Fun fact: GPS uses 10 bits to store the week. That means it runs out... oh heck – April 6, 2019 Older satnavs and such devices won't be able to use America's Global Positioning System properly after April 6 unless they've been suitably updated or designed to handle a looming epoch rollover. GPS signals from satellites include a timestamp, needed in part to calculate one's location, that stores the week number using ten …
Seriously, in this day and age?
Stop messing about and use 64- or 128-bit for everything, with a reserved bit that - when present - signifies an extended format follows which you can use to add in additional information (but would be ignored by devices that don't support it). And then when you define that extended format.... make that have a reserved bit...
"In 1978 64K was 'all you will ever need', or 48K for Apple II."
In 1978, 64K RAM was more than the typical home microcomputer buyer would want to spend. RAM was pretty pricey in those days. Don't forget also that those early micros commonly used ROMs for OS and programming languages: Apple ][s had 16 bit addressing just like all the rest of the early 8 bit crowd and could therefore address 64K memory.
There were any number of single-board hex keypad LED output microcomputers of that era which used about 1K RAM and more advanced machines like the TRS-80 and Apple ][s were offered with 4K RAM (admittedly, they both ended up being kitted out with more than that as standard). But if you had much (much!) deeper pockets than the typical hobbyist had and could splurge on a really big machine like a Cray 1 (first delivery of a trial machine in 1976), you could be looking at a 64 bit processor (data side, anyway) and over 8 megabytes of RAM - and a very big electricity bill.
> Yes, and 640K is all you will ever need....
"It's pointless giving hobbyists more than 1k, they wouldn't know what to do with it".*
- Andrew Hewson in a magazine (Sinclair User?) column, on the release of the first ZX80 RAM pack.
Hewson later went on to form Hewson Consultants, which became a moderately successful 8-bit games company thanks to some of those hobbyists. They not only knew what to do with it but also sent him the proof to populate his company's catalogue.
*Probably a paraphrase. It's pushing 40 years since I read it.
"Zilog Z80 chip it ran on was the processor for all sorts of much more serious machines. And it is true to say that its registers were mostly 16-bit (apart from the 8-bit accumulator used for arithmetic...)."
Sort of - the 6 general purpose 8 bit registers B+C, D+E, H+L (and their counterparts B'+C', D'+E', H'+L') can be used in those pairs as 16 bit addressing registers.
Commands such as ADD HL, BC treat the HL register pair as a 16 bit accumulator: add with carry the contents of register pair BC to HL, and place the result in HL (also works on DE and SP).
The two indexing registers IX and IY are 16 bit, as are the program counter and stack pointer (PC and SP). The accumulator and flag registers (and their A' and F' counterparts) were 8 bit. It has a lot of 16 bit stuff going on inside, does the Z-80, but it's also basically designed around 8 bit data words.
The ADD HL, BC instruction is rather slow: it takes 4 machine cycles (15 T states) according to my ancient "Programming the Z80" book here (Rodney Zaks, 1980, pub. Sybex. The paper is still white and strong, the paperback spine holding together almost as good as new - I'm impressed with that side of it at least). One would assume in any case that the Z80's ALU is not 16 bits wide and I wondered how many bits it operated on. Four, apparently:
<http://www.righto.com/2013/09/the-z-80-has-4-bit-alu-heres-how-it.html>
Every time I read about the details of one of those early 8 bit CPUs, I find out that it was designed by some jolly clever engineers who made apparently strange decisions which nevertheless worked out rather well in practice.
The ZX80 was a miracle of cost reduction. I was working in the labs of an electronics firm. One of the teams pulled one apart and "obtained" a circuit diagram. There was some head scratching understanding how it managed to work. I recall that rather than there being a couple of chips to scan the keyboard switches they used the processors upper address lines, presumably by having a bit of code that jumped to "funny" addresses that would light up the upper address lines.
Well in 1978 I was using a Apple II where I considered it the height of luxury to have 32K. Having upgraded from a PDP 8E which had 12K allocated to one of the ASR 33 teletypes, 4K to the other one.
Pretty sure every microcomputer in 1978, the 8 bit'ers, had a 16 bit address buss so max memory was 64K without bank switching. Although I think I remember some odd ball system based on a 2900 bit slice that had 24 bit addressing.
"Tell kids today that you only need a 1 bit processor and they'd not believe you"
I visited a supercomputer centre a little while ago and in their timeline display showing the various machines they've had over the years and their specs, was one listed as 24,000 1-bit cores (or something to that effect), which left me a bit puzzled.
I suspect they were talking about magnetic core memory. It would have been 24,000 tiny magnetic rings strung on a mesh of set and sense wires. Each ring/core can store 1-bit of data. http://www.ricomputermuseum.org/Home/interesting_computer_items/magnetic-core-memory
"The ZX80 was limited even for its time, to keep the cost very low. Other home computers at the time had 4k or more RAM."
Indeed it was, but the ZX80 could be expanded with a 3KB RAM pack which I assume provided 4KB RAM in total - or later on, it could take the 16KB RAM pack developed for the slightly more advanced ZX81.
A design classic, that RAM pack wasn't. I used multiple layers of double-sided foam sticky tape to ensure that my 16KB RAM pack didn't wobble and cause crashes when plugged in to my ZX81.
That was because Sinclair reused the ZX80 RAM pack mould instead of Rick's ZX81 RAM pack design.
Puh. In the year 1067, my great^N grandpa made a NAND gate out of twigs and moss, and used insects to implement the multiphase delay lines. It was Turing complete. He wrote the first video game, but the insects got bored, and the moss dried out before the first line of code was fully executed.
"Grow your own veg? Kill your own cows? Quarry your own stone? Distill your own petrol? You're Exxon's bitch."
Yes. Can do all of those, choose not to for convenience and pragmatic reasons (you'd be amazed how often I don't need to quarry stone) but, if the need arose, I could (although I'd probably brew and distil alcohol instead of refine petrol, petrol isn't great, even with ice and a slice)
I have maps and can use them, have even done an orienteering course, but for the "convenience and pragmatic reasons" you mention, I use a sat nav and believe it or not I have never driven into a river or down an impassable road because just like using a map, you still have to apply common-sense.
"I have never driven into a river or down an impassable road"
Common sense .... so true! Couple months ago my GPS sent me over a cow-pastured ridge on a dirt road, down an impassable road completely covered with 3 inches of water and ... I stopped when it wanted me to go through the river blocking the route. Finally realized that the GPS can be wrong, especially in the rural Tennessee/Virginia border area. Common sense finally prevailed & I didn't get stuck. With my wife and one of the kids in the car, it would have been a bit awkward... BTW, when I checked, Google Maps also had the same road as my GPS, it's wrong also, although the satellite view shows accurately the small river that was the road, and the larger river I decided not to chuck the SUV in.
Finally realized that the GPS can be wrong, especially in the rural Tennessee/Virginia border area
It used to be the practice of map companies to intentionally introduce errors in their maps, in order to catch anyone re-using the maps and reselling them as their own. I wonder if the sources for GPSes and Google Maps has the same intentional errors?
"but, if the need arose, I could (although I'd probably brew and distil alcohol instead of refine petrol, petrol isn't great, even with ice and a slice)"
You have to distill the octane from the crude oil before you refine it, which involves getting rid of as much octane (and replacing it with such catchily titled compounds as (2,2,3)-trimethylpentane, although (2,3,3)-trimethylpentane will burn better). I'm saying you can't do either yourself.
Did you make the device you typed your comment on yourself?
@DavCrav.
'Petrol' might be a bit grandiose, but I can and have recovered distillates from crude which will run a petrol IC engine, grammar school chemistry lessons were great.
For your second point, no, I didn't design and build this computer, I have however designed and built others which were used on bulletin board systems and early incarnations of the 'net in the UK, notably the JANET via MCC's dialup portal and out onto the wider Internet before the web was a twinkle in Tim Berners Lee's eye but I fail to see how that's relevant to the points I answered from your first comment?
"....If ya can't do it yourself you is someone's bitch."
Grow your own veg? Kill your own cows? Quarry your own stone? Distill your own petrol? You're Exxon's bitch. ..."
--
YUP! That is EXACTLY what we do down here! We have big gardens, do kill our own cows (humanely with the bolt gun!), cut and finish our own rocks and granite for the fireplaces and kitchen countertops AND we make our own e85 gas or even wood grain 100% methanol which goes into the diesels! Were all self-sufficient down here and as hardcore preppers, we KNOW how to take care of ourselves!
"I use a map."
I remember the bad old days too, map scrunched up on the steering wheel while you desperately try to work out if the motorway exit just disappearing is the one you really needed.
As a baby engineer, I once got stuck on the Coventry ring-road after visiting Marconi. Every turn I took eventually brought me back onto the bloody ring-road. I began to seriously suspect I may die there, endless circling 'picturesque' Coventry...
Now I just tell the sat nav where I want to go and it does it flawlessly with zero effort on my part, even taking into account the dynamic traffic situation. I'd never go back to paper maps willingly.
The full quote from Dirk Gently's Holistic Detective Agency is:
He had extracted himself from the Cambridge one-way system by the usual method, which involved going round and round it faster and faster until he achieved a sort of escape velocity and flew off at a tangent in a random direction, which he was now trying to identify and correct for.
