I can hear the software vendors right now. "Oh, sure, I'm going to label my software as 'pop-up', that'll bring in the customers, oh, yeah!" More likely, they'll fight it on the grounds of anyone who ever made or makes use of the Yes/No dialog box -- "That's a pop-up, too, make them label their software." Totally meaningless.
Anyway, did anyone else read this and think immediately of the Evil Bit? The whole thing has got to be a joke, right?
I wasn't really trying to slam AMD here. They certainly have their place assured in the legendary Hall of Space Heaters (especially the Thunderbird core,) but their trophy is the same size as Intel's. And I have nothing against AMD, either. The last five CPUs I have personally purchased have all been AMD, but I've also had hot Intel (and Cyrix) chips, too.
Let's just say that none of them will ever be invisible on a thermal scan. That reminds me -- next time my firefighter buddy has the Cairns-Iris around, we'll have to look at the PC through it.
Anyway, my biggest problem with these cases is that they don't look particularily well ventilated. Of course, if you can afford one of these, you can also afford a watercooler, so it shouldn't make much difference. Hmmm...I'm not so sure I want a case that costs more than the computer it houses...
"Rodger Cosgrove, president of Entremedia, a direct marketing firm and a member of Reason's board, assisted in coming up with a program that allows the subscriber list to be integrated with satellite photographs. He also worked with Xeikon, the manufacturer of the printer that made the endless customization possible."
It might be useful for showing John Q. Public exactly how powerful these systems have become
Except it's only happening on the cover of Reason.
It's a libertarian magazine. Nobody of any significance to the American political process would be caught dead reading it. Sorry to be the troller of bad news, but there just isn't enough support between the donkeys and the elephants to make any difference at all.
I'd like to thank everybody who had suggestions, they're all good. And I'd implement them in a heartbeat exccept for the important fact that I'm not the DBA.
I'm just the poor schnook presenting the front end, the guy who has to make sense of this crap data and simplify it so an average underpaid employee can find bedsheets for our customers. When we first encountered this rotten data, we went to the DBAs who told us "hey, these other folks are responsible for the data in these tables, yell at them." And those folks really have no idea how they got into this situation in the first place -- it took us a long time to track down the origins of the bad data. And once we identified their problems, they said, "Too bad, we can't fix all that data now, not in the budget, deal with it as is." Our client who is representing the end users of this data said he couldn't push those other people into fixing their data, and so we'd just have to be smarter in how we presented our screens.
So that's my sad tale. We wrote nasty, inefficient code to sniff around the tables to try to present the most logical choices to our clients. It mostly works, and the customers mostly find their king-sized bedsheets in the style of their choosing. But it shouldn't have to be like this, and it wouldn't have been if the originators of that data had been forced to follow a schema, rather than simply adding new values to existing ones.
But it's not just simply handwaving a few sanity checks into existance. The SIZE field contains data for all merchandise, not just bedsheets. SIZE might contain '60X48' for a tablecloth, or 'XXL' for a T-shirt, or '11W' for a pair of shoes; all situations where 'KING' would have no meaning. And if we add a new line of merchandise, say bicycle tires, the merchant would need to add new sizes from his spreadsheet, such as '26X1.375' (even though the schema has only a 5 char field for SIZE.) Do we blindly add his new size, truncating all his tires to '26X1.'? The search for 26x1.825 tires is going to return a lot of crap.
Look at the case I mentioned earlier. In that case we had someone adding a line of merchandise completely unrelated to bedsheets that came in King and Queen sizes, but they had 'KNG', 'QUN' in their spreadsheets (they probably chose a 3 letter wide column in their spreadsheet to fit their screen.) So someone imported the data, adding KNG and QUN to the list of valid sizes. So now the SIZE field contains KNG and QUN as options, and the next merchant buying bedsheets very reasonably selected KNG and QUN from their dropdown list, missing the very existance of KING and QUEEN.
We got screwed because someone imported data from a spreadsheet where they had the opportunity to invent their own sizes, free from any schema or data model restrictions. Had they been entering the data into the database in the first place, they would have seen KING and QUEEN as options and not felt the need to invent their own.
For you? For your personal application? A spreadsheet, hands-down, no question. It fills all your requirements, gives you quick analysis, pretty pie charts, etc.
For a business? For tracking the bicycling habits of the downtown commuters who use the racks at the various garages, looking for ridership trends during the bus strike, or to determine if we need to spend tax money on bicycle lanes or trail improvements? A database, hands down. It allows spotters on 10th Street to enter data without interfering with the spotters entering data on 11th Street. It allows the planning commissions to perform complex joins, or to compare commute data to weather data or to bicycle theft rates. And if they need pretty pie charts, they can still use Excel as a front end to an ODBC data source to prepare them.
