I heard of exactly this same phenomenon in the machine tools industry, back when a lot of places were switching from manual tools or semi-automated ones over to more fully computer controlled CNC ones.
Guys who had worked for years learning how to do complex machining tasks quickly (and if you've ever seen a skilled manual machinist working, it really is a black art sometimes), how to build jigs, etc. suddenly saw themselves being made obsolete. As a precaution, a lot of people who knew of ways to either make the set-up processes (getting an automated machine ready to run the CNC program that will make a part) or the programs themselves more efficient, either didn't share their knowledge or didn't apply it fully. Certainly it wasn't passed up the chain of command to be implemented widely -- why would somebody want to do that, when their little 'tricks' might be the only think keeping them employed?
I expect that this is probably still a problem, and I don't have any easy solutions. In the programming world, where the coding itself is -- at least to managers -- at least as much of a black art as machine operation, and there is perhaps even more danger of having your job outsourced to India (in the machine tools industry it's China), there's a strong incentive to create miles of uncommented, byzantine code.
It's not until managers understand the reasons behind this behavior and start rewarding documentation with advancement and most importantly job security, and build a relationship of trust between employees and themselves, that they'll get the best products their employees can produce.
I agree with much of what you said, but in terms of a resource, an actual person is many times better than even the best, most well-written documentation. Sometimes, if the system is complex enough, I could see the retaining of an additional person being justified, even if their only function was to act as a 'living encyclopedia' and help other (more junior) employees troubleshoot the system.
I can't count how many times I've had a problem with a well documented system, and even after wasting hours of my time (and in many cases, hours of other people's time who just have access to slightly different documentation) only to finally find someone who has intimate knowledge of the system in question, and get an answer in five minutes. There are limits to how good documentation can be, even when it's searchable, indexed, and cross referenced.
No amount of documentation in even the best information storage and retrieval system can compare with the power of a person who actually understands a system intimately, and then applies that understanding to other people's problems as needed.
I disagree, I think we're clearly on step 27, here:
A minor, rarely occurring flaw in the device begins to be discussed in the Apple support forums. Whiny, artistic types post lengthy diatribes about how this terrible design flaw has made the device unusable and scarred them emotionally. Electronic petitions are created demanding that Apple replace the devices for free, plus pay for counseling to help traumatized users overcome their emotional distress.
Just out of curiosity, how many of those books are there, actually written by Laumer? I've only ever seen the 'Annals' one in stores, and those were all used copies. I just ran a quick search and it seems there are other books in the same series but written by other authors. I thought 'Annals' was pretty good (I hope that's the right name, it's a series of short stories), but I've never wanted to get any of the other authors' stuff without knowing if they're just rip-offs that the publisher had done later on or something.
Since you're the first person I've come across who's mentioned the book(s), care to make a recommendation?
Sure a sentry gun is cool. But isn't a sentry gun in the form of a talking rabbit that much cooler? You could animate its facial expressions, and give it pithy one-liners to say as it engages targets.
Now if only we could make it walk around... nah, that's just going to end badly.
... is the names of their products. I mean "Nabaztag the WiFi Bunny"? It sounds either like a cartoon villain, or a new kind of pharmaceutical. Perhaps a failed idea for Pfizer's mascot? And "Le Pad Osmooze"... I'm going to hope that sounded better in French. The only thing "Osmooze" brings to mind is 'osmosing ooze.' What the hell was on that focus group's mind?
Anyway, the products are mildly interesting, but their applications are weak. It seems like any time a company comes up with a peripheral, the first thing they do with it is find some way for it to notify you when you have email. For God's sake stop it, there are enough email notifiers out there already. There's got to be something better you can do with a 95-euro, 23-cm tall, talking, WiFi enabled, suspiciously Pokemon-esque talking bunny.
Isn't there?
WordProcessor Recovery Possible
on
The Digital Dark Age
·
· Score: 2, Informative
I'm not sure about Wordstar, but I'm fairly certain with MS Works that if you open up that file in your favorite text editor that your data will be in there, and you should be able to just copy it out. Naturally you'll lose all the formatting, and probably have a lot of crappy unreadable characters introduced, but I'll bet you'll recover the bulk of it.
I had a similar issue once with a very nice (but very dead) word processing program that I used to use called WriteNow -- where the developer has stopped work and sales on the product but refuses to release it even in binary form to the public -- I had a bunch of disks of documents, but none of the original program disks and no computer with it installed anymore. I was pretty stuck until I just opened up the files one day and searched through them until I found the plain text. A few minutes of cleanup in BBEdit later, and I was ready to go.
Needless to say, they're all stored as.txt files now. Anyway, I wanted to say that I can definitely feel your pain with a problem like that, but that there are solutions to at least one of the problems you described.
Actually many punch cards which were hand-entered on keypunch machines have the text printed across the top of the cards in a typewriter-like human readable form. Otherwise it would be hard to detect whether an error had occurred during the process of keying the card.
I'm not sure whether this is specific just to early punch cards, and went away with later, higher-speed systems where the cards were being produced by computers at much higher rates, but I've seen a few punch cards and they had numbers at the top which corresponded to the punched holes. It was complete gibberish as far as I could tell, but it was human-readable in the most basic sense.
I'm not sure though that this goes for all punch cards. Maybe it was a feature specific to the system for which the cards I saw were for.
Re:hardware is much, ah, *harder* than software
on
The Digital Dark Age
·
· Score: 1
I think you've got a good point about the interface. I didn't really think about that when I was thinking about including the hardware, but you're right, including a IDE or ATA CD-Rom would be pretty frustrating for some reader down the line, if that format no longer existed.
I was thinking about old hardware that I own, and what I'd put in there if I had to build a time capsule. I'm pretty sure I've got a parallel port CD-Rom drive somewhere, that would probably be it. Unfortunately I don't have any documentation on the control commands for it, but it got me thinking.
What would be really nice for a time capsule would be some kind of ruggedized reading device, with a very simple interface and simple machine-readable output. I'm imagining a CD-Rom drive with a little panel (or knobs, even) where you could put in a filename and press "Go," and it would begin to spit it out a serial or parallel interface, one bit or character at a time. Maybe an 8-bit parallel interface, with additional pins for clock, logic high, low and ground, at some fairly low bitrate. Even if you didn't have any other old equipment, it'd be pretty elementary for someone probing around with an oscilloscope to figure out what was what. You could even do some pretty rudimentary pictograms for clock (square wave symbol), high (1), low (0), that would help the interpreter. Unless they've transcended electricity and binary logic altogether, they ought to be able to puzzle that out, and route it into whatever data storage equipment they'd rather use.
