 years ago) and running before that in 1989.

Not that long ago, I found a binary compiled on AIX 3.2.5 (probably compiled on a 7013 model 530) which ran on AIX 5.3, and I don't see any reason why this particular binary would not work on AIX 7.2.

These systems are not members of the Linux only POWER family, according to the link to the brochure. They will run AIX and IBM i/OS as well as Linux.

And I think that you may find that some 360 mode programs no longer run on the latest zSeries, as they've slowly been retiring some of the older execution modes on the later mainframes. I think that everything 370XA mode or earlier will not run without being re-linked.

Pluto's emitting X-rays, and NASA doesn't quite know how

Peter Gathercole Silver badge

Re: Star Surgeon @DNTP

This is the second James White Sector General reference in as many weeks.

One of my school friends claimed that James was his uncle. Never found out whether it was true or not.

HP Inc's rinky-dink ink stink: Unofficial cartridges, official refills spurned by printer DRM

Peter Gathercole Silver badge

Re: My printer not HP's @wordmerchant

Well, if a NutriBullet wouldn't do it, I don't know what would. Vicious things, they are. Designed to smash seed kernels.

Emacs and Vim both release first new updates in years

Peter Gathercole Silver badge

@prinox - 4-function calculator

Well, calculator. Emacs has one of those (ESC-X calc-keypad).

Has more than four functions, however, and is an RPN calculator by default.

Pass the 'Milk' to make code run four times faster, say MIT boffins

Peter Gathercole Silver badge

Re: OpenMP ... does not have a compiler @Frumious

Mpicc is a wrapper around various C compilers, and is normally used with gcc.

It's really not a compiler in it's own right, more like a pre-processor.

The MPI component will take a number of inline directives, and generate some C, unroll some loops into parallel threads, do map reduction and some other optimisations, and add some glue code and hooks to library routines to handle passing data between local and remote threads.

Having done this, it then passes the resultant intermediate source to the backend (real) compiler which generates the linkable code which is then passed to the linker to resolve all the library calls.

If you remember, the earliest C++ 'compilers' worked in exactly the same way as a pre-processor to a C compiler, but I would suggest that C++ is more of a complete language than OpenMP.

Peter Gathercole Silver badge

Re: Software? Or maybe hardware. @Brewster

Fair enough. I missed the line on "common algorithms" in my own reference. No excuses there, but, again, I will wait for the final paper to see which algorithms these are.

I've worked with people doing serious work with HPC systems, and provided technical support for those systems. They put a lot of effort into trying to make sure that data is already in an appropriate place before it is needed. Generally this is done by localizing data in chunks where much of the required data is close to other related data, in blocks aligned to a common block sizes. They try to make sure that data is used such that it can be fetched in regular ways so that the pre-fetch and cache hardware will have it lined up for when it is needed. They make sure that the minimum amount of data is requested between cores and systems over the interlink.

When you are iterating over a loop many millions of times, saving a few instruction per iteration, and avoiding cache misses and context switches can save you a huge amount of resource.

Whilst I can see that there is the possibility that Milk could make savings for some particular types of problem, all it is really doing is eliminating efficiency problems from non-optimized code. It's no substitute for experienced programmers, but held out as a carrot for organizations wanting to reduce the skill set of their programmers. IMHO, having worked with some very skilled programmers, it will take a long time before this can be realistically achieved.

It will be very interesting to see whether adding Milk to the Unified Model for weather forecasting will result in faster code.

Peter Gathercole Silver badge

Re: Software? Or maybe hardware. @Brewster

OK, I'll wait for the full paper to be published, but there's a number of things in the quotes in the article that make little to no sense (although it could be that the journo writing the original article has not fully understood it).

Let's start with "when a core discovers that it needs a piece of data, it doesn’t request it"

Um. A core. A physical processor. How is this conditioned by the Milk compiler without wrapping the load instruction with a whole load more code, because that is all a compiler can do!

Then we've got "adds the data item’s address to a list of locally stored addresses".

and then what. Puts the thread to sleep? Hello. Expensive context switch. How's that going to improve latency and throughput.

and "and redistribute them to the cores"

The compiler can sort this out? It's going to have to know a huge amount about the shape of the system the code is going to run on before it generates the code. And most systems leave the placing of data in memory to the kernel and the hardware memory translation mechanisms. Milk's going to be able to control all of this, simply?