As you were probably trying to point out, you need to be realistic in your assessments.
The central point of the whole article above is that if individual commuters had spreadsheets like yours, they would be of no value to a business or government trying to use the spreadsheet-based data.
The point where a business analyst can make a difference is in spotting these spreadsheets floating around a business (and taking on importance of their own,) and replacing them with a database.
You write this like it's a bad thing that you can't import your non-sanitized data directly into a database.
Obviously you have never dealt with trying to give the clients what they want after someone else has polluted the database with their crappy imported data.
Here's a real example: our clients wanted to find all the KING size bedsheets. We looked them up and found that we've got those in 'KING ','King ','KNG ', 'K ', '76X80', and ' KING'. Sure, we had ten KING sized bedsheets that matched their request. But when our providers complained that their clients couldn't find the king sized bedsheets in these other styles, we had to point out that they filled the database with non-normalized values. Their solution was not to normalize their data ("that is too much typing for poor us, boo hoo,") but rather to tell us to give the clients a pull-down with every size of bedsheet we carry so they can pick it themselves. So we did, and now all the clients have to sift through literally 20 different abbreviations for four standard sizes. And God help the poor customer who just wants a king-sized bedsheet.
In the case you mentioned, you are a provider of data, not a consumer. As such, you are responsible for providing valid data; that is, data that will work for the consumers. In many (most?) cases, the data providers are not the data consumers. So there is a burden of responsibility on the providers to use data that will make sense to the clients. In our business, we have hundreds of providers who are assumed to be knowledgable and responsible for this data. We have tens of thousands of clients who are just trying to do their jobs, and finding things like king sized bed sheets is just one tiny, untrained, unrewarding aspect of their day. If you give them a search box and they type 'KING' and get many results (but not the one they want) they will rightly assume they did everything right and that the merchandise doesn't exist. They made no mistake, other than trusting that the data made sense.
And you whine because your DBAs won't let you import an unchecked spreadsheet. Cry me a river pal, but get your ass typing. Your unverified data is worse than worthless. It makes the real clients look stupid in front of the customers. And if that's not enough to make you care, look at it this more selfish and pragmatic way: if the clients can't find your merchandise, they can't sell it. You'll be the one justifying to your boss why nobody sold any ' KING' sized sheets.
Granted my needs in this department run towards correcting the colors from my slide scanner, and I'm using them to compensate for the problems in the original color films (dye fading, crappy cheap and/or old film, variations between Ektachrome and Kodachrome, that sort of thing.) And, I'm using the ICC color profiles that came with Vuescan (the best scanning software I've used yet) and so they're corrected before I load them up in Paint Shop Pro. And as far as my digital camera goes, I set the color balance before shooting (taking a shot of a gray card is pretty damn simple these days) and the camera itself adjusts the colors quite adequately.
So, I'm not using color management from within the GIMP or Paint Shop Pro on the images taken directly from my digital camera, but you didn't ask that. You asked which home users need color management. My answer is anyone who cares about the images they produce, and that's not the exclusive domain of the professional photographers. The disks I'm producing are simply family photo albums on DVD for my wife's family, and will probably never be seen by more than fifty people. But they're the most important audience I can imagine, and I don't want to give them crappy images.
The courts have already pretty firmly established that once you go out in public, you are out in public and do not have a reasonable expectation of privacy in many ways. For example, you can be arrested if a cop sees you standing on a sidewalk smoking crack.
However, when you're in your house the courts are on your side. They have established in several cases that the government cannot "scan" your house looking for evidence of illegal activities. They can sift through your trash, and peer through open windows (with high powered telescopes) but they cannot use infrared detectors to search for a basement full of grow lights (and the supposed pot growing beneath them.) I suspect that case would be good precedent for preventing the police from "war-scanning" neighborhoods looking for RF tags belonging to reported shoplifted merchandise.
If they had a warrant, however, they could scan whatever they pleased.
Actually many of the RF security tags in use today have a fusible component. When the item is sold, the cashier runs it over the "burn pad". It emits a substantially higher power level of RF signal than the door readers. The high level of RF being pumped through that tiny antenna causes a fusible link to actually burn out, rendering the tag "dead." No logic involved, and it's ultra-cheap off the shelf technology that's in widespread use today.
These RF security tags are recognizable as square paper-backed foil stickers about 3cm on a side. They sometimes have a fake barcode printed on them as camoflauge.
There is another completely different RF technology security tag that uses a rectangular plastic housing about 1cm x 4 cm x 2 mm thick. It contains a series of metal foil plates that are arranged to resonate at the RF frequencies of the door transmitters. They are deactivated by bringing a magnet to the tag causing the foil in the housing to shift to an adhesive target. This separates the metal plates which prevents it from resonating at its original frequency.