If all you had on the storage discs was 1-byte text code like ASCII (heck 7-bit BAUDOT would be OK too), it ought to be pretty simple for someone to do a frequency analysis and figure out what's what, even if (the horror) they've lost or forgotten ASCII. After that, figuring out our language would be up to them and the Rosetta Project people.
This is true -- I know a person like this. (If anyone cares to read my other posts, it's the same guy who made a set of Helmholtz coils that encompassed a small room.) I know for a fact he has several 8" floppy disk drives with their associated controllers, tucked away "just in case." I'm also pretty sure he has a punch card reader. Whenever anybody had data stored on some sort of archaic format, he was the go-to person on campus: if he didn't actually have a machine to read it, he probably knew somebody who did, or where you could find one.
The only problem -- and frankly the more pressing problem as far as I'm concerned than data loss due to formats becoming obsolete -- is that the people who have real-world technical expertise with older systems and formats are retiring, and in some cases dying off. It's my sincere hope (perhaps through efforts like Wikipedia,) that people who have knowledge that could be useful to future users will write it down, or that other people will try to write it down for them, before they're gone.
It's that loss of human experience and expertise -- real working nuts and bolts knowledge -- that to me seems far more serious in terms of creating a 'new dark age' than whether or not we can access our Quicken records in 25 years.
I agree. But there should be something available to store stuff that's not quite important enough by itself to warrant printing out, but might be worth storing anyway.
Also, there are those things which gain importance by being a complete record. For example (and this is a weak example, I admit) take all my email. It's far too much to print out, and it wouldn't be worth the paper anyway. However, that's not to say it's unimportant: if I could keep a complete record of every email I ever wrote, for my entire life, that repository in toto could be important or worthwhile. It's important though only because it's a complete record, and useful really only if it's searchable and indexed (or sorted). Very few messages are by themselves important, but it would be neat to have the ability 30 years from now to go through and see what was on my mind, at any given point in time.
For things like this, you really need a storage medium that's both random access and supports a high data density. To go with that, it's probably acceptable for it to have a shorter lifespan than the "time capsule" applications.
Frankly I think for video, your best bet for archival purposes is some sort of analog format if you want to store physical artifacts. Actually, if you can get film stock that's archival and not unstable like the old celluloid is, I'd use film. I'm not sure if they still make K-14 process movie film, but if they do that's the stuff you want. Not E-6, it's dye based and fades. Otherwise get some silver-halide black and white. I know that's still available.
Put all your video on film, with the audio in the visible-light modulated stripes on the side, and pack that up in your time capsule with a few projectors and a bunch of spare bulbs, and I think even 1000 years from now -- provided the film base doesn't degrade, which is a big IF -- the people who find it ought to be able to figure out how to play it back.
I suppose that solution might not appeal to geeks, but if I was really going to build a time capsule for the future, I think that's the sort of solution I'd be looking at. Simple, reliable, obvious (to comprehend on the far side).
Re:if you expect to have to reverse engineer it
on
The Digital Dark Age
·
· Score: 2, Insightful
This is a pretty good point.
If you're thinking that your data will be carried forward electronically, then there's no reason why a set of specifications or source code in a commonly-understood language (I can't imagine that any reasonable programmer of the future wouldn't be able to at least puzzle out some well-commented Pascal or C) showing how to decode your data. However, you'd have to hope that whoever is 'carrying forward' your data isn't lazy or cheap, because this would be the kind of thing that would get cut off or removed for being redundant if you started running short on storage space. It would be unfortunate, but not hard to believe, if some space-consious bureaucrat somewhere were to keep only the data and not any of the sample decoders, especially if it was in a repository with a lot of other people's data, and each one had its own set of example decoders. I guess the decision of whether to trust the archivists in question would vary with each situation.
In the case of physical media, I think it's important to bundle each set of data-containing artifacts with either an actual reader device which produces some kind of easily understood output, or schematics for same. For instance if you were going to bury a vault of CD-Rom type discs, it would make sense to put at least a CD-Rom drive in (actually in a vault, you might as well put a whole computer in too). In something smaller, at least include a decoder schematic, or at least some kind of minimal diagram. I'm thinking the bare minimum would be something like what NASA put on the top side of the Voyager probes' "Golden Records."
Actually if you want a good example of a data-storage 'time capsule,' I think the Golden Records pretty much are the 'best practice.' Engraved into metal, built to last practically forever, each one comes with a cartridge and some minimal schematic instructions. If I'm not mistaken too, the very beginning of the recording is a triangle-wave test tone, which is represented on the front, so you know you've gotten it right. I've also read that they included analog slow-scan TV images on there too, although how they expected people to puzzle those out I don't know.
Paperback popular books are especially bad: I've even heard some people say that they're made intentionally like this, so that they don't get passed around too many times before they fall apart. Thus people buy their own copies. While I'm not sure that's true (I think it's more likely just cost-cutting on the part of publishers), and I suppose future generations probably won't suffer too badly by losing all the $6.99 copies of "The Davinci Code," it is distressing when I see the supposedly 'archival' copies of important documents, or textbooks, printed on crummy paper.
It's not as if you can't buy 100% cotton rag paper anymore, it's just a lot more expensive than the Staples $1.99/ream crap that people use most of the time. Granted, there's a lot of stuff that's being printed out every day that doesn't merit saving, but when you want something to last, you need to have the right material to put it on.
Personally, I do all my personal (dead tree) correspondence with an IBM Selectric II on Strathmore Legacy 24 lb., 100% cotton, bright white, Monarch size, laid executive stationary. I bought about a thousand sheets of it from a stationary store that was going out of business (another peeve of mine there) and suspect that it'll last me the rest of my life if I use it judiciously.
Whether the people I write to decide to keep my letters or throw them away is their business. But at least I know that if they decide to keep them, they'll always be readable. If you want it to last, you can't beat carbon black pounded into cotton.
I was thinking of how you could store data that would really stand up to the test of time. History provides us some examples: things cut into stone seem to do pretty well. Paper isn't bad, providing you store it well. Animal skins, not so good. Celluloid isn't either (evidenced by the old movies and cartoons that are degrading).
