Posts by Martin Gregorie

Re: Pets? PETS?

.. and some of them enjoy sliding down corrugated iron roofs at night.

Re: Back in the early 90's

... and I thought the WIMP acronym came from Xerox PARC, along with the first graphical windowing scheme, as opposed to partitioning a text screen into separately scrollable areas and using text-only pop-up boxes as Word For DOS did.

FWIW, the first graphical windowing system I saw was an ICL PERQ, which sported a portrait-orirnted black and white screen. That was at an internal Future Technology presentation run by the BBC in 1980. The PERQ was nice, but I was totally blown away by the NTSC(?) Sound Surround system - capable of driving up to 8 channels, so it could provide any soundscape from stereo to a full 3D experience with nice features like varying the ambiance as you used the control that shifted your audio position in a concert hall toward or away from the orchestra. It completely blew the socks off Stereo FM as well as making the Sony surround sound system an obvious also-ran. Sounded wonderful on headphones too.

The second graphical system I remember seeing was a Xerox Alto system in 1984, followed by getting my hands on an Apple Lisa later the same year.

Megastrutures can save astronomy but microsats can't

However, if orbital megastructures are built they will have some saving graces:

1) There are unlikely to be many of them

2) If they are REAL megastructures the astronomiers can live in them and put their telescopes on them

3) Said megastructures will almost certainly in geostationary orbits because they'll cause major damage if their orbits should ever decay and they'll be so big and heavy that regular orbit-boosting like ISS requires would be neither practical or affordable

For decent sized examples of orbital megastructures, see Neal Stephenson's "Seven Eves"

Re: "[integrated combustible engine] WRONG"

About the last thing I'd want in any vehicle is a >>combustible<< engine.

Combustion should remain inside an internal combustion engine, which is itself incombustible. It this condition is not met, an external conflagration is the expected outcome.

Re: Fun with punch cards

My first exposure to assembly language programming involved an ICL 1901 and punch cards because, apart from the operator's control typewriter (a KSR 33 teletype) its only input device was a card reader.

We also had 12 key hand punches which we used the transcribe our program from coding sheets to card decks.

Bit of a shock to the system, since I'd previously learned to write Algol 60 for the university's Elliott 503. The Elliott used paper tape for input, which we prepared and edited on a Friden Flexowriter - a wonderful device.

Re: I predict

Same here, except that my email store is a PostgreSQL database.

There is a good reason for that: I also run SpamAssassin locally. It queries the email database, and whitelists email from any address that I've previously sent mail to. This lets me safely use a set of fairly aggressive antispam rules.

The final safeguard is that all spam gets quarantined for a week. Each morning I get a daily report of any spam that was quarantined during the previous day so I can easily retrieve anything that was incorrectly treated as spam: this is quite rare and is almost always messages from a new service I've just signed up to offline or over the phone.

Subarashī seems like a good match with Zuk's appparent megalomania.

Re: No print? No buy.

Better: stick to B&W laser printers, especially if you're a low volume printer user.

Laser printers don't have:

- a built-in unreplacable roll of head-wiping ribbon (thank-you Epson Stylus)

- ink to dry out and go gummy long before the cartridge is empty (all inkjets)

The last printer I bought is a black and white HP laserjet, purchased for the above reasons 4 years ago and still working without complaint on the smaller-than-standard 'trial' cartridge it came with.

Re: Weight of passengers

Standard procedure in a glider, where those on-board ARE the ballast needed to put the balance point (Centre of Gravity, CG) in the right place, so everybody onboard *must* know their weight while wearing a parachute and a W&B check is done before flight.

If you're flying a single seater its up to you to know your weight with parachute, to check it against the W&B table before flying, and add additional nose weight if you're below the minimum pilot weight.

If the glider is a two seater the same applies with the weight of the person in the rear seat included only in the total pilot weight: only the weight of the bod in front is used to check whether the glider is balanced. BTW, the reason the instructor always sits in the rear seat is because that is on or close to the CG, so when a new pilot is sent solo the instructor can get out without affecting the glider's balance.

Re: Hmmm....

With the wonders of MCAS floating, like scum, to the surface yet again I'm left wondering how long it will take for NASA to realize that Boeing's Starliner, given its history of poorly tested software, faulty fuel control valves and late delivery, may not be the safest or cheapest way for astronauts to commute to the ISS.