Some of the tags being offered for sale have a rewritable storage component; some even store 1K or more! They all have a static identification, however, that remains an unchanging unique identifier.
The retailers aren't as keen on the rewritable tags, primarily due to the cost. And they don't need it -- they can tie the data in their systems to the static IDs without having to worry about people forging the contents of the dynamic memory.
How about this perfectly legitimate desire of the police to use RFID?
Picture RFID scanner at the doors of a bank, recording every RFID tag that passes through them. The bank is robbed at gunpoint. The surveillance cameras come up with a blurry photo that reveals nothing more than a guy wearing a Bill Clinton Halloween mask. But the RFID scanner recorded tennis shoes purchased from the Buffalo, Minnesota WalM*rt (credit card #12345), jeans purchased from the Buffalo Target (same credit card), underwear recorded as missing from a Gap store, a shirt custom embroidered "Dan" and sold to Bob's Bowling Team in Rockford, Minnesota. It also recorded the RFID tags bundled with the money that was handed to the thief.
The cops will be waiting at Dan's house before he makes it home.
Is this a "legitimate" use of RFID? Is this a privacy violation, or is it simply good police work based on the side effects of a technology being used for purposes other than which it was intended?
Now, put the scanner on the "other door." OK, so now the cops bust Dan for bank robbery, find him with the money PLUS a pound of drugs. They ask the Buffalo and Rockford convenience stores to turn over their surveillance records for the previous week, and see Dan walk through the doors with Joe. Perhaps the gas station cameras from that time even reveal Dan and Joe exchanging a handshake. Can this be used as evidence to go to Joe's house, and perhaps find more drugs?
Isn't it true that most scenarios currently being considered by retailers involve removing/disabling the tags at checkout?
No, it's not true.
Retailers are exploring the potential for returning items based on the RFID tag. That requires the tag to remain active while in the customer's possession.
The benefits of using a durable tag are obvious: the retailer won't require a receipt for the return, as it can simply look up the history of the item, figure out how much you paid for it, and whether you paid cash, check or credit card, and return your money correctly.
The drawbacks are unknown (or at least known only to some privacy wonks who are routinely lumped in with the tinfoil hat brigade,) and that's what Senator Leahy says he wants to explore. Right now, major U.S. retailers are looking to invest lots of money in RFID. Once that expensive infrastructure is in place, they will fight hard to keep it. Senator Leahy wants to make sure that these retailers start out with a long-term acceptable solution, rather than wage a battle later.
I find myself mostly agreeing with the Senator Leahy on many issues. He's certainly the most tech-savvy Senator in the nation, and he appears "geek-friendly" in my eyes. I just wish he was the Senator from my state.
The real question is whether we'll actually see a benefit to the introduction of RFID-- ie, actually keeping things stocked that I *want*, particularly sale items.
This is a common misconception that seems to be attributed to RFID tagging of merchandise. Retailers already have inventory systems, they track every piece of merchandise sold. They know when the Punxatawnee Walm*rt is running low on Gillette Mach III razors, the automated replenishment systems kick in already without RFID tags.
The benefits to RFID tagging (to the retailers) are:
1. Source tagging means less retailer handling at all stages of product handling
2. RFID means built-in theft-deterrent systems
3. Unique item IDs allow return systems to implement fraud prevention
1. As a retailer, I don't have to unpack this large cardboard box of stuff to know that it contains 28 packages of razors, 52 cans of shaving cream and 10 bottles of shampoo. I can scan it when the truck arrives at the loading dock of the store, verifying the merchandise arrived. I can accept drop-shipments from vendors straight to my stores (without having them delivered to a warehouse, and then having to truck them myself to the stores.) The stuff I do pack up in boxes to put on trucks can be all mixed together in a single shipping container, and I don't have to break the boxes open to count the merchandise.
2. Since the tags are unique, if a thief is caught with a tagged item a retailer can produce an invoice showing that unique item was purchased from Nike delivered to the Punxatawnee store on March 23rd. I can also show that single item was never paid for through a cash register. That way the thief can't claim, "I bought it at the Brooklyn store, it's already mine, here's the receipt."
3. As a retailer, the uniqueness of each item tag means I can now tie my return transactions to the sale transactions. There is a direct link showing that this particular item was sold at this time to this person. If it's returned, I can verify that it was sold, and at what price (even without a receipt.) Without unique tagging, I either trust a receipt (which may have been forged, or previously returned on) or I don't give the customer cash. Without a receipt I won't return full value for merchandise now on clearance. If it was purchased with a coupon, I can reprint that coupon for the customer when they return the merchandise. If it was purchased with a discount, I can return the discounted price.
