Posts by Alan Brown

"the computer cannot necessarily understand what that object is, why it is there, or what it is doing. It certainly cannot accurately infer what the object is about to do next."

I've seen a LOT of humans in charge of large chunks of moving metal who have similar issues (or take the point of view that they are more important than everything else around them)

One of the more important things about self-driving vehicles is that it sets a minimum skill level which humans should be able to attain before being issued with a license

My opinion for several years has been that as soon as computers are shown to be better/safer than humans, insurance actuaries will start forcing the issue by way of premiums - and if government-issued licenses are low quality they're likely to start insisting on higher level testing to qualify for lower premiums when manually controlling the vehicle

"I know how much can go wrong."

What worries ME is that it's still less than can (and regularly DOES) go wrong with wetware at the controls

Self-driving cars don't need to be perfect, just statistically better than humans at the job - and that's a pretty low bar to reach

Re: Nobody told me I wasn't allowed to do it.

The problem isn't the bad engineers. It's the MBAs and accountants with NO engineering or QA training at all

Re: Nobody told me I wasn't allowed to do it.

"In my experience, far too many developers only think about the "golden path" through the software, where everything goes exactly as expected."

Case in point in the old days: giving "restricted accounts" which only let people do email.

Do you know how many ways there were to break out of Pine (or most other "email programs") into a shell?

I set up a reward system. Any user who could send me a message after breaking out of the mail system got $10. I thought I'd nailed it down before letting people loose on it but ended up shelling out a few hundred dollars anyway (some didn't bother with the message, but enough did that the holes were picked up quickly)

Re: Nobody told me I wasn't allowed to do it.

I used to get told off for writing this kind of checking by default at high school as "unnecessary" - until it turned out it wasn't

Re: Nobody told me I wasn't allowed to do it.

This happens a lot in aviation and the testers are called "pilots"

God knows why they try weird stuff but they do

Re: Sound?

"beats upon the brain"

Re: Clop - clop - clop - clop - ....

Cargo swallow

Re: Silence please!

Seeing as it's not April 1st, and lots of reserach has already been done into this

The original idea of noisemakers was to only work up to about 10-20km/h (where tyre noise is enough to do the job anyway) and be relatively quiet. The intent is to stop them sneaking up on people in carparks, not warn them 200 metres out

EU regulations on EV pedestrian alerting noise emitters have only existed for 9 years and already cover this. Stand in front of a Leaf sometime. It's fairly unobstrusive

I still like the idea of a jetsons flitter, but the dictated noise is basically shaped "white noise" - which is better for allowing people to stereolocate the direction than single tones are (this was discovered 40 years ago when trying to find better ambulance tones. A burst of white noise worked far better than bare "blues and twos" in allowing drivers and pedestrians to work out where the thing was coming from and getting out of the way)

The problem is that whilst every "failure" advances knowledge, it also means the politicians and "biznizmen" keep slashing funding because "it doesn't work"

Pitching tokamaks, holoraums and stellerators as something other than "power" or weapons and actually getting funding would go a long way towards sustainably gettnig research continued

fusion on this planet has been happening since 1932 - and with a net positive result (for short durations) occasionally since the 1950s

It's long duration positive results without R(U)D that's the hard part

Re: ... demonstrating self-heating plasma ...

all the other parts are stagecraft. You just need a freshly served apple pie

Re: It is courageous to assume it will work

"In order to make fusion work, it needs to generate 300%-500% of the energy input to be viable."

More than that.

What goes into the chamber is less than 1% of the input electrical energy. The absolute best you can hope to extract as useful work on the output side is 35% of the heat energy produced

if you can't maintain 100k-1M over unity then it's not going to be viable enoug to overcome losses inside the system (unless tremendous efficiency gains are made on the input side)

The good news is that the fusion isn't really the hard part. The hard part is maintaining contained plasmas AND keeping things fusing at the same time.