Re: I'm confused...

What you're suggesting should work well, seeing that is pretty much what I've been doing with Fedora Linux updates for the last 15 years or so:

- make a backup with rsync onto a USB HDD

- disconnect the USB drive

- run dnf to download and apply updates

- reboot

I've not had any problems so far.

Re: We've all done it

Back in the early 70s I had a one day round trip from Wellington to Auckland just to convince a George 3 sysadmin that his system config was wrong, explain why, and fix said config. Annoying, because I'd already sent them what was substantially the same fix and had it ignored. Total working time onsite: under an hour, plus a discussion about why it was really silly (and slow) to run tape sorts using simulated mag tapes in the filestore when a perfectly good disk-based sort was a one of the standard utility programs.

On the other hand, it did get me a a couple of rides in one of the, then shiny new, B737s which had replaced Viscounts on that route. so it wasn't all bad.

Yes.

If the device can't explain how it arrived at the answer it just gave you, then it obviously isn't intelligent.

Re: speaking of Ekronoplanes lately

so how did ground effect alter characteristics of plane.

Mostly by reduced drag if you're down in ground effect, which is when you're less that half your wingspan above the ground.

A lot of glider pilots used to use ground effect as a form of competition finish if the fields next to the airfield were flat.

I've done it in an SZD Junior (single seat training glider) at the end of a day's flying. We'd been launching from the far end of our airfield from the hangar and clubhouse, so at the end of the day i offered to fly the Junior home. I took a winch launch, and flew a normal circuit as if I was going to land where I'd taken off, but instead left the airbrakes shut and flew the approach at 70 rather than 55 kts, flattening out at 15-20 feet. I stayed at that height for 680m along the main runway. At that point I'd only lost 10 kts of airspeed, popped the airbrakes and touched down 200m further on, rolling to a stop near the hangar.

A Junior has a claimed glide ratio of 36:1 at 45 kts, so at typical flying height it would have lost just over 60 feet in flying the same distance at 45 kts, or 150ft at 70kts: Juniors are draggy little beasts and airframe drag increases as the square of the flying speed.

Re: H2

.... and don't forget that reforming natural gas, i.e. producing so-called 'Blue Hydrogen', produces a shedload of CO2, virtually all of which is dumped into the atmosphere because that's the cheapest way of disposing of it - never mind what all that waste CO2 is doing to global warming.

The bottom line is that, unless all the carbon that has to be removed from natural gas (methane, CH4) to produce 'Blue Hydrogen' is permanently buried, running a vehicle on 'Blue Hydrogen' emits MORE CO2 into the atmosphere than using petrol or methane would, because of both the energy inputs to the reforming process and the fact that the waste carbon emerges as carbon dioxide, CO2, which needs still more energy to be converted into a solid that can be buried or to compress it sufficiently to be re-injected into the oil well it originally came out of.

But, don't just take my word for it, read this:

https://arstechnica.com/science/2021/08/blue-hydrogen-pushed-by-gas-companies-harms-climate-more-than-coal-study-says/

Re: Awaiting the day

And you LIKE the sound of delivery drones flying past your window every day?

IIRC the first Amazon drone delivery trial in a low rise development got cancelled because noise complaints from neighbours.

Re: Infosys has over 750 staffers working on the project

With 750 staffers yanked in to fix it, something tells me that the project is irretrievably stuffed and is unlikely to be fixed any time soon, if ever, simply because the team is far too large.

Back in the mid 70s I was one of the 80 freelancers pulled in the fix the Naval Dockyard Project, which was a several million over budget and 3 years late at the time, mainly because its development methodology was straight out of "Those Magnificent Men in Their Flying Machines". You know it: "There is nothing a German Officer cannot do" except in this case the 'German Officers' were a group of civil servants who had mostly never seen a computer before they were sent on systems design and COBOL programming courses.

We turned the Naval Dockyard Project round and had it working in 18 months thanks largely to the first class group of project managers from Dataskill who wholly replaced the MOD management team, but, including the best of the MOD programmers, who were retained, I doubt there were ever more than 120 analysts and programmers in the team. Any more and the project would have collapsed from sheer unwieldiness.

