Lots of people purchase those external USB drives, plug the cable in, press the "backup" button, then put the drive back on the shelf until the next backup. I don't know if you consider that "static" or not, but they're not running 24x7.
As far as mag tape goes, virtually every tape used in for computer backups has been used as a digital storage medium. The only analog exceptions I can think of are the audio-cassette tapes used by some of the earliest home computers (Commodore VIC-20s and C-64s, TRS-80s, Color Computers, and Apples,) which modulated the audio input to a tape recorder like a modem; and a weird VHS-based system I remember seeing ads for that stored a lot of data in the video signal. Even the 9-track reel-to-reel tapes you might see as props in old movies stored their data digitally.
Tape has a couple of things going for it. It's a well-understood medium with a long track record; and it's about as cheap as you can get. Otherwise tape is relatively slow because of the limitations of sequential access, and tapes have become bulky compared to hard drives. Tape drives themselves are also expensive; they're finicky about things like alignment and environment, and they're not exactly widely-consumer-available items. They are still workhorses found in many mainframe environments, though, so they're not quite dead yet.
OK, this is my last post; if you post after this one you win.:-)
Yes, today USB dongles are available for Centronics parallel as well as RS-232. But they're not available for ST-506 hard drives. If you want to get data off one of those old drives, you'll need to find an old box with an ST-506 card and somehow get the data off of it. Perhaps you could rig up a Kermit, XModem or uucp link over RS-232, or you might get lucky and find an ethernet adapter plus the correct device driver software for whichever version of DOS you're running, and find a TCP/IP stack and an FTP client. To me, that's not a plan for an archival system, that's a crapshoot.
So in 20 years, when we all have fiber-attached holographic crystal storage, and all this SATA nonsense is long behind us, do you know if you'll be able to find a SATA-to-holofiber dongle? What if absolutely every component goes wireless, and mass storage goes 802.11-DD even inside the box? Do you know if you'll be able to buy an RF-to-SATA dongle for your old drives? How about in 40 years?
My point is that if you apply what we've seen happen to standards over the last 20 years, you'll find that very, very little of it has survived. Will RS-232 remain relevant for another 20 years? In that time it's slipped from a dominant personal computer interface to a quaint 9-pin jack on the back of yesterday's PC. What will happen to USB and/or Firewire? Just because we think they're ubiquitous today is no guarantee that they will be 20 years from now.
I think the right answer is to say that there is no good long-term archival solution for computer data. There are good medium term solutions, but the long term requires an active migration schedule, meaning a data curator will be required to port the data to the "medium of the moment" every ten years or so. You can't just stick a CD in a museum exhibit and hope to still have access to a working CD-ROM drive a hundred years from now.
And as far as buses go, I looked in the wwwikipedia and found this entry for Computer bus, which has this unambiguous second sentence: "Unlike a point-to-point connection, a bus can logically connect several peripherals over the same set of wires." (And you can check the "history" page, I had nothing to do with that definition!:-) Yes, RS-232 is a collection of related electrical signals, but it's not a bus, it's a point-to-point connection specification. USB qualifies as a bus (just barely, because they put the word "logical" in their definition) because the standard specifically defines a "hub" and a means of device discrimination. Although you can buy an RS-232 "hub" (OK, so it's a 25-pole, n-throw switch,) the RS-232 standard itself has no such specification for using or controlling such a device, so it doesn't fit that category.
Sort of. While RS-232 is still fairly common, and PCI serial cards are in the $5-$15 price range, many modern motherboards no longer come with on-board serial or parallel connectors. They are fading slowly, but they are fading. And I think you'd find it difficult to transfer the contents of a 250GB drive over an RS-232 connection. 230kbps is the fastest I've ever seen it pushed to, and my understanding is anything over 19.2kbps is already faster than the official published standard supports. So extrapolate that forwards 20 years. Ordinary drive sizes could be 1000 times bigger than they are now, in the 250 TB range. Will USB-2 or Firewire still be useful then, or will they be fading just as RS-232 is fading today?
( And just to pick a pointless fight, I don't think RS-232 qualifies as a "bus" as it's really only an electrical interface between a single DTE and a single DCE device. "Bus" implies (to me, anyway) that it supports more than one simultaneous device, and without some kind of add-ons or non-standard protocol RS-232 doesn't fit that definition. )
The G.P. was talking about storing material today and retrieving it 20 years from now. A hard drive is actually a very poor solution for archival of material. First, drive technology has advanced dramatically in the last 20 years. It continues to advance dramatically -- compare the commonly produced sizes now to those made just 5 years ago, let alone 20 years.
Electrically, the interfaces have changed, too. You'd be very hard-pressed to find a modern machine that's capable of reading from an old Winchester drive -- the ST-506 interface is dead and buried, and the adapter cards that spoke ST-506 were made only in the era of 8-bit ISA bus machines; you won't find a PCI card that supports them. ATA came along, and has advanced to ATA-6. Now we have SATA, which begs the question of how long PATA will live. Do you want to bet your future retrieval of the data to finding an ancient machine that can read SATA on the 2026 equivalent of eBay?
There are plenty of physical reasons not to use hard drives as an archival medium, too. You'd probably be hard-pressed to find an old Winchester drive that could spin up today after sitting idle for 20 years. Drives manufactured back then suffered from stiction, which was caused by lubricants that sat idle for too long. Do you know what's wrong with the longevity of data on drives manufactured today? I don't. Will today's lubricants still flow freely in 2026? Will the platters, heads and mechanics survive the years uncorroded? Will the electrolytic capacitors still hold a charge? Will the connectors have shifted due to thermal expansion and contraction? Will the magnetic fields of some bits have dissipated due to their proximity to other bits? Will the adhesive holding the media to the platters have broken down?
