Buying a game is like buying a car. DRM is like the key to that car.
If I buy a car, I do not expect to have a significant percentage of the cost of that car used to make the key. Especially when the justification for that is that the key cannot be copied by key-cutters and third-party garages.
Yes, this make the car more "secure" - for a while - because official keys cannot be fabricated without the manufacturer's co-operation. FOR A WHILE. But there isn't a car in production that has an "uncopyable" key, or that can't be broken into without any key at all - the fact that manufacturers can make a key in the first place tells you that.
However, in blocking out this "unauthorised key copying" industry (some of which is actually legitimate - not everyone who takes their keys into a key-cutting place is intending to break into someone else's house), the manufacturer is spending more time designing more and more elaborate keys (all of which can be copied "unofficially" at any garage within a matter of days of the car being produced), charging me more and more for the privilege (by moving some of the value of the game into the cost of the key itself), and in the process giving me a car that I can't always drive, sometimes won't open, that I must NEVER lose the keys to (because in a couple of years it'll be almost impossible to get an official key ever again, and in the meantime the only option for replacement is to buy an entirely new car direct from the manufacturer in order to get "another" key) and where the key weighs 12 kilos, cannot be put onto a ring with other keys, and comes only in flourescent day-glo orange (just in case someone wants to run off with it to make it work on another car of the same model) - with holographic iris-identification-over-IP built in to the key just in case you try to lend your car to someone else.
Software copyright infringement is a problem. So do something about it that a) hurts the people doing it to you and b) doesn't hurt the people who aren't. Current DRM solutions do NEITHER of those.
I never stopped playing AOE (specifically AOE2:Conquerors). I *DID* stop playing it online because MS just sucked the life out of the multiplayer aspect by locking it to a single vendor for online matchmaking and then destroying that facility when they got bored of AOE.
So, what's here for *me*, someone that wants to play AOE but was forced by Microsoft's enforced-obsolescence to stop playing it online unless I wanted to faff about with third-party software or entering IP addresses? I won't believe it won't happen again, and I don't believe that a new MMO "reboot" will be anywhere near as good as the AOE2:Conq. And are we talking about a monthly subscription model or can I actually *OWN* the game (or at least my copy of it) forever?
In the meantime, playing the classic version over a private VPN it is.
Maybe it was just me but I found it a bit lacking. I'd heard all the hype, which I avoid paying attention to, and then when the free giveaway came around, I thought "Why not?".
There's a 100-something games on my Steam account and I got through Portal in about 3.1 hours (at least one hour of which I'd left it running trying to get the falling-far achievement while I had dinner) - that's probably the least I've invested in a Steam game, with the possible exception of some of the free stuff that's thrown at me in the bundle deals and I never even load up (and DogFighter, which I deleted after less than an hour just the other day). It was short, too short. The puzzles weren't that tricky and it was more a case of being able to aim to the spot you *knew* you had to and/or doing it inside the time necessary (like the long-double-jumps where you have to move the portal mid-way through a jump). I didn't see it as a "puzzle" game at all - it was usually obvious what you needed to do and just a matter of doing it. I think the only levels where I had to have more than one go were the ones that required strict timing or complete accuracy.
I know that there are a lot of extra puzzles and other achievements but I didn't find it very exciting or challenging at all. I'm so glad I didn't pay for it, it got deleted almost immediately after I hit the ending. Trying to tie it into the HL universe, etc. just taints an otherwise good series.
I appreciate the *technology* of being able to have extra windows onto an existing world, including recursive views, but that novelty wore off very quickly (I've seen Quake Mods that achieved the same thing), because about the only thing you can usefully do with it is something like Portal. I'm afraid I consider it in the same gasp as something like a novel indie game, or something like that. And, frankly, the whole plot/story/background/character got so annoying that I switched off all sound after about 2-3 puzzles - I always switch off music, I've never had to switch off the sound before.
I didn't think it was very original or creative beyond the use of actual recursive windows onto a 3D world. The rest of the game basically stemmed automatically from that premise. I consider Portal to be more of a tech demo - a bit like the Lost Coast thing. That's about the level it's at, about the right length, about the right kind of "storyline" added to it. I won't be buying Portal 2 unless I know it's going to evolve from Portal, not just be more of the same.
Er... yeah, if you leave the default memory cache enabled - Opera does its own in-memory caching where some other browsers rely on the underlying filesystem to cache for them, and Opera loads QT which counts as "memory used" on Windows but not under the vast amount of Linux distros that already have it in memory to be shared. There are a million and one ways to tweak Opera, which is another plus for it, including disabling quite a lot of functionality that you wouldn't want active on low-memory machines.
Opera on my computer, as an upper bound, never takes as much as an equivalent FF process with the same windows open. I have memory (in fact, all) caching disabled.
Pretty accurate - and that's from a long-term Opera user.
Shame that Opera sees such little take-up. It has 99% of the functionality of the common addons for Firefox already built-in (and has for years), it is a damn sight faster on low-end machines than Firefox, it's cross-platform, it's got a built-in mail client that is more than good enough for the average joe (with super-fast searching for EVERYTHING), and it's normally first with any innovation (WebM, Acid-compliance, HTML5, etc.) - hell, for the last version they discovered myriad websites with a common javascript bug that preventing them providing a 10.x version number in the user-agent, so they had to stick with 9.8 and some extra gumfph elsewhere to tell you the real version number. No other browser's spotted that yet.
If someone could tie Pidgin into Opera, I'd never need another bit of software again.
Your contract is with the retailer who gave you a receipt. Anything else doesn't affect your statutory rights to take the damn thing back to them and get a replacement/refund. You do *NOT* have to accept if they say otherwise. To be honest, you do not need the receipt either but try convincing a small-store manager of that.
If in doubt, send a stroppy legal letter to the RETAILER, not the manufacturer - they are legally obliged to deal with it and they know it (store managers tend to hide behind "company policy" which does not accurately reflect the law. You can't be expected to deal with a guy in China who you have no business relationship with in order to get a working product - the guy who sold you the device and gave you a receipt is the one who has to sort it out.
Air-pressure monitors on your tyres can get stolen if they are external (i.e. valve caps), or require special tyres if they're internal. Kids round here steal valve caps just because it's funny. Sod having to replace a miniature wireless pressure sensor each time, or add the cost of a non-standardised wireless hardware to the price of each wheel / tyre (at the moment - it's cheaper for me to replace the entire wheel than just the rubber tyre and I can get a whole set for less than the price of a Wii). Also that requires dashboard indicators, wireless communications, etc. or some very clever mechanics to keep it wired (and an air-line to your tyres is a silly idea because it would be almost impossible to design a leak-less, rotating valve and if you could, again, you've just quadrupled the cost of a new set of tyres and any minor repairs to your car). Besides, a lot of top-end cars DO now have the pressure sensors on their wheels. You have to buy the manufacturer's tyre/wheel if they break or some of them require batteries, etc.