However glass is really good, and while it might not have the proven track record that stone tablets to, it can also support a much higher data density. For example, Ansel Adams original glass plate negatives are in some cases just as sharp as the day they were shot, and they should stay like that for the foreseeable future providing they're well taken care of. But even they are dependent on the chemicals used in processing -- whether the silver sticks to the glass over time, etc.
So here's what I was thinking: what if you used some sort of photographic process to physically etch a pattern of bits into glass: use a fairly strong acid and get the etching pretty deep, or maybe etch the bits at the bottom of phonograph-like grooves so that light surface touching wouldn't destroy them. If you could make something like this that could be read with a regular CD Rom, that would be even better.
I think some sort of process like this is used on metal (or is it actually glass?) to make the dies for stamping CDs. Basically I'm suggesting just make and retain the masters, but don't degrade them by stamping anything.
I take a little issue about AIFF. It's basically just a container format, generally for an uncompressed 44.1kHz/16-bit audio stream of audio data.
Seems like that would be the very best format to store it in. At least someone could cut away the file headers and have a shot at decoding the audio data pretty easily -- I can't say the same for MP3 or MP4, if you've lost the decoding algorithms.
Uncompressed digital audio, stored in some kind of easily dissectable container (AIFF or WAV) seems by far to be the best way to it. After all, once you figure out the sample rate and depth, you really just need to run it through a DAC (or a software equivalent). It's essentially the same as a bitmap for graphics. And if you wanted to make the sample rate known, you could begin the recording with a blast of test tone, and note its frequency and relative volume somewhere where the finder would see it.
Realistically, in the near-term I think the best choice for audio is the plain old Red Book compliant Audio CD, just because of the market penetration and amount of time the format's been around. I'm pretty confident that as long as that form factor sticks around (12cm dia by 1.2mm thick) drives will probably have CD-DA backwards compatibility, unless there is some radical change in how the data is stored away from using lasers. I know Red Book has some error correction data over and above what uncompressed DA would, but the format is so widely known that I think the advantages of the format being well known outweigh the disadvantages that someone would have trying to 'unpack' the data by hand, bit by bit.
But as a data format only -- independent of the physical media you choose to store it on -- uncompressed formats are the answer. For text, ASCII. For graphics, Bitmaps. For sound, as close to the uncompressed digital audio as you can get, in as simple a container as possible.
I agree, I've always thought this was pretty terrible.
It's so bad that if I have the option of mounting something as a NFS volume or FTP virtual volume in the Finder, or accessing it through 'ftp' using the terminal / CLI, I'll use the CLI every time. Having to type 'put' and 'get' beats the hell out of having my whole system (or at least the Finder) lock and need to be rebooted because the internet connection got dropped, or the server went down, or any number of other things. Plus it just sucks to have to stare at the spinning beachball while it connects to a slow server and wonder whether it's going to connect or keep spinning forever.
Actually, I've found the best and easiest way is just to use 'rsync -avz' to mirror a folder between my server and workstation, and make the modifications to the local volume and then re-sync them when I'm done. I keep the rsync command as a textclipping on my desktop, and when I'm ready to leave at night I just open a terminal window, drop the text clipping into it, hit enter, type my password, and away it goes. No interaction is needed on my part, and the connection shuts down when it's finished.
Plus, if you're using 10.4 and you use rsync between two Macs, you can use the -E switch to copy the resource forks as well as data. Obviously this errors if you're rsync'ing to anything other than another Mac (in which case you lose the resource forks).
I'm not necessarily saying you're wrong about Skype making security a selling point, but I do think that there's a difference between a company's advertising and marketing rhetoric, and what people actually use it for. I don't know anyone who actually uses Skype for "security." I'm sure there are some people out there, but I'm willing to bet it's pretty rare. And those people are dumb.
Most people use Skype because it's a lot cheaper than the regular phone company, and doesn't require a monthly service fee like "real" VOIP service like Vonage does. It's a great way to talk to friends across the country, and a lot better implemented than the various IM services' voice chat features. It also seems to work its way through firewalls and NAT routers better than the IM service voice protocols do, too, although I can't figure out why.
Skype may hype its own allegedly secure protocol, but in reality I doubt it's a selling point. What sells is the price -- free or at least very cheap -- and the fact that it just works out of the box without any more setup than AIM. To most users the security is an afterthought, and I doubt that if it disappeared tomorrow that many of them would really care.
If the perceived 'secureness' of Skype does anything at all for most users, it just serves to counteract the general vague uneasiness that many people have with security over the internet in general. It provides them a sense that their internet phone is now as secure as their conventional phone, without really understanding how secure or insecure that really is. (Given that many people's conventional phones involve a low power, unencrypted FM radio transmitter, so that any idiot in the neighborhood with a RadioShack scanner can listen in, this isn't saying much.) In short, it's like a kind of "wired equivalent privacy" for internet telephony.
If a government, large corporation, big NGO (or really anyone with an IT department) was using Skype for secure communications, they're insane. There are real 'secure' alternatives out there, sure none that are easy to use probably, but they offer real protection. Security shouldn't be a selling point for Skype to anyone with half a brain or a job to defend. In that, I think we are in absolute agreement.
I believe Phil Zimmermann is doing you one better. (He's the guy who did PGPhone, back in the day.) His zPhone project is an end to end encryption system for IP telephony, using the RTP or SIP protocols. According to the site, it will work in unencrypted mode with a regular device, and do transparent encryption with another zPhone-capable one.
So if it actually materializes -- and I think it will, Zimmermann has pretty much always delivered the goods to the community in the past -- it'll be a whole lot better than just an update of PGPhone. And the source is going to be open for community review, unlike some past versions of PGP when it was owned by NAI.
As a sidenote, they're currently looking for a better name for the final product than 'zPhone.' The winner gets recognition, lifetime licenses for themselves and 10 friends, and their PGP key signed by Zimmermann. Pretty sweet deal.
You used to be able to buy (or if you're up for some soldering, build) an outboard box that would filter the Macrovision signal off of a composite video stream -- basically an AGC in a box. Then you could run it into a regular modern VCR.
I guess those boxes are probably illegal now by the DMCA, although they were never actually marketed for the purposes of removing Macrovision per se, it was always "video clarification." I'm sure if you google you can probably find something on them.