And Google for "maxtor sucks" if you want to read horror stories of people losing data due to the death of a hard drive.
I'm not saying Super-8 is the way to go, but it's still possible to get the data from it. Will the same be true of floppy discs, ZIP disks, CD-Rs, CD-RWs, and all the burnable variants of DVDs including +/-, DVD-R, DVD-RW and DVD-RAM?
Hey, I'm just now getting my techno-friends converted to DVR believers. And it's hard -- nobody understands how great they are until they spend an evening watching TV through one. A two-minute demo is just not impressive. It always starts with "Pause and rewind? BFD, I could do that with my VCR." But pause live TV while you answer the phone or go pick up another six-pack, and it becomes much cooler. Then when you play it back after you return and skip the commercials, they suddenly recognize it's the greatest invention since the remote.
The other problem is everybody knows me as a geek who just knows all this stuff, so they immediately assume it's out of their ballpark to understand. But when my wife starts recommending them, that's when it really seems to hit them that it's something they could use.
He hinted at it in TFA. Social networking sites "encourage" such behavior via peer pressure. "I gotta get a hundred people on my friends list!" or "I have to have more friends than Jane just to make her jealous." And one proven way to get people to notice you is to be more shocking on your page, via nudity, stunts or whatever.
Can't have a vintage event without an Apple ][+, although I'm guessing your budget won't allow it. Perhaps you could borrow one?
One thing popular during that time were timeshare computers. You might be able to find an old Teletype or Silent 700 terminal with an acoustically coupled modem, and have those dial into a machine hosting a few games such as Hunt the Wumpus, tic-tac-toe, or global thermonuclear war. Dumb terminals of that era can be had for pretty cheap -- a buddy of mine just picked up a Teletype 43 for $40 off of eBay (beware of shipping costs, though.)
Actually, that sounds a lot like a phrase from Weird Al's Your Horoscope for Today
Taurus
The stars predict tomorrow you'll wake up, do a bunch of stuff, and then go back to sleep.
#11. Incredible run-on sentences that are in a difficult-to-read font and are not punctuated and sometimes written in the second person familiar and sometimes they changed tense and ended illogically disconnected from their premises even though you read them through to the end.
I also note that Silverfast works with ICC color profiles, and can be calibrated with an ICC target. This is the weakness of software like VueScan.
Actually Vuescan comes with ICC profiles for lots of different models of scanners, and if you have an IT 8 target you can create a profile calibrated to your scanner. (If I were to do this again, I'd try to calibrate it every thousand slides or so to compensate for the lamp fading over its lifetime.) Plus, Vuescan includes film profiles for every type of film and process I've ever shot (not that I have a particularly wide range, my experience being limited to the ordinary films made by Kodak, Agfa, Fujifilm and Ilford.)
And you're right, the GUI is as ugly as a dog, and is not particularly friendly either. I can't argue that point. But once I got my workflow set up, it did run me through the scanning process quite efficiently.
One other thing to consider: what's the destination of the images being scanned? If they're going to a printer, that's one thing. But if they're just going on a DVD, a precisely calibrated scanner is total overkill for an NTSC image.
Personally, I can't imagine how much bandwidth costs Microsoft. Distributing service packs, updates, hosting time servers, all that has to cost a fortune. Of course, they have a large fortune, so it's obviously worth it.
I don't really have a problem with sites like AutoPatcher, because I don't care that much about Microsoft's TOS on that page. I just thought his belief that it was all on the up-and-up was hilarious. But imagine the sh!t that would hit the fan if he distributed a virus-infected patch. Lawyers, feds, trials, it would break his piggy bank forever. At least he's not charging money for the service, he's just begging for bandwidth.
As for people who use AutoPatcher, can you "swear" that you trust him? Imagine what would happen if he deliberately wrote a password-sniffing Trojan and included it in one of the autopatch images? Microsoft may be the Evil Empire, but you know they have too much to lose by inserting identity-thieving code. This guy is small, and he could decide "hey, the Mob is offering me three million bucks to put out this zombie-creating patch, I think I'll take the money and head to the Cayman Islands!"
I recently scanned all of my in-laws 35mm slides (over 3000 slides) and burned DVDs for them for holiday presents, and I learned a lot of lessons in the exercise.
I spent a lot more time than I had to because I scanned them all at 20MB raw image size (the jpegs averaged roughly 6MB each when I was done.) My intent was to keep good-quality archival copies of the slides. However, these large files meant that every action with them was slow: transferring them from the scanner via USB 1.2 took like 40 seconds per image, loading and saving them in an editor to rotate and crop them was slow, importing them into Arcsoft to produce the slideshow was slow, and so on.
Decide what you intend to do with the digital images first. Are you going to archive them as I did? Then accept that it will be slow. An archival quality scan of the medium format film that you describe will take hundreds of megabytes per image. But if you're just going to burn a DVD for the family and discard the scans as intermediate files, scan them at DVD resolutions and you'll save a ton of time throughout the process.
Invest in some good scanning software. The out-of-the-box stuff I got from Minolta was slower than molasses. It took it 20 seconds to autofocus each slide individually, and that was prior to the scan itself! I purchased Vuescan from Hamrick software and it sped the process considerably. They support many dozens of scanners, film profiles, etc. It automated the process of scanning a full carrier of slides. It was worth every penny to me.