Noticably absent from this list are customer loyalty/tracking programs. Why? Because loyalty systems already exist without RFID. RFID doesn't confer more magical benefits than already exist with a customer identifier married to the receipt of purchased merchandise.
There should be no argument that RFID won't benefit the retailers -- it will. The real questions are "will it harm the consumers?" or "given the current privacy concerns surrounding RFID, will it drive away potential customers?"
Really? Exactly which law did they break? Please, cite it. And once you have the citation, please describe the actions that they took that you believe violates that law.
You may not like what they're doing, (or if you're like me you're probably amused as hell,) but your rant isn't valid until you have specifics. Right now, as far as I can tell they've approached the level of "practical joke." Any legal claim, such as "invasion of privacy" or "bait and switch", has specific legal tests that have to be met. You really need a lot more information to make this kind of claim.
Until then, I'd avoid downloading any UT-Keygen programs from the p2p networks if I were you. Or, at least install Zone Alarm to keep them from tracking you if you download the wrong warez.
Except for the recent announcement by AT&T wireless that since they acquired Cingular they were diverging from the three standard bands and adopting 850 MHz. Not 900, not 1800, not even 1900. There are no 'non-AT&T special' 850 MHz phones in existance yet (and I mean 'special' in a short-bus kind of way.)
So, if you want GSM with AT&T you now have your choice of one of three feature-poor phones (no Bluetooth, no PalmOS, and no cameras (although I don't personally care about cameras)); or your choice of a different GSM provider. And no dual-mode analog feature (although that's just a pet peeve of mine, I am unaware of any GSM phones that have dual mode analog) although with AT&T's extremely poor GSM coverage analog is your best bet for getting a call out of anywhere. But they'll thoughtfully downgrade your existing phone to one of their chunks'o'junk for "free" when you sign up. Whoopee.
So, add another non-standard standard to the mix. Again, and for strictly anti-competetive reasons only. Sigh.
Ooo, thanks a lot! EAC is very, very slick indeed!
I'd been using a hodgepodge of stuff before, but this is very nice. And I like that it uses freedb. (I haven't forgiven them for gracen0te yet, and I get kind of bitter that way.)
This is the only point I think I could argue in Microsoft's favor. I imagine that in the not-too-distant future that Microsoft will have a video-ripper that would produce these digitally protected WMVs, and (with a few advances in hardware) might be able to rip them in real time or even faster.
But other than that I agree totally. I already found that just ripping my existing CDs is already more bother than it's worth to me, so I just use an old-fashioned FM tuner and put up with the crap on the radio. I'm certainly not going to bother ripping video.
The result is the customer is very clear on what they are actually paying or not paying for and MS probably gets less then they should.
Except that's true only to a moderately computer-literate buyer.
There are many, many Joe Sixpacks who go to those shows just to buy the cheapest PC available so they can have a PC. They know nothing about the equipment, they may (or may not) have a technically adept friend who steered them towards these shows for a cheap PC, etc. They know that if they go to Best Buy, a PC will cost them $400, but if they go to a show they can get one for $200.
Joe Sixpack doesn't know anything about "pirated" copies of software. As far as he's concerned, he legitimately bought it because he gave the guy at the show $200. And the transactions typically aren't itemized; it's not like there is a receipt indicating he's paying $50 for the case, $50 for the hard drive, $50 for the mobo, and $50 for the CPU. So there isn't even the omission of a "Windows 98 - $33" line to tip him off that he received Windows without paying for it.
If you already know enough to ask questions like these, then yes, you already know you're getting more than what you're paying for. As you pointed out, both parties will nod and wink at each other when the "test software" remains on the hard drive. Neither party has an active interest in talking about the origins of this software -- so the booth vendor can unfairly compete on price with machines that have bundled legal copies of software, and the booth customer gets a computer full of software for dirt cheap.
While I think Microsoft is 'ethically challenged' in oh-so-many ways, it is still not appropriate for me to respond by making copies of their software in protest. If I think their software is overpriced; or if I think they are hiding API calls from me as a developer; or if I think they are predatorily acquiring competitors, I do have legal recourse through the court system. Acquiring an unlicensed copy of Windows claiming "stick it to da man for overcharging!" is not a form of protest -- it's a form of stealing.
However, I think the point of this discussion is that some "low-end" computer owners acquire a 3D game, play it for a while, and get to enjoy it. Then one day they see it running on a 3GHz box with a top-end graphics card on Tech TV or at a friend's house, and say "Wow, you mean this is what it looks like on a fast computer? I gotta get me some of that!"
They then run out and buy a new computer. If they choose one based on the P4, they then drop several dollars into Intel's coffers.
Remember who authored the original article: Intel.
