March 29, 1996, Afternoon
KAREN: Thank you, Connie. The other members of this group besides myself and Walter Henry are Steve Gong, Steve Seid, Margaret Norton, Joan Lefkowitz and Helene Whitson. I'd now like to introduce Paul Messier. Paul is going to be presenting on behalf of the working group on Changes in Technology and Practice. Paul is an art conservator, specializing in the conservation of photographic materials. He's conducted a lot of advanced training and research, and teaching in the area, and is a founding member of the Boston Art Conservation, a partnership of professional conservators which provides conservation services in the Boston area and throughout the country. Paul, also, if you could introduce your working group please.
PAUL MESSIER: I'd very much like to introduce my working
group, and I feel badly that I didn't bring the names up with me. I
don't want to go through the names and miss somebody, so I'll just
skip it. Well, fortunately, I have the flyer here now. The original
working group members included Peter D'Agostino from Temple
University, Susan Dowling from New Television, David Douglas from
Harvard University, Bob Doyle, who I believe is here, from
Experimental Television Center, George Fifield, who is here from the
Decordova Museum and Sculpture Park, Mary Ide, who is here as well,
from WGBH, Katherine Jones-Garmil, from the Peabody Museum, which is
affiliated with Harvard, Leslie Kopp from Preserve, Inc. - its
particular focus is dance archives - Elizabeth Morse, from the
Harvard University Library Conservation Lab, Gregor Trinkhaus
Randall, from the Massachusetts Board of Library Commissioners, and
Tim Vitale, who is a conservation consultant now in Glasgow,
Scotland. I'm looking at new technologies, and a good rule of thumb
when dealing with new technologies is that if you've heard of it, it
isn't new. Thank you very much. It's sort of a self-negating
premise, but let's talk about some new technologies, and as soon as
you hear of them, you can be assured that they're not new. This is a
work in progress, and if I'm out of line in this talk, I really want
you to let me know. In fact, I think you owe it to the group to let
me know.
In terms of new technologies for video preservation, one thing that - a lot of the working group's time and effort ended up getting focused on digital video, particularly as a preservation medium, so I crafted this talk around a gimmick, and the gimmick is assessing digital video as a preservation medium. We're going to follow that theme throughout. The context for finding a new preservation medium, or any preservation medium, is the tidal wave of magnetic tape degradation. Forties and fifties, in particular, material is at risk, but of course, we heard today that even material from as early as 1985 is deteriorating significantly.
I do not want to ignore analog to analog conversion. That's not what the talk is about, but there are definitely some advantages in terms of maintaining the native format. Again, we're talking about a fundamental change when we're talking about going from analog to digital, and I'll discuss that in a second. But, I think there is some value to maintaining the native analog format. The primary disadvantage to analog to analog is this multigenerational loss that was discussed earlier. You go from the master to the copy, and you have a fall-off in image quality. Analog to digital - definitely some advantages here, but also some disadvantages. Multigenerational loss is more or less eliminated. When you're copying, it's a bit to bit copy, it's a clone, essentially. The primary disadvantage as far as I can tell is that the analog signal is only sampled. And what do we mean by that? Let's take a look at what we're talking about when we're talking about digitization, and this could be digitization of any analog signal - it doesn't have to be video.
There are three major steps that we're talking about. We're talking about sampling, which is measured in megahertz, quantizing, which is bits, and encoding. In terms of sampling, the value of the analog signal is read at evenly spaced time intervals, so there in yellow, that waveform, is our analog signal. On the X axis, we have time proceeding in constant intervals, and those white dots would represent, then, our samples of the analog signal taken at specific time intervals.
The next step is quantizing. And quantizing - you take those samples and you convert them to a value. And those values are bits. A bit is a binary integer, of course, it's either a zero or a one. For video, we're talking about 8 to 10 bits. More bits, the more precision, the more subtlety you're capturing. A low bit description of color might be to describe the stop sign as red. A higher bit description might be the stop sign is located precisely here, on the CIE/LIB color space, on the Munsell color chart. That's really what we're talking about: precision, subtlety, more bits, more memory ultimately required to store all that information, but higher quality - digital output.