Use a dust brush on each and every negative before scanning it. A cheap squeezy rubber-bulb brush will clean up most dust and hairs nicely, and they're only like $5.00.
Don't bother printing them unless you actually want the prints of the pictures.
Find a good program to help rotate and crop the images, clean up dust specks, and fix colors. I used Paint Shop Pro, and eventually got pretty fast at it. Later, I found RPhoto (freeware! on the web) that enabled me to whip through rotating and cropping at high speeds.
Figure out in advance how you want to organize the images you scan. I built a directory structure by year, and scanned the images in rough chronological order. If there is no organization to your media, be sure to take the time to tag them at some point in the process (probably the time you crop and rotate them.) Names, places and dates are all good searchable data. I used a short description for the file names, but I wish I'd edited the EXIF data when I had the chance.
Regarding medium format film, ask about flatbed scanners at a good photography shop. When I was shopping for newer film scanners, I found an Epson flatbed with a "negative attachment." It consisted of a backlight-box that had a snap-in film carrier on the bottom that would hold 2 five-frame 35mm filmstrips. You could remove the film carrier and use a larger frame to hold your negatives in place (the adjustable carriers that you use in enlargers to hold medium format film comes to mind.)
Once you figure out what you're doing, take a few minutes and write up an instruction sheet. You'll probably go stir-crazy after scanning a thousand frames, and you'll likely want to take a break for a few months. It's nice to come back to full instructions so you can pick up exactly where you left off.
Realize that this will take a lot of your time. Check with a commercial photo house and ask about their scanning rates. I was quoted from about $0.75 per slide to $1.20 per slide. Of course with over three thousand slides to scan I wasn't about to spend that kind of money, but I did spend several hundred on a Minolta Dimage film scanner, and many, many hours scanning. That's where the instruction sheet helped -- my wife picked it up and she started scanning in her free time, too! You might want to consider hiring a photo house to use a drum scanner just for your medium format slides, rather than tackle them yourself. You'll get the best quality scans
This site should be "within the limits" of that TOS simply because they don't provide the software. He just provides a tool which you can use to download it from the official Microsoft site, and the TOS doesn't say anything about how you download them, just where you download them from.
Autopatcher, on the other hand, provides the actual software, which is explicitly prohibited by the TOS you mentioned. He has this hilarious line in his FAQ:
Q: Is AutoPatcher legal?
A: Yes, nwraptor once spoke to a Microsoft employee and apparently they know about us but dont care what we do! Now that's legal advice you can hang your hat on!
Ask your cop friend how many of those PS2s are plugged into HDTVs with surround speakers, vs how many are jacked into 25-year-old Magnavox TV sets.
Games are almost as much of a status symbol as Nikes. Yes, they'll probably churn game consoles. But how many churn TV sets to keep up with those consoles?
by the time a new medium has been produced, the old software is worthless.
Again, it's not true that your old software "is worthless". Your old copy of Office 95 didn't stop working when Office 97, Office 2000 or Office XP came out. They didn't become worthless. You volunteered to stop using it.
Microsoft is a corporation that lives almost entirely on churn. Think about their cash flow, and where it comes from. Sales of new products is the bulk of their money, with a relative trickle from their professional services. Microsoft.com isn't a pay-as-you-go web site. They're not like IBM who licenses mainframe software on an annual basis. Your copy of XP stopped generating them revenue the moment after you bought it; you don't pay a subscription fee for it. Same with Office. Think about Word -- what features did they add to Word to make you need to buy the latest version? I promise you that "Now with advanced Tabs and Rulers!" isn't a slogan designed to drive Office fanbois into the stores.
Microsoft is somewhat afraid of the near future because their biggest cash cow, Office, doesn't require upgrades at the same rate as their operating systems do. The only reason I upgraded my home version of Office 97 was I needed to add Powerpoint, not because I needed "adjustable margins" or whatever they had added to Word in the previous 8 years. And that's why their long term plan is.net and Vista. With Trusted Computing, they'll be able to move you to a subscription model. Just think of it: an OS that can enforce licensing. No more selling Office licenses that are good forever. With no new features, they can "give" the software away, but cripple things like printing or saving unless you pay them per month, or even on a by-usage basis. Want to create a Powerpoint slideshow? That'll be $10, please. Now there's a revenue stream to bet your future on.
Eventually Microsoft will encourage people to not run unsigned code. "Ooo, it might contain a virus, don't run it or your Windows Warranty will be voided!" How much do you think Microsoft will charge to sign a copy of Open Office or Ghostscript? And do you honestly think they'd ever sign Exact Audio Copy? Hell, they'll probably put it in their "Pirate Tools" list of binaries that will never run.
Console generations are the epitome of churn. They add a few new features, boost the screen rez a touch, and start putting out non-backward-compatible titles. They have done this every five years for the last three decades, meaning churn, churn, churn.
Think about the word "obsolete" for a minute. Does it mean your old console is worn out, eroded by time and usage? Did it break? Did your N64 games stop working when the Game Cube came out? Did your Game Cube stop working on the release date of the Wii? Did Super Mario Kart expire, or did Bowser refuse to come out and play? No, it's obsolete because you were the victim of successful marketing to your own greed. "Own the shiniest video game! Your old console sucks because we have a new one! Don't be the chump with last year's console!"
Nothing went wrong with your existing system, yet you replaced it on the whim of a corporation. Churn.
