In March, I wrote a post on Fault Tolerance for the Hiperwall blog. I’ll link to it here since my name didn’t get attached to it there.
|
|||||
|
At Hiperwall, our tagline is “See the big picture” and we do that well. In applications from scientific visualization to control rooms and operations centers, being able to see lots of information in great detail allows our users to understand more clearly and make important decisions quickly. We’ve shown billion pixel images on Hiperwall systems, but haven’t had anything larger until now. One of our developers used a capability provided by an NVIDIA library that the developers of Witcher 3: The Wild Hunt included in their video game to take an enormous game capture. He chose a resolution of more than 61,000 pixels by 34,000 pixels, so 2 gigapixels altogether. It took an absolute beast of a computer about a minute to render and save the resulting 1 GB file. Once the image was imported into our Hiperwall system, we could see how amazing it looked. Here we’re showing the image fully zoomed out on our small 24 megapixel Hiperwall, so we can see Geralt on his horse. Click on the photo to see it in more detail.
 Geralt in a 2Gpixel image shown on Hiperwall When we zoom in to see his face, however, we can see the amazing detail in the rendering. We can see details of his witcher eye and the links on his armor. This image shows a zoom level of 1.0, so every pixel on the Hiperwall shows one pixel from the image.
Finally, we animated zooming in on the image, so you can see how smoothly we can manipulate images, even those with 2 billion pixels. (Sorry about the canned music – we had people talking in the room as I shot the video.) We had a heckuva storm on Friday and a bit more on Saturday, and it really roiled the ocean. We went to Treasure Island Park in Laguna Beach early this morning and saw lots of crashing surf and splashing waves. I don’t recall ever seeing the sea so active in Laguna, at least not recently. The pictures don’t really do it justice, but you can get some idea of what we saw. The video shows the waves on some rocks – I liked watching the water drain off the rock as the wave subsides. Almost every software product has a click-through End User License Agreement (EULA) that we typically just click OK on so we can actually use the software. Perhaps we shouldn’t be so quick to click, because we’re often giving up our privacy or agreeing to other restrictions, but the software companies have us over a barrel, so we dutifully agree and move on. This is a real problem. EULAs were originally intended to protect the software company from liability and to prevent us from ripping them off. More recently, however, they’re throwing in binding arbitration clauses (which can be argued are legitimate protection for the company, but it does restrict our rights) and privacy clauses, which is the subject of this article. I believe these EULA problems can only be fixed through legislation, because competition and free enterprise don’t apply after we’ve already bought the product. I have a Sleep Number bed and bought the SleepIQ add-on. SleepIQ is a $300 system in the pump (the mattress is filled with air) that measures your breathing, heart rate, and sleep. While it probably isn’t really worth $300, it is pretty cool, so when you buy a Sleep Number bed (and you should – they’re good), buy SleepIQ too. The sensor in the pump talks to an app on my iPhone, and the app was recently updated, which required me to click through a EULA. Well, I read this one, and it stinks. They threw in that they can do nearly anything they want with my private personal information that they’ve collected, including using it for marketing (not a surprise) or selling it to their business partners (bad). If I didn’t agree with the EULA, I needed to disable the SleepIQ system, thus throwing out the $300 I paid. It did say I could continue to use the bed – how generous of them! This sort of thing isn’t any different than if you bought a car, and when you got it home, the entertainment/navigation system popped up a EULA that required you agree to all sorts of abusive terms or they’d disable all the features. That likely wouldn’t fly in cars, because consumer protection for cars is pretty good (lemon laws) and the consumer outrage would torpedo sales from that manufacturer. Sadly, with most other products, we’re stuck with whatever they want to give us. What should be done? I think EULAs should be restricted to EULA things, like defining the protections for the software company and restricting copying and other things. Agreeing to such a EULA could still be required before the user can use the software. Privacy statements have no place in the EULA, and I think protection of our personal data should be a legislated default. In no way should we be forced to give up protection of our private data in order to use the service. Of course, the company can offer us incentives that make the service more effective if we opt in to sharing our information, but it should not be mandatory. There are some services whose whole purpose is to share our data with someone – well they should get our permission explicitly, as well. The problem is, companies aren’t going to do this voluntarily. There’s profit in our names and email addresses and personal preferences. Selling that information is essentially free money for a company that makes software or other products. Since consumers don’t often know what we’re signing away and realize that we have no choice in doing so, there’s little market pressure to stop this practice. That’s why (I hate to admit it) government should get involved and protect the rights and privacy of its citizens. We need to stop companies forcing customers to agree bad things in order to use services we pay for. This sort of behavior should not be accepted or allowed. For those of us that have used Windows 7 for a long time, the Windows 10 login mechanism is annoying, and that’s doubly so when your machine is locked. So I’ve made a handy guide that describes the login process for a locked Windows 10 system.