Batteries - this would require another very-heavy-duty battery that is charged over a split-load charger along with the main battery. Not a problem but several hundred dollars worth of equipment, plus the place to put it (although a lot of cars are designed to have the battery on either side for their other-hand-drive models), plus twice the replacement costs (not to mention the specialised charger, which CAN be just a huge capacitor / diode I think). Additionally, you have to have extra cabling and several kg's more weight (lead-acid batteries are not light - go pick one up). To compensate for what is currently a little wiring that cuts everything but the side-lights and hazard flashers when the key is out of the ignition. Most cars solved this problem decades ago by cabling accessories through the ignition and providing a headlight-warning alarm. Cheaper, simpler, more efficient on your fuel, better for the environment.
Mobile phone charger / Bluetooth speakerphone - upgrade your stereo, or your car. I bought the cheapest stereo I could find recently (to replace a tape deck in my car) and it came with USB ports on the front and Bluetooth functionality - I can charge anything with a USB cable. And providing a slot for a mobile phone to sit in would cause merry hell with some countries regulations which often cite that such things have to be out of reach of the driver when they are driving. That's why a lot of sat-navs have their own custom sucker-window things, and for pocket-change you can get a mobile phone holder that plugs into your heating vents or similar.
And all of the above suggestions don't actually HELP your car efficiency at all. The batteries mean you are constantly (over-)charging two horribly poisonous batteries where one would do, the mobile phone / Bluetooth thing is just convenience - where you charge your phone does not affect your driving range and personally I'd rather you weren't answering ANY calls while you were driving, hands-free or not. The wireless pressure monitors put inordinate amounts of engineering, possibly more horrible-chemistry batteries, extra weight, wiring and display into a car to do something that just LOOKING at a tyre can tell you. That's not productive for efficiency.
It's been over a hundred years with these machines - and in all that time you've been required to ensure your car is roadworthy. Believe it or not, a little blinking light on your dashboard CAN'T tell you that - you have to inspect the tyre yourself. While you're there, check the pressure by sight/hand/gauge, you lazy git.
And that the switches weren't blocking DHCP from anything but the authorised DHCP server, and that it took 3 days to track down a rogue DHCP server (not hard, you usually get the MAC address in seconds, trace that to a port, disconnect the port and see who shouts that their network connection isn't working - if it's a remote switch on the end of that port, go to that switch, rinse and repeat).
Hell, it would take less that an hour if you just pulled cables at random until that MAC disappeared.
Like most of the things in the story - incompetent admins and IT setups allow human error to be amplified. Seriously, one of them is basically a hospital network not using spanning-tree.
According to the all-knowing Wiki: "phase I of the GSM specifications were published in 1990"
So, depending on your point of view:
- it's taken 20 years to implement something that had a published standard and worldwide, cheap hardware examples used by millions of people every day.
- the standards took 20 years for an outsider to be able to implement them independently.
And we're still only talking alpha code with specialised hardware.
- Software patents are not considered valid in the majority of the world, precisely because they get in the way of perfectly reasonable actions. - Software patents on an "open" standard might not be enforceable if those patents are required to implement that standard, and are the only sensible way to do so. Interoperability with an existing product can't be protected by a patent if that patent is the only (or only sensible) way to do things. - All the mentioned patents have prior art except one, which is so far worded towards Java only use that it falls foul of the previous statement (I didn't think you *could* patent something that specific to a particular product). - The Oracle patents are particularly weak, most of them re-iterating 1980's knowledge of programming. - Google probably has one of the largest patent profiles ever, especially in the area of collating huge amounts of data into a database - this is commercial suicide for Oracle who could well see a retaliatory attack that they just can't afford to defend against (yes, THAT many patents). Google's patents are likely to be MUCH more substantial than these Oracle ones. - Sun never had a problem with IP protection. You don't need to protect your IP when "Java" is in everything from mobile phones to servers - basically Sun *WAS* Java and not much else before it was taken over, and saw no need to sue anyone at all substantial over patent infringement when it could have done at any time for even more cash. - Going for Google first is commercial suicide - there will be other, smaller, players using third-party Java VM's. - Suing immediately is a sign of desperation. Much more conducive to receiving compensation would have been quiet negotiations (there hasn't been ANY time for that since the Oracle takeover) and/or asking them to work around the patent at least. The path chosen is the most stupid and expensive. - The lawyers here are Boies, Schiller & Flexner - the same ones that handled the SCO case's IP side. That went well for them. *fall into fits of derisive laughter*.
Much more likely is a quiet settlement involving cash, or Google saying "Go for it" and filing a counterclaim for a whole host of patents they own. Google can pretty much take Oracle to the cleaners if it wants. It makes me wonder why Oracle has set itself up to be that target.
It depends on your country, and the exact laws. I wouldn't be so certain. In the EU, for instance, a photograph of a car breaking the law is enough to send a ticket to the house of the owner of that car. They are then legally required to identify the driver at the time of the incident - not doing so runs you into all sorts of "obstruction of justice, harbouring a criminal, etc." laws and also a lot of court time even if you just "forgot" who was driving that day - you have to basically put the court into the same doubt as yourself about who was driving, which isn't easy or cheap.
And a vast proportion of the cameras in the EU are forward-facing to catch faces. Those that aren't can normally capture enough of the inside of the car to prove male/female just from the silhouette (my father-in-law found this out when he said it was either himself or his American female friend driving, but he couldn't prove conclusively which it was at which time, so they sent another image from the same incident which proved which it was - they even chased her down to the US from the UK and sent her a ticket).
In the US things might be different, but if someone is driving - with permission - in your car and breaks any law, and the police ask who was driving at the time of the incident, you're in deep legal water if you start lying or saying you don't know. Same as if you'd knocked over a traffic light, been chased by police and were both found in the back seat of the car - they won't just let you off because they couldn't prove who was driving, they'll charge you both with obstruction. It's obvious that the legal owner of the car must know who was driving, or he'd report the car as stolen. If he does report the car as stolen, you're driving a stolen car. If he doesn't, you were driving with his permission and therefore he's obliged to identify you.