And anyway, The article talks first about a new language, and then talks about modifications to OpenMP. Which is it? OpenMP is not a language in it's own right, and it does not have a compiler, it's more like a pre-processor that expands a number of in-line directives in the code into something that the following compiler (Fortran or C) can generate linkable code.

I don't know whether you've ever used OpenMP, but it already does significant data reduction. It sounds like this "Milk" language is merely a modification to OpenMP, and not a language in it's own right.

Ah. I found the original MIT release. One of the things it says is "manage memory more efficiently in programs that deal with scattered data points in large data sets". It also says nothing about "common algorithms". It talks about a rather generalized, sparse data problem that is not actually really suited to the type of computing tha HPC systems and OpenMP are really suited to.

But I'll look for the full presentation, but I doubt that it will claim the type of panacea that the Register article claims.

Peter Gathercole Silver badge

Software? Or maybe hardware.

So instead of just fetching the data, you're going to catch the request (in software?), and defer the code waiting for the data until the data can be more efficiently fetched.

Just how many more very expensive context switches will this generate? And where are the other threads that can be dispatched once all of them are waiting for an 'efficient data fetch'. And how will that affect the latency of individual threads?

I'm sure that there are some highly threaded applications with unpredictable data flow where this could be a benefit, but on the brute-force codes that make up most HPC applications, which mostly process data in predictable ways, especially Fortran code where the standard dictate how data is stored in arrays, this is likely to be completely unneeded extra code that can only slow the total throughput.

I think I'll let the hardware cache pre-fetch hardware provide all the speedup most real 'Big Data' requires.

Using a thing made by Microsoft, Apple or Adobe? It probably needs a patch today

Peter Gathercole Silver badge

Re: iPatch

Previous history. In the late 1970's and 1980's AT&T Exptools (and probably other tool packages - V7 Addendum tape springs to mind) had a utility to edit i-nodes on a UNIX file system that was called ipatch. I probably have a paper copy of the man page somewhere.

Delete Google Maps? Go ahead, says Google, we'll still track you

Peter Gathercole Silver badge

Re: eh? @RAMChYLD

Unfortunately, cell phones have to advertise where they are and be tracked, so calls can be routed to the cell where the phone is. There's no way the 'phone system could probe the cell network of the whole world to locate a phone.

So it's axiomatic that a functioning mobile phone can be tracked without GPS, WiFi, NFC or Bluetooth.

The difference is that the cell location information is normally limited to the service providers running the cell network, and agencies with legal access to that information. The combination of WiFi/Mobile Data and GPS/AGPS makes this type information available to all apps with some tracking function.

I run will all comms except the phone disabled, but mainly because of battery life.

'Oi! El Reg! Stop pretending Microsoft has a BSOD monopoly!'

Peter Gathercole Silver badge

Re: Machine Operating System @davidp231

Thanks for the link to RetroClinic, and the detailed startup information.

My BEEB has an issue 3 board, ordered on the first day that they accepted orders, and was delivered with OS 0.1, and still has Basic 1 (the OS was upgraded when the disk upgrade was fitted). I never really used B+ or B+128s, but I did have access to one of the first Masters that was available, with a 3xAA battery holder for normal batteries, not a single battery, but no longer. I also ran an Econet Level 3 fileserver with a 10MB hard disk for a network of machines with many of the available peripherals.

I think the DataCentre may make it onto my Christmas list! I hope it doesn't clash with the ATPL Sideways ROM board, but I suppose I could take that out.

Peter Gathercole Silver badge

Re: Machine Operating System @8271.

Yes. How memory plays tricks.

Peter Gathercole Silver badge

Re: Machine Operating System

My thoughts are with Martin here, although I'm not sure that a machine running 1770 DFS had a battery (a model B+ or B+128 probably, but could be a B with a 1770 controller rather than the 8270 that was more common on model Bs. Probably not a Master or Master Compact because they came with ADFS).

The BEEB *ALWAYS* booted from ROM, but could be set to run an exec file called !BOOT from the floppy (or hard disk, but this is not ADFS) when it was started. To my mind, this looks like this is what it is trying to do. I would want to look at the access light on the drive to be sure.

Model Bs had space for a DIP switch on the keyboard rather than non-volatile memory to set the startup options, but it was possible to solder wire links rather than using a switch.

If it is that the diskette has worn out, I hope they have a supply of the soft-sectored, 5.25" double-sided, double-density diskettes, because I tried to locate some to buy a couple of years ago to re-write my collection, and totally failed to find any, even on the auction sites!