China's claimed 15 minutes of plasma containment (no fusion). The best fusion runs so far are still less than a minute. Things have a tendency in R&D to hang around at one level for decades or suddenly start improving when someone has an ephiany (Blue leds being a case in point)

For all we know, Tokamaks (torus or spherical) are a dead end and stellerators with better computer control were the answer after all. Right now we don't know what we don't know and we know vastly less than the blind men exploring the elephant.

It would be nice to see fusion in 20 years but it's better to be harshly realistic and say that it wil take at least a century of hard work. That way we can be pleasantly surprised if it takes less instead of moaning about it all being a fraud and physcists being full of snake oil

It also means that they can get properly planned budgets instead of having them slashed and burned when they fail to get there. The story from funding bodies has frequently been much along the lines of: "I'll give you a billion dollars a year to work on fusion" "I'm still giving you 800 million dollars a year for fusion, why haven't you gotten it working yet?" "I've been giving you $500 million a year for the last 3 years, where is it?" "If you don't get fusion working next year I'll stop giving you the $50million per year"

Re: All aboard the next part of the gravy train trip

The biggest problem with fusion is that every time there's a setback, funding gets cut

This is bleeding edge discovery, almost pure physics. Each "failure" - isn't. It's invariably a new discovery which opens up more possibilities and questions (but backers only interested in $$$ don't see it that way)

Until relatively recently Fusion has been funded at a a"fusion-never" level. It was the EU and China who kicked it back into high gear

It's WORTH pursuing. It's also WORTH noting that any commercial outfit which does so is virtually guaranteed to be throwing money into a black hole - but then again so were the 6 different teams chasing the Transistor after that was proven to be theoretically possible by Heisenberg in the 1920s - and despite AT&T gett8ing there first with the point contact device it was the thin film transistor that Philips demonstrated 6 weeks later which ended up being the basis for virtually all modern electronics

Fusion will probably work eventually, once (as Edison famously pointed out) all the methods which don't work are eliminated. In the meantime the research has given us a whole new bunch of knowledge about subatomic physics we didn't previously have along with VAST experience in handling and controlling HOT plasmas (which will be highly valuable for spaceflight)

Molten Salt is a stepping stone to the future - and it should be noted that in 1961 light water technology was regarded by nuclear scientists as a very temporary stepping stone along the way due to its many drawbacks. If any popped up today they'd be asked why the hell we're still using it (and where's their flying car?)

Thorium has a bunch of benefits over uranium - mostly that it gets away from the weapons cycle(*), but thorium is a really easy fuel to obtain (watse product of raw earth mining) and doesn't need enrichment.

Seaborg et al tried to weaponise thorium and failed every time. U233 bombs would be really cheap if they worked, but there are fundamental reasons it can't happen and why enriching natural uranium (actually, depleting it to produce very pure plutonium to make bombs - enriched uranium is a waste product of weaponsmaking) is always going to be vastly more economic for military systems than other processes. The problem is that military systems have been allowed to exert total control over the nuclear cycle (tail wagging the dog) instead of being muzzled hard.

Re: We need to replace Vlad's Gas

Quite frankly I'd prefer panda nuclear to russian gas if we really had to make that choice

China's been the world's production/technology house for 3500 years. They'd rather sell you stuff than invade you

FWIW, a large part of the reason the USA is scared of China and is waving the military boogeyman around is because China owns enough of the USA to invoke "Pax Morporkia"

"If you fight, we'll call in your mortgages. And incidentally that's my pike you're pointing at me. I paid for that shield you're holding. And take my helmet off when you speak to me, you horrible little debtor."

On top of that, moving to a nuclear world (less oil dependent) makes the world less likely to continue using USD for international trade and once a hegmonic currency loses that position, any accruded debt becomes a liability for the host country (this is what happened to britain aftare WW2, quite apart from the lend lease debt) - the USA has stacked up _quadrillions_ of dollars of debt and may not be able to repay it all of that happens

Re: To be Prufrockian

Fusion is the easy part

Holding a plasma for meaningful periods is the hard part. 15 minutes is the record

Achieving unity is another target which has only been beaten for a few seconds at a time and only at chamber energy in/out levels - not gross energy in/recoverable work. What's needed isn't unity in the chamber but 100k:1 to 1M:1 ratios and for sustained periods (hours, not minutes)

