Posts by Peter Gathercole

Re: Damn I'm old!

I'd already got over my arcade phase by the time this came along.

Whilst I remember the first appearance of Space Invaders, Galaxians and Asteroids when they were new, and even the cabinet Pong and whatever the two player racing car game was called (was it Skida?) it was the arrival of Missile Command and Battle Zone that mostly got me over it, especially after I came across someone who clocked my regular Missile Command system (where I was normally first or second on the score board) in front of my eyes! This guy didn't even finish the game. He walked away with cities still stacked across the screen.

I later tried to master Tempest, and got some success (able to get on the score board at the pub in South Shields I went to just to play Tempest), I was in reality merely mediocre at it.

But I was crap at Defender and Centipede, so probably would not have been able to play Robotron very well.

A decent real pinball machine will still get my attention, however.

Re: And if you do need to keep them both live @me

I replied to the wrong comment. This should have bee about the next comment off the main one! Read on, and it will make sense.

Re: Poland

Poland is not a good comparison. Within human memory, Poland's history includes occupation by Germany, an authoritarian state, and association with the USSR, an authoritarian regime. Both of these situations mandated having identity documents.

Implementing a more modern alternative just seemed like normal for the Poles.

The UK has never previously had a permanent ID system, except during periods of war. It's not a normal situation for us, and trying to impose such a system is foreign. I think a lot of people around the world don't understand how this would be something new and unwelcome in the UK.

I personally can see the benefits of a government backed trusted ID system, but it must not be mandatory, and it must not do anything more than allowing people to bypass the completely stupid and untrustworthy methods we use now. But there is no way that once introduced, it would not be used as an instrument of control over the people.

Re: Wait until he finds out...

The relay is an ordinary common or garden low cost, low voltage relay. They are mechanical devices which have operational duty cycles measured in 10's or maybe 100's of thousands of operations. There was no expectation that they would last forever, but for the job they were procured for, they were fine.

Any mechanical system will fail if operated repeatedly to a point beyond it's design lifetime. A higher priced device would have lasted longer, but would have cost more. It's just that being in a computer, ir could be abused more quickly.

Re: How accurate to busses have to be?

MSF has sub-second accuracy, and the edge(s) of the final synchronising pulse is probably accurate to a millisecond or less (I used to know, memory is failing). The whole time 'packet' takes a minute to transmit, true, but it is the final pulse that actually signals the point in time to sync to.

A correctly configured MSF radio clock should wait for three consecutive good time packets before setting their internal clock.

Re: Downvote - really?

No. I think that one click on the opposite arrow cancels the original. To reverse it, you have to then press a second time.

Re: Prompts?

I think you should be asking whether the results are reproduceable, not that they are deterministic.

If we had total undetstanding of the universe, everything will be deterministic. Until then....

Re: After Trump's complaints about foreign states stealing American IP

I think the US military are exempt from those regulations.

Re: T60?

Huh. Ain't that a shock.

The T60 I'm looking at has a Core Duo T2400. That's a 32 bit processor! No wonder it tops out at no more that 3GB. I always thought it was a 64 bit system. Must check the one my wife is using.

I wonder whether the MoBo is capable of taking a Socket 479 Core 2 Duo. Or even whether it is worth doing it (although it does have the 1400x1050 screen, so that may be a reason).

Re: 2028 sounds like a death sentence for Hubble

Did you read about the failing gyroscopes? Without any functioning, it becomes impossible to aim Hubble. Not that much use after that.

If the shuttle was still working, it may have been worth replacing the gyros, but although Crew Dragon is rated for EVA, any significant rebuild in orbit from a Crew Dragon would be difficult.

I do of course know about Voyager 1 and 2, but even these will stop functioning once their power source drops below the minimum threshold to run the comms and stop them freezing. Like the Voyagers, Hubble has already exceeded it's design lifetime.

All things must pass.

Re: COBOL is easy...

Whenever I consider discussions like this, I ask myself the following question.

If you use AI to re-write the code, do you then ditch the Cobol, and continue supporting the AI generated code? Or do you keep the Cobol around, and everytime you need to change the code, make the changes in Cobol and then re-translate?