Mind you, my retirement fund is based in large part on people like you continuing to churn video games and the like. Feel free to continue your participation in capitalism.
Re:Sony's dumb decision, with historical precedent
on
No Love For The Blu-Ray
·
· Score: 2, Interesting
Because they have no way of seeing the loss of primary hardware sales. There is no measure of how many Sonys were unsold due to a lack of CF or SD support. They don't know how many Betamax VCRs they didn't sell because they didn't support VHS. All they know is if they put out a line of proprietary crap some people will buy them. And the more they put memory stick support in their Vaios, Sony-Ericsson phones, and other random bits of hardware, the more they delude themselves into believing they've developed a viable standard.
Sony arrogance has lost me as a customer for life, and all the friends and family who rely on me for tech support as well. Not that they give a crap about some random guy in Minnesota who maybe goes shopping with five or six people per year, but I can list off perhaps $10,000 in sales in the last two years that specifically did not go to Sony due simply to my opinion of them. At least four cameras, three big screen TVs and a handful of cell phones adds up to a few dollars.
Maybe that's what the web needs: a list of "lost sales". Imagine an honest (ha!) tabulation of the purchases of everyone who specifically rejected Sony products because of the company. It might surprise me to see how small the list is, or it might surprise Sony to see how badly they've judged us.
I think the problem companies like Sony and Microsoft have is that they recognize their profits come from churn. Come out with X-Box 360 and the loyal X-Box owners will buy it. Come out with PS2 and loyal PS1 owners will buy it.
But that only works for a limited set of people with enough money, and that's not even half of America any more. It is a great theory for products like games and laptops, because they're already owned by the rich half who can afford to play the churn game. But it doesn't work as well on mass market items such as TVs and DVD players. Look at how long it's taken to replace VCRs with DVD players. Most people consider a TV a "big-ticket" item, and expect them to last 20 years or more. And nobody with a non-HD TV has any reason to consider an HD-DVD or Blu-Ray player.
Churn is great for cell phones, where they can continually "upgrade" them by adding more and more crappy features, and give them away (with expensive contracts.) But churn is not going to sell HDTV sets to everyone across the nation, and HDTV is a prerequisite to selling HD players.
a lot of their stuff is written in languages like C and C++ where you can pass a buffer to a method without its bounding information
And it took them until Visual Studio 2005 for them to bite that last bullet. In case you haven't used it yet, they've added a set of "safer" routines to the standard C runtime library. They They are not backward compatible. Now, instead of sprintf(), you call sprintf_s(), which takes the traditional output buffer pointer plus a new parameter indicating the length of the output buffer. It also validates the format string, although if you let the users modify that you'll still have problems (and the documentation now specifically warns the developer that this is a security issue.) And they've added a new set of warnings to the compiler to let you know if you're still using the deprecated old routines. You can turn them off, of course, but Microsoft recommends you fix your code instead.
This was really a huge step. With it, they are forcing programmers to at least consider what happens when they use pointers. Sure, a programmer can work around these new routines, they can code sloppily, they can hard code incorrect values for buffer lengths, they can ignore error codes. Nothing will ever prevent a C programmer from being able to shoot him or herself in the foot. But now there's a better toolset that can help people identify these problems before the code is released.
Actually, it's against the law in many places to carry a device designed to aid in shoplifting with the intent to use it to shoplift. The code here in Minnesota states:
609.521 POSSESSION OF SHOPLIFTING GEAR. (a) As used in this section, an "electronic article surveillance system" means any electronic device or devices that are designed to detect the unauthorized removal of marked merchandise from a store. (b) Whoever has in possession any device, gear, or instrument designed to assist in shoplifting or defeating an electronic article surveillance system with intent to use the same to shoplift and thereby commit theft may be sentenced to imprisonment for not more than three years or to payment of a fine of not more than $5,000, or both. History: 1975 c 314 s 1; 1984 c 628 art 3 s 11; 1986 c 444; 1Sp2001 c 8 art 8 s 26
This statute is used to charge people who carry aluminum-foil lined shopping bags to try to beat the detectors. It's interesting to note that the statute doesn't say the device actually *has* to be effective, only that it has to be *designed* to aid in shoplifting.
As far as I know these statutes are usually applied only to people accused of shoplifting. Intent is very hard to prove on its own. If you carry an aluminum shopping bag into a store and leave again, they probably won't blink twice. If you are caught shoplifting AND have an aluminum-lined bag, they'll probably throw both charges at you. Oh, and don't forget that shoplifters are also forced by the courts to pay restitution to the victims. A typical investigation takes from one to two weeks worth of paperwork and other processing at a cost of about $40 per hour. Plus court costs. If they hire an expert to prove this is a shoplifting device, you'll pay for that as well.
Please note: I am not a lawyer; this is not legal advice; and if you do try to use this equipment to shoplift in one of my stores I sincerely hope you get the full three-year sentence and $5,000 fine you deserve.
Yes, I'll agree that the inductive systems do indeed have a very limited range. And that it's important to know which particular technologies we're dealing with before pronouncing them "dead" or "hacked" or "unbreakable".
As far as the crypto goes, there's no reason they couldn't employ the same sort of challenge-response that satellite decryption cards use. Even with a lab setup and full access to the electrical exchange between the cards and the receivers (including the ability to interfere with the data), it took a remarkable amount of effort to break the encryption on a single card. An unreliable RF snooping wouldn't stand a chance if the cryptography were of equivalent design.