Sometimes it's not worth the hassle. Sometimes it hits a million and one legal trip-ups. But to just claim that a "real cop" must see you do something is barmy.
Maybe it's just me but I've always treated any petition in exactly the same way:
- Classmates "petitioning" to be allowed not to wear uniforms. - Facebook "petitions" to honour some mass murderer who went on a rampage with a gun then shot himself (Incidentally: Ah, diddums...) - "Petitions" to stop Hotmail charging for their service. - People who stop me in the street and expect me to be as riled about their cause as they are. - Petitions to the No10 website, or direct to MP's.
They are all a 100% waste of time. Most of the things they demand are never going to happen, or weren't going to happen as they stated anyway. Most of the times their facts are completely wrong, their opinions far too blinkered (very self-involved, most petitions) and their solutions pathetic. The petitions website for 10 Downing Street is hilarious... basically everyone asking them to dissolve government, or give them a million pounds, or implement a compulsory Naked Thursday, and they get hundreds of thousands of names - your anti-abortion, or whatever, petition with a couple of hundred names at the bottom is much more deserving of attention than sitting in that list, or in some MP's office.
To paraphrase Terry Pratchett - anything that someone has had time to write down and send to you isn't important - it's when you have people standing in front of you, banging your desk with their fist and shouting that you REALLY have to do something. I don't agree with the specifics of that, but the mood is about right. No amount of silly anonymous memos will motivate me to do something but if someone comes and takes the time to ask, describe their problem and understand the consequences, then they get ALL my attention immediately.
I'm afraid that even hearing the word "petition" puts me in mind of schoolkids "petitioning" to scrap homework, or irate working-class mothers yelling on street corners about "having a paedo that lives down the road" while holding a clipboard. I *automatically* know that I don't give a shit about it. Electronic petitions - hell, I now give even less of a shit because any idiot can click a million times on "Stop them charging for Facebook!!!!" buttons. MP's probably do have a duty to listen to all this gumpfh - and, seriously, get yourself a spam filter and a secretary - but what good do you think it will actually DO?
ID cards, ContactPoint, Heathrow's extra runway, etc. were not affected by petitions for YEARS. What stopped them is voter apathy (and thus people voting with their feet), focus on public spending wastage (not done by petitions to stop paying MP's), and people actually providing valid reasons why not (because the schemes are useless and a waste of time, for instance). In some cases even ignoring the entire scheme works more wonders than any amount of petitions (ID cards - how many people actually signed up to the "trial"? After that, it's almost impossible to actually implement the plans with a straight face).
The population aren't, on the whole, as stupid as some people would think. Nobody wanted ID cards, nobody wanted ContactPoint, there has NOT been a big fuss over scrapping them. Showing contempt for a scheme and ignoring it generates more results than silly petitions. Poll tax in the UK died because something like 30% of people refused to pay it even when it was a legal requirement and enforced heavily. Even protests carry infinitely more weight than a petition but they are just as useless - anyone who's used Trafalgar Square for the past however-many Wednesdays knows the hassle caused and the absolutely mute response from the government because of the pettiness of the issue (how many people that don't use Trafalgar Square actually knew what that was about?). The anti-war protests outside Parliament got moved on after a long time because, basically, everyone was bored of seeing them there. They achieved precisely zip. Rioting etc. solves even less problems - in fact, it just makes the cause that the rioters are fighting for seem ruthless, unca
A lot of modern cars have or have had installed Bluetooth OBD. This means it's NOT required to have physical access - you can be several km's away with a good antenna. And it also means that such tricks would work in a virtually-evidence-free way (i.e. drive past your target of, say, a princess driving through a French tunnel - turn the car's brakes off remotely by breaking the dodgy "security" on such things, and carry on driving - in the opposite direction, a mile down the road, wherever there's a relatively clear line of sight).
And when the crash investigators look, all they see is that the ECU disabled the brakes and the car crashed.
Er... yeah, seriously... you're being taken for a ride. Minimum wage is nearly double that and a legal requirement. I assume you're either a) lying, or b) taking into account your net profit after tax, which is something else entirely. Either that, or c) working in "IT" for less than you can get at McDonald's, sweeping the streets, giving out leaflets or licking envelopes. If you are genuinely working for a large and well known corporation, time to name and shame them.
Yeah, the "learning styles" things baffles most people. You're not a good teacher unless you can come up with several ways of teaching the same thing. Some people only learn by SEEING you do something - that's a hard category to cater for sometimes because it's tempting to treat them as someone who learns by DOING SOMETHING THEMSELVES. Others learn by working around similar problems, or by watching other people tackle the same problem, or by being left alone with their thoughts on the problem. Some people can only learn by analogy, even.
Most geniuses are actually people who learn by all of the above equally, hence they are able to view all sorts of solutions and everything they are exposed to becomes an education.
I'm trying to teach myself to read music at the moment - something that's evaded me for 30 years. The recommended way is to "do it a lot" and memorise that the note on the same line as the middle of the clef is whatever. You could also sit with an instructor, or write out flashcards, or try to identify the note by the sound and imagine what the note looks like on paper, or you could just try to memorise the images of notes on paper, or you could try a million different ways.
You know what works for me? I look at the notes on the paper, interpret them using a printed sheet that came with a book (takes ages) to get a key on the piano (I already can link the keys to their note names - that was automatic for some reason), then I play the tune slowly until I recognise it, then I go to bed. I guarantee that when I get up, I know the tune, and the hand positions but still not the notes in front of me. So I play the tune over and over again while reading the notes - the hand motions are by now automatic, I don't have to interpret anything and playing the tune while looking at the notes beds them into my head better than any hard-core memorisation. I can now play about 20 little ditties off by heart and I know which note is which by certain parts of those melodies (and thus new pieces of music are literally - "oh, that's the fourth note from this tune, and that's the note at the start of this tune, etc." - I do those thoughts in realtime and the music can be read seamlessly). If I haven't played a melody with a particular note in it, I don't "know" that note by sight and I stumble. Now imagine trying to find out that an 8-year-old learns best that way, or teaching a lesson in that same style.
For me, personally, I'd probably do infinitely better if my university course had been "these is all the course material you'll ever need, now learn how you like". Things like coursework and holding back what information would be taught the next week actually hindered that part of my education. I still passed but I *know* my results would have been better if I could have just downloaded a 200Mb PDF with all the required material, hints, links and everything else and audio recordings of a handful of the lectures I had. Equally, I also know that dozens of students around me at the time would have failed miserably under that system because they would have "thought" they were learning when actually they were just reading and the final exam would be something they were never prepared enough for.