Answers to this dilemma depend very clearly on two things. Firstly, is the generated code maintainable? and secondly, does the generated code adequately cover all of the system requirements and all of the changes to them that have happened over time?

My worry is that if a business decides to ditch the Cobol, and forward fix the AI generated code, who understands the original requirements, and are those requirements sufficiently clear in the resultant code to not introduce a whole raft of corner cases going forward?

I can see a scenario where if the code base switches to the new language, a problem, or even a modification becomes impossible to work into the code, because nobody really understands the generated code to make further changes, and the rerquired changes may be too subtle to allow an AI loose to rework it.

Many of the businesses that still buy mainframes do so because they don't want to answer the above questions, because if they get the answers wrong, the cost and reputational risk is more than the cost of keeping the applications running on mainframes.

Re: And probably delicious, radiation levels be damned.

They effectively sand it off before it is cured. And it would not surprise me if the resultant sludge makes it into budget pork sausages, or anything that needed some natural flavouring.

For larger cuts of pork, they can cut the skin off and deep fry to make pork scratchings

You could stop it making the rasher curl by cutting the rind several times on the rasher (or score it before the rashers are cut). It then fries up forming extra crunchy bits on the rasher.

One common use for unwanted bacon rind cut off before cooking used to be to put it out on bird tables in winter.

Re: Nostalgia meets modern hardware

Just be a little careful. Although the VT100 is regarded as a single terminal, it was actually a family of terminals.

The VT100 was the original, built with several boards (ever wondered why the screen unit was asymmetrical, there was a card cage on the RHS) , and had options (such as the Advanced Video Option AVO) and some options regarding the amount of memory, which determined whether the 132 column mode could use all 24 lines.

What most people refer to as a VT100 was actually a VT102, which was a re-implementation on a single board with new firmware, but no upgrades. This is what most vt100 termcap/terminfo entries implement.

As a consequence, there are minor differences between the behaviour, and if you look at either /etc/termcap, or the various terminfo files, there are different entries for different models and setups.

I thought I was well versed in termcap/terminfo and the vt100 entries, but the differences tripped me (and bash) up when I was building the Peter Hizalev version of Geoff's VT100 kit, I found out that the original vt100 and the Hizalev version of the code is accurate to an original vt100, and this has one unexpected feature, in that a NULL (ASCII x00) is translated to a space by the VT100 (documented in the vt100 operators manual or programming card. But the GNU implementation of terminfo and it's vt100 entry used NULL as the timing padding character, which the terminfo entry will use at higher serial speeds.

Later models of the vt family do not print a space when a NULL is received, so don't suffer the same problem, and this is not seen when an xterm or xterm derived emulator, as again, NULLs do nothing, and is not a huge problem most of the time (although spurious spaces in the output from some commands can bee seen), but bash does cause a problem.

In a traditional old-school UNIX system with the Bourne shell or Ksh, the handling of rubout is left to the TTY line discipline, but bash does something different. It puts the terminal into a mode that turns off the line discipline (mainly so you can use the arrow keys for command line editing), and handles rubout itself. Instead of doing a "backspace-space-backspace" sequence, actually does something like "cub1" followed by a "el", which deletes to the end of line, but at high speeds will actually cause padding characters to be generated, which on an original VT100 or close-to-perfect emulation, actually causes space characters in the line you're typing when deleting characters.

This does not happen if you are using a shell other than bash (I have checked with the official ksh and some other shells), and also does not happen if you either modify the terminfo entry to include the capability "npc", either in the global terminfo for vt100, or in a local terminfo like .terminfo or pointed to by the TERMINFO environment variable.

I really ought to document this as a bug in the ncurses-term package, or possibly bash. It probably won't affect anybody outside of the retro community, as how many people put VT100 terminals on serial lines to modern Linux systems?

Re: Gullible!

That's clearly a tall story. Everybody knows that the fastest thing on the road is an unmarked white van!

Any van, as long as it's white.

The exact colour and texture of the cases of BBC micro changed over time, both as built, and as they aged.

The very early ones had a quite rough texture, and were quite yellow/beige in colour. After about a year, as Acorn scaled up production and included manufacturing in the far east, the cases were changed, becoming slightly more robust (although still not strong enough to put a monitor on), a smoother texture and a more cream colour than beige.

