That was my first reaction: somebody at IBM is in deep shit.
It seems like they had a lock on the last few big DoE supers (and supercomputer sales in general); now all of a sudden we see Cray getting back in there. I wonder if IBM stepped on somebody's toes and got given the boot on this one (it's small, maybe this is just a spanking), or if they've gotten behind in the research and power/dollar worlds because they were doing so well for so long? Or is this just the government trying to spread the love around, giving a small project to somebody else for a change?
Reminds me a little of the whole Thinking Machines business a few years ago; they were the real darlings of the govt.-contract world, and then Cray and IBM started to get upset that TM was eating out of their rice bowl and lobbied Congress to even things out. Given that they're not around anymore, I think we can all figure how how that went.
Well I don't know about the scheduling part of it, it admittedly doesn't seem like it ought to be that hard, but then again people probably though the same thing about putting in your monthly resupply order to the local warehouse, 10 years ago.
Now, any business worth its salt, particularly one that deals in perishables, has automated or semi-automated systems for managing a supply chain, and helping to ensure that the right amount of inventory is kept on-hand at various places. A big fast food restaurant probably has a system that's linked from the individual cash register transactions all the way up to their big corporate ERP system; when Jose the cashier punches through an order, it's debited from local inventory, and when local inventory reaches a low level, a resupply order is initiated from the warehouse, which similarly tracks its own inventory and orders from distributers.
Is there really anything in the whole supply chain that's "rocket science?" Nope. People did just fine for hundreds of years with pencils and paper. But that hasn't kept it from being the focus of probably billions of dollars of research and investment and computerization over the last 10-20 years.
This is really just an extension of that: once you've got your physical supply chain optimized, and you're not paying for any transportation or warehousing that you can get rid of (to the limits of your optimizable parameters), what else can you start to work on? Salary is a big expense for most businesses, so it makes sense that that would be the next thing.
There are definitely some parts of HR planning that a computer system could help. Not all people are equally good at picking out trends, or noticing what ought to be obvious -- during times when you have high sales volume, you need additional staff. But how many additional staff, and at what point does it make sense to add an additional cashier in order to reduce customer wait times? Those are pretty big questions for a shift manager at a burger joint to answer; an 'ERP-like' system for HR might be able to answer some of those questions. (Other things it could tell you: which employees are quantitatively better at which tasks? Knowing this, you can easily schedule one 'old hand' or two rookies to run the fryolator, or know which and how many employees are required to replace someone, if they leave or call in sick. Also, there are obvious performance-based benefits that you can do, although you have to be careful not to turn the system into something that can be easily gamed.)
There's also the question of staff turnover and top-down management. There's probably a big demand from higher up in the corporate structure in these organizations to know what's going on down where the rubber meets the road; if they can get reports as to how many excess employees each location has, maybe they can see if employees from one location are willing to transfer to another, understaffed one. And rather than having each new manager at a location (which probably change fairly frequently) redo the hours planning, all they have to do is implement the plan that's handed to them by the HR computer. From an executive's perspective, it one less unknown to worry about in every restaurant.
Could stuff like this be a real time- and money-waster? Sure, particularly if you're computerizing stuff that humans already do just well at, for no reason. But a lot of ERP stuff was a boondoggle, in the beginning (and some people still think it is); that's now been accepted as good business practice.
I think that the perceived advantages are so great, you're going to see more and more of this sort of automated management in the near future, like it or not.
we need more people who can manage and work with their company's talent
Talent? We're talking about fast food here.
The only reason they have people working in the back of a McDonalds/Zaxbys/whatever is because people are cheaper than machines. It's tough to program a robot to assemble burgers effeciently (dealing with mis-shapen patties, etc.).
The only reason that any of those people have jobs is because the cost of the machine that would replace them, costs more than the stream of cash that they're paid. (That is to say, the present value of the income stream which is their salary, is less than the upfront purchase cost plus maintaince costs of a machine.)
When machines get better at doing things, so that they're the cheaper option, they do the jobs instead of people.
What's ironic here is that it's the manager's job that's being computerized before the burger-boy's one.
A long time ago, I worked as a basic hourly wage-slave in a photo finishing shop. We were (until it got bought by a national chain, but that's another story) a fairly high-end place. We charged more, we took longer, but we did much better work than your average minilab. In general, we had very satisfied customers. But every once in a while we had people that were just chronic complainers.
Usually the worst would be somebody who fancied themselves to be Ansel Adams, but in reality was so incompetent they could barely use their own equipment. They'd come in with a roll of horrible photos ("what do you mean, I shouldn't use 100-speed film indoors without a flash at night? But I have this expensive camera!"), which we would duly attempt to print, and then naturally we'd be accused of gross incompetence when they got back green images. (Green being what happens when you under-expose paper in an attempt to pull the faintest traces of an image off of the negative.)
Our cue was that whenever someone started bitching that we weren't any better than the drug store across the street, we'd hand them a complimentary roll of film and invite them to use the drug store in the future.
Most of the time, people who actually tried them would come back to us; other times they wouldn't, and we'd say good riddance.
The "average customer" isn't worth a whole lot, to be honest. Once a person starts taking up a ton of employee time, hurting employee morale, or making a scene in the store and causing grief to other more reasonable customers, it's probably beneficial to just get them to go away in whatever way is easiest and involves creating the least bad blood. Pouring more effort into their problems just isn't worthwhile -- it's like continuing to double-down on a roulette wheel; it might seem like the right thing to do, but in the long run you'll probably end up losing more than you'll gain. There are always exceptions, of course, and many national retailers have gone too far the other way, but saying "The Customer Is Always Right" is terribly naive.
I think you need to make a distinction between logical and physical centralization.