"I remember the bad old days too, map scrunched up on the steering wheel while you desperately try to work out if the motorway exit just disappearing is the one you really needed."
Well thats just bad driving. You should have memorised that part of the route when you stopped at the services.
My satnav is only used in combination with a map (that may be google maps). I plan my route on the map and program the route into the satnav. Its job is to prompt me on the route and to get me back onto the route should I get diverted for some reason.
Last year I let the satnav do all the work when planning my route from Gt Yarmouth to Bedfordshire. Instead of taking me under Norwich it decided to take me on a tour of the best housing estates Norwich has to offer. I thought, ok thats cool, its having me avoid traffic hot spots, its supposed to do that. Till it tried to have me drive down a road that ended in a wall, expecting me to drive through the wall and turn right onto an A road.
This was a TomTom with the latest map. I had to park up in one of the housing estates, bring out the map (combined with google maps to zoom in on some bits) and plot a route using bloody waypoints (why are they called waypoints? I'm not playing Homeworld) to get me on what turned out to be the main A road that it was trying to avoid taking me down, which also turned out to be pretty quiet anyway.
Anyone who does not at least have a large road atlas in the car is asking for trouble.
And just to point out, I'm not 65 and learned to drive in an Imp. I'm 38 and work as a third line IT Systems Technician. I designed a 6502 based computer when I was 12 and cut my teeth on GNU/Linux by booting it off a floppy disc distro. I live in the commandline and the GUI is there to watch youtube and manage the windows the commandlines happen to be in. I have a degree in computer science, have experience programming in everything from C64 basic up to Java, C# and Haskell. I fully understand that satnav on my windscreen yet I happily chose a superior, recent paper map (and google maps lol) over it.
When I chose a map I know exactly why I chose it. The satnav is a tool. It was allowed to dictate to me once, and it blew it big time.
odd isnt it , that our satnaz machines are constantly demanding updates to maps , just in case theres been a new estate built in eastern europe or something . The pickiest of these bots demand to be online at all times when navigating ....
yet in the old days the Ordanance survey maps were updated about once a decade , and people still managed to avoid driving into rivers
This was a TomTom with the latest map
Well, there's your problem, right there.
TomTom is good in some places, but where it is bad it elevates "bad" to a whole new level.
They removed the map for Malta because it was frankly so out of data it must have been scouted about a century ago, and when you then switch to the European map to have at least some help, it was routing you through Malta as if you were driving on the right - no, really, I grabbed screenshots of it. And, of course, routing remained as bad as it used the same database.
I seriously dislike giving Google any data, but it tends to work as long as you have a data connection. The argument for using TomTom is that it's mostly offline other than for traffic data where you share your location to receive the benefit of traffic holdups and automatic reroutes - and its adoptive ETA estimates tend to be reasonably accurate. Its speed advisories, however, are not spectacular either - best just pay attention to the road signs.
However, my work demands redundancy. I do have maps in the car, and I know how to use them. I often plan a route on a map, simply because it gives me the headlines and an overview so I can also plan rest stops and (if it's really long) places to sleep and maybe add a small detour to have a nicer view or come across somewhere I haven't been before. A GPS gets you mainly "just there", a map allows you to make the trip more interesting because zooming out on your GPS isn't quite the same - it's like the difference between reading a multi-page document on screen or in print.
I thought that in Malta they don't drive on the left or the right; they just drive in the shade
Come to think of it, it may be difficult to map because there are more holes than road. It is mildly amusing to see people driving around in noisy low slung sports cars. There isn't a place where you can do more than 80 legally, and the underskirts must grind down pretty quickly after a drive over Maltese roads, grin. You're better off in a Range Rover - that has at least the ability to ignore the potholes because of its large wheels.
Some potholes are only undeep because you're actually driving over the roof of your predecessor :).
Yeah, Google seems to cope best with dynamic routing (apparently all that data slurping is good for something besides their balance sheets), but you can also use it in offline mode as well, you just have to remember to pre-load the map slices you'll need and yer golden in fact I'm midway through updating the maps from Dublin to the in-laws' in Sweden and the map sizes seem manageable so far - 5 gigs or so for most of the UK, Northern France ...
TomTom has idiots working for their route algorithm. Both of ours used to insist on entering Manhattan when going past it.
Once I set waypoints across the Goethals Bridge, Staten Island and the Verrezano Narrows Bridge in order for us to travel from New Jersey to Long Island, NY. The bloody TomTom started yelling about making a left turn, but I told my wife to ignore it. Then I grabbed the device and zoomed the map out so I could see the route, and, yes, it wanted us to go across the bridge, back into NJ, then take the Holland Tunnel into Manhattan, turn right around and take the same tunnel out of Manhattan and the same bridge back onto Staten Island before resuming the route I had planned.
Several times it tried to get us to drive into Washington DC while I was driving around it on I295. Once I was tired enough to let it persuade me and it took me into a road blocked off with concrete blocks and chainlink fence. It looked like Beirut after a particularly successful urban uprising.
There was also that beyond stupid "Keep left ... then ... turn right" thing on multi-lane highways in rush hour.
I switched to Garmin after my wife did when TomTom announced it would no longer support map updates on our models and urged us to buy a new device.
Last week I got a survey from TomTom asking what they could do to improve.
I gave them a point-by-point list, starting with "break the fingers of whoever came up with the route planner".
"it wanted us to go across the bridge, back into NJ, then take the Holland Tunnel into Manhattan, turn right around and take the same tunnel out of Manhattan and the same bridge back onto Staten Island"
Wow! The worst my Garmin has ever done is try to route me off a motorway junction, over the top and back onto the same motorway again. Admittedly, this was shortly after they opened the new section of dual carriageway somewhere between Perth and Aberdeen some years ago and I'd not recently updated the maps data (or it hadn't yet made it into the map data. Then there was that time the A1 was re-routed/upgraded west of Ferrybridge instead of East and the SatNav showed me driving through fields until the next map update. Oh, yes, and the new Queensferry Crossing West of Edinburgh where it thought I was cruising over the Firth of Forth. But these are all minor things considering that the roads and the road signs should take priority for the driver, but the SatNav has never done the sort of silly things yours seems to have done. TomTom seem to have a lot more foibles than Garmin.
I've never had much difficulty with TomTom driving around Europe. I have had Google Maps take me to the wrong place on several occasions.
Waze, which I believe is based on Google maps, somehow contrives to do a good job but I find the TomTom verbal instructions and lane guides (on the newer devices) to be superior.
My experiences in the UK with Tom Tom have been overwhelmingly positive. I prefer it to google maps - which often seems to tell me to leave a roundabout just AFTER I've done it, for example. The ETA predictions from Tom Tom always seem to be accurate and when I've been in a convoy with other cars that "know better" they always seem to arrive about 20 minutes later.
I had that happen just last week. I was heading up 192 in Kissimmee Fl trying to get to Coliseum of Comics. I was about a quater mile from having to make the U-Turn and pull into the car park when the until-then Trusty Garmin ordered us to make a left now. Intrigued, we did.
At one point on the single track road we saw a sign "Road Narrows".
"Where do you keep this GPS?" I asked, after we had finished laughing hysterically.
My wife said "Next to the old TomTom".
"Well there you go! The bloody TomTom has infected the Garmin with Stupid Route Syndrome!" I yelled.
We navigated the chicane with one wheel in each roadside ditch, and drove slowly past what looked like a Buddhist commune c/w outbuildings gilded to eyewatering levels. As more statuary and architecture became apparent we became less sure of the Buddhist attribution and entertained the possibililty of vile cultists a-la HP Lovecraft.
Eventually we reached the place we wanted to be after a needless diversion through banjo-player country, swamps and non-euclidean architecture. The route had, I estimated, cost us at least ten minutes.
My Toyota nav system isn't too bad, but it does seem to lose the plot if it's doing a detour. I manually left the programmed route to avoid a motorway collision I'd caught on the radio, before the RDS had sent the update through. The Sat Nav insisted on routing me back to one junction down from where I had left the road, instead of taking me back on three junctions down, which minimised the junction count as I was on one of those weird parallel-to-the-motorway A roads.
"The Sat Nav insisted on routing me back to one junction down from where I had left the road, instead of taking me back on three junctions down, which minimised the junction count as I was on one of those weird parallel-to-the-motorway A roads."
To be fair, I'd expect any SatNav to do exactly that in the same circumstances. It's routing you based on your destination and settings, eg fastest, shortest etc., absent any other information such as traffic info for the route. You can't really blame the SatNav because the traffic provider hasn't yet got or updated for the accident.
The algorithms don't weight "staying on the same road" or "keep heading in the same general direction" strongly enough, I reckon.
The Toyota system isn't actually that bad, but as I said, it has an annoying tendency to lock your next navigational waypoint if you deviate from the path. But it will recalculate the whole route some times. It's a bit weird like that.
So if it was intending to tell you to stay on the motorway at Junction 12, and you came off at Junction 11, onto a parallel A road that next meets the motorway at Junction 15, it will often attempt to send you back to Junction 11 rather than Junction 15, because J11 is closer to J12.
Not according to recent practical experiment, it don't.