I love how everyone hears rumour of magical, mystical facial recognition and automatically believes them to be true in all circumstances.
The accuracy of an industry standard facial recognition system is roundabout 54% in fairly-positive conditions. That's assuming a perfect, high-resolution image of the person involved from multiple angles, perfect reference images and everything else. Additionally, although you are captured 300 times a day on CCTV in London, London police are hastily and speedily backing away from CCTV as anything over than a deterrent - why? Because they're budgets have been cut? No. Because they want to give the public more privacy? No. Because about 80% of CCTV of "known incidents" is basically useless for identification or criminal prosecution and without getting there quick and nabbing the bloke red-handed, you can't actually use the CCTV for much else other than deterrence, detection and a seamless link between a person and earlier incidence (assuming you CAN keep the link between all your cameras seamless). It's human-work all the way down to the cop arresting some guy, not automated magic. Police CCTV is all human-operated, for a number of legal, technical and practical reasons.
Additionally, for a camera to capture an image that can do facial recognition you're looking at a minimum of 100 pixels square to do anything vaguely useful (and even then, you're chances aren't high). So to capture an image in Heathrow and have it automatically capture everyone, you're probably looking at a several-dozen-megapixel-image in realtime. Not impossible, but improbable given current technology, storage and analysis. And that's to do something quite basic. For anything "sensible", you're needing to zoom into every individual or get a VERY good photo of them as they pass through a doorway, hopefully one at a time, looking straight at the camera, in perfect lighting. Not impossible but certainly not in place at UK airports.
The "state-of-the-art" in CCTV-Britain is actually being able to read your car number plate from a handful of meters away, by focusing on the road lanes directly, one camera per lane, looking for a black-on-yellow standardised number plate and then performing bog-standard OCR on it (which can take as long as it likes once a simpler algorithm has flagged the frame as "I think this contains an area that looks like a number plate"). The accuracy of ones that are OCR'able is nowhere near 100%, even after human intervention on the tricky ones, and there are no published statistics on exactly how many images they capture are useless for performing OCR (even human) on. So the start of the art for a quite simple job that every petrol station's been doing for the last 20 years in the UK (with about the same accuracy - probably less than 70% - push it to 85% with human assistance on every frame) is a standard camera on a pole that can't do anything more than an off-the-shelf computer and webcam.
Then you get into facial recognition - determining just the outline of a face from a full-colour image is a task approximately as difficult as identifying a number plate and then OCR'ing it. To then find keypoints, adjust for camera angle, motion, to store and analyse in a statistical test against millions of other keypoint-recordings, you'll looking at an enormous amount of computing power and cleverness that's beyond most modern PhD computer vision specialists. Limited tests are always impressive but, by definition, limited. We can *just* about keep a track of which blob is what on a heat map and have some chance of following the right guy based on an analysis of blob-shapes and blob-motions. Even that isn't perfect.
We can also *just* about do something similar for genes and cells under a microscope - recognising the shape of abnormal cells/genes based on purely an image of them. It's *still* not as accurate as a human doing the same job but it saves an awful lot of time. However, identifying simple-to-spot cells robotically spread and prepared perfectl
"show us 14 photos of yourself and we can identify who you are"
I highly doubt that. I assume we're talking about a globally unique identification of a single individual. I call crap, given that we can't even do that with anything at all - fingerprints, DNA, or anything else. No biometric is that good. And, besides, if you have 14 photos of me, you know who I am anyway - I'm the guy who's in the photo. It doesn't exactly prove much at all, or help you out unless the photo shows me doing something illegal and I need to be traced. I *guarantee* you that other humans will catch me from my photo in a newspaper before any computer-based system does, and probably with much smaller margins of error.
And 14 photos is a HELL of a lot. And it depends on their quality, and your clothing, and the lighting, and the angles, and the focus, and anything obscuring the picture, and the resolution. Otherwise you're magical "14 photos" system could be used on 14 frames of any CCTV footage and instantly pinpoint the criminal. See what a ridiculous assertion that is?
Either used fixed-point (yuck), symbolic calculations and then only finding the decimal expansion at the last stage, or rewrite your formula to avoid any possible lack of precision (i.e. any division).
"I can think of only one movie where they had a robot that was clearly Man's physical (as well as mental) superior ; the panther like military robot in "Red Planet""
Apparently never watched the Terminator movies then. Those robots were vastly superior to a human physical form, and only fail because of their vulnerability to large explosions, liquid nitrogen, liquid metal, multiple huge weapons and/or human intellect outwitting them.
Robots are already, and have been for many years, physically superior to a human. That's why we use them for everything from packing sweets to building cars, to disabling bombs, to exploring other planets - they can do things we can never do as a species, and that's why we use them. Their advantages have ALWAYS come from absolute precision, perfect timing and movement, tireless working and the ability to do things quicker than any human ever could. If they were faster, stronger, more accurate, more consistent etc. than us, we'd never bother to use them at all. This is why you have "robots" in your kitchen - your blender, for instance. If I give you dough to mix, or something to chop you cannot ever do it faster than one of those could.
Robots, by nature of their construction and engineering, have always and will always be "better" at a physical task they are designed for. That it's taken this long for a simple wall-climbing robot to catch up is kinda embarrassing because they aren't limited by muscle-strength, can be constructed of very strong but extremely light materials, don't need to keep themselves "alive" as well, and there's enormous tactile and sensor-based response to work from - most wall-climbing robots don't have to do much computer-vision work, if any. When I say to you "learn to climb this wall", ingenuity gets you so far but if you're lacking in muscle and holding a lot of weight, that's your biggest obstacle. With a robot, it's only a matter of cost and construction time. And our ingenuity applied to robots is only ever "make it work like a human" (so *WE* can understand how it does it), "make it work like some animal adapted for that task", or "that's not a robot, it's just a giant wheel that goes up the wall".
Robots in a physical form have always and will always crush us, purely because of the materials they are made of. The beauty is, though, that they are only ever order-followers. They can only do what they are explicitly told to do, only sense what they are explicitly told to sense, and only react how they are explicitly told to react. They may "appear" human but they can't "think outside the box" even when the box is very simple. Otherwise, we'd already be dead.
Slashdot Mirror
Buying a game is like buying a car. DRM is like the key to that car.
If I buy a car, I do not expect to have a significant percentage of the cost of that car used to make the key. Especially when the justification for that is that the key cannot be copied by key-cutters and third-party garages.