However, there are still drawbacks to RF ID cards. While you may not be able to read them, you might be able to detect them, and possibly even identify their nature. The example I frequently use with RFID tags is to picture such an antenna set up at the entrance to the Bada Bing! club from the Sopranos. In this case, imagine a reader mounted in the seat of a barstool. If a reader could determine "what's in your wallet?" without your knowing, what else could a bad guy find out? They be fronting a "man-in-the-middle scam" where a radio-connected henchman is exchanging signals, charging expensive jewelery in a store located in another city? Many things are possible if the exchange can take place without the user's involvement, even without breaking the encryption. And too many people want to steam-roller RFID forward while ignoring legitimate questions about security.
Physical contact readers are the only sure way ordinary people have to prevent surreptitious communications. While the advantages of RF are numerous (convenience, sanitation, no moving parts, no contacts to get dirty, no problems with static electricity) the removal of control from the user is a huge weakness that opens many avenues of exploitation.
I don't know if this is the case. Everyone seems to assume you can "intercept" the RFID information from many meters away. I guess I'm not sure which technology is used in credit cards, but if it's anything like ISO 14443 standard or even ISO 15693, the max distance is only going to be 1.5 meters or less.
Anyone stating "max distance" for RF is creating limits where none exist. With a correctly-sized transmitter, a sensitive enough receiver, and a large enough antenna, there's nothing preventing reading over much greater distances.
The "hacker" world distance record for reading RFID tags (not necessarily the same technology that's in these credit cards) was set at Defcon in August 2005. It was 69 feet, or over 21 meters. You can see the Make Photo Blog pictures of the gear used. While the kit may look bulky, 69 feet would allow you to have it in a van parked outside a store shooting in through the windows.
Regarding the correctly sized antenna, the WiFi shootout that year scored a record 125 miles for an unamplified 802.11 link. 125 miles from a pair of hundred-milliwatt transceivers chatting at 11 mbps.
And don't assume it's not worth the trouble, either. You don't know what dollar values may be transacted via RFID, nor what thefts may be possible with the intercepted data.
That's not to say that encryption isn't capable of rendering the data useless to an eavesdropper. We don't know if it is or isn't good encryption, but that's immaterial. Don't rely on distance alone to protect you. It won't.
I wouldn't say you were "forced" to do anything. You're perfectly free to cancel any accounts with them and open a different card at a different bank. Not that anyone at HSBC will shed a tear to see you leave; after all, you're just a "privacy kook" in their eyes.
But maybe someday us "privacy kooks" will leave in statistically significant numbers, and eventually someone might notice.
Slashdot Mirror
As far as mag tape goes, virtually every tape used in for computer backups has been used as a digital storage medium. The only analog exceptions I can think of are the audio-cassette tapes used by some of the earliest home computers (Commodore VIC-20s and C-64s, TRS-80s, Color Computers, and Apples,) which modulated the audio input to a tape recorder like a modem; and a weird VHS-based system I remember seeing ads for that stored a lot of data in the video signal. Even the 9-track reel-to-reel tapes you might see as props in old movies stored their data digitally.
Tape has a couple of things going for it. It's a well-understood medium with a long track record; and it's about as cheap as you can get. Otherwise tape is relatively slow because of the limitations of sequential access, and tapes have become bulky compared to hard drives. Tape drives themselves are also expensive; they're finicky about things like alignment and environment, and they're not exactly widely-consumer-available items. They are still workhorses found in many mainframe environments, though, so they're not quite dead yet.
Yes, today USB dongles are available for Centronics parallel as well as RS-232. But they're not available for ST-506 hard drives. If you want to get data off one of those old drives, you'll need to find an old box with an ST-506 card and somehow get the data off of it. Perhaps you could rig up a Kermit, XModem or uucp link over RS-232, or you might get lucky and find an ethernet adapter plus the correct device driver software for whichever version of DOS you're running, and find a TCP/IP stack and an FTP client. To me, that's not a plan for an archival system, that's a crapshoot.
So in 20 years, when we all have fiber-attached holographic crystal storage, and all this SATA nonsense is long behind us, do you know if you'll be able to find a SATA-to-holofiber dongle? What if absolutely every component goes wireless, and mass storage goes 802.11-DD even inside the box? Do you know if you'll be able to buy an RF-to-SATA dongle for your old drives? How about in 40 years?
My point is that if you apply what we've seen happen to standards over the last 20 years, you'll find that very, very little of it has survived. Will RS-232 remain relevant for another 20 years? In that time it's slipped from a dominant personal computer interface to a quaint 9-pin jack on the back of yesterday's PC. What will happen to USB and/or Firewire? Just because we think they're ubiquitous today is no guarantee that they will be 20 years from now.
I think the right answer is to say that there is no good long-term archival solution for computer data. There are good medium term solutions, but the long term requires an active migration schedule, meaning a data curator will be required to port the data to the "medium of the moment" every ten years or so. You can't just stick a CD in a museum exhibit and hope to still have access to a working CD-ROM drive a hundred years from now.
And as far as buses go, I looked in the wwwikipedia and found this entry for Computer bus, which has this unambiguous second sentence: "Unlike a point-to-point connection, a bus can logically connect several peripherals over the same set of wires." (And you can check the "history" page, I had nothing to do with that definition! :-) Yes, RS-232 is a collection of related electrical signals, but it's not a bus, it's a point-to-point connection specification. USB qualifies as a bus (just barely, because they put the word "logical" in their definition) because the standard specifically defines a "hub" and a means of device discrimination. Although you can buy an RS-232 "hub" (OK, so it's a 25-pole, n-throw switch,) the RS-232 standard itself has no such specification for using or controlling such a device, so it doesn't fit that category.