It's possible (as your Google example points out) to have a physically decentralized system which is logically "centralized," at least insofar as it can be made to look like a monolithic system.
It's this sort of thing which seems to have a lot of possibilities in the future. Having all your eggs in one basket is just asking for trouble (just ask Napster, or the people who had their websites run out of New Orleans datacenters last year). But having to deal with a distributed/decentralized environment is hard, and it's limiting. Networks which can self-organize, and then present a unified front to the outside world, are really the future. It's sort of a reversal of the old client server model: instead of taking a single server and creating lots of little virtual server volumes, and showing them to many clients, you instead take lots of processing/storage nodes and abstract them into a single VM, and then present that to the user or the user's applications as a whole.
I'm not saying it's a magic bullet -- having a distributed system that's logically centralized makes it almost as vulnerable to malware as a true centralized system (because by centralizing it, you provide an avenue for a virus to spread or affect large amounts of data) -- but it does solve a lot of the problems inherent in old-style centralized systems while retaining some of their advantages.
While I too have heard the factoid about most people only using around 5k words per day in their working vocabulary, I think it's wrong to assume that everyone's 5k-words is the same, or really even all that close. There are obviously going to be parts which overlap (otherwise we wouldn't be able to easily talk to each other), but everyone's working vocabulary is going to be different. This is particularly true in emails, if a high percentage of your email volume is with people you work closely with, and are familiar with the same concepts. (It probably is less true for a public-relations department that fields a wide variety of questions and has to respond to them -- but they probably don't want a lot of incoming mail filtering anyway.)
If I work doing ERP implementations all day, my vocabulary and certainly the content of my emails is going to be very different from someone who's working in a law firm.
The only problem is that you don't want to rely on those factors too much, because if I get an email from Legal one day, I probably don't want it being routed to the trash can automatically. (Well, maybe I do, but that's another story.)
I'm not sure about this actually. Most home users I know have some form of personal spam filter, namely whatever's built into either Apple Mail or Outlook Express. I'm not sure how smart those systems are (or even what kind of logic they use) compared to the SpamAssassin-type mailserver filters, but they're very common.
I can't think of a decent email program these days that doesn't provide some level of automatic spam filtration; usually they work by having you manually separate out / earmark spam messages for a while, and then once the system gets trained, it starts moving them to a separate "Spam" box for you.
Although systems like that still require individuals to download all the spam to their local systems, wasting bandwidth, I think they're some of the best solutions overall, because they end up having more-unique filters. Also, it's easier to recover a false-positive from your local machine's spambox, than it is to retrieve it from your ISP's file. In concert with the X-Spam headers provided by most ISPs, I think intelligent filtration at the client level is probably one of the most viable near-term 'solutions' for spam.
We need to send out some spam for some sort of unbelievable scheme -- free enlargement pills, or whatever, so that these folks give us their names and addresses, and then we need to send mixed teams of underemployed programmers and Spetsnaz commandos to go teach them about safe computing practices.
The fact that anyone buys stuff from some of the spam messages I get is a sad testament to humanity.
Well then instead of using it as a rear-view mirror, they could use it for dashboard-type instrument cluster displays. That would be fairly realistic, since you have to remove your eyes from the windshield to view them in reality (unless you have a HUD).
Instead of the common (but fake) driving displays, where the dashboard is magically visible at the bottom of the driver's field of vision, perfectly in focus and clear, make the TV picture nothing but the view out the window, and then put the speedometer/tach/shift-indicator on the controller.
You could even do stuff like have the controller display blink colors, so that you'd catch it in your peripheral vision, just like you'd notice an instrument panel light in a car or plane.
Based on what I know of the car-stereo manufacturers, they will run a cable from a proprietary connector on the back of the head unit, to the glove compartment, where it will terminate in an entirely different, proprietary connector. Then, you will be able to order short pigtail adaptors, at exorbitant cost, to go from the radio manufacturer's proprietary connector, to Apple's proprietary connector. These adaptors will be available aftermarket if you want an additional one, or one other than what your car came with (for Sirius/XM/Nomad/whatever), but the cost would be slightly more than just replacing your entire head unit with another one from Crutchfield that has an analog audio input.
You know, they make an external RF remote for the iPod now.
It's not manufactured by Apple -- I'm not sure who makes it, actually, although when I'm done writing I'll try some Google searches -- but I've seen them in person. It's a little black receiver box that mounts flush to the bottom of the iPod (via the Dock Connector), and then has a pass-thru connection on its bottom, so that the 'pod can still charge.
Then it has a small remote control that you can put anywhere in the vicinity, and adjust the volume and do track skips and FF/REV. The remote isn't exactly ergonomically designed for steering-wheel mounting (if they were smart, they'd make one that was), but you could definitely use a little self-sticking Velcro and mount it on your dash, or on the back of your steering wheel, and achieve almost the same effect.
I can't personally vouch for the product having just come across it recently, but it seems pretty darn slick. I have a 40GB 3G iPod that I use almost exclusively in my car, and I agree that being able to skip forward and backward through tracks would make the whole experience a lot nicer. (Frankly that makes it as good as a CD Changer interface for me, since most of them don't show track names, and that's the next improvement you can make once you have track skip and volume.)
I understand your point, but I think that what you're asking for doesn't really exist yet. It would be nice if there were some standardized connector that combined analog audio plus some form of control I/O (and it would not surprise me if there were, somewhere, interfaces and connectors designed for this purpose, for other applications). However there isn't one right now.