Long Island, Belt Parkway, Verrazano Narrows Bridge, Staten Island Expressway, Goethal's Bridge, I95 South to Sunny Florida, Land of Giant Anthropomorphic Mice. No suggestion that this journey would be improved by a trip into scenic lower Manhattan for an hour at the hands of the NY Traffic Cops (who at rush hour have the job of madly waving you away from the signposted routes to the way out of Manhattan so you can fully savor the chaos).
Return trip unblighted by cries of Garmin GPS desperate to go on sightseeing tour of TriBeCa and SoHo at night.
So: Shenanigans.
You think you had it rough, mate?
I mean, you DID have it rough, but let me drop some hot perspective on you:
The total land area of Great Britain - that means (for the benefit of my fellow Colonialist readers,) England, Scotland AND Wales, together the largest contiguous island making up the bulk of the UK - has a total land volume of 80,823 square miles (209,331 square kilometers.)
The State of New York has a land volume of 54,555 square miles (141,300) alone.
New England - which is (for the benefit of you limey readers,) collectively the States of Maine, Vermont, New Hampshire, Massachusetts, Rhode Island and Connecticut, which is the region of small(er) states to the northeast of New York - has a land volume of 71,992 square miles (186,458 square kilometers).
Pennsylvania, the state to the immediate west of New York, 46,055 square miles (119,283 square kilometers).
Any two of those regions (New England and New York collectively, or NY and PA together,) handily outsize Great Britain. Even if we're generous and add on the entire island of Ireland, 32,595 square miles (84,421 square kilometers,) GB and Ireland together have 113,418 square miles. New England, New York and PA altogether, 172,602.
A long-haul truck driver (lorry driver) stationed in mid-New York State would consider servicing all of those a reasonable task; and that's not even adding in the nearby local states like New Jersey, Virginia, West Virginia, Ohio, Delaware, Maryland, and spaghetti forbid he have the appropriate licenses and requirements to drive up into Ontario.
So yeah, you had it bad. You had the additional complicating factor of all those squirrelly, fiddly medieval roads that just got paved over, but when it comes to road area?
We have road atlasses as thick as an Encyclopedia Britannica, and if you're lucky they cover the major state and county roads.
Australia has a total land area of 7,659,861 sqkm (around 2,969,607 sqm)!!!
Australia's land area is surprisingly close to the total land area - i.e. not counting surface water, of which there's another couple hundred thousand square miles - of the contiguous United States (i.e. the contiguous 48 states plus Washington D.C.).
Obviously if you add in Hawaii and Alaska there's a bit more land area in the US. Really Hawaii is within the margin of error; Alaska adds about another 20% to the US land area.
Of course there are a lot more road miles here than in Australia. Particularly if we count the "forest roads" built and maintained by the US Forest Service, which has around 375,000 miles of them. (That's nearly 8 times as many road miles as the US Interstate system.)
But ya don't have the Ordinance Survey.
You can get maps almost, I say almost as good, but you have to special order them.
Every contract I worked I bought an OS map of the area, even in the outskirts of London where the full beauty of the form cannot be properly realized, from a local WH Smiths or equivalent.
This post has been deleted by its author
As a baby engineer, I once got stuck on the Coventry ring-road after visiting Marconi. Every turn I took eventually brought me back onto the bloody ring-road. I began to seriously suspect I may die there, endless circling 'picturesque' Coventry...
Driving to the football in Coventry with a mate one Saturday we had satnav problems. I knew the route having done it before many times and was directing him. He insisted on keeping the satnav on and was concerned when it kept saying 'recalculating' as we drove on. He stopped when we started getting turn around commands. He looked at the screen and could see according to it we had left the road a few tens of meters ago. We were actually driving through what had been the old Peugeot factory (now a new development) and his shiny new car had outdated info. He didn't like the idea we'd gone off the 'map' and was going to turn round before I persuaded him to go on. As soon as we reached what the thing referred to as road it stopped whining. I said he'd follow the bloody thing into the sea if it told him to. I was only half joking.
You can't have been much cop if you couldn't glance at a map of the Coventry ring road area (like the A-Z) and simply count off the roundabouts as you drove until you got to the one you needed.
And the only way to get off the ring road and end up back on it again on Gulson Road, Fosehill Road, Radford Road or the road that went past the Memorial Park whose name I forget (it's been 30 years since I was there) is if you stupidly made a u-turn or several consecutive right or consecutive left turns.
Bah and double bah on whiny people with no sense of direction.
Yeah, I remember the bad old days of having to reply on mapping that had rivers, roads, contours, woods, forests, canals, pylons, quarries, parks, built-up areas, tide marks, railways, bus stations, railways stations, cuttings, embankments, bridges, footpaths, bridleways, marshes, lakes, steeples, all in high contrast with colours picked to be distinct from each other at a glance. Thank god we got rid of all that.
> I've since heard that it's a pretty common thing. Or was, before sat-navs became widely used.
yep, did the same somewhere in kent (I forget where) before the days of ubiquitous sat navs. The ex and I were lost for about an hour, then found the exit we wanted. 2 miles from where we'd started
"As a baby engineer, I once got stuck on the Coventry ring-road after visiting Marconi. Every turn I took eventually brought me back onto the bloody ring-road. I began to seriously suspect I may die there, endless circling 'picturesque' Coventry..."
I drove on the Coventry ring road once. Never again. That thing is a nightmare.
Indeed and I can carry maps in my head. I once drove my aged mother from here in Dundee to my cousin's place in the countryside south of Macclesfield entirely on having looked up the route and used Google Earth to Street View the actual house so I would recognise it. Worked very well, there, and back again.
I frequently put maps in my head. On any number of occasions I have run a new running route by memorising the map, including in foreign cities. It is hard to run while looking at a map on your phone and I refuse to use earphones while running and often I don't even take my phone with me (extra weight, the arm wallet is uncomfortable).
Apart from mistaking a frozen stream for a path in a very snowbound park in Ljubljana I have not gone wrong.
We have had any number of cautionary tales of relying on a satnav instead of using our brains. If you never navigate by your own wits and knowledge that ability will atrophy. Google Earth is damned useful but I have a drawer full of paper maps which I also use as they encode details Earth often doesn't.
Which means having a defined topology. Now I prefer to "hop" from "router" to "router". I know my destination, with varying degrees of precision, so I just head to the first local gateway node which puts me onto the correct physical connection, displaying a "local" route table, and pointing to the pathway where I can find another source of authority (e.g. The North, The South).
Using known points of reference to trace a longer route is called "piloting", and is one of the two forms of navigation used since navigation was invented (the other one is called "dead reckoning" and is how ships discovered new places by crashing into them before the advent of the chronometer and an established meridian baseline against which to calibrate it).
It has been around for centuries, and you've been doing it since you could walk. Neat, eh?
No routers needed.
Piloting still needs a list of waypoints for the entire journey, and needs a map that covers everything.
With a suitably formatted address, one could navigate the UK road network based on purely that alone. OK, finding your source of authority once you've reached locale of small suburb of town on the outskirts of city in the county/region means that you DO have to stop and ask someone for directions... and therein of course lies the problem. Who likes asking for directions?
We used to call those waypoints "road signs" when I used to drive all over the UK in search of work and drink.
If you need an address that works always, you need (in the UK) an OS Map and the grid reference of where you are going. Or you could use Lat. and Long. and use a compass, sextant and chronometer (at which point I reckon the GPS is actually the better deal given the cost of a serviceable sextant and the cloud cover this time of year).
For most car-accessible places you don't need such things. You takes your compass and do the longest bits by poor-man's dead reckoning (I'll drive south-west until I see something I know is near where I need to be) then pilot using terrain features pulled off the map until you are there.
If you need an internet metaphor to do the job, no problem, it's your head you are working with.
Hey adnim: I always backup my GPS with a map generated by Google. That way when the GPS tries to take the shortest route in Italy you are not trying to drive up or down a donkey trail in the mountains or an alley built to contain an invading enemy in the heart of a city. I own some real paper maps as well. It's fun to actually see the big picture.
"I use a map."
There's an old Joke: "How do you drive an engineer crazy? Tie them to a chair and force them to watch you fold a map the wrong way."
Last week, for the very first time, I had to explain that joke to a young engineer that did not know what a fold-up street map was. When the zombie apocalypse comes he'll be staring teary-eyed at his dead GPS while us old farts have stolen enough fuel to drive to Sanctuary.
"I use a map. Was a mobile service engineer in a previous life... have A-Z of most UK cities. I don't own a satnav. My 17 year old car don't have one either. If ya can't do it yourself you is someone's bitch."
I still am a mobile service engineer, after 20 years (I *like* the job, so shoot me!) and I used to have streetmaps for most cities and towns in the UK, collected over a few years. Anyone still using maps for that sort of job when SatNavs are available is an idiot as far as I'm concerned. Obviously I mean a proper, dedicated device with regular map updates and live traffic information, not some cheap bit of tat that doesn't even have a proper postcode database and the maps are 5+ years out of date.
Interesting.
Point is, if it all goes tits up... I will know where I am, where I am going and how to get there.. Unless I am on a plane, or a boat or in some other place that requires GPS to know where it is at.
