Never used a Transputer, but saw a demo or two.
It was impressive all right: I've seldom seen revolutionary new hardware vanish so fast or so completely.
541 posts • joined 10 Apr 2007
Never used a Transputer, but saw a demo or two.
It was impressive all right: I've seldom seen revolutionary new hardware vanish so fast or so completely.
... all UK news sources refused point blank to carry any election broadcasts or propaganda until the party concerned had published a fully costed manifesto and its balance sheet.
I like Java, but their premise of "There would be no banking if Java didn't exist" is ridiculous.
Exactly. Banking runs largely on COBOL.
Grace Hopper deserved far more time than she got:
- first to show that a computer could handle text
- wrote the first assembler (before she did that you programmed entirely in numbers)
- involved in the development of MATH-MATIK and FLOW-MATIC (forerunner of COBOL)
At least they got her involvement in COBOL correct.
As others have said, C should have replaced Java in the series and, given that this was a British radio series, there should have been at least a nod to BCPL, which was developed at Cambridge University, because the C genealogy as described by Brian Kernighan is:
For sure all the curly bracket block-structured languages have C in their ancestry just as all the block structure languages that use begin..end can be traced back to Algol 60.
A $10M fine on a profit of $770M is just over 1%.
Thats just a brush with a feather duster that they'll never notice. If the FCC was serious they'd bong them 25% of last year's profit.
...and impractical for a few other reasons other reasons, such as COST and time constraints.
Assume that a drone with a book dangling from its claws can manage 60 mph and has the typical drone endurance of 30 minutes. This sets two physical limits:
1) it can only deliver within 15 miles of its take-off point because its range at 60 mph is 30 miles and it is flying an out-and-return mission.
2) The delivery flight will take 15 minutes from take-off to delivery. This leaves just 15 minutes to receive the order, pick and pack it and get the parcel to the drone port. Otherwise they can't meet the 30 min order-to-in-your-hands target. For this to work reliably they'll need a warehouse at every drone-port.
So, far from letting them sack their current delivery drivers, this cunning plot is likely to need even more people to staff the matrix of warehouses and drone-ports covering the country: these will include delivery drivers to keep the drone-ports stocked with best-selling items, a crew of pickers and packers plus drone operators and mechanics at each drone-port and, of course, a management team for each drone-port.
Doubling or quadrupling the drone's range and/or performance doesn't make a lot of difference to the practicality of this delivery method.
So now we're going to be deluged with encrypted spam!
Could this saga turn out to be a longer version of Damon Knight's short story "To Serve Man".
Ray Bradbury had something to say about this too.
See "Fahrenheit 451" with its thought police and the suppression of all fiction apart from Hemmingway-style realism and all entertainment apart from mindless 24 hour soaps.
For those who can't or won't read there's a passable film version....
Why does the bar chart show .org as bigger than .uk while the pie chart has them the other way round?
Were they sampled at different times or are the numbers just plain dodgy?
I'm entirely in favour of copyright, but why should anybody get compensated for something they created 43 freaking years ago? Especially if it isn't their own work, but merely something they bought.
IMO copyright should be like a patent: it should have a lifetime of, say, 25 years or until the creator's death, whichever is longest, and should be non-renewable. I fail utterly to see why anybody or company should continue to benefit from somebody else's work after that.
Why is it always near midnight? ...in this case because, though things may not get nasty for some time, we're very close to the point where the our fate becomes inevitable no matter what we do in the future.
We're probably already past the sustainable human population, but is anybody taking notice apart from, amazingly, the Pope? And, even he's not about to admit that two kids would be a lot better than three or to promote contraceptives.
I still don't see any serious attempt to de-carbonise energy production outside of China, and even there they're only doing it because their hand has been forced by a country-wide smog problem that rivals Victorian London. Carbon sequestration is a bad joke due to its appalling overall energy efficiency and a severe lack of very long term guaranteed non-leaking storage. No serious attempts at producing enough low or zero carbon energy, such as desert-based solar-electric or thorium nukes, to replace our current sources are evident, which means they're 20-30 years away at best.