History is littered with promising technology which never made it out of the laboratory and internal combustion engines which only made barely enough power to spin themselves

Every "failure" is a learning experience (something accountants and politicians fail to grasp), but I think commercial fusion power is still 60+ years away from being a reality. In the meantime China and others are forging ahead with SAFE fission power systems such as molten Salt technology (Wuwei and others)

(Anyone who thinks molten sodium is safe needs to have their head examined. That's the technology that actually has the "cracking pipes" and serious corrosion issues erronously attributed to salt tech and a coolant which explodes if it contacts water/burns furiously in contact with air isn't a bright idea even in the non-nuclear side of the equation - as they found out at ~Monju. Nuclear proponents need to learn the mantra of "Just because you CAN doesn't mean you SHOULD" and take chemical safety into consaideration)

Re: re bad printing on fuses

Same here but there's definitely a problem with a lot of electromechanical parts when you hold up older vs newer ones side by side to compare

Barely readable stampings aren't anything new, but clearly marked parts often aren't obtainable anymore from the usual (or official) suppliers

The funny part is that you can usually find high quality well marked parts from china - but official channels won't buy them as they cost 1% more than the absolute cheapest they can lay their hands on

Re: It never gets old

"It's a fuse, innit? A fuse is a fuse is a fuse!"

Yes, I've run into this response

Re: Normal operation

One assumes (hopes?) the checklist got updated?

It's annoying when this happens. Doubly annoying when it KEEPS happening

Quality British Customer Service - the kind that made British Leyland a worldbeating sucess

This is why commonwealth customers beat paths to everybody ELSE'S front doors as soon as they had an opportunity

But of course, it's all everyone else's fault and our company needs protection from those EEEEEVIL foreigners and their national-security-threatening ability to get things done right first time and/or not gaslight the customer

The problem with BTOR is that the guys who come out are contractors paid to come out - but only paid if they close the fault and either NOT given past information or don't bother reading it (despite assurances given to the customer (the ISP) that the techs have been briefed before coming out)

What they do is close the fault regardless as soon as they step offsite, meaning the whole cycle has to be started over. The entire system is unfit for purpose and essentially being milked hard by unscrupulous contractors due to the very low per-rate payments

OFCOM are utterly useless. because regulatory capture

I'll bet the last pair of blockes were actually BTOR employees, unlike all the previous ones

Re: Several times...

The number of times I've found a IEC plug "only just" hanging in makes me wonder about a lot of people

And yes, they've been the cause of intermittent issues where "pushing the plug in properly" has solved it

Re: Printer Cartridge

I've heard a similar story from BBC micro days about a teacher who managed to install a cassette into a floppy drive slot

In her defence some 5 1/4 inch drives had flip front doors but it's still impressive

Printer sellers use the razorblade model - subsidised printers, expensive ink

If they changed what printers really cost then the devices would cost about twice as much in most cases

Lasers are so cheap that the only reason to justify buying inkjets is photo-quality printing

Re: Interesting ...

"However, it really doesn't like being inactive for a couple of weeks"

This is standard problem #1 with _ALL_ inkjets.

If you're not printing every other day, don't buy one.

Lasers don't have this problem, are only slightly more expensive and are generally MUCH cheaper to run

Re: Interesting ...

get yourself a rubber kit. The standard problem with all printers that are experiencing paper jams is worn out rubber pickup/friction rollers/worn springs and almost all of these are replaceable

The lifespan of most paper drawer pickup rollers is around 50k sheets. If they're smooth/shiny, they need replacing

search term 'Brother [model] consumable kit' (or roller kit or refurb kit) is your recommended starting point

Re: Playing the "Enviromentally Responsible" card

"If the price is anything like HP - I assume they end up in a bank vault as some sort of currency reserve."