( And just to pick a pointless fight, I don't think RS-232 qualifies as a "bus" as it's really only an electrical interface between a single DTE and a single DCE device. "Bus" implies (to me, anyway) that it supports more than one simultaneous device, and without some kind of add-ons or non-standard protocol RS-232 doesn't fit that definition. )
Electrically, the interfaces have changed, too. You'd be very hard-pressed to find a modern machine that's capable of reading from an old Winchester drive -- the ST-506 interface is dead and buried, and the adapter cards that spoke ST-506 were made only in the era of 8-bit ISA bus machines; you won't find a PCI card that supports them. ATA came along, and has advanced to ATA-6. Now we have SATA, which begs the question of how long PATA will live. Do you want to bet your future retrieval of the data to finding an ancient machine that can read SATA on the 2026 equivalent of eBay?
There are plenty of physical reasons not to use hard drives as an archival medium, too. You'd probably be hard-pressed to find an old Winchester drive that could spin up today after sitting idle for 20 years. Drives manufactured back then suffered from stiction, which was caused by lubricants that sat idle for too long. Do you know what's wrong with the longevity of data on drives manufactured today? I don't. Will today's lubricants still flow freely in 2026? Will the platters, heads and mechanics survive the years uncorroded? Will the electrolytic capacitors still hold a charge? Will the connectors have shifted due to thermal expansion and contraction? Will the magnetic fields of some bits have dissipated due to their proximity to other bits? Will the adhesive holding the media to the platters have broken down?
And Google for "maxtor sucks" if you want to read horror stories of people losing data due to the death of a hard drive.
I'm not saying Super-8 is the way to go, but it's still possible to get the data from it. Will the same be true of floppy discs, ZIP disks, CD-Rs, CD-RWs, and all the burnable variants of DVDs including +/-, DVD-R, DVD-RW and DVD-RAM?
The other problem is everybody knows me as a geek who just knows all this stuff, so they immediately assume it's out of their ballpark to understand. But when my wife starts recommending them, that's when it really seems to hit them that it's something they could use.
You're probably right. At least until holodeck interaction becomes common.
*Firefox's spell checker suggested BluRay should be spelled "blurry". So much for HD.
He hinted at it in TFA. Social networking sites "encourage" such behavior via peer pressure. "I gotta get a hundred people on my friends list!" or "I have to have more friends than Jane just to make her jealous." And one proven way to get people to notice you is to be more shocking on your page, via nudity, stunts or whatever.
One thing popular during that time were timeshare computers. You might be able to find an old Teletype or Silent 700 terminal with an acoustically coupled modem, and have those dial into a machine hosting a few games such as Hunt the Wumpus, tic-tac-toe, or global thermonuclear war. Dumb terminals of that era can be had for pretty cheap -- a buddy of mine just picked up a Teletype 43 for $40 off of eBay (beware of shipping costs, though.)
The stars predict tomorrow you'll wake up, do a bunch of stuff, and then go back to sleep.
#11. Incredible run-on sentences that are in a difficult-to-read font and are not punctuated and sometimes written in the second person familiar and sometimes they changed tense and ended illogically disconnected from their premises even though you read them through to the end.
Actually Vuescan comes with ICC profiles for lots of different models of scanners, and if you have an IT 8 target you can create a profile calibrated to your scanner. (If I were to do this again, I'd try to calibrate it every thousand slides or so to compensate for the lamp fading over its lifetime.) Plus, Vuescan includes film profiles for every type of film and process I've ever shot (not that I have a particularly wide range, my experience being limited to the ordinary films made by Kodak, Agfa, Fujifilm and Ilford.)
And you're right, the GUI is as ugly as a dog, and is not particularly friendly either. I can't argue that point. But once I got my workflow set up, it did run me through the scanning process quite efficiently.
One other thing to consider: what's the destination of the images being scanned? If they're going to a printer, that's one thing. But if they're just going on a DVD, a precisely calibrated scanner is total overkill for an NTSC image.
I don't really have a problem with sites like AutoPatcher, because I don't care that much about Microsoft's TOS on that page. I just thought his belief that it was all on the up-and-up was hilarious. But imagine the sh!t that would hit the fan if he distributed a virus-infected patch. Lawyers, feds, trials, it would break his piggy bank forever. At least he's not charging money for the service, he's just begging for bandwidth.
As for people who use AutoPatcher, can you "swear" that you trust him? Imagine what would happen if he deliberately wrote a password-sniffing Trojan and included it in one of the autopatch images? Microsoft may be the Evil Empire, but you know they have too much to lose by inserting identity-thieving code. This guy is small, and he could decide "hey, the Mob is offering me three million bucks to put out this zombie-creating patch, I think I'll take the money and head to the Cayman Islands!"
I spent a lot more time than I had to because I scanned them all at 20MB raw image size (the jpegs averaged roughly 6MB each when I was done.) My intent was to keep good-quality archival copies of the slides. However, these large files meant that every action with them was slow: transferring them from the scanner via USB 1.2 took like 40 seconds per image, loading and saving them in an editor to rotate and crop them was slow, importing them into Arcsoft to produce the slideshow was slow, and so on.