Yes, this make the car more "secure" - for a while - because official keys cannot be fabricated without the manufacturer's co-operation. FOR A WHILE. But there isn't a car in production that has an "uncopyable" key, or that can't be broken into without any key at all - the fact that manufacturers can make a key in the first place tells you that.
However, in blocking out this "unauthorised key copying" industry (some of which is actually legitimate - not everyone who takes their keys into a key-cutting place is intending to break into someone else's house), the manufacturer is spending more time designing more and more elaborate keys (all of which can be copied "unofficially" at any garage within a matter of days of the car being produced), charging me more and more for the privilege (by moving some of the value of the game into the cost of the key itself), and in the process giving me a car that I can't always drive, sometimes won't open, that I must NEVER lose the keys to (because in a couple of years it'll be almost impossible to get an official key ever again, and in the meantime the only option for replacement is to buy an entirely new car direct from the manufacturer in order to get "another" key) and where the key weighs 12 kilos, cannot be put onto a ring with other keys, and comes only in flourescent day-glo orange (just in case someone wants to run off with it to make it work on another car of the same model) - with holographic iris-identification-over-IP built in to the key just in case you try to lend your car to someone else.
Software copyright infringement is a problem. So do something about it that a) hurts the people doing it to you and b) doesn't hurt the people who aren't. Current DRM solutions do NEITHER of those.
Less hole-punching.
I never stopped playing AOE (specifically AOE2:Conquerors). I *DID* stop playing it online because MS just sucked the life out of the multiplayer aspect by locking it to a single vendor for online matchmaking and then destroying that facility when they got bored of AOE.
So, what's here for *me*, someone that wants to play AOE but was forced by Microsoft's enforced-obsolescence to stop playing it online unless I wanted to faff about with third-party software or entering IP addresses? I won't believe it won't happen again, and I don't believe that a new MMO "reboot" will be anywhere near as good as the AOE2:Conq. And are we talking about a monthly subscription model or can I actually *OWN* the game (or at least my copy of it) forever?
In the meantime, playing the classic version over a private VPN it is.
Maybe it was just me but I found it a bit lacking. I'd heard all the hype, which I avoid paying attention to, and then when the free giveaway came around, I thought "Why not?".
There's a 100-something games on my Steam account and I got through Portal in about 3.1 hours (at least one hour of which I'd left it running trying to get the falling-far achievement while I had dinner) - that's probably the least I've invested in a Steam game, with the possible exception of some of the free stuff that's thrown at me in the bundle deals and I never even load up (and DogFighter, which I deleted after less than an hour just the other day). It was short, too short. The puzzles weren't that tricky and it was more a case of being able to aim to the spot you *knew* you had to and/or doing it inside the time necessary (like the long-double-jumps where you have to move the portal mid-way through a jump). I didn't see it as a "puzzle" game at all - it was usually obvious what you needed to do and just a matter of doing it. I think the only levels where I had to have more than one go were the ones that required strict timing or complete accuracy.
I know that there are a lot of extra puzzles and other achievements but I didn't find it very exciting or challenging at all. I'm so glad I didn't pay for it, it got deleted almost immediately after I hit the ending. Trying to tie it into the HL universe, etc. just taints an otherwise good series.
I appreciate the *technology* of being able to have extra windows onto an existing world, including recursive views, but that novelty wore off very quickly (I've seen Quake Mods that achieved the same thing), because about the only thing you can usefully do with it is something like Portal. I'm afraid I consider it in the same gasp as something like a novel indie game, or something like that. And, frankly, the whole plot/story/background/character got so annoying that I switched off all sound after about 2-3 puzzles - I always switch off music, I've never had to switch off the sound before.
I didn't think it was very original or creative beyond the use of actual recursive windows onto a 3D world. The rest of the game basically stemmed automatically from that premise. I consider Portal to be more of a tech demo - a bit like the Lost Coast thing. That's about the level it's at, about the right length, about the right kind of "storyline" added to it. I won't be buying Portal 2 unless I know it's going to evolve from Portal, not just be more of the same.
Thanks guys. I always thought I was a geek before I read the comments here - all for 56 seconds of "extra" content for a 27-year-old film.
I'd now like to hand my geek-card in to someone who has a greater need for it.
Er... yeah, if you leave the default memory cache enabled - Opera does its own in-memory caching where some other browsers rely on the underlying filesystem to cache for them, and Opera loads QT which counts as "memory used" on Windows but not under the vast amount of Linux distros that already have it in memory to be shared. There are a million and one ways to tweak Opera, which is another plus for it, including disabling quite a lot of functionality that you wouldn't want active on low-memory machines.
Opera on my computer, as an upper bound, never takes as much as an equivalent FF process with the same windows open. I have memory (in fact, all) caching disabled.
Pretty accurate - and that's from a long-term Opera user.
Shame that Opera sees such little take-up. It has 99% of the functionality of the common addons for Firefox already built-in (and has for years), it is a damn sight faster on low-end machines than Firefox, it's cross-platform, it's got a built-in mail client that is more than good enough for the average joe (with super-fast searching for EVERYTHING), and it's normally first with any innovation (WebM, Acid-compliance, HTML5, etc.) - hell, for the last version they discovered myriad websites with a common javascript bug that preventing them providing a 10.x version number in the user-agent, so they had to stick with 9.8 and some extra gumfph elsewhere to tell you the real version number. No other browser's spotted that yet.
If someone could tie Pidgin into Opera, I'd never need another bit of software again.
Same in the UK.
Your contract is with the retailer who gave you a receipt. Anything else doesn't affect your statutory rights to take the damn thing back to them and get a replacement/refund. You do *NOT* have to accept if they say otherwise. To be honest, you do not need the receipt either but try convincing a small-store manager of that.
If in doubt, send a stroppy legal letter to the RETAILER, not the manufacturer - they are legally obliged to deal with it and they know it (store managers tend to hide behind "company policy" which does not accurately reflect the law. You can't be expected to deal with a guy in China who you have no business relationship with in order to get a working product - the guy who sold you the device and gave you a receipt is the one who has to sort it out.
Cost, cost, cost and, er, cost.
But let me go into more detail:
Air-pressure monitors on your tyres can get stolen if they are external (i.e. valve caps), or require special tyres if they're internal. Kids round here steal valve caps just because it's funny. Sod having to replace a miniature wireless pressure sensor each time, or add the cost of a non-standardised wireless hardware to the price of each wheel / tyre (at the moment - it's cheaper for me to replace the entire wheel than just the rubber tyre and I can get a whole set for less than the price of a Wii). Also that requires dashboard indicators, wireless communications, etc. or some very clever mechanics to keep it wired (and an air-line to your tyres is a silly idea because it would be almost impossible to design a leak-less, rotating valve and if you could, again, you've just quadrupled the cost of a new set of tyres and any minor repairs to your car). Besides, a lot of top-end cars DO now have the pressure sensors on their wheels. You have to buy the manufacturer's tyre/wheel if they break or some of them require batteries, etc.