Over time, especially if subjected to sunlight, the older cases darkened quite substantially, in some cases almost going brown (like my Issue 3 motherboard system). You can get them a little brighter with IPA, but all this really does is get any grime of of the surface. I haven't (and won't) tried retrobrighting the cases of my systems.

From what I remember, the colour of the cream cases was more stable, and if anything, went lighter rather than darker with age (my issue 7 BBC micro that I inherited from my father is currently in an inaccessible storage area, so I can't check).

I do not know why the beige colour was choosen, but I would say that it was designed just after the 1970's when oranges and browns were common colours for home decor and real wood was more common, so maybe it was just designed to blend in to the home. It didn't seem out of place at the time. Plus, there was much more diversity across the board when it came to the style of home computers. The PC has a lot to answer for when it came to homogenising computer design.

Re: From available evidence below...

It needed to be this shape for coin verifiers and sorters in vending and other machines. These normally work by measuring various properties like weight, diameter, thickness and sometimes by magnetic signature. But most start with diameter and then check weight, so a constant diameter shape is essential for coins.

Re: Connect TRITON

At the IBM AIX Systems Support Centre in the UK (aka Call-AIX) in the 1990's, the AIX servers were all named after types of whales and other cetaceans, the AIX/PS2 desktop workstations were named after cartoon characters (mine was Foghorn), and the slightly later X-Stations were named after American Indian peoples (I think mine was Algonquin, although I could be wrong).

When we later put RS/6000 model 43Ps on some desks, we started using professions, although I chose Magician (not strictly a profession, I suppose), but as I started that naming convention, I got to choose! I don't think anyone chose Hooker!

The TS/Channel Support group upstairs named their systems after Greek gods.

Are you sure? BlackbeltBarrister said on a short video that if they have suspicion that you've been involved in a crime, such as if they've identified you as someone they're interested in for a previous crime (flagged up by surveillance video maybe), that Section 50 of the Police Reform Act 2002 can be used, and under that section, refusing to provide a name and address, or giving false information can lead to arrest.

Looking at that section online, that seems to be mainly for suspected public order offences, but BBB states that it may be for other already known offences. And like so often, if you verbally abuse an officer, then that can automatically trigger this section, landing you in hot water.

I agree that if they have no cause, then they should not be able to demand identification, but in the case of being flagged, albeit incorrectly, by a surveillance system as someone they're looking for could well be considered a just cause. They need to know whether you're the person they're looking for, and if you refuse to identify yourself, they might just assume that you are who they are looking for.

I am not a lawyer. If this affects you, don't rely on this, and consult proper legal representation.

Re: A very healthy sign

Googling the risks, age becomes a meaningful contributory factor according to an National Institute of Health document I found at https://pmc.ncbi.nlm.nih.gov/articles/PMC12294447

Re: "we never loaned any of our tools to any of the non-IT staff ever again"

You know, even marketing people have lives outside of work, which may involve things like car maintenance.

And there was a time when even academic leaning schools had wood and metal working on the curriculum, so anybody coming out of school before I guess about 1980 would have come across basic tools and their use. At my senior school, I learned how to safely use a lathe, milling machine, mechanical saws, pillar drills, as well as most hand tools. And I was never going to go into a profession that would need them.

So, I never really used the skills with powered workshop tools in my work life much beyond drills and handheld tools, but I do know how to use hand tools in non-work related endeavours (but I hasten to add, I am not a marketing person).

... except for the ones that come from the Isle of Wight.

Re: Home Office targets small boat crossings

Reference please.

I also thought I had seen that the boats are deflated by being slashed to make them less useful, but I don't immediately have a reference for that.

In case you hadn't noticed, everything has an ecological side to it with UK Government issues. They have to look at that side of things by law.

Re: Not surprised

Where I live, we still have a chinese takaway that does not accept cards, and there is another takeaway that prefers pick-up orders to be paid by cash. I could use other takaways if I wanted to, but I'm happy with the food, and don't mind paying by cash.

Whether this is because they're avoiding bank charges (although I know that cash handling at banks is also costly), or accountability for tax reasons I cannot say, but they've both been running like this for long enough that I'm sure that HMRC would have checked on them if there was a suspicion that they were avoiding taxes.