So that’s it. If we’re used to Windows 7 just having the password box on the lock screen and just being able to type our password to unlock it, it takes typing the password 3 times under Windows 10. Sometimes form does not triumph over function. My wife just got a new, modern car with a fancy Ford Sync 3 entertainment system, and I’m a little jealous of that. This is the first car we’ve had with Bluetooth and built-in navigation and the ability to play MP3s and such. But these systems are far from perfect. Every review of pretty much every car other than the Teslas complains about the entertainment systems and how fiddly and unintuitive they are. Somehow Tesla gets it right, while the traditional car companies are terrible at this sort of thing. Or perhaps the reviewers are so impressed by being pushed back in their seat by the Tesla’s massive acceleration and the giant iPad-looking screen is so pretty that they can’t think straight. Or they’re afraid Elon Musk will cancel their Tesla order if they say anything bad. In any case, I don’t find the Ford Sync 3 to be hard to use, but it is much farther from being good than it should be to fit in the center of the dashboard of an automobile. This version ditches the Microsoft influence for a version built on the QNX real-time operating system bought by RIM a couple years back. QNX is a good choice to build a real-time interactive entertainment system on, but the Sync software on top of it needs a lot of help. The very first day we got the car, I went out to learn about the Sync system and the damn thing locked up within seconds of me touching it. I had force it to reset, which wasn’t too hard, but messes things up (more below). This happened again this weekend as we were driving. I was in the passenger seat and wondering why it couldn’t find the iPod I stored a bunch of music on (more on that later too), and the touch screen locked up. Once I reset it, it behaved, but having to reset it twice in just over a week tells me it isn’t very reliable, and it needs to be. Anything that could distract or stress a driver should be eliminated as the highest priority, and having your entertainment/navigation system lose its mind is a good way to distract and stress a driver. Speaking of navigation systems, this Sync 3 system has a proprietary nav system that is reasonably friendly and competent, but not nearly as nice as the old Garmin Nuvi my wife was using. The Sync nav system is quite out of date (Nordstrom has not been in the South Bay Galleria for quite a while now), and Ford says we can purchase map updates from the dealer. Well, it is a new car, so the map should be updated, but isn’t. And my wife’s old Garmin had lifetime map updates included. The biggest thing my wife misses about her Garmin though is the current speed display on the GPS screen. This independent speed display is quite handy if your speedometer is hard to read or not very precise, or you just don’t want to look down often. I can see why Ford wouldn’t want this, because it might show a slight difference between GPS speed and the speedometer, and they wouldn’t want anyone to think their car is reading the wrong speed. But I’ll trust GPS speed any day, and it is a shame it isn’t an option. Now the biggest problem with resetting the system is that it loses the correct time. That’s right, it forgets the correct time! And if you ask it to set according to the GPS time, it can’t handle Daylight Savings time. Seriously. This should not be a problem in a fancy expensive computerized entertainment/nav system. The Garmin unit that I have and the one my wife had handle Daylight Savings time just fine, so the fact that a modern system can’t is just sloppy/lazy programming. Another problem is that it keeps locking up the iPod I put a bunch of song on. It seems to me that iPods should be a pretty well known device to be able to handle well, and maybe the problem is on the iPod side, but it shouldn’t lock it up. When that happens, the iPod can’t be found in the source list and has to be re-scanned after I reset the iPod. Yet more distractions that shouldn’t ever happen. The system has voice commands that are mostly OK, but it is so tedious to use, we just use Siri from our iPhones to make calls and read texts and such. The built-in stuff isn’t nearly as good. Luckily, holding the voice command handle for a couple seconds activates Siri through the built-in sound system. All this leads to the problem: the car makers want to own this stuff and charge too much for it, so they make their own software and interfaces, and none of it is as good as what Apple and Google make. While CarPlay and AndroidAuto are available in some cars, they are not common yet, and they will still require some interaction with the lousy built-in systems to work. It could be that Tesla has all of this solved, but Ford sure doesn’t. I never write about politics or economics, because I’m not an expert in these things, but California is about to pass a huge increase in the minimum wage (over time), and that’s something I have an opinion about. Talking about this is a little scary, particularly in California, where my opinion won’t be popular, but my opinion is for a different reason than you think. Lots of noise has been made about how this will hurt small businesses, and more noise has been made about the need for a living wage, and there is truth on both sides, but that’s not what I care about here. I think the wage rise is unfair and demoralizing to the very people it is supposed to help. In college, I worked a couple of minimum wage jobs. Well, they started at minimum wage, but quickly I took on new responsibilities or showed my value and got raises, which made me very happy. Now if the minimum wage was raised so the new hire makes as much as I do, well then I’d be unhappy. And there’s no way I’d be able to “catch up” to the wage curve the new hire is on. The effect of these rapid increases over time is that they will compress the salaries of many workers to the minimum wage. Employers