An plain-old analog input to the car's audio system would be nice, and in fact a lot of aftermarket stereos have them (my Aiwa headunit in my old car, circa 2000, had one), and I think some Pontiac cars do in the factory system. However, that really doesn't "do it" for a lot of people. Really, it's not much of a step up from a cassette adaptor, except for perhaps an increase in fidelity (which most people won't notice anyway, because they're tone deaf, as evidenced by the fact that they're listening to 128kbps MP3 files). A whole lot of iPod-owning people would get in their cars with their iPods, plug them in to their stereos, and then wonder why the hell they still have to go fishing around in the cup holder to pause/rewind/fast-forward the song. Not much of an improvement over their old car, they'd think -- and that's not good if you're an auto manufacturer, and you're using the $10 Dock Connector as a sales point for a $20,000+ automobile.
In order to get real "integration," you need to have some way of controlling the iPod from the factory head unit. In order to do this, you need to go through the Dock Connector. Since Apple controls the Dock Connector, it's pretty much guaranteed to only work with Apple gear.
Without control functions, even the most brain-dead, consumerist sheeple is going to realize that they're being overcharged for a "feature" that their 1988 Honda Civic provides already, via its cassette deck, and that the only reason this new feature exists, is because the manufacturers have removed the cassettes just as people seem to have found a use for them after a hiatus of 10 years.
I suspect, at some point down the road, some enterprising individual will make a "Dock Connector Breakout Adaptor," a little bricklike box that you can plug into the Dock Connector interface in your car, and then use to connect an 'unsupported' audio device in to your car's system. This is already common practice for the myriad proprietary CD-Changer connectors, so I don't see this as being any different, except that it'll be easier since I'm sure there are already factories in Hong Kong tooled up to make unlicensed, reverse-engineered Dock Connectors.
Would it be better if there were some agreed-upon, open standard for portable MP3 player hardware (audio/power/IO/control) interfaces? Damn straight it would be. But in the absence of that (don't hold your breath), having a single de facto standard is sometimes obnoxious, but as long as the lawyers don't get too much in the way, the market will provide the breakout boxes and other gadgets that an enterprising individual will need to connect whatever they want.
While I agree with you in principle, I really don't think it's Subversion in particular. Unless you have information that you feel substantiates that, I think that this is more like the old VMS file system than any type of SV/CVS/code-version-control scheme. It seems to be implemented on a much lower level, and with a greater degree of system integration than an SV-based system would provide.
Plus, using SV just doesn't seem consistent with other stuff they've done in the past; this seems like an in-house project. Apple has a tendency to only use existing codebases occasionally, and when they do, they make a big point of it. Since nothing has been mentioned, I think it's way more likely this is something proprietary they've cooked up (unless HP licensed them the old OpenVMS stuff).
Mod parent up. This was my immediate question as well, and I still haven't heard it answered.
If you want to encode information into the delay between key-press packets, then you need to make the delay significantly longer (at least a few standard deviations) than the average difference between two keypress packets.
People don't type at exactly the same rate, so if the delay in between keypresses varies (I'm making up numbers here) between 100 and 150 ms, then you need to make the introduced delay greater than 50ms.
Alternately, you could buffer all of the incoming keystrokes in the computer, and send them out at a constant rate (say exactly 100ms apart); then you'd only have to add a small delay to them in order to encode information. But unless the packets are being buffered and sent out in such an orderly fashion by the host system already, it seems like this kind of behavior could be easily picked up on, because it would cause a delay of at least a few keystrokes in an interactive system (if there's one packet per keystroke and you're queueing and buffering a few packets at a time). I'm sure there's probably some nice mathematical formula for the amount of transit time you'd add (from the time the key goes down to the time it's received by the host system) as a result of buffering out all the variation in the timing between packets... I just can't think of it right now.
Ultimately though, I don't see any defense against an attack like this. If someone can compromise your hardware, particularly your input devices, you're quite screwed. I've always seen it as an extension of the 'local console root' rule: if someone can get to the CPU, then they have root. I guess we've got to extend this to keyboards, mice, and monitors as well: if you don't know where everything that you pass unencrypted information through was last night, maybe you shouldn't be using it.
Messing with the delay is only one of many ways that someone could sneak information out of an area -- it's neat, technically, but there are a lot of low-tech ways that would work just as well (including the audio recorder trick from a while back, where you can determine a typed password by listening to a recording of the keypresses).
If you only wanted a system that would work once, you could build a more powerful keystroke-recorder into a keyboard. Instead of having it mess with the delay, make it wake up the computer in the middle of the night (logging on -- it's not hard to grab your password on a Windows box, since it's nicely defined as the first thing you type after pressing Ctrl-Alt-Del and before return), and then executing a macro that emailed a recording of everything that had been typed recently to a dead-drop.
Difference is whether the civilian population itself is a target. Both are accomplishing a goal which is related (a sort of psyops/morale "victory"), but in the now-prohibited WWII-style of attack, this is accomplished by intentionally killing civilians. The Baghdad airstrikes were an attempt to achieve the psychological effects without the mass casualties, and I'm not sure if history will say that it was particularly successful. (In retrospect I think it's like trying to scare someone by shaking your fist, but when they know you won't really hit them.)
In this case it's really the means, not the end, which is important. There are many ways of accomplishing the same end (morale damage to the enemy), some of which are more or less reprehensible than others.
The fine line between actual 'terrorism' and 'psychological warfare' depends almost entirely on whether you actually go out of your way to kill innocent people to accomplish your goals.
I think it's important that we not broaden the meaning of "terrorist" to include so-called 'economic terrorism' and other politicized uses, because that weakens the actual meaning. I think there's also an important distinction between 'guerilla warfare' and 'terrorism,' namely that guerilla warfare involves forces using their environment, which may include the civilian population, as cover, but directs its attacks at a defined opposing force; terrorism attacks the civilian population directly, and may or may not use it for cover.