Standard 360KB or 1.2MB 5.25" IBM disks do not work.

I think I've seen some enterprising person build a SD card adapter that looks like a hard disk on the 1MHz bus for a BEEB, but that would need ADFS installed!

Maybe time to re-code for a Raspberry Pi.

Phones exploding in kids' hands, shares tanking – but it's not all good news at Samsung

Peter Gathercole Silver badge

Re: Removable battery

But it's strange. The three phones that I've taken apart that supposedly have non-removeable batteries (Sony Xperia SP, Nexus 4 and HTC Desire 626) are surprisingly easy to take the batteries out! The only thing that makes it difficult is the double-sided tape holding it in.

All this "it makes the phone slimmer" argument a bit lame.

IBM lifts lid, unleashes Linux-based x86 killer on unsuspecting world

Peter Gathercole Silver badge

Re: HP beat em already?

Remember that IBM has produced some of the densest servers already with the 9125 F2C Power7 775, aka p7IH (the 'failed' Blue Waters machine, that still got sold to serious HPC shops).

When Power 8 came along, with the drop of the GX++ bus, there was not an p8IH, because of the redesign work necessary for the Torrent hardware (AFAIK), but the last time I spoke with people in the know, they were intending to take a shot at a p9IH server for HPC, but that is no guarantee that a product will be see the light of day.

You've also got to look at the density of the BlueGene/Q, now pretty old, but still scattered through the Top500 list, and peaking at position 4 in the June list.

Peter Gathercole Silver badge

Haven't checked, but...

If this is a member of the Power family, even though it won't (yet?) run AIX or IBM i, it will be LPARable like all the rest (assuming you buy the entitlement).

You've been able to LPAR Power servers, even the Linux only ones, for longer than Unisys have been using Intel processor in Clearpath servers.

IBM pretty much wrote the book on partitioning servers.

Latest Intel, AMD chips will only run Windows 10 ... and Linux, BSD, OS X

Peter Gathercole Silver badge

Re: @fandom

OK. I was confusing 386 and 32-bit x86. Sorry.

Peter Gathercole Silver badge


...will drop support... There. FIFY.

It's not happened yet, and when it does, mainstream Linux distro's will remain supported in i386 for several years to come, because they will keep the older kernels in their LTS repositories for quite a while (Ubuntu will drop support first in 18.10)

I estimate that I'll have retired all my i386 boxes way before Ubuntu 18.04 drops out of support in 2022 or 2023.

It's time for humanity to embrace SEX ROBOTS. For, uh, science, of course

Peter Gathercole Silver badge

Re: Escort - Reminds me of an old joke

Q. What's the difference between a Skoda and a sheep

A. You're less embarrassed getting out of the back of a sheep!

Of course this goes back to before VW bought Skoda. Maybe it should be a Dacia now.

Watch SpaceX's rocket dramatically detonate, destroying a $200m Facebook satellite

Peter Gathercole Silver badge

Re: molepeople @Prst.

Nice Bromeliad Trilogy (Truckers, Diggers, Wings) reference.

Have an upvote, but hold off on the poetry, please.

Kindle Paperwhites turn Windows 10 PCs into paperweights: Plugging one in 'triggers a BSOD'

Peter Gathercole Silver badge

Re: I remember @Charles (again)

I just read your post again.

Is the support really not there? Have you checked to see whether the chipset is not supported, or that the system just does not have an entry in the hwdata lookup-table for a particular a manufacturer and device ID.

If it is the latter, you need to add an entry in either /usr/share/hwdata/usb.ids or pci.ids from one of the repos. on the Internet. Not that difficult.

Peter Gathercole Silver badge

Re: I remember @Charles

It is quite true that there are unsupported devices. I have a PCI wireless card made by ASUS that seems to use some form of Broadcom chipset for which the drivers don't exist for Linux.

But equally, the drivers don't exist for Windows 7 or later, and they were pretty difficult to find for Windows XP!

So, not the OS provider (of whichever sort) at fault.

But Linux has another trick up it's sleeve. If you can find a usable WinXP driver, there exists a method of wrapping the Windows driver up so that it can execute on Linux. This is described as ndiswrapper, and if you really need to get a device working, this can provide a solution.

But for most people, the solution is, unfortunately, to discard the device and get another.