The only other project I've been on that approached the MOD job in staff size ran for a similar time before having its plug pulled without producing anything useful.

Re: Wrong nail?

If you're writing a lot of Java windows that are essentially rows and columns of labels,data values and a few Buttons, i.e. similar to the type of screen layout that used to be common on greenscreen mainframe terminals or using 4GL packages to program PCs, try using the se.datadosen.RiverLayout layout manager.

If RiverLayout is a good fit with what you're trying to do, you'll find it makes laying out a Panel much easier than it would be with a standard Swing layout manager.

Re: It's most likely dry mud from silt settling out.

Yes indeed, and mud at the bottom of a lake on a low gravity planet that dried out in an atmosphere thats already getting a bit too thin to blow dust about isn't likely to get compacted into something very strong.

Consequently, when some robothing tries to cut cores from the lake-bed using a hollow drill fitted with teeth big and sharp enough to cut through hard rock, it follows that the core is likely to disintegrate into something rather like fine sand rather than just standing there.

I do wonder if they tried using a Perseverance test vehicle to drill holes in dried-out lake-beds in the Mojave or some other, even drier, part of the south-west USA or central Australia. But surely they must have tried that. . . mustn't they?

Re: I think Audacity has had it, then.

It certainly looks as if Audacity maintenance has ground to a halt - Fedora, and so other RH distros as well, are already at least one release behind Audacity's current state, IOW Nobody at Muse can be arsed to do a proper job of distributing binary updates.

AFAICT work on fixing open bugs has stopped too.

Re: Gopher

Sure, when Gopher was a thing and "The Whole Internet Guide" was a must-have in your technical book collection, but I stopped using it when the first web browsers appeared.

I long ago lost or binned my copy of "The Whole Internet Guide", but my copy of "UNIX Systems Programming for SVR4", which is a similar age, is still useful and contains stuff that's relevant for writing C on a Linux box.

Any planet can migrate when fitted with enough spindizzys: their use isn't limited to small objects such as cities.

Spot on!

Slightly off-topic for this thread, but if you want to know who did what and when between 1895 and 1909 when the basics of controlled flight were being worked out, read "Kill Devil Hill" by Harry Combs - he was the Learjet chief test pilot and, so by far the best qualified author to have written about the early days of flight. This book shows that, while its not hard to make a machine fly under its own power, making it fly stably is quite a bit harder and making it both easily controllable and stable is at least an order of magnitude harder still. It describes the thought and experimentation that went into achieving all this. Finally, not getting killed while doing all that takes exceptional engineering and problem analysis skills.

That Harry Combs manages to cover all that in a pretty readable book shows that he is a pretty good author too. While his book is primarily about the Wrights, it also covers all all the work done by other pioneers in Europe, UK and North America

.

Re: Next logo design

Never mind the logo - that's just window dressing.

What about doing something positive, such as only using:

- recyclable or compostable packaging materials

- couriers with electric delivery vehicles.

Re: The future is coming

Exactly so. Including a low wattage conductor pair within a fibre cable is currently known and available technology. There are no extra development costs since the capacity to provide power to a subscriber phone is already a standard part of every exchange.

Indeed, having a copper pair in the cable should even save on trouble-shooting costs by keeping cable-break testing and location as simple as it is today: just use a bridge megger to locate the fault. IANAE but it seems that fault location in fibre optic cable isn't so easy, as it depends on the break reflecting a strong signal.

Re: Clapham?

Don't ever think that comms lines follow anything that even faintly resembles the shortest route from A to B. They don't.

Back in 1980, when I was on contract at the BBC, our development team was in an office in Cavendish Square and the main BBC computer Centre was about 800m west of Shepherds Bush. We lost all contact with our mainframe for a couple of days when some berk with a digger took out the cable - at Acton!

For those who don't know London, Acton and Shepherds Bush are both west of Cavendish Square - but Acton is about twice as far away from Cavendish Square as Shepherds Bush and almost directly in line with it, so that cable not only appeared to double back on itself, but was at least three times as long as it needed to be.

Re: Why?

Pretty much any non-trivial project will be over budget, late and buggy if the project manager and the technical people who design it, manage development and run unit, integration and acceptance testing are not part of the bid team and their inputs are ignored rather then used to determine pricing, staffing and timescales.

If the project is an in-house one, then a senior user should be part of the design and management team..