Batteries - this would require another very-heavy-duty battery that is charged over a split-load charger along with the main battery. Not a problem but several hundred dollars worth of equipment, plus the place to put it (although a lot of cars are designed to have the battery on either side for their other-hand-drive models), plus twice the replacement costs (not to mention the specialised charger, which CAN be just a huge capacitor / diode I think). Additionally, you have to have extra cabling and several kg's more weight (lead-acid batteries are not light - go pick one up). To compensate for what is currently a little wiring that cuts everything but the side-lights and hazard flashers when the key is out of the ignition. Most cars solved this problem decades ago by cabling accessories through the ignition and providing a headlight-warning alarm. Cheaper, simpler, more efficient on your fuel, better for the environment.
Mobile phone charger / Bluetooth speakerphone - upgrade your stereo, or your car. I bought the cheapest stereo I could find recently (to replace a tape deck in my car) and it came with USB ports on the front and Bluetooth functionality - I can charge anything with a USB cable. And providing a slot for a mobile phone to sit in would cause merry hell with some countries regulations which often cite that such things have to be out of reach of the driver when they are driving. That's why a lot of sat-navs have their own custom sucker-window things, and for pocket-change you can get a mobile phone holder that plugs into your heating vents or similar.
And all of the above suggestions don't actually HELP your car efficiency at all. The batteries mean you are constantly (over-)charging two horribly poisonous batteries where one would do, the mobile phone / Bluetooth thing is just convenience - where you charge your phone does not affect your driving range and personally I'd rather you weren't answering ANY calls while you were driving, hands-free or not. The wireless pressure monitors put inordinate amounts of engineering, possibly more horrible-chemistry batteries, extra weight, wiring and display into a car to do something that just LOOKING at a tyre can tell you. That's not productive for efficiency.
It's been over a hundred years with these machines - and in all that time you've been required to ensure your car is roadworthy. Believe it or not, a little blinking light on your dashboard CAN'T tell you that - you have to inspect the tyre yourself. While you're there, check the pressure by sight/hand/gauge, you lazy git.
And that the switches weren't blocking DHCP from anything but the authorised DHCP server, and that it took 3 days to track down a rogue DHCP server (not hard, you usually get the MAC address in seconds, trace that to a port, disconnect the port and see who shouts that their network connection isn't working - if it's a remote switch on the end of that port, go to that switch, rinse and repeat).
Hell, it would take less that an hour if you just pulled cables at random until that MAC disappeared.
Like most of the things in the story - incompetent admins and IT setups allow human error to be amplified. Seriously, one of them is basically a hospital network not using spanning-tree.
According to the all-knowing Wiki: "phase I of the GSM specifications were published in 1990"
So, depending on your point of view:
- it's taken 20 years to implement something that had a published standard and worldwide, cheap hardware examples used by millions of people every day.
- the standards took 20 years for an outsider to be able to implement them independently.
And we're still only talking alpha code with specialised hardware.
- Software patents are not considered valid in the majority of the world, precisely because they get in the way of perfectly reasonable actions.
- Software patents on an "open" standard might not be enforceable if those patents are required to implement that standard, and are the only sensible way to do so. Interoperability with an existing product can't be protected by a patent if that patent is the only (or only sensible) way to do things.
- All the mentioned patents have prior art except one, which is so far worded towards Java only use that it falls foul of the previous statement (I didn't think you *could* patent something that specific to a particular product).
- The Oracle patents are particularly weak, most of them re-iterating 1980's knowledge of programming.
- Google probably has one of the largest patent profiles ever, especially in the area of collating huge amounts of data into a database - this is commercial suicide for Oracle who could well see a retaliatory attack that they just can't afford to defend against (yes, THAT many patents). Google's patents are likely to be MUCH more substantial than these Oracle ones.
- Sun never had a problem with IP protection. You don't need to protect your IP when "Java" is in everything from mobile phones to servers - basically Sun *WAS* Java and not much else before it was taken over, and saw no need to sue anyone at all substantial over patent infringement when it could have done at any time for even more cash.
- Going for Google first is commercial suicide - there will be other, smaller, players using third-party Java VM's.
- Suing immediately is a sign of desperation. Much more conducive to receiving compensation would have been quiet negotiations (there hasn't been ANY time for that since the Oracle takeover) and/or asking them to work around the patent at least. The path chosen is the most stupid and expensive.
- The lawyers here are Boies, Schiller & Flexner - the same ones that handled the SCO case's IP side. That went well for them. *fall into fits of derisive laughter*.
Much more likely is a quiet settlement involving cash, or Google saying "Go for it" and filing a counterclaim for a whole host of patents they own. Google can pretty much take Oracle to the cleaners if it wants. It makes me wonder why Oracle has set itself up to be that target.
God, it's like SCO & IBM again.
Stepping on the toes of just one the world's largest corporations not enough for them?
It depends on your country, and the exact laws. I wouldn't be so certain. In the EU, for instance, a photograph of a car breaking the law is enough to send a ticket to the house of the owner of that car. They are then legally required to identify the driver at the time of the incident - not doing so runs you into all sorts of "obstruction of justice, harbouring a criminal, etc." laws and also a lot of court time even if you just "forgot" who was driving that day - you have to basically put the court into the same doubt as yourself about who was driving, which isn't easy or cheap.
And a vast proportion of the cameras in the EU are forward-facing to catch faces. Those that aren't can normally capture enough of the inside of the car to prove male/female just from the silhouette (my father-in-law found this out when he said it was either himself or his American female friend driving, but he couldn't prove conclusively which it was at which time, so they sent another image from the same incident which proved which it was - they even chased her down to the US from the UK and sent her a ticket).
In the US things might be different, but if someone is driving - with permission - in your car and breaks any law, and the police ask who was driving at the time of the incident, you're in deep legal water if you start lying or saying you don't know. Same as if you'd knocked over a traffic light, been chased by police and were both found in the back seat of the car - they won't just let you off because they couldn't prove who was driving, they'll charge you both with obstruction. It's obvious that the legal owner of the car must know who was driving, or he'd report the car as stolen. If he does report the car as stolen, you're driving a stolen car. If he doesn't, you were driving with his permission and therefore he's obliged to identify you.