I did not say GPS was useless nor am I denying GPS has many uses. I think it is pretty awesome. However, I don't need or use it. It is another thinking bypass for most people. I prefer my thinking NOT to be bypassed.
It seems 63 people missed my point. I am not a sailor. I do not provide location services to others. I don't need GPS.
For a site that is for IT professionals, I am surprised that my post was so misinterpreted by so many.
Meh.
My faith in the future of IT and the people that will provide services based on IT is now at an all time low.
So, you can use a map etc. Fine. Good for you. Doesn’t mean it’s inconceivable you’d be affected. GPS is also used as an extremely accurate method of timing in industries that require it, such as banking and the financial sector.
Sadly , while I am not the sort of person that reads things like this and immediately thinks about investing everything I own in canned food and shotguns, and I am confident the finance industries have taken the relevant steps to mitigate this already (while they do seem to have let the IT on the consumer side get old, The IT used in the trading systems is usually state of the art (after all, if a stock transaction is delayed by even a millisecond, it can lose the company tens of millions of pounds) it is still possible that any member of the population could be affected.
Why on earth would you send 128 bit data packets for GPS? Complete waste of bandwidth. Real issue is that people are using data for things it isn't really for, but software and firmware was designed to handle this many many years ago - you'd have to be relying on some ancient kit for this to be a real issue.
Cool. An 8-bit header. With a reserved bit that indicates there's another bit to follow.
Then you can pack any number of bits. 1 packet of 7 to start, an extended packet taking you to 14 (beyond even their CURRENT plans to change GPS - or were you reading the article and took away that they can't change the protocol?), then maybe another. If you follow any kind of otehr long-established protocol, they then have often fast-bursts. First 7-bits with an indicator in the last biy, which says that the next byte is at, say, double-rate, or compressed. Old devices ignore it, get 7-bit accuracy. New ones read it and get twice as much.
The most inefficient kind of 7-bit packing would give you 7-bit accuracy in the first 8-bits. 14 in the first 16. 21 in the first 24. 28 in the first 32... 56 in the first 64-bits.
It's not about the specifics of GPS - it's about how you go about designing the protocol for future expansion rather than do the literal "640k" gag. And people were doing this WAY BEFORE the 70's because they had seen some of the fastest advances with the tiniest hardware and it became necessary.
It's not about whether GPS *does* it. It's about why it wasn't designed to do it, why you would continue to allow it (if you don't have a single "reserved for extended use field", you are literally stuffed - even TCP did that for ECN), and why - if you ARE extending it - you don't allow for further extension to your own.
"Seriously, in this day and age?"
I hear what you are saying but we are talking about ancient tech designs here. Still use a mobile phone to make calls and send SMS's etc? You'd be surprised just how 1990's the call/text encryption on those things are, yet we are sold them for a grand.
Do you listen to DAB radio? Lovely bit of 80's digital tech that is (not).
After a while you get used to it and realise that nothing we use today is bleeding edge. The stuff that is bleeding edge is the stuff that dont work properly or is still being talked about in universities and showed off at conventions.
Seriously, in this day and age?
Why not? Arguably, the field is too big, so it isn't wrapping sufficiently frequently to catch out lazy software developers before products actually ship. It's only going to be an insurmountable issue if the satellitle clock differs by 10 years, at which point you won't know whether it's a decade behind or ahead. That's more than enough margin, I'd have thought. Those who can't deal with wrapping integers should confine their professional activities to making web pages more annoying.
If you are going to rollover there is something to be said for _less_ bits and rolling over more often. Imagine a system with 3 or 4 digit day counters that rollover. Systems with 27 years until the first rollover are more likely to end up with some broken code, perhaps written by consultants, or "temporary hacks" that need fixing before rollover. Even if they dont have bugs they have 27 years of code without testing the rollover so when it looms they will have a y2k type effort to be on the safe side.
Those with 3 year rollovers will have code tested earlier, more often: less bugs and cheaper to fix.
"Seriously, in this day and age?
Stop messing about and use 64- or 128-bit for everything"
But GPS originated in the 1970s, and also has the constraint of needing to transmit data from around 20,200 km altitude to down here using not very powerful transmitters. That means there's an engineering need to minimise the amount of data transmitted, and also one must consider backwards compatibility.
Going up to 13 bits from 10 bits in this case makes sense: it extends the week counter from around 20 years life to 8 times that, i.e., around 160 years, by which time I dare say the current GPS system of operation will be long gone. Yes, you could make it 128 bit and that would last long beyond the heat death of not just this universe, but some beyond (although I suspect relativistic effects involved in the collapse of one universe followed by a subsequent big bang might well invalidate all assumptions regarding the passage of time). I'm not convinced that's entirely useful.
For older devices, however, a firmware patch is going to be necessary to handle the week epoch rollover, and GPS.gov recommends anyone who is unsure about their readiness for the turnover, particularly enterprises, consult the manufacturer to make sure they have the proper updates and protections in place.
What do you think Garmin and TomTom are going to do, put out updates or sell the most they've sold in years?
I stopped using Garmin because the maps are STILL about 15 years out of date in my area, and the company is a complete Apple-style we-know-better-than-you dick.
I stopped using TomTom after my device with lifetime maps stopped functioning because the maps got too big, and the response from TomTom support was basically "sucks to be you, buy a new device" and they refused to re-section the maps to fit... so "lifetime" ended up being about 11 months.
Maybe they picked an Apple and removed the SD slot in recent models. With phones reducing their sales, I'm quite sure they were tempted to plan more obsolescence into the devices. Anyway, they should be careful about how long it takes for a device to become "obsolete" - if they're greed and kill devices too early, they will reduce their sales even more.
Definitely possible on the Garmin at least - I still keep an old Garmin 200-something as a backup in the car running OSM-based maps of most of Europe that I could grab pre-packaged from several sites online... Also, I guess this Nuvi's going to the landfill, because I have the sneaking suspicion Garmin won't be releasing any updates for a '08 device
@gene "Garmin because the maps are STILL about 15 years out of date in my area"
i complained to Garmin about this, whilst driving on a motorway in Spain that didnt exist on my 'garmin with lifetime-update maps(tm)', their response was extremely silly
"We are not responsible for errors caused by our Map Provider, "We simply provide the hardware platform for their maps" (praphrased slightly - but I got the message to foad)
So I now use Slurp, sorry Google, as at least it is somewhat useful to use all the stolen (crowdsourced) data from all the android handset tracking, with a bit of an overlay.
I stopped using TomTom after my device with lifetime maps stopped functioning because the maps got too big, and the response from TomTom support was basically "sucks to be you, buy a new device" and they refused to re-section the maps to fit... so "lifetime" ended up being about 11 months.
That is pretty disgraceful. My TomTom did NOT have lifetime maps, so I wasn't quite so shocked as you must have been when they told me, recently, that it was never going to be updated again because the maps had got too big. I just thought "That's why I've been using an SD card instead of the internal storage for the last several years -- I thought that's why they made these things expandable?"
Now I realize that they may have been trying to make me upgrade to one that could count more than 1k weeks, without admitting the laxity of their programming. Time will tell ...
Absolutely they are used.
My phone has a GPS, and also gets FM traffic updates, so it doesn't require data. It's also 5" in size, and has a couple of features cell phones don't seem to have.
But more importantly, it (a) is sunlight readable, and has a non-reflective screen, something that no phone seems to have. And (b) it works based on pressure points, so it doesn't require skin contact. That may not sound like much, but in a Canadian winter, the ability to use your GPS with your gloves on is a selling point.
And that's just consumer GPS. Things like specialty trucker GPS don't really have cell phone apps that can replace them yet.
I assume they mean SatNav, but clarification from the OP would be welcome...
Sigh, yes. Can think, can't write to save my life, it appears.
Yes, the Satnav (a TomTom model) is 5", has a sunlight readable screen, does not require a data plan to update traffic, and can be used with gloves. None of which is true for my cell phone.
I paid real money (only time so far) for an Android mapping app called CoPilot. Around 50 notes to give routing based on length, height, width, weight of your vehicle.
I bought this for use with a caravan. A dedicated SatNav with the same features costs around £350. They do a trucker version as well IIRC.
Free map updates as well which means I won't be paying for any more map updates for my venerable TomTom.
Intelligent routing is a must have when you have a large outfit. Being stuck down a narrow lane which isn't wide enough to turn in is a fate usually reported for articulated lorries with dodgy (cheap car?) SatNavs but can't be fun with a caravan either.
I'm not sure a road atlas or paper map gives you road width, bridge height, or turning radius for tight bends either.
I have checked (briefly) and you do get different routes depending on setup.
TL;DR they do trucker SatNav for phones.
Does anyone still use them?
I do. When I head off into the hills for a few days walking and wild camping my primary navigation is map and compass but I stick my GPS in the rucksack just in case of cloud, white-out or navigational incompetence. No need to worry about keeping it charged cos it keeps charge for ages and I carry spare batteries (AA). No need to worry about enough signal to get a map. No need for a subscription for decent outdoor maps. It's water resistant, drop proof and I can work it with thick gloves on and see it in bright sunshine or pitch blackness. It carries a full set of 1/25000 UK maps and the compass always points north - unlike my iPhone's which points in random directions.