Unfortunately, civil servants and other people running governments and the like never staff projects this way and have no clue why it would be a good idea to do so.

Re: spraying pink spunk doesn't solve the problem

I think you're assuming the "snot cloud" is closer in consistency to superglue than it needs to be.

Not at all: what I was thinking of was something more like the epoxy-based gunk that's used to seal chimney-stack flashings etc and make sure they stay stuck down. That looks like thick, grey gravy and contains a lot of chopped glass fibre.

To do what the video shows, you need two components: something fibrous to get wound round the drone and its rotors and some form of contact glue to instantly make everything into a single mass. Superglue doesn't actually stick that fast, but contact cements do.

So, keep the glass fibre strands but replace the epoxy gravy with some variety of contact cement in a fast-evaporating carrier and there's your anti-drone system: as the gloop-cloud expands, the carrier evaporates and the fibres become sticky rather than slippery, so they stick to the rotors and each other until the initially spread-out mess gets thoroughly tangled together around the rotors and drone body, at which point the rotors can no longer spin and the drone drops out of the sky as the video showed.

I think this explains why the almost-invisible cloud didn't do much at first, but then quite rapidly congealed into a dark mass stuck to the falling drone at about the time that the rotors stopped.

Re: Nutters

Boeing's software typically includes some NASA-like 'Paranoid Programming Practices'.

Yeah, like running the flight-critical 737MAX MCAS system off a single computer connected to a single Angle Of Attack sensor. I know that the MAX carried two MCAS computers, each with its own AOA sensor BUT:

- only one MCAS computer and AOA sensor was active during a flight.

- each MCAS computer could only access its own AOA sensor

So, no cross-checking ability provided and the small red MCAS malfunction indicator was an optional paid-for extra, so not generally fitted because bean-counters.

IOW, Boeing was applying almost exactly the opposite of 'NASA Paranoid Programming Practices' as well as ignoring generally accepted aviation principles of multiple redundancy for known unreliable but vital sensors.

FWIW the aircraft I fly have no redundant flight instruments. However, part of the pre-solo training curriculum is to fly a circuit and land without either ASI or altimeter. I have subsequently done just that, but this time for real, when the ASI failed.

Re: Not new

ICL were doing something similar back in 1973 when a 60MB removable disk was a big deal, but it was a software, rather than a hardware, solution.

Their improvement was to maintain two ordered request queues per drive plus a vector showing where the heads were and which direction they were moving. Requests that could be satisfied by keeping the heads moving in the same direction went on one queue and the rest went on the other. The queue pointers were swapped when the 'ahead' queue was emptied.

This made the heads float gently in and out rather than banging madly back and forth across it. The result was roughly doubled throughput and reduced wear and tear on the drive mechanics. It was a standard feature of the George 3 operating system from, IIRC, Mk 6.4 onward.

I can't see any reason why something like this couldn't be implemented within the disk drive: the extra queue storage would be relatively minimal and modern disks already contain a microcontroller.

Re: Recall what Deja Vu was...

I'll take your word for that, sir.

However, my ongoing experience with the Fedora version (0.146) shows a tendency to crash when shutting down and from time to time it looses contact with the current posting profile: a 'do nothing' profile update restores its ability to post - until the next time it happens.

The only reason I'm still using it is that IMO its still the best of the bunch for those of us running Linux and it doesn't trip over these bugs very often.

That said, if Forte Agent was ported to Linux, my hand would be in my wallet like a flash but maybe it will run under Wine: I haven't tried that.

Re: Lego and origami

In my case, Meccano, rather than Lego was my starting point, rapidly followed by balsa, glue, model aircraft kits and then building from plans.

Ever since I began to design my own flying models and (briefly *) flew RC models I've been able to visualise new models in 3D before I designed and built them. That's both complete models and partial assemblies. This ability has vastly helped in designing lots of stuff, not just new model aircraft.

* I quickly realised that, for me anyway, flying my own free flight designs in competitions was much more fun than driving RC models round the sky. Better exercise too, since you get to chase and retrieve your model after every flight.