Sometimes it's not worth the hassle. Sometimes it hits a million and one legal trip-ups. But to just claim that a "real cop" must see you do something is barmy.
Maybe it's just me but I've always treated any petition in exactly the same way:
- Classmates "petitioning" to be allowed not to wear uniforms.
- Facebook "petitions" to honour some mass murderer who went on a rampage with a gun then shot himself (Incidentally: Ah, diddums...)
- "Petitions" to stop Hotmail charging for their service.
- People who stop me in the street and expect me to be as riled about their cause as they are.
- Petitions to the No10 website, or direct to MP's.
They are all a 100% waste of time. Most of the things they demand are never going to happen, or weren't going to happen as they stated anyway. Most of the times their facts are completely wrong, their opinions far too blinkered (very self-involved, most petitions) and their solutions pathetic. The petitions website for 10 Downing Street is hilarious... basically everyone asking them to dissolve government, or give them a million pounds, or implement a compulsory Naked Thursday, and they get hundreds of thousands of names - your anti-abortion, or whatever, petition with a couple of hundred names at the bottom is much more deserving of attention than sitting in that list, or in some MP's office.
To paraphrase Terry Pratchett - anything that someone has had time to write down and send to you isn't important - it's when you have people standing in front of you, banging your desk with their fist and shouting that you REALLY have to do something. I don't agree with the specifics of that, but the mood is about right. No amount of silly anonymous memos will motivate me to do something but if someone comes and takes the time to ask, describe their problem and understand the consequences, then they get ALL my attention immediately.
I'm afraid that even hearing the word "petition" puts me in mind of schoolkids "petitioning" to scrap homework, or irate working-class mothers yelling on street corners about "having a paedo that lives down the road" while holding a clipboard. I *automatically* know that I don't give a shit about it. Electronic petitions - hell, I now give even less of a shit because any idiot can click a million times on "Stop them charging for Facebook!!!!" buttons. MP's probably do have a duty to listen to all this gumpfh - and, seriously, get yourself a spam filter and a secretary - but what good do you think it will actually DO?
ID cards, ContactPoint, Heathrow's extra runway, etc. were not affected by petitions for YEARS. What stopped them is voter apathy (and thus people voting with their feet), focus on public spending wastage (not done by petitions to stop paying MP's), and people actually providing valid reasons why not (because the schemes are useless and a waste of time, for instance). In some cases even ignoring the entire scheme works more wonders than any amount of petitions (ID cards - how many people actually signed up to the "trial"? After that, it's almost impossible to actually implement the plans with a straight face).
The population aren't, on the whole, as stupid as some people would think. Nobody wanted ID cards, nobody wanted ContactPoint, there has NOT been a big fuss over scrapping them. Showing contempt for a scheme and ignoring it generates more results than silly petitions. Poll tax in the UK died because something like 30% of people refused to pay it even when it was a legal requirement and enforced heavily. Even protests carry infinitely more weight than a petition but they are just as useless - anyone who's used Trafalgar Square for the past however-many Wednesdays knows the hassle caused and the absolutely mute response from the government because of the pettiness of the issue (how many people that don't use Trafalgar Square actually knew what that was about?). The anti-war protests outside Parliament got moved on after a long time because, basically, everyone was bored of seeing them there. They achieved precisely zip. Rioting etc. solves even less problems - in fact, it just makes the cause that the rioters are fighting for seem ruthless, unca
A lot of modern cars have or have had installed Bluetooth OBD. This means it's NOT required to have physical access - you can be several km's away with a good antenna. And it also means that such tricks would work in a virtually-evidence-free way (i.e. drive past your target of, say, a princess driving through a French tunnel - turn the car's brakes off remotely by breaking the dodgy "security" on such things, and carry on driving - in the opposite direction, a mile down the road, wherever there's a relatively clear line of sight).
And when the crash investigators look, all they see is that the ECU disabled the brakes and the car crashed.
You need to buy an alarm clock that has a light. Problem solved.
Er... yeah, seriously... you're being taken for a ride. Minimum wage is nearly double that and a legal requirement. I assume you're either a) lying, or b) taking into account your net profit after tax, which is something else entirely. Either that, or c) working in "IT" for less than you can get at McDonald's, sweeping the streets, giving out leaflets or licking envelopes. If you are genuinely working for a large and well known corporation, time to name and shame them.
Yeah, the "learning styles" things baffles most people. You're not a good teacher unless you can come up with several ways of teaching the same thing. Some people only learn by SEEING you do something - that's a hard category to cater for sometimes because it's tempting to treat them as someone who learns by DOING SOMETHING THEMSELVES. Others learn by working around similar problems, or by watching other people tackle the same problem, or by being left alone with their thoughts on the problem. Some people can only learn by analogy, even.
Most geniuses are actually people who learn by all of the above equally, hence they are able to view all sorts of solutions and everything they are exposed to becomes an education.
I'm trying to teach myself to read music at the moment - something that's evaded me for 30 years. The recommended way is to "do it a lot" and memorise that the note on the same line as the middle of the clef is whatever. You could also sit with an instructor, or write out flashcards, or try to identify the note by the sound and imagine what the note looks like on paper, or you could just try to memorise the images of notes on paper, or you could try a million different ways.
You know what works for me? I look at the notes on the paper, interpret them using a printed sheet that came with a book (takes ages) to get a key on the piano (I already can link the keys to their note names - that was automatic for some reason), then I play the tune slowly until I recognise it, then I go to bed. I guarantee that when I get up, I know the tune, and the hand positions but still not the notes in front of me. So I play the tune over and over again while reading the notes - the hand motions are by now automatic, I don't have to interpret anything and playing the tune while looking at the notes beds them into my head better than any hard-core memorisation. I can now play about 20 little ditties off by heart and I know which note is which by certain parts of those melodies (and thus new pieces of music are literally - "oh, that's the fourth note from this tune, and that's the note at the start of this tune, etc." - I do those thoughts in realtime and the music can be read seamlessly). If I haven't played a melody with a particular note in it, I don't "know" that note by sight and I stumble. Now imagine trying to find out that an 8-year-old learns best that way, or teaching a lesson in that same style.
For me, personally, I'd probably do infinitely better if my university course had been "these is all the course material you'll ever need, now learn how you like". Things like coursework and holding back what information would be taught the next week actually hindered that part of my education. I still passed but I *know* my results would have been better if I could have just downloaded a 200Mb PDF with all the required material, hints, links and everything else and audio recordings of a handful of the lectures I had. Equally, I also know that dozens of students around me at the time would have failed miserably under that system because they would have "thought" they were learning when actually they were just reading and the final exam would be something they were never prepared enough for.