I'm not claiming conspiracy, or anything else nefarious here. I'm just pointing out that it's in Intel's best interests to get as many people as possible hooked on beautiful graphics, because that translates directly into bottom line sales. And I can tell you from personal experience that getting people hooked on the low-end graphical version of a great game does directly translate into hardware sales. Unfortunately for Intel, I upgraded my AMD chips to faster AMD chips, but the idea is the same.
I guess maybe I've written too many COM objects to think hard about it anymore.
I found a lot of the problems were in keeping up with the Microsoft AOTD (Acronym Of The Day.) COM itself is quite elegant. And ATL provides a decent wrapper layer around most of the ugly stuff. But if you go starting to play with IDispatch, or OLE, or trying to use the MFC wrappers instead of the ATL wrappers (or worse, try mixing and matching MFC and ATL) you're asking for a world of headaches. And DCOM (or COM+) has been completely useless to us -- the authentication schemes hog-tie you into uncontrollable access problems that we can't afford. But just plain, simple COM objects are great on a Windows box. I have C++, VB and ASP clients all sharing my objects. They just work.
Not that I don't have troubles now and again, mind you. It's just that once you've got it, it's not a big deal. The ATL smart pointer macros wrap the interface nicely, to the point where once you've acquired a pointer to one, you don't even have to know you're working with a COM object anymore (unless you choose to.)
Of course, the COM architecture is not as elegant as STL, and it's not as simple as stdio. But it can be powerful.
<real_reason>
Besides, they pay me money to do it. So what am I supposed to do, turn 'em down?:-)
</real_reason>
Slashdot Mirror
Now, if I could just ASCII-ART up a cockroach ...
Anyway, did anyone else read this and think immediately of the Evil Bit? The whole thing has got to be a joke, right?
Let's just say that none of them will ever be invisible on a thermal scan. That reminds me -- next time my firefighter buddy has the Cairns-Iris around, we'll have to look at the PC through it.
Anyway, my biggest problem with these cases is that they don't look particularily well ventilated. Of course, if you can afford one of these, you can also afford a watercooler, so it shouldn't make much difference. Hmmm...I'm not so sure I want a case that costs more than the computer it houses...
I liked the black walnut burl. But none of them look like they have adequate ventilation for the AMD systems I run.
"Rodger Cosgrove, president of Entremedia, a direct marketing firm and a member of Reason's board, assisted in coming up with a program that allows the subscriber list to be integrated with satellite photographs. He also worked with Xeikon, the manufacturer of the printer that made the endless customization possible."
Except it's only happening on the cover of Reason.
It's a libertarian magazine. Nobody of any significance to the American political process would be caught dead reading it. Sorry to be the troller of bad news, but there just isn't enough support between the donkeys and the elephants to make any difference at all.
I'm just the poor schnook presenting the front end, the guy who has to make sense of this crap data and simplify it so an average underpaid employee can find bedsheets for our customers. When we first encountered this rotten data, we went to the DBAs who told us "hey, these other folks are responsible for the data in these tables, yell at them." And those folks really have no idea how they got into this situation in the first place -- it took us a long time to track down the origins of the bad data. And once we identified their problems, they said, "Too bad, we can't fix all that data now, not in the budget, deal with it as is." Our client who is representing the end users of this data said he couldn't push those other people into fixing their data, and so we'd just have to be smarter in how we presented our screens.
So that's my sad tale. We wrote nasty, inefficient code to sniff around the tables to try to present the most logical choices to our clients. It mostly works, and the customers mostly find their king-sized bedsheets in the style of their choosing. But it shouldn't have to be like this, and it wouldn't have been if the originators of that data had been forced to follow a schema, rather than simply adding new values to existing ones.
Look at the case I mentioned earlier. In that case we had someone adding a line of merchandise completely unrelated to bedsheets that came in King and Queen sizes, but they had 'KNG', 'QUN' in their spreadsheets (they probably chose a 3 letter wide column in their spreadsheet to fit their screen.) So someone imported the data, adding KNG and QUN to the list of valid sizes. So now the SIZE field contains KNG and QUN as options, and the next merchant buying bedsheets very reasonably selected KNG and QUN from their dropdown list, missing the very existance of KING and QUEEN.
We got screwed because someone imported data from a spreadsheet where they had the opportunity to invent their own sizes, free from any schema or data model restrictions. Had they been entering the data into the database in the first place, they would have seen KING and QUEEN as options and not felt the need to invent their own.
