Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment Re:Unforseen? (Score 2) 108

Alright, you're the current leader on this thread. All the rest are tripe and nonsense (identify poisonous snakes, my arse).

I've used Google Translate to do exactly this for my Italian girlfriend and her family. Shall I tell you the problem, however? The translation still sucks.

And making it same colour and "augmenting" it into the picture of the menu is nowhere near as useful as just translating it. Honestly, the movement and positioning and losing the translation just as you slide the device over to them, and the translation shifting and jumping all over the place and even changing the words when you're holding the phone relatively still - it doesn't help at all. And though it does a fairly clever job of trying to put the text in the right colour at the right place, it makes a mess of it.

Honestly, after two seconds of "gimmick show" that we did with it, it was then back to human translation and/or just getting the one word to translate without having to type it in.

But, in terms of something that 30 years down the road might be potentially useful if we solve all the other problems with it, you're currently in the lead.

As always, however, the simpler service of just recognising the text and putting up a list of the translation as static text rather than trying to superimpose it on a real-time camera image will still probably win in terms of effort, legibility and speed once the initial gimmick wears off.

Comment Re:Unforseen? (Score 1, Interesting) 108

What is the potential of AR?

Because, to be honest, I can't really see it.

You can project shit over your car windscreen. Or you could just have a pane of glass and have all the shit you need to look at slightly below it. Thus, you can see all the road AND all the instruments. Why do you want or need one obscuring the other?

I think the same of AR. I can point my phone at the Eiffel Tower and see... what? The Eiffel tower behind lots of shit that I may or may not want, that I could still read if I just pressed the button and held the phone facing down like a normal person reading stuff.

What does AR give you over anything else? So far the biggest usage is imprinting "hidden" cartoon characters over your high-res image of the thing ACTUALLY IN FRONT OF YOU.

I work in schools. They have "AR" books. In reality, it's an app that shows a 3D object hovering over the top of a bar-code like thing in the book. It would be easier, cheaper and more useful to have a) the 3D object showing in an app and no book, or b) a book and no 3D object depending on the information.

I'm honestly struggling for what information you'd want to overlay over reality that you can't get quicker, easier and less obtrusively by just holding the same device in your hand and looking at it.

Comment Re:"Serious Crime" (Score 3, Informative) 48

"However, before the trial was set to start, a fresh charge was entered, to which Cooper pleaded guilty."

Game over. Nobody convicted him but himself. It didn't even need a trial, it's automatic and then he's automatically punished. No judge stood there and THOUGHT that he was guilty. He admitted guilt from the very start of the trial.

If you don't think that what you did overstepped the mark, you don't plead guilty, especially if you're a midwife whose career is on the line. And then you could appeal to the court based on the circumstances and might be FOUND guilty. But admitting it is NOTHING to do with courts being overly harsh. Someone says you committed an offence. It goes to court. You admit that you did. Game over. No evidence needs be heard.

Chances are that he was way over the mark, knew it, and didn't mention that in any of his posts.

Don't get your news, or your opinion of the justice system, from a one-sided, uninformed puff-piece from the media. Even the BBC.

Comment Re:NY State Inspections (Score 3, Funny) 123

The test is different - run on a specimen model of vehicle - and the pollution monitored. It's highly prescribed in terms of how it's run. This happens, then that happens, then that happens.

The software is built to detect this series of unusual speeds, consistent runs, in a predictable order, and modify its parameters.

It's the most disgusting kind of "cheat the test" mechanic, which is why they're being screwed to the wall for it. It's not a question of anything being activated, or the software going into a certain mode automatically. It's literally designed to detect the test sequence and then cut its performance on detection so that it passes the test. There is no other reason for that code to exist, and it NEVER activates during normal driving (so the test is entirely useless, yes, but also it's been gamed quite deliberately).

It goes unnoticed because there's been no retesting of vehicles after the initial certification, except your normal emissions test, which is designed to take account of older, more polluting cars. Nobody is saying the car itself isn't legal to pass emissions tests - they just committed deliberate fraud to get it classed as lower emissions vehicle than it EVER could be in normal use.

New out of the factory, it's more polluting than it ever is tested to be, even on the regular emissions tests (which are incredibly short-lived anyway), and will only get worse with time and by the time it does fail tests, you are out of warranty and they can just say it's because of how you used the vehicle or bad luck.

But on the road, under normal usage, it's giving out 20-30 times what it does on the regular emissions tests or on the specimen model tests when the model is generally assessed for taxation, etc. based on its pollution.