The complexity only happens because I've added a second display device and audio system, and those are just me doing things because I could; neither is inherent in the concept of a HTPC. (In fact, they wouldn't happen except that I want to use the same computer for home theater and as a regular PC.) The compexity of a HTPC is no greater than a regular computer, and probably less complex than the usual mess of video/audio/RF stuff that anyone with a cable box, VCR, DVD player, and TV right now. The complexity is not the issue stopping people.
I think the reason media centers aren't more successful is because the average person doesn't care about doing the things they provide, and thus isn't going to spend the money. I don't think anyone (myself included) wants to do computer-ish functions like web browsing or email on their televisions. (The "WebTV" era, where you could sell that to people in lieu of a computer, ended now that practically everone has a computer already.)
There aren't any 'killer apps' yet that really motivate people to go out and spend the money that HTPCs cost. Why would you want to have a $500-1k HTPC to do DVR functions, when for a few bucks a month, the one from your cable company will do the same thing?
Most 'media centers' are a solution looking for a problem that the average person doesn't have.
Pretty much. I question why the City of Hoboken ever involved themselves with these people. It sounds like a real boondoggle; you assumedly pay for the garage/machinery, but then once you get it, you can't operate it yourself, you have to agree to let them come and "manage" the whole thing for you, forever.
And what do you do if they decide to raise their rates? You're S.O.L. -- if you don't like what they're going to charge you to manage your robotic garage, then your garage magically stops working.
It's like the ultimate no-bid contract. There can't ever be any competition for the operation of that garage, because only that one company can do it.
What a load of crap; if I were a resident who had my tax dollars spent on a lock-in like that, I'd be pretty annoyed.
I use -- and have for quite a few years, since I gave up my standalone TV because it took up too much space and was just another thing to move around -- my computer as a TV. It takes a little bit of prior planning in how you set up your room, but I just always make sure to put my computer desk on a wall opposite some sort of seating (couch, chairs, bed, etc.). When I want to use it as a TV, I just turn full-screen on, kick the chair out of the way, and watch it.
If you have anything less than a 19" or 20" monitor this isn't going to be much fun, but it's completely workable if you have a reasonably-sized monitor and your rooms aren't large (so that the distance from your seating to the display isn't too long). I also have two sound systems that I switch between, a small set of computer speakers that I use when I'm sitting close, and a much larger set of HiFi speakers that I change to for movie-watching.
Really you just need to think ahead a little bit, and not stick your computer off in a corner somewhere, where it's impossible to see the screen.
The other thing that I think is overlooked, is the ease with which you can attach a projector to most computers. When I ran into a little extra money a while back, I decided instead of getting a standalone TV (I do have room for one now, if I wanted it), to just get a projector and attach it to the secondary monitor connector on my computer. It's smaller and less obtrusive than a big TV (cieling-mount, projecting on a painted wall) when not in use, and I can change between watching something on the CRT monitor and on the projector just by turning the PJ on, and dragging the viewing window into the alternate display's desktop space. That way the computer handles the upscaling to the PJ's native resolution, without any thought on my part.
I end up using the computer to watch TV/movies in basically three ways: when I'm sitting right in front of it, I'll put the TV or movie in a window so I can multi-task, when I want to just casually watch, I'll make it fullscreen on the regular monitor and push back my chair a bit, and when I really want the full-on home theater experience, I put the video on the projector, turn down the lights and turn off the regular display.
About the only thing it's missing right now is a remote control, but that's just because I haven't bothered to get one and I'm waiting for Apple to release Front Row for Power Macs -- based on yesterday's announcement, that'll happen with the new version of OS X.
Try again. The CD is model-specific (and potentially revision), it's not machine-specific. If you went out and bought two Minis on the same day, you can use either one's restore CD to reinstall software on the other. You just can't use either one's CD to install software onto your iMac/MacBook/whatever.
This is nothing new. In fact, I remember back in the OS 9 and early OS X days, there were lots of semi-sketchy utilities around for "spoofing" the machine code that your system reported to the installer, so you could use a newer machine's Restore CD to pirate a copy of the OS (or install it on a machine it wasn't necessarily designed for, i.e. onto a Beige 'Serial Port' G3). It's been a while since I've been involved in any such shenanigans, but it sounds like they're just doing the same sort of stuff.
The bundled software that comes with Macs has been very loosely copy-protected -- actually, I wouldn't even call it copy-protected, it's more 'pass-around-to-all-your-friends protected' -- for quite a while. I remember first encountering it on my Performa 6400/200, which must have been late OS 8 or early OS 9.
Retail install CDs, to the best of my knowledge, have never been and are not currently specific to any particular model or unique machine, outside of the normal checks for compatibility. (If you try to install Tiger on my old 6400, wherever it is today, I expect you'd have problems. But one set of install discs will let you install onto as many compatible machines as you can put them in and reboot, if you're one to flout the law.)
The only really new thing that Apple has done differently with their sales strategy, as far as copy protection is concerned, is made it intentionally difficult to run certain applications on otherwise-supported hardware (i.e., making it hard to put Front Row on your G5 Tower, even though there's no technical reason why it ought not be able to run). But in terms of the system itself, I think Apple still has some of the most relaxed copy protection that I've seen on big-name commercial software -- probably because you can only run it on their hardware, so the piracy concerns are more limited than Microsoft's are.
Slashdot Mirror
That was my first reaction: somebody at IBM is in deep shit.
It seems like they had a lock on the last few big DoE supers (and supercomputer sales in general); now all of a sudden we see Cray getting back in there. I wonder if IBM stepped on somebody's toes and got given the boot on this one (it's small, maybe this is just a spanking), or if they've gotten behind in the research and power/dollar worlds because they were doing so well for so long? Or is this just the government trying to spread the love around, giving a small project to somebody else for a change?