For a business? For tracking the bicycling habits of the downtown commuters who use the racks at the various garages, looking for ridership trends during the bus strike, or to determine if we need to spend tax money on bicycle lanes or trail improvements? A database, hands down. It allows spotters on 10th Street to enter data without interfering with the spotters entering data on 11th Street. It allows the planning commissions to perform complex joins, or to compare commute data to weather data or to bicycle theft rates. And if they need pretty pie charts, they can still use Excel as a front end to an ODBC data source to prepare them.
As you were probably trying to point out, you need to be realistic in your assessments.
The central point of the whole article above is that if individual commuters had spreadsheets like yours, they would be of no value to a business or government trying to use the spreadsheet-based data.
The point where a business analyst can make a difference is in spotting these spreadsheets floating around a business (and taking on importance of their own,) and replacing them with a database.
Obviously you have never dealt with trying to give the clients what they want after someone else has polluted the database with their crappy imported data.
Here's a real example: our clients wanted to find all the KING size bedsheets. We looked them up and found that we've got those in 'KING ','King ','KNG ', 'K ', '76X80', and ' KING'. Sure, we had ten KING sized bedsheets that matched their request. But when our providers complained that their clients couldn't find the king sized bedsheets in these other styles, we had to point out that they filled the database with non-normalized values. Their solution was not to normalize their data ("that is too much typing for poor us, boo hoo,") but rather to tell us to give the clients a pull-down with every size of bedsheet we carry so they can pick it themselves. So we did, and now all the clients have to sift through literally 20 different abbreviations for four standard sizes. And God help the poor customer who just wants a king-sized bedsheet.
In the case you mentioned, you are a provider of data, not a consumer. As such, you are responsible for providing valid data; that is, data that will work for the consumers. In many (most?) cases, the data providers are not the data consumers. So there is a burden of responsibility on the providers to use data that will make sense to the clients. In our business, we have hundreds of providers who are assumed to be knowledgable and responsible for this data. We have tens of thousands of clients who are just trying to do their jobs, and finding things like king sized bed sheets is just one tiny, untrained, unrewarding aspect of their day. If you give them a search box and they type 'KING' and get many results (but not the one they want) they will rightly assume they did everything right and that the merchandise doesn't exist. They made no mistake, other than trusting that the data made sense.
And you whine because your DBAs won't let you import an unchecked spreadsheet. Cry me a river pal, but get your ass typing. Your unverified data is worse than worthless. It makes the real clients look stupid in front of the customers. And if that's not enough to make you care, look at it this more selfish and pragmatic way: if the clients can't find your merchandise, they can't sell it. You'll be the one justifying to your boss why nobody sold any ' KING' sized sheets.
Granted my needs in this department run towards correcting the colors from my slide scanner, and I'm using them to compensate for the problems in the original color films (dye fading, crappy cheap and/or old film, variations between Ektachrome and Kodachrome, that sort of thing.) And, I'm using the ICC color profiles that came with Vuescan (the best scanning software I've used yet) and so they're corrected before I load them up in Paint Shop Pro. And as far as my digital camera goes, I set the color balance before shooting (taking a shot of a gray card is pretty damn simple these days) and the camera itself adjusts the colors quite adequately.
So, I'm not using color management from within the GIMP or Paint Shop Pro on the images taken directly from my digital camera, but you didn't ask that. You asked which home users need color management. My answer is anyone who cares about the images they produce, and that's not the exclusive domain of the professional photographers. The disks I'm producing are simply family photo albums on DVD for my wife's family, and will probably never be seen by more than fifty people. But they're the most important audience I can imagine, and I don't want to give them crappy images.
However, when you're in your house the courts are on your side. They have established in several cases that the government cannot "scan" your house looking for evidence of illegal activities. They can sift through your trash, and peer through open windows (with high powered telescopes) but they cannot use infrared detectors to search for a basement full of grow lights (and the supposed pot growing beneath them.) I suspect that case would be good precedent for preventing the police from "war-scanning" neighborhoods looking for RF tags belonging to reported shoplifted merchandise.
If they had a warrant, however, they could scan whatever they pleased.
These RF security tags are recognizable as square paper-backed foil stickers about 3cm on a side. They sometimes have a fake barcode printed on them as camoflauge.
There is another completely different RF technology security tag that uses a rectangular plastic housing about 1cm x 4 cm x 2 mm thick. It contains a series of metal foil plates that are arranged to resonate at the RF frequencies of the door transmitters. They are deactivated by bringing a magnet to the tag causing the foil in the housing to shift to an adhesive target. This separates the metal plates which prevents it from resonating at its original frequency.
Some of the tags being offered for sale have a rewritable storage component; some even store 1K or more! They all have a static identification, however, that remains an unchanging unique identifier.
The retailers aren't as keen on the rewritable tags, primarily due to the cost. And they don't need it -- they can tie the data in their systems to the static IDs without having to worry about people forging the contents of the dynamic memory.