Alternately you can try and get an old, pre-Macrovision VCR, of which there are many but they generally are mono and you'll have problems finding replacements for the consumable parts. Also, there are a lot of professional VCRs that you can find on eBay which have AGC circuits and are thus immune to Macrovision.
Also, at least the last time I read up on the subject, Macrovision only exists on the composite video circuits; an S-VHS recording on the S-Video in or a Beta SP machine (if you can afford one) on the component in.
I thought I'd point out that something like this was actually mentioned on NPR (yesterday morning I think, around 10:30 or 11am, on the station here in Washington DC). Some pundit or analyst they had hauled in was engaging in some wild speculation about Google buying AOL and what that could mean for Google/Microsoft relations and the computer-using world in general.
They were basically asking -- and I'm pretty sure he actually used the word 'kill' -- "Is AOL the key to Google Killing Microsoft?" As soon as I saw this article, I thought it was ironic that a day later, Slashdot asks "Is AOL the key to Microsoft Killing Google?"
Obviously it's a little farfetched to think that Google could "kill" Microsoft, with or without AOL (despite how wonderful that would be, ain't gonna happen). But they did have a pretty good point about how Google would be one of the few companies that could buy up AOL's userbase (which really is all AOL has to offer to a potential buyer) and actually do something interesting with it.
Basically the analyst's opinion seemed to be that AOL is currently dying -- it's losing some thousands of customers a month -- and there aren't too many companies that would want to have anything to do with them. Microsoft is one, Google is another. But while Google seems to have a lot of creative ideas and a (admittedly very short) record of success, Microsoft doesn't seem to be doing much new and different with the internet lately, and probably wouldn't be able to retain AOL's customers any better than they're currently doing.
Actually, I think he makes a good point. Perhaps not about the canteen, but definitely about the radio.
In a combat situation, to an infantryman, a radio is probably the most powerful weapon he'll ever carry or employ. With a rifle or any other organic weapon at the platoon level you can certainly do some damage to an enemy, but it's nothing compared to the hurt you can put on the other guys by calling in an artillery barrage, or close air support.
And that doesn't even account for the damaging strategic or tactical effects that the information itself may have. If you see enemy where you didn't know they were, or more critically don't see enemy where you think you ought to, and that information gets passed up the chain of command and becomes part of a tactical decision, it can influence a lot more than a few rounds out of an M-16 is going to.
Also, you'd better believe that if you're walking around in a combat zone and talking on a radio that you might get shot. Maybe not outright, if you're not otherwise threatening than you'll probably just end up answering a lot of very pointed questions at gunpoint. But unless you know with certainty who the person is and who they're talking to on the other end of the radio, it damn well is a threat.
Or imagine you were a commander about to begin an ambush, and you had the option of which enemy you wanted to open the ambush by shooting first, you'd better bet that it's going to be the guy carrying the radio.
I can see the point as to whether these things out to be called "weapons," or whether some other term should be applied. In the military, they fall under the vague general category of 'force multipliers,' things that allow force and weapons to be used more effectively. However their importance, whatever we label them, should not be understated, as they are more than anything else the keys of modern warfare.
So, pardon me for intruding on the argument, but what's the point in this context?
If we use that broad a definition of 'weapon,' then it's ridiculous to attempt to bar them from space (or any other realm), unless we want to competely de-orbit everything we've ever put up there.
Communications, navigation, even weather satellites have huge military uses. They're force multipliers: they don't change the actual balance of troops on the ground, but they might make one side a lot more effective than the other.
In fact there probably isn't any major technological development in recent history -- perhaps in all history -- that was without a military application, directly or indirectly.
The ban on 'space based weapons' makes little sense given a broad definition and understanding of what weapons are. It only applies, and IMO was designed to be applied, to those weapons which actually exert some sort of force directly on an enemy. Given that we've had imaging satellites in orbit for nearly as long as the capability has been available in order to spy on other countries for military and strategic political gain, and nobody has really expressed a large problem with this (at least no one who's opinion, on the global stage, matters), I think it's safe to say that the queasiness with space-based weapons refers only to a narrow subset of war-fighting technologies. After all, the GPS system was undeniably a military technology to begin with, and it is more a weapon than a spy satellite is (after all, the GPS system is used to automatically guide cruise missiles to their targets), but it has a host of legitimate uses and is rapidly becoming an essential component of global commerce.
Personally, I think the ban on space-based weapons is sort of a noble though, but ultimately naive and doomed. Human beings have taken war to every other realm we've ever explored, from the upper atmosphere to the deep sea; war is not so much a science in itself as a sort of 'meta science,' a combination of all other areas of endeavor. To arbitrarily say that such a vast frontier is off-limits to what is perhaps our oldest past time and obsession -- killing each other -- is laughable. When we progress off this planet, there is no doubt in my mind at all that we will bring our weapons, in physical form or as knowledge, with us.
The anti-space weapons people appear to me like a parent trying to shelter their child from the realities of the world. Noble, perhaps, but ultimately doomed. Still, that's not to say that they should give up, but they also need to realize that they have in large part already lost, and that their fight is hopeless in the long run.
Actually, if you really want to drive Amazon's preference engine mad, just let several people with wildly different tastes use the same account. Or buy some Christmas gifts for friends while you're also buying books for yourself. You get some strange recommendations from then on.
This really can't be blamed on the preference engine though, since it's just a form of the old Garbage In, Garbage Out principle. The preference engine is mostly filled with data from single-consumer accounts. If you then go and create an account and let four people use it, the recommendations won't work very well, because it's not a case that the engine is designed to deal with very elegantly. So you get all sorts of strange predictions, or ones which can only possibly appeal to one person's tastes. (In other words, you don't get any help in finding a book or movie that the whole family would like -- instead you get recommendations that are obviously person-specific.)
More generally, my point is that I'm not really concerned about these preference engine type devices because they're not as generalizable. They only make predictions based on the data that's being put into them by other people, therefore they're only as 'good' as people's situations are similar. So if people get similar predictions and as a result buy the same things from Amazon, it's really only a reflection of underlying similarities in tastes. They are a symptom of groupthink, not a cause.
Personally, I like Amazon's system. I've never bought anything based on a recommendation, so I don't know how successful a business tool it is, but I get a kick out of navigating through their site sometimes and following the links, seeing what comes up. What I have found useful though are the links which show actual purchase data, for example while looking at a WiFi router it says "35% of people who looked at this item eventually went on to buy *this one*"... this is a great tool to look for products that are really popular (which translates usually into a wide user support base) or that suck terribly and have better alternatives.