There's one feature you see in poor societies that this, otherwise very interesting piece, doesn't explain. That is the extreme hospitality shown to passing strangers in poor societies, often in circumstances where the chances of reciprocality are very limited. This kindness is seldom or never matched among westerners, but is a happy memory from travels in the Middle East, Iran and the Indian subcontinent.
Errrm, quoting voltage tells us nothing about the power this ATM uses when running. Power is measured in watts, not volts.
Since low power consumption is evidently a spiffy new feature in the ATM world, how about an update telling us how much power this new design uses? Including current ATM power consumption as well would be nice, so we can tell whether the new machine is truly wonderful or merely germicidal.
I too had a Mamod stationary steam engine but, as others have said, once you're run it a few times and you found out that you couldn't really do anything with much with it except watch it run, it got boring. I got much more fun from the clockwork Magic Motor that came with a red and green Meccano No 6 set and let me make tanks and tractors that actually worked.
However, the thing that really got my interest going in designing and building things that could be made to work was model aircraft. Almost every kid I knew at the time made and flew models at some point. Initially control-line models were the thing to have because you could fly them anywhere there was a grass patch 50m square and you couldn't lose them. The favourites among my friends were either a Keilcraft Phantom kit with a Mills 1.3 diesel up front or a Veron Beebug with the smaller Mills 0.75 diesel pulling it. Building and flying them taught you a lot, from operating the engines without getting your fingers hit, through building (and later repairing) the models to eventually learning how to fly them without crashing.
Then, a few years later, the excellent Cox TeeDee glowplug motors appeared - such wonderful precision engineering that running them in consisted of running the engine rich for 30 seconds and then leaning it out and hearing that lovely scream.
There's also the fact that example.com, example.net and example.org are required by the IETF to be unresolved, which is something I've used in tests in OSS projects. Tests that turned out to fail on Verizon fibre connections, because ISPs getting search revenue is more important than working applications.
All three resolve domains and ping from here as 188.8.131.52, which is registered to a resident of Santa Monica, CA so is unlikely to be anything to do with Virgin's munging. If you look at them with a web browser, they all show the text:
This domain is established to be used for illustrative examples in
documents. You may use this domain in examples without prior
coordination or asking for permission.
And, as a link on this page says they are IANA reserved domains and adds that "These domains may be used as illustrative examples in documents without prior coordination with us. They are not available for registration or transfer", I think you're wrong in claiming that they should not resolve. Did you check your assumptions before posting?
the NHS had failed to properly consult on the privacy concerns of sharing data
doesn't begin to describe the problem. "Some twunk thought they could flog all our data to life insurers and other lowlife and we'd all be to stupid to notice and too apathetic to do anything if, by some mischance, we did smell a rat" is a bet closer to the truth.
I, for one, don't trust them even slightly and won't support any care.data revival until the management up to and including the Minister for Health have been sacked. They're all guilty of attempting to sell data they don't own.
Its an ASK-13 dual seat training glider flying out of Booker, and now on the ground there. They usually aero-tow, so that was probably somebody taking a high tow for the last flight of the day. BTW, did you notice that you were seeing a FLARM trace rather than radar?
I agree with the first poster in this thread: the flashing as I slide my mouse over the second level menu bar on my way to the top-level bar is very irritating. I also dislike having pictures associated with the top-level menu bar hide the second level bar.
However, my main beef is that the contrast in colour between an unread article and one that has been read is far too small. It doesn't have to be as big a change as the old blue/scarlet switch, but the current black/dark grey switch is almost impossible to see at a glance: black/dull red (seen already) would make scanning for new articles a lot easier.
In your dreams. The lakes are methane as the article said and ethane, not ethanol as the article misprinted, plus (much) smaller amounts of assorted higher molecular weight alkanes. These are flammable, or would be if there was any free oxygen in Titan, and definitely not intoxicating.