Decide what you intend to do with the digital images first. Are you going to archive them as I did? Then accept that it will be slow. An archival quality scan of the medium format film that you describe will take hundreds of megabytes per image. But if you're just going to burn a DVD for the family and discard the scans as intermediate files, scan them at DVD resolutions and you'll save a ton of time throughout the process.
Invest in some good scanning software. The out-of-the-box stuff I got from Minolta was slower than molasses. It took it 20 seconds to autofocus each slide individually, and that was prior to the scan itself! I purchased Vuescan from Hamrick software and it sped the process considerably. They support many dozens of scanners, film profiles, etc. It automated the process of scanning a full carrier of slides. It was worth every penny to me.
Use a dust brush on each and every negative before scanning it. A cheap squeezy rubber-bulb brush will clean up most dust and hairs nicely, and they're only like $5.00.
Don't bother printing them unless you actually want the prints of the pictures.
Find a good program to help rotate and crop the images, clean up dust specks, and fix colors. I used Paint Shop Pro, and eventually got pretty fast at it. Later, I found RPhoto (freeware! on the web) that enabled me to whip through rotating and cropping at high speeds.
Figure out in advance how you want to organize the images you scan. I built a directory structure by year, and scanned the images in rough chronological order. If there is no organization to your media, be sure to take the time to tag them at some point in the process (probably the time you crop and rotate them.) Names, places and dates are all good searchable data. I used a short description for the file names, but I wish I'd edited the EXIF data when I had the chance.
Regarding medium format film, ask about flatbed scanners at a good photography shop. When I was shopping for newer film scanners, I found an Epson flatbed with a "negative attachment." It consisted of a backlight-box that had a snap-in film carrier on the bottom that would hold 2 five-frame 35mm filmstrips. You could remove the film carrier and use a larger frame to hold your negatives in place (the adjustable carriers that you use in enlargers to hold medium format film comes to mind.)
Once you figure out what you're doing, take a few minutes and write up an instruction sheet. You'll probably go stir-crazy after scanning a thousand frames, and you'll likely want to take a break for a few months. It's nice to come back to full instructions so you can pick up exactly where you left off.
Realize that this will take a lot of your time. Check with a commercial photo house and ask about their scanning rates. I was quoted from about $0.75 per slide to $1.20 per slide. Of course with over three thousand slides to scan I wasn't about to spend that kind of money, but I did spend several hundred on a Minolta Dimage film scanner, and many, many hours scanning. That's where the instruction sheet helped -- my wife picked it up and she started scanning in her free time, too! You might want to consider hiring a photo house to use a drum scanner just for your medium format slides, rather than tackle them yourself. You'll get the best quality scans
Autopatcher, on the other hand, provides the actual software, which is explicitly prohibited by the TOS you mentioned. He has this hilarious line in his FAQ:
Q: Is AutoPatcher legal?A: Yes, nwraptor once spoke to a Microsoft employee and apparently they know about us but dont care what we do! Now that's legal advice you can hang your hat on!
Games are almost as much of a status symbol as Nikes. Yes, they'll probably churn game consoles. But how many churn TV sets to keep up with those consoles?
Again, it's not true that your old software "is worthless". Your old copy of Office 95 didn't stop working when Office 97, Office 2000 or Office XP came out. They didn't become worthless. You volunteered to stop using it.
Microsoft is a corporation that lives almost entirely on churn. Think about their cash flow, and where it comes from. Sales of new products is the bulk of their money, with a relative trickle from their professional services. Microsoft.com isn't a pay-as-you-go web site. They're not like IBM who licenses mainframe software on an annual basis. Your copy of XP stopped generating them revenue the moment after you bought it; you don't pay a subscription fee for it. Same with Office. Think about Word -- what features did they add to Word to make you need to buy the latest version? I promise you that "Now with advanced Tabs and Rulers!" isn't a slogan designed to drive Office fanbois into the stores.
Microsoft is somewhat afraid of the near future because their biggest cash cow, Office, doesn't require upgrades at the same rate as their operating systems do. The only reason I upgraded my home version of Office 97 was I needed to add Powerpoint, not because I needed "adjustable margins" or whatever they had added to Word in the previous 8 years. And that's why their long term plan is .net and Vista. With Trusted Computing, they'll be able to move you to a subscription model. Just think of it: an OS that can enforce licensing. No more selling Office licenses that are good forever. With no new features, they can "give" the software away, but cripple things like printing or saving unless you pay them per month, or even on a by-usage basis. Want to create a Powerpoint slideshow? That'll be $10, please. Now there's a revenue stream to bet your future on.
Eventually Microsoft will encourage people to not run unsigned code. "Ooo, it might contain a virus, don't run it or your Windows Warranty will be voided!" How much do you think Microsoft will charge to sign a copy of Open Office or Ghostscript? And do you honestly think they'd ever sign Exact Audio Copy? Hell, they'll probably put it in their "Pirate Tools" list of binaries that will never run.
Think about the word "obsolete" for a minute. Does it mean your old console is worn out, eroded by time and usage? Did it break? Did your N64 games stop working when the Game Cube came out? Did your Game Cube stop working on the release date of the Wii? Did Super Mario Kart expire, or did Bowser refuse to come out and play? No, it's obsolete because you were the victim of successful marketing to your own greed. "Own the shiniest video game! Your old console sucks because we have a new one! Don't be the chump with last year's console!"
Nothing went wrong with your existing system, yet you replaced it on the whim of a corporation. Churn.
Mind you, my retirement fund is based in large part on people like you continuing to churn video games and the like. Feel free to continue your participation in capitalism.