Given that it's 3:11am, I hope this isn't completely incoherent.:)
Slashdot Mirror
I heard of exactly this same phenomenon in the machine tools industry, back when a lot of places were switching from manual tools or semi-automated ones over to more fully computer controlled CNC ones.
Guys who had worked for years learning how to do complex machining tasks quickly (and if you've ever seen a skilled manual machinist working, it really is a black art sometimes), how to build jigs, etc. suddenly saw themselves being made obsolete. As a precaution, a lot of people who knew of ways to either make the set-up processes (getting an automated machine ready to run the CNC program that will make a part) or the programs themselves more efficient, either didn't share their knowledge or didn't apply it fully. Certainly it wasn't passed up the chain of command to be implemented widely -- why would somebody want to do that, when their little 'tricks' might be the only think keeping them employed?
I expect that this is probably still a problem, and I don't have any easy solutions. In the programming world, where the coding itself is -- at least to managers -- at least as much of a black art as machine operation, and there is perhaps even more danger of having your job outsourced to India (in the machine tools industry it's China), there's a strong incentive to create miles of uncommented, byzantine code.
It's not until managers understand the reasons behind this behavior and start rewarding documentation with advancement and most importantly job security, and build a relationship of trust between employees and themselves, that they'll get the best products their employees can produce.
I agree with much of what you said, but in terms of a resource, an actual person is many times better than even the best, most well-written documentation. Sometimes, if the system is complex enough, I could see the retaining of an additional person being justified, even if their only function was to act as a 'living encyclopedia' and help other (more junior) employees troubleshoot the system.
I can't count how many times I've had a problem with a well documented system, and even after wasting hours of my time (and in many cases, hours of other people's time who just have access to slightly different documentation) only to finally find someone who has intimate knowledge of the system in question, and get an answer in five minutes. There are limits to how good documentation can be, even when it's searchable, indexed, and cross referenced.
No amount of documentation in even the best information storage and retrieval system can compare with the power of a person who actually understands a system intimately, and then applies that understanding to other people's problems as needed.
Just out of curiosity, how many of those books are there, actually written by Laumer? I've only ever seen the 'Annals' one in stores, and those were all used copies. I just ran a quick search and it seems there are other books in the same series but written by other authors. I thought 'Annals' was pretty good (I hope that's the right name, it's a series of short stories), but I've never wanted to get any of the other authors' stuff without knowing if they're just rip-offs that the publisher had done later on or something.
Since you're the first person I've come across who's mentioned the book(s), care to make a recommendation?
So ... basically ... like this guy?
Why stop at just a camera, when you could give it a camera and a gun?
/ 23/015205&tid=216&tid=159&tid=222&tid=137
http://hardware.slashdot.org/article.pl?sid=05/09
Sure a sentry gun is cool. But isn't a sentry gun in the form of a talking rabbit that much cooler? You could animate its facial expressions, and give it pithy one-liners to say as it engages targets.
Now if only we could make it walk around... nah, that's just going to end badly.
... is the names of their products. I mean "Nabaztag the WiFi Bunny"? It sounds either like a cartoon villain, or a new kind of pharmaceutical. Perhaps a failed idea for Pfizer's mascot? And "Le Pad Osmooze" ... I'm going to hope that sounded better in French. The only thing "Osmooze" brings to mind is 'osmosing ooze.' What the hell was on that focus group's mind?
Anyway, the products are mildly interesting, but their applications are weak. It seems like any time a company comes up with a peripheral, the first thing they do with it is find some way for it to notify you when you have email. For God's sake stop it, there are enough email notifiers out there already. There's got to be something better you can do with a 95-euro, 23-cm tall, talking, WiFi enabled, suspiciously Pokemon-esque talking bunny.
Isn't there?
I'm not sure about Wordstar, but I'm fairly certain with MS Works that if you open up that file in your favorite text editor that your data will be in there, and you should be able to just copy it out. Naturally you'll lose all the formatting, and probably have a lot of crappy unreadable characters introduced, but I'll bet you'll recover the bulk of it.
.txt files now. Anyway, I wanted to say that I can definitely feel your pain with a problem like that, but that there are solutions to at least one of the problems you described.
I had a similar issue once with a very nice (but very dead) word processing program that I used to use called WriteNow -- where the developer has stopped work and sales on the product but refuses to release it even in binary form to the public -- I had a bunch of disks of documents, but none of the original program disks and no computer with it installed anymore. I was pretty stuck until I just opened up the files one day and searched through them until I found the plain text. A few minutes of cleanup in BBEdit later, and I was ready to go.
Needless to say, they're all stored as
Actually many punch cards which were hand-entered on keypunch machines have the text printed across the top of the cards in a typewriter-like human readable form. Otherwise it would be hard to detect whether an error had occurred during the process of keying the card.
I'm not sure whether this is specific just to early punch cards, and went away with later, higher-speed systems where the cards were being produced by computers at much higher rates, but I've seen a few punch cards and they had numbers at the top which corresponded to the punched holes. It was complete gibberish as far as I could tell, but it was human-readable in the most basic sense.
I'm not sure though that this goes for all punch cards. Maybe it was a feature specific to the system for which the cards I saw were for.
I think you've got a good point about the interface. I didn't really think about that when I was thinking about including the hardware, but you're right, including a IDE or ATA CD-Rom would be pretty frustrating for some reader down the line, if that format no longer existed.
I was thinking about old hardware that I own, and what I'd put in there if I had to build a time capsule. I'm pretty sure I've got a parallel port CD-Rom drive somewhere, that would probably be it. Unfortunately I don't have any documentation on the control commands for it, but it got me thinking.
What would be really nice for a time capsule would be some kind of ruggedized reading device, with a very simple interface and simple machine-readable output. I'm imagining a CD-Rom drive with a little panel (or knobs, even) where you could put in a filename and press "Go," and it would begin to spit it out a serial or parallel interface, one bit or character at a time. Maybe an 8-bit parallel interface, with additional pins for clock, logic high, low and ground, at some fairly low bitrate. Even if you didn't have any other old equipment, it'd be pretty elementary for someone probing around with an oscilloscope to figure out what was what. You could even do some pretty rudimentary pictograms for clock (square wave symbol), high (1), low (0), that would help the interpreter. Unless they've transcended electricity and binary logic altogether, they ought to be able to puzzle that out, and route it into whatever data storage equipment they'd rather use.