Reminds me a little of the whole Thinking Machines business a few years ago; they were the real darlings of the govt.-contract world, and then Cray and IBM started to get upset that TM was eating out of their rice bowl and lobbied Congress to even things out. Given that they're not around anymore, I think we can all figure how how that went.
Well I don't know about the scheduling part of it, it admittedly doesn't seem like it ought to be that hard, but then again people probably though the same thing about putting in your monthly resupply order to the local warehouse, 10 years ago.
Now, any business worth its salt, particularly one that deals in perishables, has automated or semi-automated systems for managing a supply chain, and helping to ensure that the right amount of inventory is kept on-hand at various places. A big fast food restaurant probably has a system that's linked from the individual cash register transactions all the way up to their big corporate ERP system; when Jose the cashier punches through an order, it's debited from local inventory, and when local inventory reaches a low level, a resupply order is initiated from the warehouse, which similarly tracks its own inventory and orders from distributers.
Is there really anything in the whole supply chain that's "rocket science?" Nope. People did just fine for hundreds of years with pencils and paper. But that hasn't kept it from being the focus of probably billions of dollars of research and investment and computerization over the last 10-20 years.
This is really just an extension of that: once you've got your physical supply chain optimized, and you're not paying for any transportation or warehousing that you can get rid of (to the limits of your optimizable parameters), what else can you start to work on? Salary is a big expense for most businesses, so it makes sense that that would be the next thing.
There are definitely some parts of HR planning that a computer system could help. Not all people are equally good at picking out trends, or noticing what ought to be obvious -- during times when you have high sales volume, you need additional staff. But how many additional staff, and at what point does it make sense to add an additional cashier in order to reduce customer wait times? Those are pretty big questions for a shift manager at a burger joint to answer; an 'ERP-like' system for HR might be able to answer some of those questions. (Other things it could tell you: which employees are quantitatively better at which tasks? Knowing this, you can easily schedule one 'old hand' or two rookies to run the fryolator, or know which and how many employees are required to replace someone, if they leave or call in sick. Also, there are obvious performance-based benefits that you can do, although you have to be careful not to turn the system into something that can be easily gamed.)
There's also the question of staff turnover and top-down management. There's probably a big demand from higher up in the corporate structure in these organizations to know what's going on down where the rubber meets the road; if they can get reports as to how many excess employees each location has, maybe they can see if employees from one location are willing to transfer to another, understaffed one. And rather than having each new manager at a location (which probably change fairly frequently) redo the hours planning, all they have to do is implement the plan that's handed to them by the HR computer. From an executive's perspective, it one less unknown to worry about in every restaurant.
Could stuff like this be a real time- and money-waster? Sure, particularly if you're computerizing stuff that humans already do just well at, for no reason. But a lot of ERP stuff was a boondoggle, in the beginning (and some people still think it is); that's now been accepted as good business practice.
I think that the perceived advantages are so great, you're going to see more and more of this sort of automated management in the near future, like it or not.
we need more people who can manage and work with their company's talent
Talent? We're talking about fast food here.
The only reason they have people working in the back of a McDonalds/Zaxbys/whatever is because people are cheaper than machines. It's tough to program a robot to assemble burgers effeciently (dealing with mis-shapen patties, etc.).
The only reason that any of those people have jobs is because the cost of the machine that would replace them, costs more than the stream of cash that they're paid. (That is to say, the present value of the income stream which is their salary, is less than the upfront purchase cost plus maintaince costs of a machine.)
When machines get better at doing things, so that they're the cheaper option, they do the jobs instead of people.
What's ironic here is that it's the manager's job that's being computerized before the burger-boy's one.
A long time ago, I worked as a basic hourly wage-slave in a photo finishing shop. We were (until it got bought by a national chain, but that's another story) a fairly high-end place. We charged more, we took longer, but we did much better work than your average minilab. In general, we had very satisfied customers. But every once in a while we had people that were just chronic complainers.
Usually the worst would be somebody who fancied themselves to be Ansel Adams, but in reality was so incompetent they could barely use their own equipment. They'd come in with a roll of horrible photos ("what do you mean, I shouldn't use 100-speed film indoors without a flash at night? But I have this expensive camera!"), which we would duly attempt to print, and then naturally we'd be accused of gross incompetence when they got back green images. (Green being what happens when you under-expose paper in an attempt to pull the faintest traces of an image off of the negative.)
Our cue was that whenever someone started bitching that we weren't any better than the drug store across the street, we'd hand them a complimentary roll of film and invite them to use the drug store in the future.
Most of the time, people who actually tried them would come back to us; other times they wouldn't, and we'd say good riddance.
The "average customer" isn't worth a whole lot, to be honest. Once a person starts taking up a ton of employee time, hurting employee morale, or making a scene in the store and causing grief to other more reasonable customers, it's probably beneficial to just get them to go away in whatever way is easiest and involves creating the least bad blood. Pouring more effort into their problems just isn't worthwhile -- it's like continuing to double-down on a roulette wheel; it might seem like the right thing to do, but in the long run you'll probably end up losing more than you'll gain. There are always exceptions, of course, and many national retailers have gone too far the other way, but saying "The Customer Is Always Right" is terribly naive.
Quake4? Just think of the framerate you could get in glxgears!
I think you need to make a distinction between logical and physical centralization.
It's possible (as your Google example points out) to have a physically decentralized system which is logically "centralized," at least insofar as it can be made to look like a monolithic system.
It's this sort of thing which seems to have a lot of possibilities in the future. Having all your eggs in one basket is just asking for trouble (just ask Napster, or the people who had their websites run out of New Orleans datacenters last year). But having to deal with a distributed/decentralized environment is hard, and it's limiting. Networks which can self-organize, and then present a unified front to the outside world, are really the future. It's sort of a reversal of the old client server model: instead of taking a single server and creating lots of little virtual server volumes, and showing them to many clients, you instead take lots of processing/storage nodes and abstract them into a single VM, and then present that to the user or the user's applications as a whole.