Picture RFID scanner at the doors of a bank, recording every RFID tag that passes through them. The bank is robbed at gunpoint. The surveillance cameras come up with a blurry photo that reveals nothing more than a guy wearing a Bill Clinton Halloween mask. But the RFID scanner recorded tennis shoes purchased from the Buffalo, Minnesota WalM*rt (credit card #12345), jeans purchased from the Buffalo Target (same credit card), underwear recorded as missing from a Gap store, a shirt custom embroidered "Dan" and sold to Bob's Bowling Team in Rockford, Minnesota. It also recorded the RFID tags bundled with the money that was handed to the thief.
The cops will be waiting at Dan's house before he makes it home.
Is this a "legitimate" use of RFID? Is this a privacy violation, or is it simply good police work based on the side effects of a technology being used for purposes other than which it was intended?
Now, put the scanner on the "other door." OK, so now the cops bust Dan for bank robbery, find him with the money PLUS a pound of drugs. They ask the Buffalo and Rockford convenience stores to turn over their surveillance records for the previous week, and see Dan walk through the doors with Joe. Perhaps the gas station cameras from that time even reveal Dan and Joe exchanging a handshake. Can this be used as evidence to go to Joe's house, and perhaps find more drugs?
No, it's not true.
Retailers are exploring the potential for returning items based on the RFID tag. That requires the tag to remain active while in the customer's possession.
The benefits of using a durable tag are obvious: the retailer won't require a receipt for the return, as it can simply look up the history of the item, figure out how much you paid for it, and whether you paid cash, check or credit card, and return your money correctly.
The drawbacks are unknown (or at least known only to some privacy wonks who are routinely lumped in with the tinfoil hat brigade,) and that's what Senator Leahy says he wants to explore. Right now, major U.S. retailers are looking to invest lots of money in RFID. Once that expensive infrastructure is in place, they will fight hard to keep it. Senator Leahy wants to make sure that these retailers start out with a long-term acceptable solution, rather than wage a battle later.
I find myself mostly agreeing with the Senator Leahy on many issues. He's certainly the most tech-savvy Senator in the nation, and he appears "geek-friendly" in my eyes. I just wish he was the Senator from my state.
This is a common misconception that seems to be attributed to RFID tagging of merchandise. Retailers already have inventory systems, they track every piece of merchandise sold. They know when the Punxatawnee Walm*rt is running low on Gillette Mach III razors, the automated replenishment systems kick in already without RFID tags.
The benefits to RFID tagging (to the retailers) are:
1. As a retailer, I don't have to unpack this large cardboard box of stuff to know that it contains 28 packages of razors, 52 cans of shaving cream and 10 bottles of shampoo. I can scan it when the truck arrives at the loading dock of the store, verifying the merchandise arrived. I can accept drop-shipments from vendors straight to my stores (without having them delivered to a warehouse, and then having to truck them myself to the stores.) The stuff I do pack up in boxes to put on trucks can be all mixed together in a single shipping container, and I don't have to break the boxes open to count the merchandise.
2. Since the tags are unique, if a thief is caught with a tagged item a retailer can produce an invoice showing that unique item was purchased from Nike delivered to the Punxatawnee store on March 23rd. I can also show that single item was never paid for through a cash register. That way the thief can't claim, "I bought it at the Brooklyn store, it's already mine, here's the receipt."
3. As a retailer, the uniqueness of each item tag means I can now tie my return transactions to the sale transactions. There is a direct link showing that this particular item was sold at this time to this person. If it's returned, I can verify that it was sold, and at what price (even without a receipt.) Without unique tagging, I either trust a receipt (which may have been forged, or previously returned on) or I don't give the customer cash. Without a receipt I won't return full value for merchandise now on clearance. If it was purchased with a coupon, I can reprint that coupon for the customer when they return the merchandise. If it was purchased with a discount, I can return the discounted price.
Noticably absent from this list are customer loyalty/tracking programs. Why? Because loyalty systems already exist without RFID. RFID doesn't confer more magical benefits than already exist with a customer identifier married to the receipt of purchased merchandise.
There should be no argument that RFID won't benefit the retailers -- it will. The real questions are "will it harm the consumers?" or "given the current privacy concerns surrounding RFID, will it drive away potential customers?"
They are paying you for it by giving you a discount.
Really? Exactly which law did they break? Please, cite it. And once you have the citation, please describe the actions that they took that you believe violates that law.
You may not like what they're doing, (or if you're like me you're probably amused as hell,) but your rant isn't valid until you have specifics. Right now, as far as I can tell they've approached the level of "practical joke." Any legal claim, such as "invasion of privacy" or "bait and switch", has specific legal tests that have to be met. You really need a lot more information to make this kind of claim.