More worryingly, assuming this report was accurate there's nothing illegal about what they were doing. We had a presentation on drones at work as part of our attempts to figure out if we need to be worried about our aircraft hitting them.
Since you say you're in the industry, you should know that it is not legal to fly any model aircraft above 400 ft or inside controlled airspace. Drones weighing under 20kg that aren't licensed for airwork and being used for it come into the same category as model aircraft. They are covered by CAP 658:
Sounds like it would be a good idea for you to read CAP 658. Same applies to anybody else who has a drone and isn't a BMFA member and a member of a model aircraft flying club. I hope you've got third party insurance for that drone.
The main thing that occurred to us was that we could do exactly what the drone operator in this case was doing without breaking the law. - what makes you think he wasn't breaking the law? If you still think its legal, I suggest you have a word with that American idiot who put his drone up to 3500ft straight through his local class B airspace and then posted the video on YouTube. Its a safe bet he didn't enjoy his mandatory visit to the FAA. Do that here, get caught, and you'll be invited to Swanwick for tea and bikkies. You're unlikely to enjoy the experience either.
I agree with both the pevious posters - dumbed down and stuff locked away.
Locked away: when I visited the Science Museum earlier this summer I was looking forward to seeing the Difference Engine run (I was there at the wrong time on my last visit a lonh time back), but now its hermetically sealed into a glass case so that it can't be run even if the museum wanted to do so.
Dumbed down: there was only a very scrappy notice about its purpose, some summary descriptions of prototype parts and drawings and absolutely no attempt at a coherent explanation of how it worked or what the Method of Differences is all about. Humpf!
Any search that is not "biased" is not a search. That's what a search engine does, is to winnow out the wheat from the chaff.
Yeah, sure, but is it still unbiased when links to its owners other businesses appear, by magic or pure coincidence of course! near the top of the first page along with more links the all those who put money in their hands? How do you know this can't or won't happen?
... if the motor whatnots can spin it fast enough to make for a less damaging landing that pretty much guarantees that if you catch it, the phone will twist itself out of your hand and bounce down the nearest drain.
The only way to really know what you're eating is to avoid as much preprepared, manufactured food as possible because you have no control over what's gone into it or over the truth of the ingredient list written on the packaging.
Instead, get a few good cook books. Buy raw materials (meat, veg, spices, fruit and oils etc.) and cook them yourself. Its fun to do, much better for your health and probably tastes better as well.
Bacon isn't usually eaten in India or the Middle East. If you've ever seen pigs from those parts or know what they are fed on the chances are that you wouldn't want to eat bits of them either.
You'll be fine flying any drone for pleasure provided it weighs less than 7Kg, you don't fly it above 400ft AGL, within an airfield's Air Traffic Zone, in controlled airspace or anywhere close to houses or people and it must be in clear line of sight for the pilot at all times. You can even carry a camera provided its used purely for your own enjoyment: sell an image and you're doing airwork, which needs licences, type approval for the drone and a qualified pilot. These are exactly the same rules that apply to any model aircraft.
CAP 658 lays down the rules for flying model aircraft, UAVs and drones in the UK. It covers both private and commercial use. You can find it here:
or go to http://www.caa.co.uk and search for CAP658.
Looks good and certainly worth trying, but a translation or two would be useful, so:
- I wouldn't even think of starting from a British sausage - too much bread and other fillers in it, but a pack of pork mince would be a good starting point. So, what goes into an American spicy sausage? i.e what should be mixed into pork mince? Chilli powder and black pepper? Garam Masala? Something else?
- what is Bisquick? Is this just a self-raising wholemeal flour or something more complex? I assume 'C' is an abbreviation for 'cup', so you mean 1.5 cups of Bisquick.
- Shake and Bake sounds like dried breadcrumbs, but how much of it is there in a 'package'?
Exactly my experience with Experian!
So, I signed up for a credit report to see details of this supposed loan, didn't recognise it at all, and cancelled the 'free' trial account as soon as I'd saved a copy of their report.
Colour me unsurprised. If HMG is going to use that shower for identity verification, all I can say is "God help us all" because bugger all else will.