I'm not saying it's a magic bullet -- having a distributed system that's logically centralized makes it almost as vulnerable to malware as a true centralized system (because by centralizing it, you provide an avenue for a virus to spread or affect large amounts of data) -- but it does solve a lot of the problems inherent in old-style centralized systems while retaining some of their advantages.
Unfortunately, there's just not enough space in the tubes to allow us to do that.
Instead, we'll need to use a dump truck to move the data around.
While I too have heard the factoid about most people only using around 5k words per day in their working vocabulary, I think it's wrong to assume that everyone's 5k-words is the same, or really even all that close. There are obviously going to be parts which overlap (otherwise we wouldn't be able to easily talk to each other), but everyone's working vocabulary is going to be different. This is particularly true in emails, if a high percentage of your email volume is with people you work closely with, and are familiar with the same concepts. (It probably is less true for a public-relations department that fields a wide variety of questions and has to respond to them -- but they probably don't want a lot of incoming mail filtering anyway.)
If I work doing ERP implementations all day, my vocabulary and certainly the content of my emails is going to be very different from someone who's working in a law firm.
The only problem is that you don't want to rely on those factors too much, because if I get an email from Legal one day, I probably don't want it being routed to the trash can automatically. (Well, maybe I do, but that's another story.)
I'm not sure about this actually. Most home users I know have some form of personal spam filter, namely whatever's built into either Apple Mail or Outlook Express. I'm not sure how smart those systems are (or even what kind of logic they use) compared to the SpamAssassin-type mailserver filters, but they're very common.
I can't think of a decent email program these days that doesn't provide some level of automatic spam filtration; usually they work by having you manually separate out / earmark spam messages for a while, and then once the system gets trained, it starts moving them to a separate "Spam" box for you.
Although systems like that still require individuals to download all the spam to their local systems, wasting bandwidth, I think they're some of the best solutions overall, because they end up having more-unique filters. Also, it's easier to recover a false-positive from your local machine's spambox, than it is to retrieve it from your ISP's file. In concert with the X-Spam headers provided by most ISPs, I think intelligent filtration at the client level is probably one of the most viable near-term 'solutions' for spam.
Clearly then, they are the problem.
We need to send out some spam for some sort of unbelievable scheme -- free enlargement pills, or whatever, so that these folks give us their names and addresses, and then we need to send mixed teams of underemployed programmers and Spetsnaz commandos to go teach them about safe computing practices.
The fact that anyone buys stuff from some of the spam messages I get is a sad testament to humanity.
Well then instead of using it as a rear-view mirror, they could use it for dashboard-type instrument cluster displays. That would be fairly realistic, since you have to remove your eyes from the windshield to view them in reality (unless you have a HUD).
Instead of the common (but fake) driving displays, where the dashboard is magically visible at the bottom of the driver's field of vision, perfectly in focus and clear, make the TV picture nothing but the view out the window, and then put the speedometer/tach/shift-indicator on the controller.
You could even do stuff like have the controller display blink colors, so that you'd catch it in your peripheral vision, just like you'd notice an instrument panel light in a car or plane.
Really, I think Tesla would be much higher-frequency than that.
Based on what I know of the car-stereo manufacturers, they will run a cable from a proprietary connector on the back of the head unit, to the glove compartment, where it will terminate in an entirely different, proprietary connector. Then, you will be able to order short pigtail adaptors, at exorbitant cost, to go from the radio manufacturer's proprietary connector, to Apple's proprietary connector. These adaptors will be available aftermarket if you want an additional one, or one other than what your car came with (for Sirius/XM/Nomad/whatever), but the cost would be slightly more than just replacing your entire head unit with another one from Crutchfield that has an analog audio input.
What, you thought it would be convenient?
In the end, all I can do is HOPE that maybe these ass-wipes will get iPods and DROWN OUT the sound of their handset ringing.
I think the ass-wipes are all waiting for an iPhone, so they can do both at once.
Adding more technology to stupidity never improved anything, it just makes stupid things happen faster.
You know, they make an external RF remote for the iPod now.
h p/products_id/2052 I'm 95% sure this was what I saw.
It's not manufactured by Apple -- I'm not sure who makes it, actually, although when I'm done writing I'll try some Google searches -- but I've seen them in person. It's a little black receiver box that mounts flush to the bottom of the iPod (via the Dock Connector), and then has a pass-thru connection on its bottom, so that the 'pod can still charge.
Then it has a small remote control that you can put anywhere in the vicinity, and adjust the volume and do track skips and FF/REV. The remote isn't exactly ergonomically designed for steering-wheel mounting (if they were smart, they'd make one that was), but you could definitely use a little self-sticking Velcro and mount it on your dash, or on the back of your steering wheel, and achieve almost the same effect.
I can't personally vouch for the product having just come across it recently, but it seems pretty darn slick. I have a 40GB 3G iPod that I use almost exclusively in my car, and I agree that being able to skip forward and backward through tracks would make the whole experience a lot nicer. (Frankly that makes it as good as a CD Changer interface for me, since most of them don't show track names, and that's the next improvement you can make once you have track skip and volume.)
Links:
http://www.thinkdifferentstore.com/product_info.p
http://www.welovemacs.com/p21.html This is an another example, although not as slick.
I understand your point, but I think that what you're asking for doesn't really exist yet. It would be nice if there were some standardized connector that combined analog audio plus some form of control I/O (and it would not surprise me if there were, somewhere, interfaces and connectors designed for this purpose, for other applications). However there isn't one right now.