If all you had on the storage discs was 1-byte text code like ASCII (heck 7-bit BAUDOT would be OK too), it ought to be pretty simple for someone to do a frequency analysis and figure out what's what, even if (the horror) they've lost or forgotten ASCII. After that, figuring out our language would be up to them and the Rosetta Project people.
This is true -- I know a person like this. (If anyone cares to read my other posts, it's the same guy who made a set of Helmholtz coils that encompassed a small room.) I know for a fact he has several 8" floppy disk drives with their associated controllers, tucked away "just in case." I'm also pretty sure he has a punch card reader. Whenever anybody had data stored on some sort of archaic format, he was the go-to person on campus: if he didn't actually have a machine to read it, he probably knew somebody who did, or where you could find one.
The only problem -- and frankly the more pressing problem as far as I'm concerned than data loss due to formats becoming obsolete -- is that the people who have real-world technical expertise with older systems and formats are retiring, and in some cases dying off. It's my sincere hope (perhaps through efforts like Wikipedia,) that people who have knowledge that could be useful to future users will write it down, or that other people will try to write it down for them, before they're gone.
It's that loss of human experience and expertise -- real working nuts and bolts knowledge -- that to me seems far more serious in terms of creating a 'new dark age' than whether or not we can access our Quicken records in 25 years.
I agree. But there should be something available to store stuff that's not quite important enough by itself to warrant printing out, but might be worth storing anyway.
Also, there are those things which gain importance by being a complete record. For example (and this is a weak example, I admit) take all my email. It's far too much to print out, and it wouldn't be worth the paper anyway. However, that's not to say it's unimportant: if I could keep a complete record of every email I ever wrote, for my entire life, that repository in toto could be important or worthwhile. It's important though only because it's a complete record, and useful really only if it's searchable and indexed (or sorted). Very few messages are by themselves important, but it would be neat to have the ability 30 years from now to go through and see what was on my mind, at any given point in time.
For things like this, you really need a storage medium that's both random access and supports a high data density. To go with that, it's probably acceptable for it to have a shorter lifespan than the "time capsule" applications.
Frankly I think for video, your best bet for archival purposes is some sort of analog format if you want to store physical artifacts. Actually, if you can get film stock that's archival and not unstable like the old celluloid is, I'd use film. I'm not sure if they still make K-14 process movie film, but if they do that's the stuff you want. Not E-6, it's dye based and fades. Otherwise get some silver-halide black and white. I know that's still available.
Put all your video on film, with the audio in the visible-light modulated stripes on the side, and pack that up in your time capsule with a few projectors and a bunch of spare bulbs, and I think even 1000 years from now -- provided the film base doesn't degrade, which is a big IF -- the people who find it ought to be able to figure out how to play it back.
I suppose that solution might not appeal to geeks, but if I was really going to build a time capsule for the future, I think that's the sort of solution I'd be looking at. Simple, reliable, obvious (to comprehend on the far side).
This is a pretty good point.
If you're thinking that your data will be carried forward electronically, then there's no reason why a set of specifications or source code in a commonly-understood language (I can't imagine that any reasonable programmer of the future wouldn't be able to at least puzzle out some well-commented Pascal or C) showing how to decode your data. However, you'd have to hope that whoever is 'carrying forward' your data isn't lazy or cheap, because this would be the kind of thing that would get cut off or removed for being redundant if you started running short on storage space. It would be unfortunate, but not hard to believe, if some space-consious bureaucrat somewhere were to keep only the data and not any of the sample decoders, especially if it was in a repository with a lot of other people's data, and each one had its own set of example decoders. I guess the decision of whether to trust the archivists in question would vary with each situation.
In the case of physical media, I think it's important to bundle each set of data-containing artifacts with either an actual reader device which produces some kind of easily understood output, or schematics for same. For instance if you were going to bury a vault of CD-Rom type discs, it would make sense to put at least a CD-Rom drive in (actually in a vault, you might as well put a whole computer in too). In something smaller, at least include a decoder schematic, or at least some kind of minimal diagram. I'm thinking the bare minimum would be something like what NASA put on the top side of the Voyager probes' "Golden Records."
Actually if you want a good example of a data-storage 'time capsule,' I think the Golden Records pretty much are the 'best practice.' Engraved into metal, built to last practically forever, each one comes with a cartridge and some minimal schematic instructions. If I'm not mistaken too, the very beginning of the recording is a triangle-wave test tone, which is represented on the front, so you know you've gotten it right. I've also read that they included analog slow-scan TV images on there too, although how they expected people to puzzle those out I don't know.
You're absolutely right.
Paperback popular books are especially bad: I've even heard some people say that they're made intentionally like this, so that they don't get passed around too many times before they fall apart. Thus people buy their own copies. While I'm not sure that's true (I think it's more likely just cost-cutting on the part of publishers), and I suppose future generations probably won't suffer too badly by losing all the $6.99 copies of "The Davinci Code," it is distressing when I see the supposedly 'archival' copies of important documents, or textbooks, printed on crummy paper.
It's not as if you can't buy 100% cotton rag paper anymore, it's just a lot more expensive than the Staples $1.99/ream crap that people use most of the time. Granted, there's a lot of stuff that's being printed out every day that doesn't merit saving, but when you want something to last, you need to have the right material to put it on.
Personally, I do all my personal (dead tree) correspondence with an IBM Selectric II on Strathmore Legacy 24 lb., 100% cotton, bright white, Monarch size, laid executive stationary. I bought about a thousand sheets of it from a stationary store that was going out of business (another peeve of mine there) and suspect that it'll last me the rest of my life if I use it judiciously.
Whether the people I write to decide to keep my letters or throw them away is their business. But at least I know that if they decide to keep them, they'll always be readable. If you want it to last, you can't beat carbon black pounded into cotton.
I was thinking of how you could store data that would really stand up to the test of time. History provides us some examples: things cut into stone seem to do pretty well. Paper isn't bad, providing you store it well. Animal skins, not so good. Celluloid isn't either (evidenced by the old movies and cartoons that are degrading).
However glass is really good, and while it might not have the proven track record that stone tablets to, it can also support a much higher data density. For example, Ansel Adams original glass plate negatives are in some cases just as sharp as the day they were shot, and they should stay like that for the foreseeable future providing they're well taken care of. But even they are dependent on the chemicals used in processing -- whether the silver sticks to the glass over time, etc.