On the subject of wireless drivers, I've put Linux on many, many laptops and computers, and outside of the example listed above, it has just worked for pretty much all of the systems I've built in the last 10 years.

The only caveat to this is generally, I only build on older systems, not bleeding-edge ones. It is possible that a new system may require some lead time for someone to figure out what's different.

But ask yourself. Is the fact that a device manufacturer is prepared to provide Windows driver but not a Linux one a fault of Linux?

Peter Gathercole Silver badge

Re: I remember @AC

The way that Windows device driver identification works is fundamentally broken IMHO.

It requires you to have a specific driver for the USB and PCI (and I presume PCIe) device identity. This is normally provided by the manufacturer (remember all those shiny round things that came with the device).

The result is that if you upgrade a windows system, and the existing propriety driver that used to work fails the upgrade compatibility check and is removed from the configuration, the device will be left with no driver loaded. This is even if there are perfectly good drivers for that particular device on the system (this is particularly bad for network devices that are largely built from standard chipsets).

in the case of network drivers, this may mean that you can't even get to the Internet to try to find working drivers!

The Linux model, which has generic drivers for almost all of the chipsets included with the OS, and a device ID mapping file that points to the correct generic driver for a particular device, means that as long as you can identify what driver should be used, even if it is not in the config. file already, you have a fighting chance of getting it working without having to find another machine and start mucking about with USB memory sticks to copy the driver to re-install.

And you're not beholden to a device manufacture who has no real interest in providing new drivers for old hardware.

A plumber with a blowtorch is the enemy of the data centre

Peter Gathercole Silver badge

Fun with Serial lines.

Two examples, both from the same place.

First. Wired corridor in a Polytechnic for Acorn Econet (bussed RS422 serial network used for BBC Micro's). One day, the network stops. We check each of the access ports (very basic 5 pin DIN connector in a box soldered directly to the wires in the cable to reduce contact resistance in connectors). All OK. Terminator, OK. Meter across the wires shows a short. Eventually tracked it down to a staple carelessly driven through the cable to 'tidy it up'.

Secondly. Camtec X.29 PAD used as an RS-232 terminal switch. In order to get it's attention from the PDP-11, it was necessary to generate a communication break (data line connected to ground for a second or so). DZ-11 or the comms software (can't remember which) could not do this, so I created an interposer that consisted of a 25 pin male D-shell connected to a 25 pin female D-shell with soldered wires between the pins, and the two held together with long bolts with several nuts holding everything in place, and a press-to-make, release-to-to-break switch between pin 7 and pin 2 (or was it 3). One day, PDP-11 gets slower and slower, and eventually stops. Reboot, everything OK for a while, then the same thing happens. Looking in the log, it was reporting data over-runs on one of the DZ-11 ports.

Turns out some vibration had loosened the nuts, one of the D-Shells had moved, causing a short from pin 2 to pin 3 (data out and data in or vice-verca) in the wiring to the press switch. Login banner sent from PDP-11 came back as if typed from the terminal, which generated errors and a new login banner. Eventually system was so busy fielding exponential amounts of data that it ground to a halt.

Moral. Good wiring is important, good soldering equally so.

Former RN flagship HMS Illustrious to be sold for scrap – report

Peter Gathercole Silver badge

And, if you remember, they were only going to buy a single catapult, and swap it between the two ships as they entered/left refit. (Even now, it is only intended to keep one at sea at a time, which is how they intend to get away with so few planes.)

A ship of the size of these, with only a single catapult would be useless. Nimitz and Ford class carriers have four....

Whatever idiot suggested this method of working obviously thought that removing/fitting a catapult could not possibly be any more complicated than changing a car tyre!

Peter Gathercole Silver badge

Electrical energy

Well, the US EMALS system is not going to be retro-fitted to the Nimitz class of carriers because the two nuclear reactors installed on these ships do not provide enough energy. So the chances of a gas-turbine/generator system providing enough juice seems unlikely.

They're being installed on the Gerald R Ford class, which have an uprated generating capacity compared to the Nimitzs.

Peter Gathercole Silver badge

The biggest problem with the QEs is that they are powered by gas turbine/electric propulsion.

They could not have had steam catapults fitted, because there is no steam plant to generate steam (unlike the US nuclear carriers). And the electro-magnetic plane flingers were not available when the ships were first planned, need huge amounts of electrical power, and are bloody expensive (being current US technology).