are going to have to make everyone earn at least the minimum wage, but if I was earning a dollar more than minimum, I’d guess that that dollar gap wouldn’t be kept after the minimum is raised, because of the newly increased expenses. Hopefully, if the employer is nice and wants to reward achievement, I’d still be a little above minimum, but there’s no guarantee of that. So the new people catch up to previous high achievers each time the minimum wage increases. You can argue that nothing was taken from the high achievers, and, in fact, they will be caught up in the wave of increasing wages, but I think it would be demoralizing to the high achievers, and they’re the future of the business, so it’s a bad idea to demoralize them. While I was at Northrop Grumman, we had some boom times in the economy where engineers became expensive. We were hiring in new engineers at salaries that weren’t far below the salary that I was earning even after getting good raises every year. This was depressing, because the engineers in my cohort were on a much lower wage curve than these new engineers that started so much higher than we did (even accounting for inflation). The only way to change the curve is to leave and get higher pay at another job, which isn’t what any company would want. So I know how it feels to have new hires fresh out of school earning nearly as much, and it is demoralizing. I can guess this will happen near the minimum wage level as well. I don’t have a solution to propose. The increase is a done deal and will be signed into law today. But we shouldn’t be surprised if it doesn’t help as much as people think it will, and I think some of that will be the effect I described above. My initial reaction to Apple fighting the court order to unlock the terrorist’s iPhone was “Good for Apple. Doing anything that could reduce my privacy is bad!” And lots of the breathless news reports continued to make me think that way with all the talk of backdoors and such. But then I started reading articles in more technical press and I see that the issue is a lot more nuanced (and way more nuanced than Donald Trump’s absurd call for a boycott – didn’t he learn when people called for boycotts of his enterprises?). It seems that the FBI isn’t asking Apple to unlock the phone, but to give it the tools by which the FBI can unlock the phone (and only this phone). They want Apple to make a special version of iOS tailored to that phone only that drops the policy of erasing the phone after 10 failed PIN entries and they want this new iOS to accept PIN entries quickly rather than enforcing significant delays after failed PIN entries. This would allow the FBI to unlock the iPhone quickly by trying the 10,000 possible PIN combinations several per second (via a debugging interface). So once I learned that, I thought that it wouldn’t be the end of the world if Apple complied, because this custom version of iOS would only work on that phone and, if a bad guy got ahold of it and changed it to work on another phone, the digital signature wouldn’t match, so it couldn’t be loaded on the other phone. So Apple wasn’t creating a terrible security hole that could be exploited by governments, corporations, criminals, etc. And then what I think is the real problem hit me: If Apple shows they can (help) unlock this iPhone, then they will be inundated by court orders to unlock iPhones. Every time someone disappears, the police will ask a court to force Apple to unlock their phone to see if they’d been messaging someone suspicious. If grandma dies, her relatives will want her phone unlocked in case there were photos or other information they want. And you know there will be judges that will grant such orders. So if Apple goes along this once with this seemingly reasonable case, they (and Samsung and Google and everyone else) will have to do it every day. The floodgates will open and the court orders will never end. And that’s not the business Apple or any of them want to be in. So how do we solve this? Maybe there are compelling reasons that a phone needs to be unlocked, and this case is as good an example as any. But I don’t think a court order is the answer. As Apple says, we need a public discussion, and we probably need new laws that define exactly when unauthorized unlocking is appropriate (and I mean “unauthorized” because the unlocking is being done without permission of the original setter of the passcode). That could avoid the stream of arbitrary court orders. If the lawful reasons for unlocking are broad enough, then Apple can just set up a side business that unlocks iPhones for $10,000 or $50,000. That would deter common street criminals from using it, but if the information on the phone is that important, then it’s totally worth it. I registered my drone. Well, actually, I registered myself as a drone owner. Because the FAA mandated that we all register our drones, I went to the FAA UAS (Unmanned Aircraft Systems) website to register my DJI Phantom 3 Pro (which is a really nifty drone). It turns out that only US citizens can register drones. So green card holders, visa holders, visitors, and illegal immigrants don’t have to register their drones. But since I’m a citizen, I created my account, typed in my name, address, phone number, and credit card info, which resulted in a registration ID that I am to mark on my drone. There were no questions about how many drones I have, nor what type, nor even the serial number(s). What this means is that I registered myself, putting my information into yet another government database that will become public information sometime, thus allowing advertisers and fraudsters yet another way to get my info. And, of course, if the site is ever hacked, the info will get out even quicker. I had to put my credit card info in, because drone (owner) registration costs $5, though that fee is waved for a while. Rather than just not charging me and needing my credit card info, I will be charged, then refunded the $5, which seems to be the least efficient way to do