Until then, I'd avoid downloading any UT-Keygen programs from the p2p networks if I were you. Or, at least install Zone Alarm to keep them from tracking you if you download the wrong warez.
So, if you want GSM with AT&T you now have your choice of one of three feature-poor phones (no Bluetooth, no PalmOS, and no cameras (although I don't personally care about cameras)); or your choice of a different GSM provider. And no dual-mode analog feature (although that's just a pet peeve of mine, I am unaware of any GSM phones that have dual mode analog) although with AT&T's extremely poor GSM coverage analog is your best bet for getting a call out of anywhere. But they'll thoughtfully downgrade your existing phone to one of their chunks'o'junk for "free" when you sign up. Whoopee.
So, add another non-standard standard to the mix. Again, and for strictly anti-competetive reasons only. Sigh.
I'd been using a hodgepodge of stuff before, but this is very nice. And I like that it uses freedb. (I haven't forgiven them for gracen0te yet, and I get kind of bitter that way.)
But other than that I agree totally. I already found that just ripping my existing CDs is already more bother than it's worth to me, so I just use an old-fashioned FM tuner and put up with the crap on the radio. I'm certainly not going to bother ripping video.
Except that's true only to a moderately computer-literate buyer.
There are many, many Joe Sixpacks who go to those shows just to buy the cheapest PC available so they can have a PC. They know nothing about the equipment, they may (or may not) have a technically adept friend who steered them towards these shows for a cheap PC, etc. They know that if they go to Best Buy, a PC will cost them $400, but if they go to a show they can get one for $200.
Joe Sixpack doesn't know anything about "pirated" copies of software. As far as he's concerned, he legitimately bought it because he gave the guy at the show $200. And the transactions typically aren't itemized; it's not like there is a receipt indicating he's paying $50 for the case, $50 for the hard drive, $50 for the mobo, and $50 for the CPU. So there isn't even the omission of a "Windows 98 - $33" line to tip him off that he received Windows without paying for it.
If you already know enough to ask questions like these, then yes, you already know you're getting more than what you're paying for. As you pointed out, both parties will nod and wink at each other when the "test software" remains on the hard drive. Neither party has an active interest in talking about the origins of this software -- so the booth vendor can unfairly compete on price with machines that have bundled legal copies of software, and the booth customer gets a computer full of software for dirt cheap.
While I think Microsoft is 'ethically challenged' in oh-so-many ways, it is still not appropriate for me to respond by making copies of their software in protest. If I think their software is overpriced; or if I think they are hiding API calls from me as a developer; or if I think they are predatorily acquiring competitors, I do have legal recourse through the court system. Acquiring an unlicensed copy of Windows claiming "stick it to da man for overcharging!" is not a form of protest -- it's a form of stealing.
However, I think the point of this discussion is that some "low-end" computer owners acquire a 3D game, play it for a while, and get to enjoy it. Then one day they see it running on a 3GHz box with a top-end graphics card on Tech TV or at a friend's house, and say "Wow, you mean this is what it looks like on a fast computer? I gotta get me some of that!"
They then run out and buy a new computer. If they choose one based on the P4, they then drop several dollars into Intel's coffers.
Remember who authored the original article: Intel.
I'm not claiming conspiracy, or anything else nefarious here. I'm just pointing out that it's in Intel's best interests to get as many people as possible hooked on beautiful graphics, because that translates directly into bottom line sales. And I can tell you from personal experience that getting people hooked on the low-end graphical version of a great game does directly translate into hardware sales. Unfortunately for Intel, I upgraded my AMD chips to faster AMD chips, but the idea is the same.
I found a lot of the problems were in keeping up with the Microsoft AOTD (Acronym Of The Day.) COM itself is quite elegant. And ATL provides a decent wrapper layer around most of the ugly stuff. But if you go starting to play with IDispatch, or OLE, or trying to use the MFC wrappers instead of the ATL wrappers (or worse, try mixing and matching MFC and ATL) you're asking for a world of headaches. And DCOM (or COM+) has been completely useless to us -- the authentication schemes hog-tie you into uncontrollable access problems that we can't afford. But just plain, simple COM objects are great on a Windows box. I have C++, VB and ASP clients all sharing my objects. They just work.
Not that I don't have troubles now and again, mind you. It's just that once you've got it, it's not a big deal. The ATL smart pointer macros wrap the interface nicely, to the point where once you've acquired a pointer to one, you don't even have to know you're working with a COM object anymore (unless you choose to.)
Of course, the COM architecture is not as elegant as STL, and it's not as simple as stdio. But it can be powerful.
<real_reason> :-)
Besides, they pay me money to do it. So what am I supposed to do, turn 'em down?
</real_reason>