My phone is OK for finding the pub in a strange city and working out which bus to get, but I wouldn't want to have to rely on it to get off a Scottish mountain when the clag comes down.
>cos it keeps charge for ages and I carry spare batteries (AA). No need to worry about enough signal to get a map. No need for a subscription for decent outdoor maps. It's water resistant, drop proof and I can work it with thick gloves on and see it in bright sunshine or pitch blackness. It carries a full set of 1/25000 UK maps and the compass always points north
That's exactly what I require. This is a bit cheeky, but care to plug that device? Outdoor exploring is great with a map, but sometimes finding the pub becomes the overwhelming priority...
Sir, I'm surprised you can still switch on a nexus 5 without the battery dying! I recall GPS on some batches was really bad indeed, which made playing Ingress all the more fun/rage inducing.
Just checked the fairly ancient network NTP servers and thankfully are both patched and the manufacturer still exists/supports them - frankly I am staggered.
One of the reasons I bought it was that while not officially replaceable, battery replacement is fairly straightforward.
Which is fortunate. One day, when it was about four years old, the original battery began to fail and attempted to push the screen out the front of the thing.
Yes. Unlike phones, they are dedicated devices and do one thing only, usually better. For example, they are not interrupted by phone calls - which could be a real issue if you're navigating in some complex area., and have no background processes that could slow down processing.
I also like small phones, but use a GPS device with a far larger screen. When they are properly car-integrated, they can also get more data - i.e. the car speed when the GPS signal is not good enough, and also can use an external antenna for better reception.
You may also have maps for many countries, and won't need a data connection when abroad. Now most offer "lifetime" maps, if you drive quite often to new destinations, a dedicated GPS is usually better than a phone.
Yes, pilots. Many Dell Streak 5s became XCSoar gliding flight management computers because of their transflective screen. ie a readable screen even in Oz sunlight at 10,000 feet. In my limited experience only a Sony Experia gets close at cost of needing its own lithium battery in cargo storage to keep screen brightness up. So with all the brands trying to differentiate them selves, all they do is become slimmer, greasier and more fragile when falling out of pockets instead of something many people have expressed an interest in, a daylight readable screen.
As for Garmin maps, lets just say in my experience in emergency services other brands have at least located significant roads between rural towns.
As for another date of doom, damn, more hardware checking in two months to see which of my devices get trashed, Finally a reason for consumer gadget buying.
+ an up to date map because the world they represent changes over time
+ a calibrated compass because the magnetic north pole has been going walkabout
+ an accurate watch (I prefer a manual wind with no date and a sweep hand)
And yes, I can get almost where I want to be without those or a GPS but sometimes almost isn't good enough.
I don't like having "The device that stops be getting lost if the weather gets really bad" and "The device I would use to communicate with emergency services if I become lost" being the same device.
I can see how for navigating towns or driving, a phone can do the job (although I still use a dedicated GPS for driving) but for walking, a dedicated GPS is a very sensible investment.
"Does anyone still use them? Phones have been able to do the job just as well for about 10 years now."
Because smart phones are completely accessible by everyone, and coverage across the UK is perfect ... yes, some of us don't own smartphones because they live in an area with non-existant mobile signal (and my elderly relatives can work a sat nav, but not a smartphone due to ico sizing, etc).
In fairness, to mitigate signal issues you can use offline maps from various vendors. OSMAnd or Maps.me are particularly handy, but in a pinch there's HERE if you have the processing power.
Online maps are at best a so-so technology choice if you are going into the wilderness (as are electronic maps without a backup in general) but there are alternatives.
I still use one. I prefer to have a dedicated sat-nav when I'm travelling as it keeps my mobile free for playing music, taking hands-free calls and such forth. I also have one specially designed for motorcycling that can be operated whilst wearing biking gloves - no way I'm strapping my mobile phone to my handlebars!
Not knocking those who use their phones as sat-navs, but not everyone likes to rely on their mobile for everything...
Many people still use them as they are made more rugged than phones and work when there is no mobile signal.
Typically those using GPS devices like these are likely orienteering / hill walking and using the Garmin like a high tech compass to help get bearings to waypoints.
"Does anyone still use them? Phones have been able to do the job just as well for about 10 years now."
----------------- ----------------- --------------
On the contrary, phones still don't do the job nearly as well.
1. A phone plan with data will add enough to your costs to buy a new GPS every six months... while I typically replace them every five years - a 90% cost reduction.
2. Phone coverage is far from complete. A quick look at coverage maps for the three biggest national providers shows probably less than 10% coverage overall, definitely less than 20%. Long stretches of significant highways have no coverage at all for hundreds of kilometers.
3. In a major emergency, particularly one that is caused by or triggers a prolonged power failure, the cell networks will go down, as soon as the cell sites exhaust their batteries. I am not even sure that all of them have batteries, but I am sure that none will last past 24 hours. The 2003 power failure blacked out 55 million people for two days. Sooner or later a large solar flare will not miss, and a large chunk of the planetary electrical grid is almost certain to fail. A dedicated GPS, particularly one stored in a metal box, will survive. The satellites may well be sufficiently hardened. Millions of kilometers of power lines? The longer the line, the higher the induced energy. The Carrington event of 1859 gives some indication of the power of such coronal mass ejections, but mid 19th century telegraphs are several orders of magnitude tougher than semiconductor electronics, or a modern integrated electrical distribution net. Battery powered GPS units can function off AA alkaline cells, and as long as enough satellites survive (between GPS, GLONASS, Beidou-2 and Galileo, 'enough' may mean those in the shadow of the earth at the time of the peak of the event) it won't need anything else to work.
4. Using a cell phone as a GPS means reporting your position constantly to your phone network, the maker of the phone's OS, the navigation app authors, the authors of any other location aware app that is running, including browser resident scripts and sites, anyone who can access the stored location tracking data in the phone, and all their business partners and data customers,
5. The interface of a well designed automotive GPS is possibly the best interface from a functional and human engineering point of view that I have seen, for a relatively simple function.
6. A good automotive GPS is also extremely temperature tolerant, working well at temperatures from -35 to +45, and probably both colder and warmer than that, though I'd have to visit another part of the country to get below -35. I was quite pleased the first time I fired up my GPS in a -25 degree vehicle and had it light up and be ready to roll in five seconds. Ten years ago, the screen would probably not have functioned. I imagine some of the truck GPS systems will work down to -50 or so, but I really don't want to try that out. If you need to navigate in even colder weather, you're going to have to do a lot of planning, and not just about GPS navigation - I don't have a clue what you'd need to take into account, and you may be far enough north that things like aurora effects or high latitude may begin to challenge the technology, as well.
7. Hand held GPS units - the ones that usually take AA batteries - are almost all water resistant, rated to survive submersion to 1 metre for 30 minutes, if not having a higher rating (IIRC the next step is 2 metres for an hour). Extreme rain or dropping it in a puddle, or a stream you can wade, or the wet bottom of a boat is not an issue if you can get it out again. Not so much with cell phones.
8. If you need further flexibility and reliability there are several ways to charge NiMh AA batteries, some of them needing neither an electrical grid nor a vehicle power outlet. As a bonus, the same type of batteries can power your flashlight and your UHF/VHF transceiver.
Digression: Indeed, in emergency situations the co-ordination centres will have stacks of AA alkaline cells so you can arm up with spares when heading out... primarily for radios, but they'll do for the other uses as well. That's why, while you may run your radio on rechargeable packs most of the time, there should be a battery case that takes AA cells among your accessories. You need the radios because if it gets bad enough, both the cell and land phone nets will be down, as will many of the trunked radio systems that utilities and government agencies, including police, use, due to power exhaustion at the cell and repeater sites. At that point much of the remaining medium and long distance (anything farther than the reach of a site to site contact, or an untrunked repeater that has power) communication capacity will be military or amateur operators.
Who uses Garmin or TomTom GPS's these days?
They pissed off their customers by spamming them every time they turned on the device telling them to update the maps, or follow road rules, etc, etc and made you accept their T&C's before you could even start using the thing, sometimes I had to accept 2 or 3 interruption messages before I could start using my tomtom.
My Android phone ALWAYS has up to date maps, no extra charge for traffic, and it just works, even with voice controls. No stupid prompts to acknowledge every time I want to use it...
Who uses Garmin or TomTom GPS's these days?
I do. Yes, my Garmin does occasionally point out that maps are getting bit out of date, but not every time it is turned on. Yes, there is one "Do not use while driving" or something like that warning that pops up on power on. Tap once on screen and its good to go.
I can update the maps when I want (if new ones are available). It gets traffic via built-in FM etc. It even works as hands free kit for the phone. Oh, and it doesn't phone home.
It does of course keep some limited history on where you've been, average/top speed but those are easy enough to reset.
Ephemeris data required
But won't that also be using the same length of week number ? At least, I'd have hoped they'd be consistent.
If they didn't do something stupid and use different formats, then the device would just think it's at some time in the past - and so are all the satellites. The time of day may be wrong (not sure why as it would only be the week that's wrong) but the position should still be correct. Might be some problems around rollover time if the GPS uses a different format internally - but that's likely to just cause it to have to do a completely cold start and find a satellite to download data from again.