An plain-old analog input to the car's audio system would be nice, and in fact a lot of aftermarket stereos have them (my Aiwa headunit in my old car, circa 2000, had one), and I think some Pontiac cars do in the factory system. However, that really doesn't "do it" for a lot of people. Really, it's not much of a step up from a cassette adaptor, except for perhaps an increase in fidelity (which most people won't notice anyway, because they're tone deaf, as evidenced by the fact that they're listening to 128kbps MP3 files). A whole lot of iPod-owning people would get in their cars with their iPods, plug them in to their stereos, and then wonder why the hell they still have to go fishing around in the cup holder to pause/rewind/fast-forward the song. Not much of an improvement over their old car, they'd think -- and that's not good if you're an auto manufacturer, and you're using the $10 Dock Connector as a sales point for a $20,000+ automobile.
In order to get real "integration," you need to have some way of controlling the iPod from the factory head unit. In order to do this, you need to go through the Dock Connector. Since Apple controls the Dock Connector, it's pretty much guaranteed to only work with Apple gear.
Without control functions, even the most brain-dead, consumerist sheeple is going to realize that they're being overcharged for a "feature" that their 1988 Honda Civic provides already, via its cassette deck, and that the only reason this new feature exists, is because the manufacturers have removed the cassettes just as people seem to have found a use for them after a hiatus of 10 years.
I suspect, at some point down the road, some enterprising individual will make a "Dock Connector Breakout Adaptor," a little bricklike box that you can plug into the Dock Connector interface in your car, and then use to connect an 'unsupported' audio device in to your car's system. This is already common practice for the myriad proprietary CD-Changer connectors, so I don't see this as being any different, except that it'll be easier since I'm sure there are already factories in Hong Kong tooled up to make unlicensed, reverse-engineered Dock Connectors.
Would it be better if there were some agreed-upon, open standard for portable MP3 player hardware (audio/power/IO/control) interfaces? Damn straight it would be. But in the absence of that (don't hold your breath), having a single de facto standard is sometimes obnoxious, but as long as the lawyers don't get too much in the way, the market will provide the breakout boxes and other gadgets that an enterprising individual will need to connect whatever they want.
While I agree with you in principle, I really don't think it's Subversion in particular. Unless you have information that you feel substantiates that, I think that this is more like the old VMS file system than any type of SV/CVS/code-version-control scheme. It seems to be implemented on a much lower level, and with a greater degree of system integration than an SV-based system would provide.
Plus, using SV just doesn't seem consistent with other stuff they've done in the past; this seems like an in-house project. Apple has a tendency to only use existing codebases occasionally, and when they do, they make a big point of it. Since nothing has been mentioned, I think it's way more likely this is something proprietary they've cooked up (unless HP licensed them the old OpenVMS stuff).
Mod parent up. This was my immediate question as well, and I still haven't heard it answered.
... I just can't think of it right now.
If you want to encode information into the delay between key-press packets, then you need to make the delay significantly longer (at least a few standard deviations) than the average difference between two keypress packets.
People don't type at exactly the same rate, so if the delay in between keypresses varies (I'm making up numbers here) between 100 and 150 ms, then you need to make the introduced delay greater than 50ms.
Alternately, you could buffer all of the incoming keystrokes in the computer, and send them out at a constant rate (say exactly 100ms apart); then you'd only have to add a small delay to them in order to encode information. But unless the packets are being buffered and sent out in such an orderly fashion by the host system already, it seems like this kind of behavior could be easily picked up on, because it would cause a delay of at least a few keystrokes in an interactive system (if there's one packet per keystroke and you're queueing and buffering a few packets at a time). I'm sure there's probably some nice mathematical formula for the amount of transit time you'd add (from the time the key goes down to the time it's received by the host system) as a result of buffering out all the variation in the timing between packets
Ultimately though, I don't see any defense against an attack like this. If someone can compromise your hardware, particularly your input devices, you're quite screwed. I've always seen it as an extension of the 'local console root' rule: if someone can get to the CPU, then they have root. I guess we've got to extend this to keyboards, mice, and monitors as well: if you don't know where everything that you pass unencrypted information through was last night, maybe you shouldn't be using it.
Messing with the delay is only one of many ways that someone could sneak information out of an area -- it's neat, technically, but there are a lot of low-tech ways that would work just as well (including the audio recorder trick from a while back, where you can determine a typed password by listening to a recording of the keypresses).
If you only wanted a system that would work once, you could build a more powerful keystroke-recorder into a keyboard. Instead of having it mess with the delay, make it wake up the computer in the middle of the night (logging on -- it's not hard to grab your password on a Windows box, since it's nicely defined as the first thing you type after pressing Ctrl-Alt-Del and before return), and then executing a macro that emailed a recording of everything that had been typed recently to a dead-drop.
Difference is whether the civilian population itself is a target. Both are accomplishing a goal which is related (a sort of psyops/morale "victory"), but in the now-prohibited WWII-style of attack, this is accomplished by intentionally killing civilians. The Baghdad airstrikes were an attempt to achieve the psychological effects without the mass casualties, and I'm not sure if history will say that it was particularly successful. (In retrospect I think it's like trying to scare someone by shaking your fist, but when they know you won't really hit them.)
In this case it's really the means, not the end, which is important. There are many ways of accomplishing the same end (morale damage to the enemy), some of which are more or less reprehensible than others.
The fine line between actual 'terrorism' and 'psychological warfare' depends almost entirely on whether you actually go out of your way to kill innocent people to accomplish your goals.