So here's what I was thinking: what if you used some sort of photographic process to physically etch a pattern of bits into glass: use a fairly strong acid and get the etching pretty deep, or maybe etch the bits at the bottom of phonograph-like grooves so that light surface touching wouldn't destroy them. If you could make something like this that could be read with a regular CD Rom, that would be even better.
I think some sort of process like this is used on metal (or is it actually glass?) to make the dies for stamping CDs. Basically I'm suggesting just make and retain the masters, but don't degrade them by stamping anything.
I take a little issue about AIFF. It's basically just a container format, generally for an uncompressed 44.1kHz/16-bit audio stream of audio data.
Seems like that would be the very best format to store it in. At least someone could cut away the file headers and have a shot at decoding the audio data pretty easily -- I can't say the same for MP3 or MP4, if you've lost the decoding algorithms.
Uncompressed digital audio, stored in some kind of easily dissectable container (AIFF or WAV) seems by far to be the best way to it. After all, once you figure out the sample rate and depth, you really just need to run it through a DAC (or a software equivalent). It's essentially the same as a bitmap for graphics. And if you wanted to make the sample rate known, you could begin the recording with a blast of test tone, and note its frequency and relative volume somewhere where the finder would see it.
Realistically, in the near-term I think the best choice for audio is the plain old Red Book compliant Audio CD, just because of the market penetration and amount of time the format's been around. I'm pretty confident that as long as that form factor sticks around (12cm dia by 1.2mm thick) drives will probably have CD-DA backwards compatibility, unless there is some radical change in how the data is stored away from using lasers. I know Red Book has some error correction data over and above what uncompressed DA would, but the format is so widely known that I think the advantages of the format being well known outweigh the disadvantages that someone would have trying to 'unpack' the data by hand, bit by bit.
But as a data format only -- independent of the physical media you choose to store it on -- uncompressed formats are the answer. For text, ASCII. For graphics, Bitmaps. For sound, as close to the uncompressed digital audio as you can get, in as simple a container as possible.
I agree, I've always thought this was pretty terrible.
It's so bad that if I have the option of mounting something as a NFS volume or FTP virtual volume in the Finder, or accessing it through 'ftp' using the terminal / CLI, I'll use the CLI every time. Having to type 'put' and 'get' beats the hell out of having my whole system (or at least the Finder) lock and need to be rebooted because the internet connection got dropped, or the server went down, or any number of other things. Plus it just sucks to have to stare at the spinning beachball while it connects to a slow server and wonder whether it's going to connect or keep spinning forever.
Actually, I've found the best and easiest way is just to use 'rsync -avz' to mirror a folder between my server and workstation, and make the modifications to the local volume and then re-sync them when I'm done. I keep the rsync command as a textclipping on my desktop, and when I'm ready to leave at night I just open a terminal window, drop the text clipping into it, hit enter, type my password, and away it goes. No interaction is needed on my part, and the connection shuts down when it's finished.
Plus, if you're using 10.4 and you use rsync between two Macs, you can use the -E switch to copy the resource forks as well as data. Obviously this errors if you're rsync'ing to anything other than another Mac (in which case you lose the resource forks).
I'm not necessarily saying you're wrong about Skype making security a selling point, but I do think that there's a difference between a company's advertising and marketing rhetoric, and what people actually use it for. I don't know anyone who actually uses Skype for "security." I'm sure there are some people out there, but I'm willing to bet it's pretty rare. And those people are dumb.
Most people use Skype because it's a lot cheaper than the regular phone company, and doesn't require a monthly service fee like "real" VOIP service like Vonage does. It's a great way to talk to friends across the country, and a lot better implemented than the various IM services' voice chat features. It also seems to work its way through firewalls and NAT routers better than the IM service voice protocols do, too, although I can't figure out why.
Skype may hype its own allegedly secure protocol, but in reality I doubt it's a selling point. What sells is the price -- free or at least very cheap -- and the fact that it just works out of the box without any more setup than AIM. To most users the security is an afterthought, and I doubt that if it disappeared tomorrow that many of them would really care.
If the perceived 'secureness' of Skype does anything at all for most users, it just serves to counteract the general vague uneasiness that many people have with security over the internet in general. It provides them a sense that their internet phone is now as secure as their conventional phone, without really understanding how secure or insecure that really is. (Given that many people's conventional phones involve a low power, unencrypted FM radio transmitter, so that any idiot in the neighborhood with a RadioShack scanner can listen in, this isn't saying much.) In short, it's like a kind of "wired equivalent privacy" for internet telephony.
If a government, large corporation, big NGO (or really anyone with an IT department) was using Skype for secure communications, they're insane. There are real 'secure' alternatives out there, sure none that are easy to use probably, but they offer real protection. Security shouldn't be a selling point for Skype to anyone with half a brain or a job to defend. In that, I think we are in absolute agreement.
I believe Phil Zimmermann is doing you one better. (He's the guy who did PGPhone, back in the day.) His zPhone project is an end to end encryption system for IP telephony, using the RTP or SIP protocols. According to the site, it will work in unencrypted mode with a regular device, and do transparent encryption with another zPhone-capable one.
So if it actually materializes -- and I think it will, Zimmermann has pretty much always delivered the goods to the community in the past -- it'll be a whole lot better than just an update of PGPhone. And the source is going to be open for community review, unlike some past versions of PGP when it was owned by NAI.
As a sidenote, they're currently looking for a better name for the final product than 'zPhone.' The winner gets recognition, lifetime licenses for themselves and 10 friends, and their PGP key signed by Zimmermann. Pretty sweet deal.
You used to be able to buy (or if you're up for some soldering, build) an outboard box that would filter the Macrovision signal off of a composite video stream -- basically an AGC in a box. Then you could run it into a regular modern VCR.
I guess those boxes are probably illegal now by the DMCA, although they were never actually marketed for the purposes of removing Macrovision per se, it was always "video clarification." I'm sure if you google you can probably find something on them.
Alternately you can try and get an old, pre-Macrovision VCR, of which there are many but they generally are mono and you'll have problems finding replacements for the consumable parts. Also, there are a lot of professional VCRs that you can find on eBay which have AGC circuits and are thus immune to Macrovision.