I do also wonder whether for smaller sales that the fixed transaction charges don't wipe out a shop's profit. I knew one news agent who always complained that the per-transaction charges by their card merchant were too much on small transactions, but I know they've switched to another now, and he is less worried about transaction charges.

Please also remember that some people use cash as a budgeting aid. What they don't have in their hands can't be spent. It's been a way of managing money for as long as there's been money!

Re: How to destroy a respected brand, 101

Sharp is already a TCL brand.

Re: Why unixifying Windows is a very bad idea

By 1990, most commercial UNIXes had ksh which implemented command history, even if it was not the default shell for the root user. AIX included it as standard from the first release for the RS/6000 at AIX 3.1 in 1990, and I know that HP/UX and Digital UNIX had it as well, but I don't exactly know when they first included it.

Ksh was around since 1983 (at least inside AT&T) and the version that was used for most UNIXes was ksh88 (the 88 relates to the year it was available), but some actually took ksh86 as an experimental tool.

Although I wouldn't recommend it (I could never get on with it as an interactive shell), csh that came with BSD had shell history from much earlier than this, although it was less like a scrollable history, and more like a list that you could select a command from by number.

While it is obvious that passing objects between commands is clearly more powerful than plain text and pipes, there is something much more intuitive parsing something that you can see, whereas using objects require you to know the components of the object before you can process them. I'm sure it's an experience thing, but to a casual PowerShell scripter, that is something I haven't learned.

The UNIX shells were never intended as general programming languages. They were evolved job control languages, extended to provide interactive capabilities and with some programming flow control features so you could use the same tools to generate simple scripts for automation. As UNIX shells have evolved, more programming features were added, but objects were beyond what the shells could handle without losing backward compatibility. You're right when you say that UNIX shells hark back to the last century, but what has kept them relevant was the stability and commonality of the shell's adoption. The first major change in UNIX-like shells were the extensions added to bash when Linux became common. Bash has many more features than are specified in the Posix shell!

But most UNIX systems administrators got used to augmenting the features of the shell by using tools like awk to do a lot of the heavy lifting.

Other shells were tried. I quite liked REXX from VM/CMS and OS/2 which had some basic object handling, but after there were some major security issues in the OS/2 and AIX versions, it never caught on.

Trans polar routes have been understood by the US for decades. That is why they used to have bases in Greenland in the '50s and '60s. They couldn't intercept the missiles back then, but the extra thousand or so miles could give extra warning for launching retaliation, and scurrying to bunkers. But they probably wouldn't move troops or equipment across the Greenland, and wouldn't stage them through Greenland because of the (current) lack of good air bases there.

The person who probably doesn't understand it is President Trump, bearing in mind how he misunderstands the Mercator projection and the real size of Greenland!

If the US was intending to move troops into Europe in a hurry now, they'd probably want to stage them through the UK by air, but if they had warning, move the bulk by ship across the Atlantic, landing either in France or maybe Spain, but Spain is a bit too far west. If the situation was not too dangerous, you might choose the Netherlands or Belgium, but bringing troop transports through the English Channel, or round the North Sea would be a killing zone if the conflict had started.

But going back to the Cold War era, the US had a standing army in West Germany of several hundred thousand soldiers, and other NATO countries had armies there as well, because before long range missiles, the threat was that the Soviet Union would try a mechanised blitzkrieg across the German plains and into France and the Low Countries after neutralizing western airbases. There wouldn't have been a need to fight across Europe, the conflict would start in Germany where there was already a standing army.

There still appear to be about 35,000 US military personnel in Germany on 40 bases. I don't know what Trump thinks about this.

Re: No new PC for you!

That relies on you having a device that allows you to get to said virtual PC, and I don't think that a PC running Windows 10 or earlier will hack it (well, it would, but runs an unsupported OS, so...)

I'm still surprised that nobody in the Virtual PC business has produced a secure, minimal OS (something a bit like ChromeOS, or a maybe a dedicated bootable Linux USB) to replace an older Windows installation on existing PCs to get access to cloud based Virtual PCs. Would be a real business enabler in my view, but wouldn't give the PC vendors the new business boost they want.