And the ephemeris isn't stored. It's calculated. I think. Yes, pretty sure, it's calculated.
-------------
It's not. The reason for the long wait to start navigating on a cold start with older models was to download the ephemeris. Clearly the newer ones cache the latest ephemeris in non-volatile memory, as it is now a much faster startup.... a matter of seconds rather than minutes.
Isn't it transmitted as part of the signal? It contains four hours worth of position information and the precise time/date is used locally to calculate an interpolated position for that exact moment. The almanac is the thing that is stored to speed startup. That's valid for 180 days, and again the rough position is interpolated by calculations which use date/time.
You know that satellite A is 200km further from you than satellite B and 20,000km further than C and 115,000km than D and D is 88,000 away from C and 30,000 from B and B is 45,700 from C but unless you know where A, B, C and D are supposed to be at that date and time then you could be anywhere on the planet.
Geostationary satellites can only be put in a restricted "belt" around the Earth equator - just having satellites there won't be useful to calculate your position in different places around the globe - you need to "triangulate" (let's keep it simple) using signals coming from different directions to minimize errors.
Moreover, depending on latitude they would appear quite low in the sky, and their signal would be easily blocked by mountains and buildings.
Maybe. The last time this happened, apparently some devices got very confused about their ephemeris calculations when the week counter rolled over and started looking for satellites that weren't in view while ignoring the ones that were, meaning that they lost GPS lock completely. There's also some complication with the individual satellites having counters that roll over at a very slightly different time from the main GPS week counter. Devices doing a clean start after the rollover completes should be able to navigate but with a wrong date.
Also, all modern devices are meant to be able to handle the week rollover properly so long as it's less than 1024 weeks after the firmware was built. So it's mostly a question of what interesting bugs they have in their handling of it.
Yes, it was the rollover - that happened last time in 1999 as the poster said. They are not suggesting that this is the Y2K issue.
I clearly remember the GPS week rollover being well discussed (e.g. on SlashDot when that was popular) the last time that it happened, and can't believe that there are devices produced later than this that could be vulnerable. Poor effort by device manufacturers if actually the case.
Yes, I remember the panic - but it turned into a non-event. Even my old Garmin GPS II+ handheld satnavs sailed through it with no problem and I fully expect them to do the same this time.
IIRC the only GPS receivers that didn't make it through last time were all in Tokyo taxis. This was certainly bad news for them, and here's what I was told about that at the last GPS roll-over:
Tokyo was heavily firebombed in WW2, and since close-packed wood and paper houses burn really well, the city planners must have started again from scratch. The rebuild had to be fast because almost all houses were destroyed and winters are cold, so in the hurry to rehouse people, the houses were numbered in the order they were built in each block. As a result the numbers along a street are in pseudo-random order, so Tokyo taxi drivers really need a working GPS with a map showing house numbers. Either that, or they'd need years to do 'The Tokyo Knowledge', assuming there is one, before getting their license.
Martin Gregorie: Tokyo was heavily firebombed in WW2 — Those who win wars decide what are war-crimes are.
Operation Meetinghouse had 100,000 people killed in one evening. Although the Americans were trying to bring the war in Japan to a speedy end, they ended up showing the world that they were as good as the Japanese in Nanking when it came to mass slaughter.
On balance they did it from the air. They didn't get on the ground and get personal with the locals. Nor was there a recognised ceasefire zone for civilians that was promptly ignored.
Its not like we Brits are innocent either what with the firebombing of Dresden. All sides committed atrocities but some where more horrific than others.
There's a good argument for Hiroshima and Nagasaki to be war crimes. Both were committed after Japan had started looking to make peace, but prior to surrender. Documents at the time show that the real reason for Hiroshima wasn't to stop the war but as a warning to Russia. Nagasaki was to show they could do it more than once. However, the plan backfired and only spurred on Russia's attempts to build their own bomb.
Sure, Japan was ready to surrender just like Germany was when Hitler was already in the bunker - some people were trying to contact US - but they too didn't want to accept an unconditional surrender, thus were trying to obtain what was not possible, - and moreover they weren't those with much power to force the military to surrender. There was no warranty Japan would have surrendered. Remember that even after the bombs the emperor had to force the unconditional surrender, and the imperial palace was attacked to hinder it. And the Hirohito recording never said "surrender".
Add that Japanese naively tried contacts through Russia, and Stalin deliberately downplayed and dismissed the approaches because he wanted to gain as much as it could before Japan surrendered. Also, if US had been forced to invade Japan, it would have meant they would have needed to weaken the European front.
Japan had already plans to resist a US invasion on the main islands. It would have been a huge massacre on both sides.
Russia was already stealing US bomb secrets, it didn't built it because it was launched on Japan. It would have built it as soon as possible anyway - and how fast they could show how much they spied - and it had stolen already a much needed B-29 to copy.
The Hiroshima/Nagasaki bombs cut the war suddenly, and Stalin plans were broken, as he declared war on Japan "too late" - a fact that tells that even Stalin wasn't able to assess the situation and didn't believe Japan would have surrendered quickly, or would have declared war well before.They waited until August 8, which was the last available day - they were bound to declare war on Japan within three months after Germany surrender.
The Nagasaki bomb was the result of different events. Japan didn't understand quickly enough what really hit it, and on the other side, US military was eager to try the "other" bomb - remember the two bombs weren't identical - the first one used uranium, the second plutonium.
Part of the reason why the Axis powers didn't surrender until they did is because unconditional surrender on the scale of great powers is absurd; we only consider it otherwise because the Allies were willing to throw millions of extra lives into the grinder to achieve it and we like to pretend we came out smelling like roses. Especially after Italy got a deal, there was no logical reason at the time to assume the Allies couldn't be brought to the negotiating table.
This is OTT for the "First GPS Millenium", but...
Yes, I'm well aware of who firebombed Tokyo and why. I'm also aware that some other cities, among them Hiroshima and Nagasaki, were deliberately not bombed with conventional weapons so that the strategists could get a better assessment of the destructive abilities of atomic weapons.
There's a lot about this in Richard Rhodes book "The Making Of The Atomic Bomb".
If you want to understand the scientific and political background to the Manhattan Project as well as learning about the politicians, military and scientists involved and other country's atomic weapon projects (yes, Japan had one as well), this is the book to read. Another title for it could have been "The History of Atomic Physics: 1995-1945", possibly more accurate but a lot less likely to attract readers.
I think the Japanese addressing system is just a way of keeping local post offices in business, as there's usually one on every super-block with a detailed map on the wall telling you where everything is.
The moment I was walking down a street and thought to myself: "ok, cross this road and it will be Meguru 2-22" and I spotted the street post confirming I was right was the moment you pat yourself on the back and feel like a local.
It probably helped a lot that it happened in 1999 when Y2K preparations were in full force, so the GPS thing could be handled as part of that overall project (just with an earlier drop dead date)
Happening in the middle of 2019, it is going to slip through the radar in a lot more places. The good news is the next time it happens it can be included as part of the Y2038 projects :)
An earlier version of ICD/IS-200 has been in force since 2006.
The first iPhone (and Google Maps IIRC) were launched in 2007.
The vast majority of GPS-based services have been invented/launched after the initial ratification of this spec.
This means two things:
1) *probably* most devices sold post-2007 will be compliant & therefore be fine;
2) the devices that aren't compliant will have been manufactured prior to 2006, and ideally either have been end-of-line'd or upgraded in enterprise environments that rely on this sort of thing.
Odds of this affecting you personally in any way (consumer, enterprise or sysadmin): I'm pegging at roughly 0.01%.
I've got various ancient GPS receivers, usually as modules rather than complete devices.
Most of them are fine with the one or two rollovers they've already seen. Some work OK once given the current date.
They all tend to give accurate position and 1pps signal (if present) - it's only the UTC time they get wrong, and then it's out by exactly 1024 weeks. The actual GPS calculations are all done using that 1024 week cycle and UTC time is just a spinoff.
Some roll over at a date that isn't at the 10-24 week epoch : they know the data can't be before they were built, so anything in the earlier part of the cycle is automatically considered the 'next' cycle. I think NTP (where it's using GPS time as a stratum 0 reference) also adds the 1024 weeks if the receiver is giving it a stupid date.
There's a good chance there will be no problem on anything but the very oldest receivers, and if you do find one reading wrongly, set the data manually (if that's an option) - it may work it out.
The GPS system doesn't recognise or use leap seconds. If you have an NTP device giving out correct UTC info from a GPS time reference it must also have a way to get leap second data from somewhere (presumably the Internet.)
Leap seconds can't be predicted, they result from measurement of the Earth's spin. Presumably a committee then issues a decree on vellum, or something. In theory leap seconds can be added or subtracted but so far they have all been added. Last time I looked UTC was about 12 seconds ahead of GPS but it may well be more now.
(Just checked - the committee is run by the International Earth Rotation Service. I presume they have a very large brass key to wind the old girl up when she is slowing down.)
GPS.gov also notes that the new CNAV and MNAV message formats will use a 13-bit week number to solve the epoch migraine right up until the planet becomes uninhabitable via climate change or we all blow ourselves up.