Also, at least the last time I read up on the subject, Macrovision only exists on the composite video circuits; an S-VHS recording on the S-Video in or a Beta SP machine (if you can afford one) on the component in.
I thought I'd point out that something like this was actually mentioned on NPR (yesterday morning I think, around 10:30 or 11am, on the station here in Washington DC). Some pundit or analyst they had hauled in was engaging in some wild speculation about Google buying AOL and what that could mean for Google/Microsoft relations and the computer-using world in general.
They were basically asking -- and I'm pretty sure he actually used the word 'kill' -- "Is AOL the key to Google Killing Microsoft?" As soon as I saw this article, I thought it was ironic that a day later, Slashdot asks "Is AOL the key to Microsoft Killing Google?"
Obviously it's a little farfetched to think that Google could "kill" Microsoft, with or without AOL (despite how wonderful that would be, ain't gonna happen). But they did have a pretty good point about how Google would be one of the few companies that could buy up AOL's userbase (which really is all AOL has to offer to a potential buyer) and actually do something interesting with it.
Basically the analyst's opinion seemed to be that AOL is currently dying -- it's losing some thousands of customers a month -- and there aren't too many companies that would want to have anything to do with them. Microsoft is one, Google is another. But while Google seems to have a lot of creative ideas and a (admittedly very short) record of success, Microsoft doesn't seem to be doing much new and different with the internet lately, and probably wouldn't be able to retain AOL's customers any better than they're currently doing.
Actually, I think he makes a good point. Perhaps not about the canteen, but definitely about the radio.
In a combat situation, to an infantryman, a radio is probably the most powerful weapon he'll ever carry or employ. With a rifle or any other organic weapon at the platoon level you can certainly do some damage to an enemy, but it's nothing compared to the hurt you can put on the other guys by calling in an artillery barrage, or close air support.
And that doesn't even account for the damaging strategic or tactical effects that the information itself may have. If you see enemy where you didn't know they were, or more critically don't see enemy where you think you ought to, and that information gets passed up the chain of command and becomes part of a tactical decision, it can influence a lot more than a few rounds out of an M-16 is going to.
Also, you'd better believe that if you're walking around in a combat zone and talking on a radio that you might get shot. Maybe not outright, if you're not otherwise threatening than you'll probably just end up answering a lot of very pointed questions at gunpoint. But unless you know with certainty who the person is and who they're talking to on the other end of the radio, it damn well is a threat.
Or imagine you were a commander about to begin an ambush, and you had the option of which enemy you wanted to open the ambush by shooting first, you'd better bet that it's going to be the guy carrying the radio.
I can see the point as to whether these things out to be called "weapons," or whether some other term should be applied. In the military, they fall under the vague general category of 'force multipliers,' things that allow force and weapons to be used more effectively. However their importance, whatever we label them, should not be understated, as they are more than anything else the keys of modern warfare.
So, pardon me for intruding on the argument, but what's the point in this context?
If we use that broad a definition of 'weapon,' then it's ridiculous to attempt to bar them from space (or any other realm), unless we want to competely de-orbit everything we've ever put up there.
Communications, navigation, even weather satellites have huge military uses. They're force multipliers: they don't change the actual balance of troops on the ground, but they might make one side a lot more effective than the other.
In fact there probably isn't any major technological development in recent history -- perhaps in all history -- that was without a military application, directly or indirectly.
The ban on 'space based weapons' makes little sense given a broad definition and understanding of what weapons are. It only applies, and IMO was designed to be applied, to those weapons which actually exert some sort of force directly on an enemy. Given that we've had imaging satellites in orbit for nearly as long as the capability has been available in order to spy on other countries for military and strategic political gain, and nobody has really expressed a large problem with this (at least no one who's opinion, on the global stage, matters), I think it's safe to say that the queasiness with space-based weapons refers only to a narrow subset of war-fighting technologies. After all, the GPS system was undeniably a military technology to begin with, and it is more a weapon than a spy satellite is (after all, the GPS system is used to automatically guide cruise missiles to their targets), but it has a host of legitimate uses and is rapidly becoming an essential component of global commerce.
Personally, I think the ban on space-based weapons is sort of a noble though, but ultimately naive and doomed. Human beings have taken war to every other realm we've ever explored, from the upper atmosphere to the deep sea; war is not so much a science in itself as a sort of 'meta science,' a combination of all other areas of endeavor. To arbitrarily say that such a vast frontier is off-limits to what is perhaps our oldest past time and obsession -- killing each other -- is laughable. When we progress off this planet, there is no doubt in my mind at all that we will bring our weapons, in physical form or as knowledge, with us.
The anti-space weapons people appear to me like a parent trying to shelter their child from the realities of the world. Noble, perhaps, but ultimately doomed. Still, that's not to say that they should give up, but they also need to realize that they have in large part already lost, and that their fight is hopeless in the long run.
Actually, if you really want to drive Amazon's preference engine mad, just let several people with wildly different tastes use the same account. Or buy some Christmas gifts for friends while you're also buying books for yourself. You get some strange recommendations from then on.
... this is a great tool to look for products that are really popular (which translates usually into a wide user support base) or that suck terribly and have better alternatives.
:)
This really can't be blamed on the preference engine though, since it's just a form of the old Garbage In, Garbage Out principle. The preference engine is mostly filled with data from single-consumer accounts. If you then go and create an account and let four people use it, the recommendations won't work very well, because it's not a case that the engine is designed to deal with very elegantly. So you get all sorts of strange predictions, or ones which can only possibly appeal to one person's tastes. (In other words, you don't get any help in finding a book or movie that the whole family would like -- instead you get recommendations that are obviously person-specific.)
More generally, my point is that I'm not really concerned about these preference engine type devices because they're not as generalizable. They only make predictions based on the data that's being put into them by other people, therefore they're only as 'good' as people's situations are similar. So if people get similar predictions and as a result buy the same things from Amazon, it's really only a reflection of underlying similarities in tastes. They are a symptom of groupthink, not a cause.
Personally, I like Amazon's system. I've never bought anything based on a recommendation, so I don't know how successful a business tool it is, but I get a kick out of navigating through their site sometimes and following the links, seeing what comes up. What I have found useful though are the links which show actual purchase data, for example while looking at a WiFi router it says "35% of people who looked at this item eventually went on to buy *this one*"
Given that it's 3:11am, I hope this isn't completely incoherent.