Speaking with a (now long retired) university procurement guy, he urged us to consider anything except HP for our purchases

Paraphrasing his explanation, his department had run a lot of testing of printers of various types and duty cycles and HP consistently came out as the most expensive, worst quality of the big names, with running costs ten+ times higher than several of the competing makers

This was BEFORE the split into HP/HPE

It matched what I'd already experienced but it was nice to see it actually quantified

Unfortunately he'd been unable to set policy on printer purchases because politics. Attempts to ban HP purchases had met with ferocious resistance, as had attempts to ban personal printers (anything smaller than a workgroup printer has eye-wateringly high cost of operation). Some groups had individual cheap inkjets on each desk and were paying incredibly high figures comnpared to using a higher quality workgroup lase, let alone a departmental one

And of course, people buy cheap printers without considering their print volumes. When I did the sums I found that a few £10k departmental lasers were £100k cheaper over 5 years than buying several more £1k workgroup printers - and then spent 4 years being obstructed by manglement who refused to spend that much on a printer when cheaper ones were available

Re: Playing the "Enviromentally Responsible" card

"they recommend wrapping kitchen waste in newspaper - yet that ink is apparently toxic."

Newspapers have generally used soy-based inks for decades. The days of toxic inks are back in the 1980s

Re: Playing the "Enviromentally Responsible" card

Contrary to assertions, paper is one of the nastier items to manufacture (quite nasty chemicals used) and has a very limited number of recycle uses (the papermaking pul[ing process chops wood lignin fibres up and it can only get "so short" before it no longer works as paper)

Coated paper is virtually impossible to recycle into anything except building insulation

In a lot of cases it's virtually impossible to compost too, thanks to the preservatives/dyes/varnishes added

The mantra is "reduce", "reuse", "recycle" - it's greener to not print at all and jumping straight to the recycling stage is the greenwashing way of assuaging a guilty conscience

Ink cartridges, on the other hand are just bits of plastic and surprisingly easy to recycle if they can't be reused (most can be refilled)

That said, the cost and energy expenditure of recycling most plastics (or paper) is so high that you'd usually expend 10% as much oil simply burning the old stuff as fuel and manufacturing new items from virgin materials

It's complex - and "recycling" only really works for a limited range of stuff.

In most cases it's a horrendous expenditure of resources. Plastics/paper in particular are extremely low value materials with a very high recycling cost - should we REALLY be spending 20p recycling a soft drink bottle that cost 2p to manfacture and distribute (and constitutes 0.1p worth of raw materials), into another 2p bottle or is it better to find a better way of dealing with it at end of life, or finding a better value proposition for the distribution of the content?

"Cleaning" cycles

One of the more notorious "Anti-generics" tactics involved running excess cleaning cycles to drain the chipless items very quickly

It will be interesting to know how many pages people get out of these consumables vs chipped ones

Re: "Interxion have still failed to share any information and refuse to answer the telephone"

try having "geographically redundant data feeds" which run in the same duct for 3 miles, split out at the nearest town but then rejoin in another duct for the final 10 miles

Quality British Workmanship

Re: not DNS, not UPS

"usually an automatic transfer switch is."

A lot of this stuff is on cranky "Industrial ethernet" controllers which are fragile as all hell and prone to playing up

I have a half megawatt UPS on my watch which I haven't been able to directly talk to for over 5 years because the ethernet has locked up - I KNOW what's wrong and fixed the crash triggers a long time ago but getting manglement to agree to reboot it is impossible

Its compantion (800kW) had a similar problem but spontaneously rebooted 6 months ago after a similar period of downtime and "miraculously came back to life"

I'd been telling people what was needed the entire time and been shouted down - the "experts" from the vendor companies may know their UPS equipment and the Caterpilar engines but they are _NOT_ IT people(*) and never read the manuals or specifications for the interface equipment to know what takes these ($2000+) ethernet bridges out of service (Hint, never put your industrial equipment on the same VLAN as any other kit, they frequently can't cope with broadcast traffic and as we know painfully well. SCADA kit is frequently crappy as all hell)

(*) Seeing a field tech spit in a serial port plug to clean the pins is the kind of thing that really inspires confidence. I mean, really?? And you wonder why you have a corrosion problem?