--- Given it was designed in the 70's they probably thought the same about a 20 year cycle :)
Even better than adding 3 bits would be to take 3 bits away, because then it would roll over every 2.5 years and that's the sort of frequency that would force people to handle the roll-over properly, with tests, so a system that rolled over every 2.5 years would probably work for ever.
A device that had been stuck in a cupboard for a long time, or had a factory reset, would not be able to tell the user the date unless the user first told it the year. But it could still navigate without knowing the date. Adding bits is a stupid temporary fix compared to handling roll-overs properly.
If I was the developer, then I would figure out that a timestamp can never be older than the device itself (if you buy a TomTom today, it will never receive dates from Jan 2019 or earlier, so we can just use a window and hope the device doesn't last for 1023 weeks :-) Probably changing the ROMs going into manufacturing once ever year or even every five years is fine.
But even without this, you'd only get into trouble if some satellites are in week 1023 and others in the next week (week 1024 but transmitted as 0). But that state lasts for less than a tenth of a second.
Glad I have my own verified compliant GPS NTP time server in my home.
I wonder how many government agencies, car navigation systems, and aircraft navigation systems will go Tango Uniform (TU) on 4/6.
What about the banking industry? At least it’s a Saturday so the stock market won’t crash.
I guess you don't use a GPS based time server or understand location determination is critical to determining GPS time. Good to know in case your resume comes across my desk.
https://www.masterclock.com/company/masterclock-inc-blog/finra-clock-synchronization-regulations-are-you-in-compliance
...actively started working on this about a decade ago.
I worked for a geo-mapping company 5 years ago that was deploying the associated software updates and sending out customer bulletins during that time. For the major customers that were heavily GPS-reliant (law enforcement, fire services, etc), we knew what kit they had using our software and provided specific advice whether or not their devices would be affected.
Here in Australia, the regulations make it difficult or sometimes impossible to use the GPS app in your phone for navigating whilst driving. It is illegal in most states to touch a smartphone whilst you are in the driver's seat or the key is in the ignition. This includes anything to do with the GPS app. So, unless you key in the address, route and all other information before getting into the car, you risk a fine. Plenty of drivers have been prosecuted when they stopped in a parking space to set the GPS or to answer the phone, but neglected to first remove the ignition key and get out of the driver's seat. I have been told by a Vicroads official (the government body controlling road use in Victoria state) that if my phone was to catch fire whilst I was driving, then I am forbidden to touch it and must continue driving until I find a safe place to pull off the road.
It's even more strict for probationary drivers. They are forbidden to use a smartphone for any purpose whatever whilst driving. Thus they can be prosecuted even if they set up the GPS before starting, or get a passenger to look after it. (They're still "using" the phone by listening to to the directions from theGPS)
Why these crazy rules? In most states, fines from drivers are an important source of state revenue and they will do anything to keep up this source of free income (justifying it on "safety" grounds). Apparently they don't get enough money by allowing bushes to grow up in front of speed limit signs, whilst ensuring that the traffic camera further down the road has a clear view.
So much inaccuracy here. It is not illegal to touch a smartphone if it's held securely in a proper holder (by proper, I mean not some homemade bodge). Fully licenced drivers can use mounted smartphones as "navigation aids" e.g. GPS, while driving, including touching it. Probabationary drivers (i.e. for the first 3 years of independent driving), cannot use a smartphone at all, (mounted or in hand), including as a GPS. But they can use dedicated GPS units.
Have to correct you on this. It is illegal (at least in Victoria) to touch a smartphone in any way whilst in control of a car (i.e. being in the driver's seat, or having keys in the ignition). That applies even if it's in a proper holder. If it's not in a commercially-provided holder (i.e. not a home-made one) then you can't use it at all. I have had this confirmed by Vicroads. Only last week, there was a report in "The Age" of a Uber driver who was fined $400 because he pressed the "accept" button on the Uber app whilst in his car. (The newspaper was campaigning for a law change to allow Uber drivers to do this). If you've been touching your phone whilst driving and haven't been caught, then you're just lucky.
Regarding probationary drivers, anything that can connect to the mobile network is deemed to be a mobile phone. So your dedicated GPS unit would only be exempt if you could prove that it had no means of connecting to the mobile network (e.g. for updates etc). You'd probably need to take a high-powered lawyer and a technical expert into the courtroom with you to prove that. Good luck trying to explain that to the cop at a roadside stop.
This is getting way off-topic, but since you keep referring to Vicroads, here it is, straight from the horse's mouth (https://www.vicroads.vic.gov.au/safety-and-road-rules/driver-safety/mobile-phones-and-driving):
"Fully licensed car drivers"
"Using a mobile phone while driving is prohibited, *except* to make or receive a phone call, to use its audio/music functions or *perform a navigational (GPS)* or intelligent highway vehicle system (in vehicle warning system) function:
* is *secured in a commercially designed holder* fixed to the vehicle, or
* can be operated by the driver without touching any part of the phone, and the phone is not resting on any part of the driver's body." (Emphasis added).
This is all consistent with ARR I quoted (it's rule 300), and the subject of your first post. The Uber app is not GPS related, and is banned.
There are plenty of GPS units with no other function. I have a friend on her P's who's using one no problems. She resented having to buy a separate device, though, when her smartphone would work fine. This rule seems arbitary, but there is always the enforceability criteria; if a road-side cop pulls you over, it's hard to prove what you were using your device for, so a GPS only device rule is the way to be sure for P-platers.
So your dedicated GPS unit would only be exempt if you could prove that it had no means of connecting to the mobile network (e.g. for updates etc).
I don't think the courts work how you think they work.
As the ones prosecuting a criminal case (traffic offences if taken to court are criminal proceedings), the burden of proof is on the police/DPP. They are the ones that have to prove that the GPS device was or is capable of connecting to the mobile network. As a defendant, you just have to cast reasonable doubt over the prosecutions case, not proof of anything (well, unless you are attempting an affirmative defence (e.g. insanity) that is, in which case you have a burden of proof of that defence).
You'd probably need to take a high-powered lawyer and a technical expert into the courtroom with you to prove that. Good luck trying to explain that to the cop at a roadside stop.
Errm, how about just providing the product manual that lists the device's capabilities and pointing out (if it is the case) the lack of mention of mobile network connectivity in the device's specifications?
"You do realise modern smartphones all support voice control ?"
You realise that turning on voice control requires you to permit the google apps to have access to ALL the data you carefully denied them access to months ago when you became aware of the privacy implications?
I cant even use voile dialing on my android as they killed off the on-phone voice recognition that worked fine and replaced it with a cloud service that if I turn it on, to tell it to call someone, it requires access to my web history. WTF has my web history got to do with a contact name?
This post has been deleted by its author
This post has been deleted by its author
Well, you say "bought", but you only bought most of it. You're leasing the battery.
That's why the numbers don't add up for me on the Zoe - I don't do enough miles to justify the monthly cost of the battery.
I wonder if that means the 10-15-year-old Ford Motor Group cars will be affected. They used Volvo satnav, which was horrible to operate, and had some shrill woman do the voice. You were in so much trouble when you took the wrong exit off a roundabout. Might be fun if that's your thing... ->
"I suspect all TomTom based satnavs may be vulnerable."
My relatively old Garmin had updates waiting for it when reading this story reminded me I'd not checked in a while. There was a new map update, some timezone updates and a language update. No mention of rollover issues or firmware updates in general.
How I found this news story too (also a Zoe driver) :).
To the person commenting about not owning the battery with the ZOE, if you have a "ZOE i" model (like I do) then you do.. have been able to buy it all rather than just lease the battery since Nov 2015 (car was released 2013).
Links:
Is My Renault affected? : https://easyconnect.renault.co.uk/
Is My TomTom affected? : https://www.tomtom.com/en_gb/updates/
Renault has recently released a patch for their R-Link satnavs to fix this, and it was applicable to my wide’s car which is not even 4y old.
But at least Renault offers a web-based satnav update service. Audi asks 300€ for each SD card update for my car! No wonder Carolay and Android Auto are so popular...
If manufacturers haven't learned lessons from the last time this happened, then... I guess I wouldn't be the least bit surprised. The way modern Tomtom kit handles UK postcodes, for example, compared with how the equivalent Tomtom kit from a dozen or more years ago did it, suggests that remembering lessons of history is not something these people do.
UK postcodes are a lot more detailed that other countries though. The ones here in Switzerland are roughly equivalent to the [A-Z]{2}[1-9]{1,2} bit*. People in other countries for some reason have problems getting their head around each street or even building having an individual postcode.
*Yes I'm aware there are edgecases
Every 10+ year old BS that we use has an external "professional level" GPS device to maintain the long term RF clock accuracy, we don't care about the time/date.... But still they all reported errors the first time this happened, but likely they've all been swapped out for newer by now... I hope... :)
What purpose does a week count actually serve in computing a position? I'm old enough to remember pre-GPS times[1], and we already had 32-bit timestamps. Is that final week purely coincidental with other silly traditions and tragically stupid one-off events, or might this be made a scapegoat for something?
[1] In fact I even spent some time in the 1980s working on a pre-GPS tracking and positioning system for vehicles, where fitting the entire onboard software into a 128k ROM was one of the issues we had to deal with.