it. Someone is losing money on those two transactions, whether it is the government or the credit card company or both, but such things aren’t free. So now that I’m registered and when I mark my drone, what good will it do? Unless someone crashes their drone while doing some forbidden activity and the police get the ID number, there’s no way to track a rogue drone to its owner. Most of the cases we heard about on the news where drones were flying near firefighters and emergency personnel (or spying into windows) didn’t have the drone being recovered by law enforcement. Instead, the owners flew the drones away, so nobody could see the ID numbers. So in my opinion, the current registration scheme doesn’t do much to match drones with owners, except in cases where the drone crashes, but it has added a new bureaucracy at the FAA with user fees to help support it. There are much better alternatives to this ineffective system, and I’d bet drone makers and owners would be better served by them. I have suggestions: Most modern fancy drones have one or more radio transceivers as well as GPS positioning in them. The drone receives commands from the controller and sends back status information. Many drones also send back camera video, which is a reasonable high data rate. So they have fairly powerful radios that can send for half a mile or more. I suggest that every few seconds, that radio could send out identifying information as well as the drone’s position and maybe even the takeoff/landing coordinates on a designated frequency. This would be sort of like IFF beacons on aircraft, and would let law enforcement know which drones are nearby, where they are, and where their operator is. That’s information that can be used if the drone enters airspace it shouldn’t or causes a nuisance. Perhaps this could be added to existing drones via a firmware update, but could certainly be mandated for all drones sold after a certain date. Of course the drone manufacturers would have to be on board, but I think it is in their best interest to make sure the hobby maintains a safe and lawful reputation. If the drone manufacturers think making changes to the in-flight radio is dangerous and could compromise flight safety, the handheld control unit also has powerful transceivers and could do the job nearly as well. It sends commands to the drone and receives telemetry as well as the video stream, so its radio could send the ID/position beacon without interfering with flight safety. In either case, a technical solution would provide information that is much more useful. Even better, it would do so for all drones, not just those owned by US citizens who have registered. I think the drone hobby is fun and always take care when flying my drone, but I can understand the need for regulations. I just think the regulations should make sense and be effective. I got my new Microsoft Surface Book last night and have decidedly mixed impressions of it. The hardware seems top notch, while the operating system and software are much worse than expected. If only Microsoft were a company that primarily made operating systems and software, I’m sure this sort of thing wouldn’t happen. Oh, wait… So the first thing I noticed is how heavy the thing is, at least in its shipping box. It turns out that the Surface Book itself is not very heavy (on the order of 3 pounds, so heavy for a tablet, but not bad for a thin laptop), but the power supply brick, while small, must be lined with lead, because it is heavy! The pen is also surprisingly substantial, and feels much more solid than the plastic pen of my Surface Pro that I’m writing on now. Even the box the unit came in weighs as much as the Surface Book, it seems. So the Surface Book is a very well crafted piece of hardware, with its nifty hinge and detachable tablet, it is very clever and very solid. I was very impressed to get an Apple-like experience in taking it out of the box and setting it up. Then I had to find the power button. It is hidden at the top left of the screen, rather than on the keyboard like most laptops. I realize that the tablet (really the computer) is separate from the unit, but I think it would have helped to have an additional power button on the keyboard section. (Now we’re talking like Star Trek: The Next Generation’s Enterprise which had the dockable saucer section and engineering section – the Surface Book has a dockable tablet and keyboard section.) Once I found the power button and did the simple setup for the OS, I set a PIN, since that’s what Microsoft seems to prefer rather than a password. I don’t know if I think it is a good idea, but since it is my personal machine, OK, I’ll do it. That’s when things went south. I also renamed the machine and rebooted. Well, it froze when I tried to log in. Doing a hard power off got control of the machine again, and it seemed to work OK. But logging in is really slow compared to my several year old Surface Pro. The new machine should be superior in every way, but Windows 10 on my old machine seems snappier. Then I undocked the tablet portion and tried to use the pen. Windows knew the pen was there and where it was pointing, but completely ignored any input from the pen, including taps, presses, right clicks (there’s a button the pen for that), etc. I imagine there’s some sort of dumb setting in Windows that says “Use the pen for art only, ignore attempts to manage Windows with it” but I haven’t found it yet. Now, pen control of Windows is something that Microsoft has been doing very well for 10 years, starting with the original Tablet PCs, which I used and loved. For this simple thing not to work correctly out of the box is truly stunning. So now the machine is sitting at home and downloading updates and firmware fixes that will hopefully address these issues. I look forward to setting it up to be a useful machine over the weekend. I am disappointed with the poor experience the software provided out of the box. Microsoft is still behind Apple in that respect. |
|||||
Comments are closed 