With enough redundancy...
They've told us how much redundancy they use - just two copies of the data plus the ZFS checksums which, unlike RAID 5 checksums, can't be used to recreate a corrupt block or the content of a dead disk though they do guarantee to spot corruption. Given that stored data is merely duplicated, there's no point in more than duplicating the rest of the supporting hardware and power supplies.
Like others, I find 9^18 reliability claim incredible: the probability of simultaneous failure of the disks holding both copies of a piece of data has to be much higher than that. Put it another way, even the best fault tolerant kit, with every component at least duplicated and the system configured as a geographically distributed system don't claim more than 9^6 reliability, so whats going on here?
I guess you weren't around in the '70s when the fledgling Apple company (3 guys) were literally building new stuff, hardware and software, in their garage
Well I was and I remember seeing commercially built microcomputers (home computers if you prefer) on sale in the Computer Store on Broadway in NYC in the summer of 1976. These were machines made by IMSAI, MITS and SWTPc. All used add-on cards to extend the basic machine, all could be programmed in assembler or BASIC and some were fitted with 8" or 5 1/4" floppy drives, though they mostly used green-screen serial terminals or ASR-33 teletypes for their user interface and were based on MC6800 or Intel 8080 chips rather than the Mostek 6502 that the Apple 1 used.
All of them were on sale in 1975, so they all preceeded the Apple 1 (April 1976) by at least 4 months. All were on sale a good two years before the Apple ][ was announced.
The microcomputer an Apple invention? I don't think so!
I had a quick scan through the docs on the care.data page Andrew linked to, and found the following:
- in Pathfinder Proposal 1, section 2 (Options):
3d "what happens if [the patients] have already opted out and"
3e "their rights to change their mind at any time"
- notes from the care.data meeting on 25th of June:
Page 3 under "Fair processing principles" I found this gem:
d) Opt out/in - this is not a legal issue/right; but is a fair processing issue so has to be included
I'm going to be totally pissed off if these bloated bureaucrats think they are going to make us opt out for a third time or simply decide to ignore opt outs. Given the attention that their masters in Parliament give to public sentiment or to doing anything they promised to do in order to get elected, I wouldn't put it past care.data to do either or both of these things, seeing that they seem to be, in the words of FZ "Totally commited to the fifty bucks".
...provided that it means we get a fully supported Linux-RT kernel. However, building a proper real-time Linux, i.e. one with deterministic response times to external events, is almost certainly going to mean big changes to the kernel, not just reimplementing the process scheduler.
But surely member states have nothing to hide and therefor nothing to fear
Probably true, provided you exclude the donations to our elected reps and the parties they belong to from companies which want to hide to effect of their lobbying on the taxes we have to pay (because they don't) and the laws we have that benefit them rather than us.
This isn't a good size comparison. If you MUST compare it with terrestrial sizes. the Earth's equator is roughly 25000 miles in circumference, so "About 40 times round Earth's equator" would be better, but IMHO "would reach a third of the way round the Sun if stretched out" would be best. The Sun's circumference is 2.74 million miles.
NONE of the existing mainstream linux partition managers are 'good enough'. fdisk and most of its ilk can't handle devices over 2GB due to the limitations of MBR.
Are you sure you mean Linux and not Windows here? The Linux fdisk utility is perfectly capable of handling bigger partitions than 2GB: I just ran it up to check. I usually use cfdisk because I prefer its user interface.
Ever see the old Sunbeam with inline crankshaft in action? Every time the rider changed gear the bike rocked from side to side, so I reckon that you'd need to open the throttle really carefully on this monster or the bike would end up sliding down the road on its side with you perched on top burning your bum on the exhaust.
Sensible bikes all have transverse crankshafts to avoid this problem.
.... use a piccy taken by Curiousity to illustrate a story about Opportunity when there are perfectly usable ones taken by Opportunity itself from its current location?
Didja think we wouldn't notice?
Word for Windows (version 2 or whatever) did NOT do the job.