Following the quantizing step: encoding. We take all that data, and convert it to a machine-readable form. Formatted, essentially. When I talk about formatting for digital material, what I mean by that is the final step of this encoding process. Formatting it for machine-readable form. For physical medium, I'm describing the physical, actual material that the digital material gets stored on. I also thought it'd be useful to look at where we are in terms of digital video. What's the state of the art? And there are plenty of other examples that I could have used. Arguably, the next home video player is going to be digital, the DVD - Digital Versatile Disk or Digital Video Disk, depending on what sources you want to read - that's going to get rolled out in the 3rd quarter of this year. Digital camcorders, laptops - there's the DVD prototype, digital camcorders, palm-size camcorders, digital editing suites, there's a laptop editing suite for news gathering, pretty nice. We've got digital broadcast satellite, DBS, which is out there. We all know about DBS - well, I think we do. There's a digital broadcast satellite - right now it's raining down upon us about one billion bits per second, falling on our heads at the moment. It's a satellite system, already for the consumer market for higher quality TV. Soon to be rolled out - that goes under the commonly marketed as Direct TV. Soon there were will be such a thing as Direct PC, which is a satellite link to computer networks, and particularly the Internet. The Hughes Corporation promises that you'll be able to download a motion picture almost immediately and MPEG 2 compressed motion picture to your computer from a web site - a World Wide Web site. And the new high definition television standard, when that gets off the ground, that's going to be a digital standard as well. So, that's all expository information, and now we're coming to the real heart of the talk. Here are the criteria that I try to measure the state of digital video against. And I'm going to blow by them because I'm going to keep coming back to them, and this is the thread that I'm going to try to develop. I want to first say that these criteria are modified criteria adapted from an article by Basil Manns, who is a research engineer at the Library of Congress. He was talking in this article, "The Electronic Image Preservation Format" - he was talking primarily about preserving two dimensional still images, and I, for the purposes of this talk and this project, adapted them for video preservation.
The first criteria: capture the image at the highest possible quality, using minimum or no compression. We can do that with digital video. The D series - D1, D2, D3 - as a curiosity, there is no D4 - D5, D6. I don't know what we're up to in the D series, but we're pushing towards ten, I suppose. What I mean by uncompressed is the gross output, the total output of the sampling, quantizing and encoding process. This is all the data , not compressed. Therefore, we're generating massive bitstreams of information, and it requires a very high level of processing power and huge data storage capacity. Therefore, there are tape-based formats. Tape is still a very efficient storage medium, especially for digital data. It is the most efficient, in terms of area to amount that you can store on it. Tape wins by far. The downside is, it's very costly. D1, which is the quality standard in terms of digital video, a videotape recorder is $140,000, going sort of more economically, D3 is around $60,000 for a deck. Compression, if we're talking about minimal compression, again, immediately ethical concerns come up when you're talking about compression. Compression is based on throwing away data, and we'll return to compression in a moment. There are a lot of different compression standards that are out there, MJPEG, and many proprietary options for digital video. MPEG, however, reigns as the most accepted standard. It is, in fact, an ISO - International Standards Organization - recognized standard for digital video compression. The others really are not standards in that sense at all. It is important to recognize that even though MPEG is the standard, within that standard there are a lot of user-defined production parameters that remain as variables, so there is no single flavor of MPEG, and that's important to point out. Even though it is the standard, there are a tremendous number of variables. I think we're up to MPEG 4, maybe MPEG 5. What we're looking at right now for high quality video is MPEG 2 - digital broadcast satellite, which is raining down a billion bits per second on your head right now, that's being delivered in MPEG 2. HDTV, when that gets off the ground, is going to be MPEG 2. The Digital Video Disks will be MPEG 2 as well. So it's finding a wide-spread use, and that's probably because, in part, that it is recognized internationally as a standard for digital media compression. This is a rudimentary explanation of what MPEG 2 is in terms of the compression scheme. The key is, it's based on identifying and eliminating redundancy across multiple frames of a video sequence. It's important to note that even though there is some operator input in the compression using MPEG 2 for the sake of efficiency, it's primarily an automated process using a combination of software and hardware. The software and hardware is figuring out what's redundant and eliminating it, which is a lot of power when you're talking about making a preservation copy. I would say that compression certainly is a controversial issue for preservation.