This post has been deleted by its author
You get the locations of satellites with respect to one another, then calculate where the satellites are right now and use those two numbers to figure out where you are. If you think it's nineteen years ago, you may come up with a different answer to where the satellites are, and thus you would have the wrong answer to where you are as well.
Also leap seconds.. whether they should be using the date to calculate leap seconds or whether there is another field for that is a question for someone who knows the GPS spec better.. but I found at least one reference online of a manufacturer doing leap seconds off the date.
> Whilst devices may not have been prepared for the rollover in 1999 (all mine were fine), I doubt anyone could not be aware of such a thing by now. If you are writing the firmware in 2014, you buy any year less than 2014 as a rollover. Duh. The sky is falling not.
Some programmers still get leap year century calculations wrong. Whatever the reason is, it's not because they weren't aware of leap years.
I was around working on low level GPS hardware at Philips Semiconductors around the last GPS week rollover.
Even then , the average handheld GPS managed to rollover and keep going.
Even a very old Garmin GPS40 single channel handheld achieved lock with fresh batteries after the rollover, with correct date , time amd position.
This was a GPS that had been stored for years in a boat. The new owner of the boat had just completed a purchase of a replacement in a yacht chandlery when I walked back in with the old pre last rollover Garmin GPS40 working properly....
Any GPS that has memory of last week number last time it had a fix and sees it going reliably 'backwards' can infer a rollover and add 1024 to the week number.. ignoring just getting a rough idea of time from an RTC ...
I had a look for info on my Garmin Edge GPS unit, and found this thread about a 'Year 2019 bug' ....... from 2015. Hopefully Garmin have made an effort to fix it since then.
https://forums.garmin.com/forum/into-sports/cycling/edge-1000-aa/97666-incorrect-date-time-in-edge-recordings-year-2019-bug
I mean the hours of UTC only go up to 23, then once you get to 23:59:59 you will have a roll over!!!!! OMG if that is improperly handled all computers will have severely wrong times. And that's just a few hours away!!!!11111!!!!!eleven!!!!!!!
!!!PANIK!!!
Too much FUD being spread here. Read the fine manual, specifically IS-GPS-200. The GPS ephemeris message (Message Type 10) has a 13-bit field for the GPS week: "30.3.3.1.1.1 Transmission Week Number. Bits 39 through 51 of Message Type 10 shall contain 13 bits which are a modulo-8192 binary representation of the current GPS week number at the start of the CEI data set transmission interval (see paragraph 6.2.4)." And no, that field wasn't added just last year. One might additionally speculate that there may be additional fields with a longer span outside the domain of IS-GPS-200 but even a rollover after some 150 years is probably adequate for practical purposes.
Yes, it's entirely possible that some poor coder relied on the 10-bit week number (there are also fields which represent the week number in 8 bits i.e., modulo 256! Panic!!!) but there's no excuse for having done so.
A bit more info for those who don't like too much light reading... The week number was mostly there to help cache ephemeris data and two bits would have been enough for that use. Each sat transmits a message roughly like "sat 1 is in virgo, sat 2 is near the moon,..." that should be discarded after a week. Each sat also sends a much more detailed message along the line of "I'm supposed to be in this orbit but I'm wobbling just a bit about this axis and this axis and this one and this one" as well as the most important one "at the tone the time will be ....". The wobble takes into account tides and how the sats are moving because of Jupiter, Sun and Moon and are a 9th order polar coordinate polynomial for those who want nightmares about complex formulas. The rough message is enough to locate them to a degree, the precision message is good enough to place them within a few meters at a given time. Modern GPS receivers know their time and calculate the "pseudo-range" to each sat it can see down to a wavelength (20 cm) or better. That is the basis for differential GPS where a different GPS receiver can send a signal saying "that sat 31 which is about 15 above the horizon is lying by 45.22322 meters". The receiver needs to have its time synchronized down to about 90 nanoseconds to even get a reasonable position fix. Its very impressive that its doing that with signals at very low baud rates and used to be broadcast at about 55 watts.
Long before the week rollover came up the first time, there was a proposal to occasionally send the full date and time in a different packet that could be added where it wouldn't interfere with normal operations. There were uses of the extra messages for military and now aviation differential signals.
This post has been deleted by its author
>Yes, it's entirely possible that some poor coder relied on the 10-bit week number
https://www.tomtom.com/en_gb/updates/
https://easyconnect.renault.co.uk/
and the Emergency Services Network AirWave references above :)
I don't think anyone is saying life will be destroyed as we know it.. I think they're saying perhaps consider checking where you're reliant on GPS (particularly for time syncronisation) and check your vendor follows IS-GPS-200 and/or has a patch for this if they need to.
This post has been deleted by its author
This post has been deleted by its author
Looks like the IS-GPS-200 documents 2 frequencies called L1 and L2. L2 fixes this but is a different frequency : http://www.catb.org/gpsd/hacking.html#y2k1 and https://www.gps.gov/technical/icwg/IS-GPS-200H.pdf (for "frequency")
"1575.42 MHz (10.23 MHz × 154) called L1;
and a second at 1227.60 MHz (10.23 MHz × 120), called L2."
CNAV is wrapped up in L2 but L2 is "pre-operational" and not to be used for critical usage.
L1 C/A (under "Legacy Signals" in Wiki has a section about 10 bits and rollover) is affected. L2C CNAV ("pre-operational") is not (blurb in wiki under "CNAV navigation message" explains increase in bits).
https://en.wikipedia.org/wiki/GPS_signals was also helpful.
Sounds like some newer receivers are on the L2 bandwagon.. but that doesn't mean there isn't a tonne of L1 only devices still around.
One very well known GPS device manufacturer I have worked with, and their firmware being deployed in many 2000-era cars and cellphones, did rely on the modulo 1024 week number. But most of this hardware is now obsolete and the problem will only happen in 2021 due to a reference date being hardcoded into the binary firmware.
in other news:
"Professor Parkinson received the Queen Elizabeth Prize for Engineering in London last night for his key role in developing the Global Positioning System or GPS, along with the rest of his team: Professor James Spilker, Jr, Hugo Fruehauf, and Richard Schwartz."
Will they have to repay their shared £1M when it all goes wrong?
It begs the question, why wasn't the field made 13 bits, not 10, in the first place or at least modified a long time ago after the lessons learned from Y2k? When Apollo was flying the skies, every bit was a big deal, but for the last 30 years or more, another 3 bits wouldn't matter a jot.
"Apparently the new protocol mentioned handles the wraparound anyway, but if not ... 157 years from now somebody is going to be very confused."
Yes, an interesting throwaway gag line, but in the real world, consider where we were technology-wise 150 years ago. Now compare that change with where we may be in another 150 years from now (assuming we aren't all either drowned or dead from the heat and pollution! (and by "we" I mean humans in general since I doubt many of us will be around then.))
After reading this I saw my TomTom suddenly lose location after doing the first part of the journey without issue.
Scary or what?
Fired up a spare Garmin which knew exactly where it is.
My (ancient) eTrex goes everywhere with me but doesn't get fired up. Just tried it and the copyright statement said 1999 to 2001 so April might be interesting.
I remember using this with Autoroute (2003 I think) and a serial/USB cable to navigate using a laptop. Happy days. When you didn't reboot with it still connected when Windows thought it was a serial mouse and much hilarity ensued (not).
I am curious to see what my Garmin does on the day. fortunately I use it more for logging than navigation.
I've had my own moment of GPS cockup when I asked Google Maps for the way to the new nature reserve and got directed to dead end private road on the opposite side of the park from the entrance gate. Fortunately Google learned but I still sanity check when possible. I also have a lot of maps and a Silva compass, JIC.
Hi am working in the GPS devices industry.
The "problem" has been known for decades since the first GPS rollover in 1999.
There are surely going to be issues but it will be mostly for very old legacy equipment.
Also, some devices I work with will be affected, but in 2021. While there is no direct event in 2021, some GPS devices have an internal reference date that they use to know where we are in time. This date is more important than the GPS week rollover date itself, and it shows that problems can happen at any time depending on when the firmware of the GPS device has been designed or built.
In addition, the newer signals broadcasted by the newer GPS satellites have a much longer rollover period, so newer devices will no longer have this problem.
As an engineer and software developped, I am much more worried about 2038. Search for Year_2038_problem on Wikipedia.
I wouldn't want to be in an airplane, elevator, or close to a major industrial plant a few days after and before that time.
A relative has been rocking a Samsung Note II for years. As of the 6th April Rollover it gets VERY intermittent GPS lock. While cycling it lost position for 14 minutes. And this repeated may times during the outing. Using a diagnostic app shows the GPS module is returning dates in 1999. In theory that shouldn't lose the position data, but it started at same time.
So far we've found a support page on T-Mobile US that confirms failure on the Galaxy S II (similar age).
There's a french android page which recognises the rollover failure, but is confused by the failure to get gps lock.
A similar report has shown up on XDA with the same symptoms as us. That guy caught diagnostics where the phone suddenly drops all satelites and then slowly rebuilds connections. So maybe the device notices the date discrepancy vs the more detailed gps frames/pages and then resets itself. Rinse and Repeat.
So we've begun shopping for a new phone :(
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