IMHO the last worthwhile version of Word was Word for DOS 4.0. It was fast, even on a 12MHz PC-AT, blindingly fast on a 40MHz 386 box, and significantly faster than Word Perfect at operations such as 'go to end of document'. But, by far its best feature was that you didn't ever need to lift your hands off the keyboard or use the mouse. You could mouse round if you wanted, but that wasn't necessary because function key usage was very well thought out, e.g. hit F8 and the current word was selected, hit it again and the sentence was selected. Third tap selected the paragraph. Fast. Simple. Memorable.
Word 5.5 slowed things down by adding crappy drop-dpwn menus which were slower than Word 4's function key system. Word for Windows completely stuffed productivity by making a grab for the mouse a mandatory and frequent distraction from keyboarding. Yes, I know it has 'keyboard shortcuts' but can you honestly say you remember or use more than a couple of them?
In fairness, Libre Office suffers from exactly the same mouse-centric problems as do most graphical text editors (gedit, I'm looking at you).
Me? I use vi when I have to and microEmacs the rest of the time because its entirely keyboard-driven multi-buffered editor and so is the fastest way to input, edit and compare text files. Its also OSS, written in ANSI C and trivially easy to port between OSes and hardware architectures: took me 10 minutes to have it up and running on a RaspberryPi.
I saw the announcement and spiffed up photo on http://mars.jpl.nasa.gov/msl/ this morning. Its now item 2 of 6 on the Whats New list.
If you go to Mission:Where is Curiosity and look at the map for day 637 you can see where the meteorite was found.
To see the photos taken by Curiosity, go to Multimedia|Raw Images and look at those taken on day 637 by the Navcams and day 640 by ChemCam and Mastcam.
But to be fair, the very nature of many-rotor aircraft means they can be engineered for practically bulletproof reliability
Maybe so, but no amount of reliability can stop it falling out of the sky after it hits a wire the operator didn't see or after some miscreant throws a missile trailing a Kevlar line through its rotor disk(s).
Speaking as a glider pilot, I don't have a problem with RC models because they are under the control of a human operator who should be able to see and avoid any full-size aircraft where he is flying. In addition, the RC model has to remain close enough to its operator for him to see not only where it is, but its attitude. Without this visual feedback its impossible to control the model. This is also why many models have different colouring top and bottom and often use assymetric colour schemes. I used to fly a bit of RC and found that using this type of paint job made the model far easier to control.
However, the thought of drones operating in class G airspace is very scary. Almost by definition these will be either autonomous or outside visual range of an operator but none of them, as far as I am aware, give the operator anything like the field of view or the fine-grained visual resolution that any GA pilot has and I don't think any of the autonomous drones have any optical see and avoid capability. In other words, current drones have little or no ability to stay clear of gliders, paragliders, microlites or balloons. These aircraft types do not usually carry transponders, so an autonomous drone that can't reliably use optical sensors to see and avoid a full-size aircraft is just an accident looking for a place to happen.
Recent reports confirm my assessment: American military drones have collided with manned aircraft (a C-130 no less), crashed because the operator didn't realise it was inverted, and had to be shot down by F-16s when the radio link failed: a whole litany of crashes and failures which all prove that drones have no place in civilian airspace or over towns, at least until the failings that led to these accidents all have proven, reliable solutions installed in every drone and subject to regular inspection and certification. Achieving this will take considerable time and is unlikely to be cheap.
It is an unfortunate truth but a lot (I'll not say "most" even though I think it is justified) open source code is, quite frankly crap.
Thats just another consequence of Sturgeon's Law which stated that 90% of everything is utter crap. Think about it. Theodore Sturgeon, an SF author, was spot on.
I've seen bad OSS code, but at least I could look at it and see that it was bad. However, I've seen much worse closed source commercial code, which carries the extra benefit that you can't see how bad it is untill you've paid good money for that dubious priviledge.