Sony arrogance has lost me as a customer for life, and all the friends and family who rely on me for tech support as well. Not that they give a crap about some random guy in Minnesota who maybe goes shopping with five or six people per year, but I can list off perhaps $10,000 in sales in the last two years that specifically did not go to Sony due simply to my opinion of them. At least four cameras, three big screen TVs and a handful of cell phones adds up to a few dollars.
Maybe that's what the web needs: a list of "lost sales". Imagine an honest (ha!) tabulation of the purchases of everyone who specifically rejected Sony products because of the company. It might surprise me to see how small the list is, or it might surprise Sony to see how badly they've judged us.
But that only works for a limited set of people with enough money, and that's not even half of America any more. It is a great theory for products like games and laptops, because they're already owned by the rich half who can afford to play the churn game. But it doesn't work as well on mass market items such as TVs and DVD players. Look at how long it's taken to replace VCRs with DVD players. Most people consider a TV a "big-ticket" item, and expect them to last 20 years or more. And nobody with a non-HD TV has any reason to consider an HD-DVD or Blu-Ray player.
Churn is great for cell phones, where they can continually "upgrade" them by adding more and more crappy features, and give them away (with expensive contracts.) But churn is not going to sell HDTV sets to everyone across the nation, and HDTV is a prerequisite to selling HD players.
And it took them until Visual Studio 2005 for them to bite that last bullet. In case you haven't used it yet, they've added a set of "safer" routines to the standard C runtime library. They They are not backward compatible. Now, instead of sprintf(), you call sprintf_s(), which takes the traditional output buffer pointer plus a new parameter indicating the length of the output buffer. It also validates the format string, although if you let the users modify that you'll still have problems (and the documentation now specifically warns the developer that this is a security issue.) And they've added a new set of warnings to the compiler to let you know if you're still using the deprecated old routines. You can turn them off, of course, but Microsoft recommends you fix your code instead.
This was really a huge step. With it, they are forcing programmers to at least consider what happens when they use pointers. Sure, a programmer can work around these new routines, they can code sloppily, they can hard code incorrect values for buffer lengths, they can ignore error codes. Nothing will ever prevent a C programmer from being able to shoot him or herself in the foot. But now there's a better toolset that can help people identify these problems before the code is released.
As far as I know these statutes are usually applied only to people accused of shoplifting. Intent is very hard to prove on its own. If you carry an aluminum shopping bag into a store and leave again, they probably won't blink twice. If you are caught shoplifting AND have an aluminum-lined bag, they'll probably throw both charges at you. Oh, and don't forget that shoplifters are also forced by the courts to pay restitution to the victims. A typical investigation takes from one to two weeks worth of paperwork and other processing at a cost of about $40 per hour. Plus court costs. If they hire an expert to prove this is a shoplifting device, you'll pay for that as well.
Please note: I am not a lawyer; this is not legal advice; and if you do try to use this equipment to shoplift in one of my stores I sincerely hope you get the full three-year sentence and $5,000 fine you deserve.
That's what I said! Booty traps!
As far as the crypto goes, there's no reason they couldn't employ the same sort of challenge-response that satellite decryption cards use. Even with a lab setup and full access to the electrical exchange between the cards and the receivers (including the ability to interfere with the data), it took a remarkable amount of effort to break the encryption on a single card. An unreliable RF snooping wouldn't stand a chance if the cryptography were of equivalent design.
However, there are still drawbacks to RF ID cards. While you may not be able to read them, you might be able to detect them, and possibly even identify their nature. The example I frequently use with RFID tags is to picture such an antenna set up at the entrance to the Bada Bing! club from the Sopranos. In this case, imagine a reader mounted in the seat of a barstool. If a reader could determine "what's in your wallet?" without your knowing, what else could a bad guy find out? They be fronting a "man-in-the-middle scam" where a radio-connected henchman is exchanging signals, charging expensive jewelery in a store located in another city? Many things are possible if the exchange can take place without the user's involvement, even without breaking the encryption. And too many people want to steam-roller RFID forward while ignoring legitimate questions about security.
Physical contact readers are the only sure way ordinary people have to prevent surreptitious communications. While the advantages of RF are numerous (convenience, sanitation, no moving parts, no contacts to get dirty, no problems with static electricity) the removal of control from the user is a huge weakness that opens many avenues of exploitation.
Anyone stating "max distance" for RF is creating limits where none exist. With a correctly-sized transmitter, a sensitive enough receiver, and a large enough antenna, there's nothing preventing reading over much greater distances.
The "hacker" world distance record for reading RFID tags (not necessarily the same technology that's in these credit cards) was set at Defcon in August 2005. It was 69 feet, or over 21 meters. You can see the Make Photo Blog pictures of the gear used. While the kit may look bulky, 69 feet would allow you to have it in a van parked outside a store shooting in through the windows.
Regarding the correctly sized antenna, the WiFi shootout that year scored a record 125 miles for an unamplified 802.11 link. 125 miles from a pair of hundred-milliwatt transceivers chatting at 11 mbps.
And don't assume it's not worth the trouble, either. You don't know what dollar values may be transacted via RFID, nor what thefts may be possible with the intercepted data.
That's not to say that encryption isn't capable of rendering the data useless to an eavesdropper. We don't know if it is or isn't good encryption, but that's immaterial. Don't rely on distance alone to protect you. It won't.
But maybe someday us "privacy kooks" will leave in statistically significant numbers, and eventually someone might notice.