Re: "One wonders what good that money could do"

despite planes being substantially less efficient than most cars

An excellent point. Its also a factor that most people never realize the importance of whenever you're transporting people and their luggage. Then seat-miles per gallon is the only important number and the mpg value of the vehicle itself is pretty much irrelevant. I worked this out in the late 70s when planning an overland London-Kathmandu return trip and deciding what type of vehicle to use.

For two people, the best answer was a Citroen 2CV van (30 mpg, so 60 seat/mpg), but there were four of us, so that was out. Next up was a VW Combi (21 mpg, 63 seat/mpg with three on board), which was OK with two or possibly three people, but add a 4th person plus gear and it was well known that its mpg fell off a cliff and it got slow, especially on hills. This was important too, because some parts of the London-Kathmandu run are quite high: 5,500ft in eastern Turkey, 7500ft on the southern rim of the Kathmandu valley.

We ended up using a long wheelbase 4 cylinder petrol Series 2 Landrover, which never in its life exceeded 16 mpg (OK, 19 mpg after I fitted electronic ignition) but four up, it hit 64 seat/mpg almost regardless of the weight of camping gear we carried. That was better than any other vehicle we could afford.

But put more people on a different vehicle and the numbers get even better: Encounter Overland ran essentially the same route driving big old RL Bedford trucks with 18 people on board and towing a steel ex-army trailer containing everybody's luggage. The truck normally did around 10 mpg, so call that 8 mpg with a trailer-full of luggage, but this still gives 144 seat/mpg - far beyond the economy anything smaller could manage despite its apparently dire mpg figure.

Altering CapsLock operation

If you're running Linux, then the command set:

xmodmap -e "clear lock"

xmodmap -e "keycode 66 = NoSymbol Caps_Lock"

xmodmap -e "add lock = Caps_Lock"

should change key assignments so that SHIFT+CAPSLOCK is needed to toggle between the capslock state and the unlocked state. Put them in a file called /etc/profile.d/capslock.sh and make it globally readable: most local versions of ~/.bashrc should find it there and and execute it as part of the login process.

Hopefully, Windows and Apple OSen will have a similar user configuration process that can reassign keys on a per-user basis.

Re: Bah!

I agree that the finance industry would never consider using floating point for storing values in any currency. Been there, done that: the rule is to hold integral values of the smallest unit used in that currency, so GBP amounts are held in pence, dollar values in cents and Indian Rupees still seem to be stored as paise (100 paise to the rupee even though the physical currency no longer has any currency smaller than a rupee.

However, COBOL's numeric COMPUTATIONAL qualifier (COMP is also valid) isn't special for currency.

All it means is that the value in a variable may contain only a sign and a string of digits, and COMP may be omitted if the variable's PICTURE contains only the characters S and 9. (unless its an edited value when +-9Z. are all usable) and it will always be handled as an integer value with the decimal point being treated as implied and/or used to align values with differing numbers of digits beyond the decimal point.

This causes issues for newbie programmers because if you have two values stored as PIC "999.99" COMP and you try to divide 7.00 by 2.00 you will get 3.00 and a remainder of 1.0 because integer division.

You can also add the 'SYNC' qualifier, which causes the computer's 'natural' storage to be used for that variable. That would be a 24 bit signed integer on an ICL 1900 mainframe, a 16 or 32 bit signed integer on a 32 or 64 bit Intel-based machine. OTOH if you were running on a low end IBM S/360 such as a model 30 you'd be using Packed BCD (Packed Binary Coded Decimal) which stored digits as 4 bits, packed two per byte, and had special instructions for dealing with larger packed BCD values stored as strings of bytes.

COBOL can also handle floating point values (COMP-1 or COMP-2 type variables, but that's outside scope for this discussion, since only a numpty or an inexperienced BASIC programmer would ever consider using floating point values to store monetary values.

Re: people should not be fraudulently using the service

Unfortunately, waving their hands in the air while feebly whining that "people should not be fraudulently using the service" seems to be typical of almost everybody who has been put in charge of NHS medical data for the last two decades. Just how big a clue stick do we need to clobber them with before they realise that a British citizen's medical data is PRIVATE and not theirs to do what they like with.

The people charged with protecting medical data are subject to GDPR penalties if they sell it, give it to PeterTheil because he's a mate or to the spawn of Google because that seems like a good idea after a few pints, so why have none of these people been charged under GDPR, which has now been incorporated into UK law, and sacked?

Bastards, all of them.