Comment OS previews? (Score 1) 149

I would say that if you do, your primary work is not important enough.

Pretty much the people I know at "the cutting edge" are people who don't really actually ever "use" their machines. Like the people who spend thousands on overclocking and so on, as soon as they've done it, they 3DMark it and on to the next build. Actually PLAYING GAMES with such a machine is secondary to their usage.

In the same way, if you can afford to install untested software "to see what it's like" on your PRIMARY machine, then you obviously weren't doing anything that might be disturbed. In a VM? On a secondary machine? For testing and development? Sure. But on the main machine that you use all the time? No.

It reminds me of the dual-boot days, when people would tell you how many OS they could boot into. All cool, I'm sure, but it just means that your primary machine isn't even decided on an OS and spends half its life rebooting into other OS.

Sure, some people might only be able to afford the one machine. But then, surely, that's the one machine that you don't want to fuck up, isn't it?

I'm happy to test the cutting edge, trial it, and have to do so as part of my job (some part of which often gets "Why aren't we on already?", so only testing can say "Because it breaks X, Y and Z"). But my machine always gets comments as to being behind everyone else's. That's because I don't upgrade unless there's a need.

If you're sacrificing time on your primary machine to test "non-working" software of any kind, then your work obviously isn't important enough, or you busy enough, to actually care about putting it at risk.

Comment Re:it's a small step, but... (Score 1) 45

Real progress would, in fact, be not having to compile for dozens of different architectures.

Such as describing a way, in a standardised language, of being able to do anything, no matter the underlying hardware. Remember those days? When languages did that for you?

Real progress would be an architecture where the source code is the program, and "first run" is actually a compile-and-cache-bytecode kind of operation. And if you move that same compiled binary to another system: a) it has the source code in it and b) it notices that the architecture isn't the same, wipes out the bytecode, recompiles on first run and off you go again.

No more reinventing the fucking wheel for every platform, no more having to compile multiple versions and formats (have a look at any program download and now imagine you don't know what Windows, Linux, Mac are, or what 64/32-bit means, or whether you want it compiled static or dynamic, just the executable, in a ZIP or a setup EXE), the best performance you can get for that particular architecture, a definitive way to force programmers to standardise on a language without #ifdef's compensating for the bits we should really be abstracting out entirely, and truly portable binaries.

I'm sure we've had it before. But commercial code almost certainly would hate the idea. I've yet to figure out why we don't quite have that on the OS platforms, though. As it is we still have multiple package repositories for each architecture (x86, x86-64, etc.), multiple packages for the one program (-dev for library and headers, another for the binary, another for the source, etc.), and even distros that still only work on one type of machine.

Comment Backup. (Score 1) 465

"is there anything -- any service -- Mr. Cooper could use to get his artwork back?"

Yes. Restore from backup in his backup software.

Whoops. You didn't do the thing that EVERYONE talks about every time they lose something, that schoolkids are taught to do, that everyone needs to use at some time in their digital lives?


Call it "temporal art", forget it existed and move on. The rest of the world already has. The only people who thought it important didn't think it important enough to save BEFORE it was lost.

Comment Sigh (Score 1) 129


I was quite looking forward to it. And then in the run-up and on the day bought nothing anyway. I briefly considered an SSD but the brand I wanted never went on sale and I'm not going to compromise just to get something cheap - I'd just buy a cheaper brand anyway.

In fact, the only thing I "bought" was a free copy of Civ IV, available to Amazon Prime users throughout the sale. It "cost" £12.99 and no doubt classes as a sale, but I never paid a penny and it was a Steam key. And then I discovered that I and everyone who might have wanted it already have it on our accounts.

Is that technically a sale for Amazon? Surely not if it's free. And everything else was pretty Meh.

Like the Steam sales and the indie packs - their day has come and gone and I only look for nostalgia's sake. The days of actually BUYING anything in them at anywhere near a discount over what I could have had them for previously? They're long gone.

Comment Re: Really? (Score 1) 726

This is a friend of mine:

Honestly, just look at the hardware involved. Not a fucking amateur or new to this stuff.

And the problem? BIOS incompatibilities when putting multiple cards into a PC. Spotted by a serious of highly-obscure and technical hints, and a lot of research.

But everyone they asked suggested power. "It's just power, it's all plug and play nowadays", in effect. And yet, it was nothing to do with that. This person isn't an idiot. And the BIOS is one of the last things I'd consider in such circumstances. And they were just very lucky that, actually, the BIOS update exists and works. I imagine a great many smaller motherboard manufacturers just don't give a shit.