If the design had been built around four or six Astute sub. reactors, they would have had either steam or surplus electrical power. But some bleeding hearts had decided that the UK should not have nuclear powered surface warships. So we have ships with limited range, reduced accommodation for crew, provisions, weapons and aircraft because of the need to have J-fule bunkers and fresh water tanks.

IMHO, they are seriously compromised ships, along with their built-around a-single-weapon system escorts, the Type 45 destroyers. I though the navy had learned this lesson after the County class and Type 81 large destroyers of the 1960's.

Peter Gathercole Silver badge

Re: Saving ships @GrumpyKiwi

We scrapped nearly all of the WWII battleships very soon after the war. All of the surviving QE class were knackered after bearing the brunt of the Mediterranean conflict, with Warspite, Valiant and Queen Elizabeth all seriously damaged at various times by mines and consequently not suitable for preservation.

The Revenge class were already in reserve at the close of WWII, because they were very slow and had fueling issues (they were built as oil/coal fired, and did not have the oil bunkerage for operations outside of the North Sea).

Nelson and Rodney were... odd. Very atypical, and would not really have been representative.

Keeping a King George V should have been possible, but it was again, these were paid off into reserve or used for training duties very quickly after the war.

Although it was not a WWII battleship, Vanguard, as the last operational British battleship (and probably the best looking example of British big-ship design - being closest to the canceled Lion class in design) would have been an excellent choice, but preservation efforts failed because it would have been so expensive (and the government in the '60s were desperately trying to cut the cost of defence).

But Belfast is not such a bad remnant. In terms of size, being the same length as the smaller British battleships, is reasonably representative of wartime cruiser design (being a stretched, or improved Southampton class), had been active in WWII and was in the best condition of all of the remaining available large ships. As such, she gives some impression of size and conditions for a large number of British ships.

Peter Gathercole Silver badge

Re: I would actually argue..

I stand suitably corrected. I was trying to avoid calling them Hawker Siddley Buccaneers.

Peter Gathercole Silver badge

I would actually argue..

..that carrier based fast jet operations actually ended with the decommissioning of the Audacious class R09 HMS Ark Royal in 1979.

Compared to that venerable old lady, the Illustrious class were relatively small.

The Illustrious class were designed with a full-load displacement of around 18,000 tons. R09 Ark Royal was designed at 35,000 tons, and evolved to over 40,000 tons full load, over twice the displacement. In addition, she embarked F-4K Phantoms, the last supersonic aircraft to fly from a British carrier, and Bristol Buccaneers.

The Mighty Ark was a hold-over from WWII armored carrier design, and by the time she was decommissioned was completely worn out through a long life and incremental modifications. In the '60s, there was a grand project to build CVA-01, a suitable replacement, but this was canned by a Labour government, who deemed that the Navy no longer needed ship based fixed-wing aircraft. As a result, the Illustrious class then on the drawing board, were re-christened Through-Deck Cruisers, and were only intended to fly helicopters for anti-submarine purposes.

It was only after trials of P1107 Kestrels and early Harriers on R09 Ark Royal and HMS Bulwark (a light fleet carrier converted to operate helicopters) proved that they could be operated from smaller ships without CATOBAR facilities that it became feasible to actually use the Illustrious class as light fleet carriers. They were completed with ski-jumps to assist takeoff, and I actually saw Ark Royal and Illustrious being fitted out on the Tyne in the '80s.

But as the Sea Harrier was subsonic, I maintain that it did not really count as a fast jet.

Google broke its own cloud by doing two updates at once

Peter Gathercole Silver badge

Re: Still planning to have these clown in your infrastructure? @Adam

... but with your own infrastructure, you can plan your own changes to avoid conflicts, and also make sure that changes don't happen during your critical business periods. You know how good the people doing the work are, you have the responsibility for hiring them!

You're also free to analyze what happened to whatever depth you want post incident, assigning the correct blame and improving future work without relying on partial, face-saving reports and promises from a service provider whose interests are not served by making the full details of their mistakes public.

I know Google have very skilled people in some places in their organization. But you think they're the people actually doing the day-to-day grind? They're probably mostly in design/third level escalation, The people doing the grunt work will be like every other company, the cheapest they can get to fill the roles.

And when it comes to suing service providers, trying to take legal action while trying to recover a business, especially against companies that employ good lawyers, is the last thing a company would need. Even if you won (after the appeal, of course) chances are the financial gain might be too late to save a business.