I think it's important that we not broaden the meaning of "terrorist" to include so-called 'economic terrorism' and other politicized uses, because that weakens the actual meaning. I think there's also an important distinction between 'guerilla warfare' and 'terrorism,' namely that guerilla warfare involves forces using their environment, which may include the civilian population, as cover, but directs its attacks at a defined opposing force; terrorism attacks the civilian population directly, and may or may not use it for cover.
The complexity only happens because I've added a second display device and audio system, and those are just me doing things because I could; neither is inherent in the concept of a HTPC. (In fact, they wouldn't happen except that I want to use the same computer for home theater and as a regular PC.) The compexity of a HTPC is no greater than a regular computer, and probably less complex than the usual mess of video/audio/RF stuff that anyone with a cable box, VCR, DVD player, and TV right now. The complexity is not the issue stopping people.
I think the reason media centers aren't more successful is because the average person doesn't care about doing the things they provide, and thus isn't going to spend the money. I don't think anyone (myself included) wants to do computer-ish functions like web browsing or email on their televisions. (The "WebTV" era, where you could sell that to people in lieu of a computer, ended now that practically everone has a computer already.)
There aren't any 'killer apps' yet that really motivate people to go out and spend the money that HTPCs cost. Why would you want to have a $500-1k HTPC to do DVR functions, when for a few bucks a month, the one from your cable company will do the same thing?
Most 'media centers' are a solution looking for a problem that the average person doesn't have.
Pretty much. I question why the City of Hoboken ever involved themselves with these people. It sounds like a real boondoggle; you assumedly pay for the garage/machinery, but then once you get it, you can't operate it yourself, you have to agree to let them come and "manage" the whole thing for you, forever.
And what do you do if they decide to raise their rates? You're S.O.L. -- if you don't like what they're going to charge you to manage your robotic garage, then your garage magically stops working.
It's like the ultimate no-bid contract. There can't ever be any competition for the operation of that garage, because only that one company can do it.
What a load of crap; if I were a resident who had my tax dollars spent on a lock-in like that, I'd be pretty annoyed.
I use -- and have for quite a few years, since I gave up my standalone TV because it took up too much space and was just another thing to move around -- my computer as a TV. It takes a little bit of prior planning in how you set up your room, but I just always make sure to put my computer desk on a wall opposite some sort of seating (couch, chairs, bed, etc.). When I want to use it as a TV, I just turn full-screen on, kick the chair out of the way, and watch it.
If you have anything less than a 19" or 20" monitor this isn't going to be much fun, but it's completely workable if you have a reasonably-sized monitor and your rooms aren't large (so that the distance from your seating to the display isn't too long). I also have two sound systems that I switch between, a small set of computer speakers that I use when I'm sitting close, and a much larger set of HiFi speakers that I change to for movie-watching.
Really you just need to think ahead a little bit, and not stick your computer off in a corner somewhere, where it's impossible to see the screen.
The other thing that I think is overlooked, is the ease with which you can attach a projector to most computers. When I ran into a little extra money a while back, I decided instead of getting a standalone TV (I do have room for one now, if I wanted it), to just get a projector and attach it to the secondary monitor connector on my computer. It's smaller and less obtrusive than a big TV (cieling-mount, projecting on a painted wall) when not in use, and I can change between watching something on the CRT monitor and on the projector just by turning the PJ on, and dragging the viewing window into the alternate display's desktop space. That way the computer handles the upscaling to the PJ's native resolution, without any thought on my part.
I end up using the computer to watch TV/movies in basically three ways: when I'm sitting right in front of it, I'll put the TV or movie in a window so I can multi-task, when I want to just casually watch, I'll make it fullscreen on the regular monitor and push back my chair a bit, and when I really want the full-on home theater experience, I put the video on the projector, turn down the lights and turn off the regular display.
About the only thing it's missing right now is a remote control, but that's just because I haven't bothered to get one and I'm waiting for Apple to release Front Row for Power Macs -- based on yesterday's announcement, that'll happen with the new version of OS X.
Try again. The CD is model-specific (and potentially revision), it's not machine-specific. If you went out and bought two Minis on the same day, you can use either one's restore CD to reinstall software on the other. You just can't use either one's CD to install software onto your iMac/MacBook/whatever.
This is nothing new. In fact, I remember back in the OS 9 and early OS X days, there were lots of semi-sketchy utilities around for "spoofing" the machine code that your system reported to the installer, so you could use a newer machine's Restore CD to pirate a copy of the OS (or install it on a machine it wasn't necessarily designed for, i.e. onto a Beige 'Serial Port' G3). It's been a while since I've been involved in any such shenanigans, but it sounds like they're just doing the same sort of stuff.
The bundled software that comes with Macs has been very loosely copy-protected -- actually, I wouldn't even call it copy-protected, it's more 'pass-around-to-all-your-friends protected' -- for quite a while. I remember first encountering it on my Performa 6400/200, which must have been late OS 8 or early OS 9.
Retail install CDs, to the best of my knowledge, have never been and are not currently specific to any particular model or unique machine, outside of the normal checks for compatibility. (If you try to install Tiger on my old 6400, wherever it is today, I expect you'd have problems. But one set of install discs will let you install onto as many compatible machines as you can put them in and reboot, if you're one to flout the law.)
The only really new thing that Apple has done differently with their sales strategy, as far as copy protection is concerned, is made it intentionally difficult to run certain applications on otherwise-supported hardware (i.e., making it hard to put Front Row on your G5 Tower, even though there's no technical reason why it ought not be able to run). But in terms of the system itself, I think Apple still has some of the most relaxed copy protection that I've seen on big-name commercial software -- probably because you can only run it on their hardware, so the piracy concerns are more limited than Microsoft's are.
It would appear Microsoft is dragging it's feet in accepting Open Source.
Was this some kind of a joke that was just too subtle for me to get? Or are you new to this planet? (If so, try the pastrami.)