Or would you like the stories I can tell of random and otherwise untraceable BSOD on gaming PCs? Change the card from AMD to an nVidia and they disappear. No, it doesn't matter what version of the drivers you are using and, no, you don't have to be using Windows 10 or anything "new". At least one of the manufacturers you list just has drivers so bad they BSOD for no discernible reason on a correctly "working" and fitted card, no matter how many replacements you try, no matter what you do to the power and card setup, but a similar-but-more-power-hungry nVidia card will "just work".

It's NOT just as simple as plug and play. It never really has been. Today it's much better, I'll grant you, but that was kind of my point in mentioning my history. But weird incompatibilities still exist, especially if you try to save on the motherboard to get a more meaty PSU or graphics cards.

Everything from PCI-E driving on the BIOS firmware, to resource management in the drivers can cause you problems. On top of the overall power, 12v rail, cooling, and other physical issues that just aren't obvious when you just buy a card and slap it into a machine.

Sure, if you choose only well-tested, reviewed and quality components, and know what you're doing, and forget about the troubles you had at installation because you're an expert and just kept everything up-to-date and tried all the combinations, everything works.

But for the average person building a PC? Forget it. Almost every one I've seen, from teenagers trying to save money to overclockers trying to show off, is far from plug-n-play.

Comment Really? (Score 1) 726

I stopped "building" PC's over 15 years ago. There's literally no point, you end up with problems and incompatibilities and extra expense and - in the end - you get a PC that you can't upgrade any further than any other.

Yet I have 1000+ games on Steam, and god knows how many on other services and discs, etc. You just buy good commodity hardware and - although not "obvious" to complete amateurs, you should know if you've ever googled - a decent video card.

Last time someone I heard of that didn't have any PC experience tried to build a PC it was a mess, but because the guy was an idiot. They phoned my technician (who's a gamer), asked what to buy, then ignored all that and tried to cobble together something themselves.

They ended up with a shit AMD card, an underpowered processor and a PSU that could never have handled it. That's NOT what my guy recommended to them, in fact it's the opposite and for much cheaper they could have got a decent PC built, certified and warranted to work for the next few years.

It's not hard to do PC gaming, it's not hard to buy a gaming PC, but it is hard to be one of those overclockers, one of those people who builds all their own PC's, and uses all the "cool" tech to get ahead of the game, especially if you're an idiot who thinks it's all just modular and plug-and-play.

As someone who grew up with ISA cards, ports, I/O, etc. I can do anything that is required. But I stopped building my own a long time ago. I never even had an "expensive lesson" because of a mistake. My builds all worked. But it's too much faffing. I don't even know (or care) what the processor sockets are any more. I just buy off-the-shelf, but decent, hardware, pre-built, with a warranty.

It's kept me running on everything from Half-Life to GTA V without any problems. And it's been upgraded several times since (SSDs, etc.).

And, out of all my Steam friends, including all the overclockers and show-offs, the person with the most play time on their machine? Me. The person who's played the most games? Me. The person who buys the most games? Me.

I'd rather spend my money on one decent machine and then spend what I would have spent on all the junk and minor upgrades over the years on actual games to, you know, play stuff. The days of having the time to piss about worrying about PSU rail draw, etc. are long gone, precisely because I just want to play.

So PC gaming isn't hard at all. Just buy a gaming machine. Or get a decent "business" machine and maybe get (or slap) a half-decent nVidia in it. Hell, if you're that worried, buy a Steam machine. Most of those will play anything you throw at them.

What's hard is being a PC gamer as a broke teenager, or trying to build something that will beat all your friend's machines. You used to have to make-do and upgrade piecemeal and make the best of what came your way. Nowadays, anything you get in the shops with the right video card is just fine, and if you are really stuck, get one of those gaming PC websites to build one for you. It'll cost more than a console, but you'll get more out of it than you ever would a console.

Comment Re:Example Not a Problem (Score 4, Insightful) 205

Because it's a band for a fucking bed.

That's not really a trademark, it's a generic term that's likely to be challenged as a trademark even in the territories that have (stupidly) allowed it.

It's like complaining that if you go searching for "sticky tape" that you get things that aren't your genuine, registered trademark "StickyTape®" sticky tapes.