How about a COBOL accounting system where all the programs were written to the same appalling standard. All the paragraph names in every procedure division were numeric though not in sequence. Section names? you must be kidding. No sections used. All the data names in every data division were of the form MT01 starting from the name of the first magnetic tape file and incrementing until the last field in the last record in the last mag tape file was reached. Same for cards (CR01,...), printed output (LP01,....) and working storage (WS01,...). Oh yeah, the code was totally devoid of comments outside the identification division. I only got to see this crap because the company I worked for had paid good money for it. It was so bad that it was unmaintainable and almost impossible to use so we junked it and wrote our own accounting package. Doing that was easier, took less time and saved us money on maintenance because we wrote it to be easily readable and well enough commented to be understandable even if the design documentation got lost or out of date - the norm in those days.
Nope - he was talking about capture, i.e. permanent data storage. IOW it doesn't matter whether all sensors autonomously send in readings or the logging system(s) poll them for data. Once the data arrives at the server that will record it, its easy to scan through the stream from each device and discard everything except the changes in a sensor reading.
Think systems don't work that way? Here's a real-life example: the switches in mobile phone cells are polled on a daily basis and their call data pulled down as via FTP as a file containing a megabyte or two of data. This is then processed in various ways, e.g. run through fraud detection kit and analysed by the network performance team before being used to populate one or more databases.
The article started off talking about stored data volumes, i.e. storing logging data, and then drifted off into sampling rates, which is all very interesting and must be considered when deciding how to get a true picture of the behaviour over time of the variable being sampled.
The answer to the storage problem, that I expected to see, is to only record the timestamped new value each time the sampled variable changes. Unless the change rate approaches the sampling rate, the storage saved by logging timestamped changes will easily exceed the overhead of recording the timestamp.
...and upgrading an ICL 1902S CPU to a 1903S required one wire to be cut, to increase the clock speed, and use of a screwdriver to replace the 'ICL 1902S' badge with an 'ICL 1903S' badge.
I'd be somewhat surprised if IBM and the rest of the seven dwarves didn't pull similar stunts.
OF COURSE it was the way you describe, and deliberately so. Star Wars was pure Space Opera, distilled from innumerable '50s pulp SF magazines and paperbacks and that was why it was magic: it left no cliche unturned. All the way from the archetypical kid from the backwoods planet making good, through the brawl in the sleazy spaceport bar to the mega spaceships and the galactic empire. When the first clips appeared at SF Cons many of the fans said it was rubbish because spaceships couldn't dogfight like WW2 fighter planes, but they forgot one thing: they do in Space Opera.
The thing that made Star Wars great was that George Lucas was obviously a pulp SF fan from way back and made the film as his tip of his hat to that genre. Subsequent films went down hill as they progressively stepped back from their origins, which is a pity because there was still a lot of unmined ore in the original seam.
Trolling and trawling are very different ways of fishing. Even Shirley kno that.
Trolling is, as described, towing something at the end of a fishing line that should look tasty to a fish. Its often shiny or brightly coloured but can also be made from feathers that undulate as they're towed. Trolling is ecologically sound because it doesn't cause collateral damage.
Trawling is dragging a huge netting bag, with a heavy frame to keep its mouth open, along the seabed behind a fishing boat. This rips up and destroys all the corals, seaweed, etc in its path and traps all the fish that don't swim out of the way fast enough. Apart from causing seabed damage, the trawl scoops up and kills a lot of unwanted types of fish which are dumped overboard. Its not even a remotely sound activity from an ecological viewpoint: fish farming is better.
GOLD because it is nice and dense, nearly twice the density of lead, and not radioactive.
There's not much thats denser than gold (SG=19.3). Osmium (SG=22.6) is the densest easily available substance and costs a lot more than gold ($77000/kg vs $27000/kg), which seems like a lot to pay for a 15% density increase. Density is important in this experiment: the denser the target, the more likely that the electrons in the beam are to hit a nucleus in that target and hence the stronger the resulting gamma ray beam. The most commonly available bulk radioactive, Uranium, is less dense than gold and half the price, but it seems likely that its radioactivity could screw up the experiment as well as making it nasty to handle or store.