Drive Bay UPS's aren't hard to come by. They only last for a few minutes normally but if it's something useful, practical and worthwhile.
Nah, you're bound to telefrag someone at some point.
I love how everyone hears rumour of magical, mystical facial recognition and automatically believes them to be true in all circumstances.
The accuracy of an industry standard facial recognition system is roundabout 54% in fairly-positive conditions. That's assuming a perfect, high-resolution image of the person involved from multiple angles, perfect reference images and everything else. Additionally, although you are captured 300 times a day on CCTV in London, London police are hastily and speedily backing away from CCTV as anything over than a deterrent - why? Because they're budgets have been cut? No. Because they want to give the public more privacy? No. Because about 80% of CCTV of "known incidents" is basically useless for identification or criminal prosecution and without getting there quick and nabbing the bloke red-handed, you can't actually use the CCTV for much else other than deterrence, detection and a seamless link between a person and earlier incidence (assuming you CAN keep the link between all your cameras seamless). It's human-work all the way down to the cop arresting some guy, not automated magic. Police CCTV is all human-operated, for a number of legal, technical and practical reasons.
Additionally, for a camera to capture an image that can do facial recognition you're looking at a minimum of 100 pixels square to do anything vaguely useful (and even then, you're chances aren't high). So to capture an image in Heathrow and have it automatically capture everyone, you're probably looking at a several-dozen-megapixel-image in realtime. Not impossible, but improbable given current technology, storage and analysis. And that's to do something quite basic. For anything "sensible", you're needing to zoom into every individual or get a VERY good photo of them as they pass through a doorway, hopefully one at a time, looking straight at the camera, in perfect lighting. Not impossible but certainly not in place at UK airports.
The "state-of-the-art" in CCTV-Britain is actually being able to read your car number plate from a handful of meters away, by focusing on the road lanes directly, one camera per lane, looking for a black-on-yellow standardised number plate and then performing bog-standard OCR on it (which can take as long as it likes once a simpler algorithm has flagged the frame as "I think this contains an area that looks like a number plate"). The accuracy of ones that are OCR'able is nowhere near 100%, even after human intervention on the tricky ones, and there are no published statistics on exactly how many images they capture are useless for performing OCR (even human) on. So the start of the art for a quite simple job that every petrol station's been doing for the last 20 years in the UK (with about the same accuracy - probably less than 70% - push it to 85% with human assistance on every frame) is a standard camera on a pole that can't do anything more than an off-the-shelf computer and webcam.
Then you get into facial recognition - determining just the outline of a face from a full-colour image is a task approximately as difficult as identifying a number plate and then OCR'ing it. To then find keypoints, adjust for camera angle, motion, to store and analyse in a statistical test against millions of other keypoint-recordings, you'll looking at an enormous amount of computing power and cleverness that's beyond most modern PhD computer vision specialists. Limited tests are always impressive but, by definition, limited. We can *just* about keep a track of which blob is what on a heat map and have some chance of following the right guy based on an analysis of blob-shapes and blob-motions. Even that isn't perfect.
We can also *just* about do something similar for genes and cells under a microscope - recognising the shape of abnormal cells/genes based on purely an image of them. It's *still* not as accurate as a human doing the same job but it saves an awful lot of time. However, identifying simple-to-spot cells robotically spread and prepared perfectl
"show us 14 photos of yourself and we can identify who you are"
I highly doubt that. I assume we're talking about a globally unique identification of a single individual. I call crap, given that we can't even do that with anything at all - fingerprints, DNA, or anything else. No biometric is that good. And, besides, if you have 14 photos of me, you know who I am anyway - I'm the guy who's in the photo. It doesn't exactly prove much at all, or help you out unless the photo shows me doing something illegal and I need to be traced. I *guarantee* you that other humans will catch me from my photo in a newspaper before any computer-based system does, and probably with much smaller margins of error.
And 14 photos is a HELL of a lot. And it depends on their quality, and your clothing, and the lighting, and the angles, and the focus, and anything obscuring the picture, and the resolution. Otherwise you're magical "14 photos" system could be used on 14 frames of any CCTV footage and instantly pinpoint the criminal. See what a ridiculous assertion that is?
Yes. Avoid floating-point.
Either used fixed-point (yuck), symbolic calculations and then only finding the decimal expansion at the last stage, or rewrite your formula to avoid any possible lack of precision (i.e. any division).
"I can think of only one movie where they had a robot that was clearly Man's physical (as well as mental) superior ; the panther like military robot in "Red Planet""
Apparently never watched the Terminator movies then. Those robots were vastly superior to a human physical form, and only fail because of their vulnerability to large explosions, liquid nitrogen, liquid metal, multiple huge weapons and/or human intellect outwitting them.
Robots are already, and have been for many years, physically superior to a human. That's why we use them for everything from packing sweets to building cars, to disabling bombs, to exploring other planets - they can do things we can never do as a species, and that's why we use them. Their advantages have ALWAYS come from absolute precision, perfect timing and movement, tireless working and the ability to do things quicker than any human ever could. If they were faster, stronger, more accurate, more consistent etc. than us, we'd never bother to use them at all. This is why you have "robots" in your kitchen - your blender, for instance. If I give you dough to mix, or something to chop you cannot ever do it faster than one of those could.
Robots, by nature of their construction and engineering, have always and will always be "better" at a physical task they are designed for. That it's taken this long for a simple wall-climbing robot to catch up is kinda embarrassing because they aren't limited by muscle-strength, can be constructed of very strong but extremely light materials, don't need to keep themselves "alive" as well, and there's enormous tactile and sensor-based response to work from - most wall-climbing robots don't have to do much computer-vision work, if any. When I say to you "learn to climb this wall", ingenuity gets you so far but if you're lacking in muscle and holding a lot of weight, that's your biggest obstacle. With a robot, it's only a matter of cost and construction time. And our ingenuity applied to robots is only ever "make it work like a human" (so *WE* can understand how it does it), "make it work like some animal adapted for that task", or "that's not a robot, it's just a giant wheel that goes up the wall".
Robots in a physical form have always and will always crush us, purely because of the materials they are made of. The beauty is, though, that they are only ever order-followers. They can only do what they are explicitly told to do, only sense what they are explicitly told to sense, and only react how they are explicitly told to react. They may "appear" human but they can't "think outside the box" even when the box is very simple. Otherwise, we'd already be dead.