I would be pretty certain that for most tiers of cloud that people are using, the terms of the contracts and their resultant SLAs that the likes of Google provide will not have clauses that provide significant redress. And did I say that they employ good lawyers? And it might even be difficult to identify which legal jurisdiction any case should be brought!

More gums than Jaws: Greenland super-sharks live past 400 years old

Peter Gathercole Silver badge

Re: New dating techniques

"...it reproduces very slowly".

Sounds like a new dating technique is required. Catch,com, maybe, or possibly eHARMony. How about EliteSinglePreditors.co.uk? GreenlandSingles.com?

'I found the intern curled up on the data centre floor moaning'

Peter Gathercole Silver badge

Re: Locked in a machine room


Neither the company I was working for at the time, nor the one in London was IBM!

Peter Gathercole Silver badge

Early hours of the morning

While I was working in the support center of one of the major IT suppliers, I was awakened by the pager in the small hours of the morning.

When responding to the call center, the operative said "I hope you don't mind, but the customer has asked who was on standby before they placed the call". Obviously, the customer had different opinions about the people who were on the standby rota. Apparently, I had passed their vetting, and I called them up.

I then spent about 20 minutes listening to the details of the problem, interspersing a few appropriate noises. At the end of this, the customer said "OK, I think I know what I need to do now".

I said, "But I've not given you any help or assistance", to which she replied "No, but you let me describe the problem to someone who would understand it, which has allowed me to think it through".

I said that I would be available if she needed to call me again, and she thanked me, and hung up. I did not hear from her again that night, so her solution must have worked. Easiest call-out I ever had.

I came across her again several years later after I had started contracting. Apparently, my CV passed across her desk for a role they were trying to recruit, and she remembered me (not just form the call described above, but from other support calls). I got a rather bemused agent on the phone, who said that she's called them unprompted to say that the role was mine if I wanted it, without any interview, and at the highest rate he'd been told to recruit at.

Unfortunately, it was London based, and I was not looking a role in the Capital. Still, it's nice to be appreciated sometimes.

Peter Gathercole Silver badge

Locked in a machine room

I was working for a large telecomms company that had outsourced development of a call billing package to a supposedly premier UK based software company. The third party were provided with a rather obscure proprietary UNIX system (made by the telecoms company) to develop on, because that was the platform the system would run from.

They proceeded to break it, and I was told to go up to London to investigate and rebuild the system for them.

After I arrived and booked in, I was taken and left (unescorted!) in a machine room in a building just off the Tottenham Court Road in London, where I discovered that they had extended the /usr filesystem over the swap space (this was when you had sys-gen'd disk partitions - it was some time ago), and diligently sorted the disk partition table, and restored the filesystem from the backup tape.

When I finished, I looked around. The machine room had no 'phones in it, and it was before mobiles were common. The door could only be opened with an electronic tag. There was nobody in listening distance of the door, no matter how much I pounded it. The only system I could log into was the system I had fixed, and there were no other users logged on.

Something like 4 hours after I had finished, someone thought to look in on me. I was cold, thirsty, and really needed the toilet. I had toyed with the idea of the fire alarm, or turning random machines off to try and attract attention, and also considered lifting a floor tile and leaving a 'present' under the suspended floor.

I cannot remember whether I received any form of apology. All I wanted to do was get out of there.

And you can guess how I felt when just two weeks later, I got a call saying that this 'premier' software company had done exactly the same thing again (after being given an explicit report of what they'd done wrong previously), and could I go up and fix it....

Vodafone bins line rental charges as it moves onto TalkTalk's turf

Peter Gathercole Silver badge

Line rental

I've never understood why there is such a huge complaint about paying line rental.

The cost of the copper/fibre infrastructure, and the ISP end equipment has to be paid for somehow, whether it be by a broken out line rental, or by having it wrapped into the monthly package.

The cost of the voice component of the phone line is minimal in this day and age, and probably most data only lines still have the hardware to do analogue calls anyway. Just think, I can buy outright a whole mobile phone, with radios, batteries, displays and everything for £10. How much does the A-D and D-A converters cost at the exchange? And the amount of digital information generated by an analogue voice call is trivial.

This means that it is extremely unlikely that a data only line with the infrastructure costs wrapped into the costs will work out any cheaper.

I'm all for making the eventual costs more transparent, however.

Biting the hand that feeds IT © 1998–2019