If you'd called your company "Joe's Shitty Products®" and someone sniped all your "Joe's Shitty Product® Sheet Fastener's", yeah, sure you have a case. But "bed band" is a description of exactly what the product is, using two generic and common English words related to that product. That's NOT what you should be trademarking.

If you want to protect a trademark you combine it with a company name that's pretty unique and which you own the trademarks in your territories and industry sectors for.

But trademarking TennisBall tennis balls is a) likely to not be allowed in the first place, b) likely to be struck down for genericity at any time and c) stupid because I don't have to be specifically sniping your trademark to have a website that scores high for searches of tennis balls that aren't TennisBall tennis balls.

BedBands is, quite honestly, one of the worst product names that I've seen. And one of the worst products that I've seen. I could make it myself, make something better, I've bought better things that do the same job, copy it in about ten minutes, and I could market it as the "best bed band product" without infringing on BedBands trademarks unfairly. Because it's a fucking band for a fucking bed.

And, to be honest, that "sit on the corner of the sheet" shite will last two seconds until I rip it off when I roll over. Most of the bed bands that actually work do so by tying the left of the sheet to the right of the sheet under the mattress, not just the corners.

I hope their blatant and unnecessary slashvertisement just waters down their trademarks even more.

This rant is trademarked by me. Nobody else can have an InternetRant internet rant but me, now.

Comment Re:Bracers (Score 2) 205

It is an invention.

But it's certainly not their invention.

I've seen straps that hold fitted sheets to beds for the last 20-something years at least. Hell, I've own a fitted sheet that came with them.

It's certainly not novel. And I think their "copies" have nothing to do with their downturn. You've been able to buy these things for decades. More likely is that people are camping on their trademark website and getting into the "related items" for their products and then consumers are realising "Hey, look, there's a cheaper version that does the same thing" and buying that instead.

But, to be honest, if you genuinely go into business thinking that product A of yours is going to sell, be successful, and repeat that success into perpetuity, even when your competitors start to clone it, then you're a damn idiot.

You have a strap for a bed. Great. You sold a lot of them. What's your plan to keep you buoyant NEXT year?

Comment Re:Meanwhile (Score 2) 392

Again, not comparable if Tesla's have, say, a limited range and therefore aren't driving on long highways in anywhere near the same amount as ordinary-engined cars.

Or, say, there just aren't enough of them to form a statistically significant average over that amount of miles, where a small fluke of careful driving would reflect much "better" than a huge amount of older, cheaper cars which *CAN* form a statistically significant number over a million miles.

Statistics is hard. And horrible. And often completely misleading. Number of accidents per millions miles is useless at the moment, while they are a niche manufacturer. My bicycle has a number of accidents per million miles that is basically zero. It doesn't mean my bike is safer, or comparable, to a Tesla or a Ford.

Comment Re:Meanwhile (Score 2) 392

Allow me to introduce the most useful maths course you may ever use in real life: Statistics.

How many Teslas are there on the road?
How many GM / Ford / etc.?

How many of those Teslas are new or only a couple of years old?
How many of the GM / Ford / etc.

How many of those Teslas are used regularly for long-haul driving trips?
How many of the GM / Ford / etc.?

How many of those Teslas cost a fortune to repair and so are driven more carefully?
How many of the GM / Ford / etc.?

How many of the people who drive Teslas bought them new and so are more likely to take care of them?
How many of the GM, Ford, etc.?

A number, on its own, without context, cannot be compared to other lone numbers, without context.

Remember when your physics/maths teachers were always yelling "UNITS, UNITS, UNITS" to how answer of "Three?". This is the extension of that. The number alone is useless. Even the number per year is useless. Comparing that to another number per year is useless.

You have to use the human part of your brain that's able to infer where those numbers may differ and why and compensate by stating them in other ways so they are more accurate.

Granted, some of the things above point at sloppy driving / maintenance, not car ownership. But that's exactly the point. If you want to see how dangerous alone the CAR is, when well-maintained, not the driver, you have to account for all the other factors.

Which your numbers do not.

Comment Re:This should be interesting. (Score 1) 202

Depends on if you care about whether you break it, or whether you have to put it back together.

Laying or removing tons of copper cables under sleepers, through bedded gravel, and under track isn't easy either. But the usual technique involves a crowbar to make a start, a winch, a van and driving off with whatever it was that you wanted - usually still attached until you come along with diamond saw to chop off how much you need.

I don't see how that same tactic wouldn't work with solar panels on a road.

Slashdot Top Deals

It is not for me to attempt to fathom the inscrutable workings of Providence. -- The Earl of Birkenhead