Getting back to our criteria: capture the image at the highest possible quality, minimum or no compression. I give it a thumbs up, a little green dot means a "go" on that. I think we can do that now, in terms of digital video. The next guideline: develop guidelines for digital conversion based on content of source material. As we just saw, digital video is quite costly and requires significant storage and processing resources. It may be useful - well, appraisal is key for the preservation of video. It's a very key first step. The question that I would pose is, do we as a culture have the same stake in preserving, say, "The Price is Right" at the same level of quality as we do for material created by video news gatherers, of a one-time newsworthy event, or video artists. If we can ethically answer, no, we don't have a similar stake in preserving those programs, those output, that information, then we need to assess where and how we dedicate our limited preservation resources. And base preservation video quality on subjective evaluation of content. And if we do that, if we're talking about making subjective judgments in our appraisal, we really need that to be guided by some sort of accepted standards of practice, and it was already mentioned, the FIAT/IFTA recommended standards, I think are a good first step in terms of this appraisal system, setting up an appraisal system. The standards acknowledge that operational and technical restrictions will influence the amount and nature of material to be retained, and then it goes on to categorize broadcast material by type, and then within each type it talks about selection criteria. So within news, you might have the news type, you will have different selection criteria than, say, the game show type. I think this is a potentially useful document when it comes to assessing video. It's a good first step and a good model. To revisit criteria, two: to develop guidelines for digital conversion based on content of source material, there really isn't anything out there dealing specifically with digital conversion. The FIAT/IFTA guidelines don't deal with that. But I give it sort of a cautionary optimistic evaluation. At least there are professional organizations out there taking steps and do seem to have the tools and the motivation, the expertise to deal with these subjective judgment issues when it comes to appraisal of video material.
Next criteria: use a format, medium and equipment that meet national and international standards. And this really relies on what you define as standard. Major digital video standard formats - again, when I'm talking format now, I'm talking about the output of the sampling, quantizing and encoding - I'm not talking about the matter, the stuff yet. I'm talking about the way the digital information is formatted on whatever medium it happens to be on. Arguably, D2 should be up there - D1, D2, D3, these are our uncompressed "standard" formats. The reason it's in quotes, again, definition of standard is kind of critical here. I would say these are de facto industry standards, since there is a fair-sized installed base within the production community. And I would say this is perhaps arguable, since these are quite expensive. MPEG 2 - we're kind of mixing apples and oranges a little bit, but I do want to point out that there is a compression standard out there, and it's not necessarily a standard in quotes, it's the real thing recognized by a seeded standards - international standards organization. There are many, many more. I could just go on. Digital videocassette is another standard for production, primarily. It's a compressed format. Digital versatile disk, or the digital video disk - that 's a new joint industry standard. So if you define standard as - if you apply a tough rule to the word standard, what's recognized by International Standards Organization, we're not doing so well in terms of standard formats, but if you're talking about joint industry standards, and de facto industry standards, then there are some choices that seem to be emerging as digital video standard formats.
I want to mention that certainly the danger of multiple competing and noncompatible formats has been recognized, and there is some effort to deal with it. Something called the open media format, which is primarily coming out of Avid, which is a digital video vendor, one of the big ones. It's an effort to create a standard for cross-platform cross-application video interchange, the theory being that any OMF formatted video from any source will work on any OMF compatible playback device. And I think this so-called standard may prevail, or at least generate some important interest in that major companies are backing it in the information technology business and the digital video businesses. And at least one professional association has recommended OMF as a working practice for data exchange. So I think there is a little bit of momentum gathering to deal with the danger of multiple noncompatible formats within the world of digital video.
Now, media. I'm talking about the stuff, the atoms, the molecules that we're saving our digital data to. Standard media - metal particle tape - we've discussed that up and down. Digital video, just because it's digital does not mean we're eliminating the dangers, the problems, the pitfalls of magnetic tape. We're still storing a tremendous amount of video on tape. In fact, it's probably the only viable storage medium right now for lots of high quality digital video, as far as I'm concerned, anyway.
Optical disks, in gray, we have CD ROM, laser disc, hard drive arrays. I think there are certainly some problems with hard drive arrays, in terms of managing it, and I don't want to get into that necessarily. But it's difficult when you think of a compressed two hour motion picture taking up about five gigabytes of hard drive space. You need a lot of hard drives, and managing all those hard drives is a nightmare.
DVD, as was mentioned earlier, is probably going to blow the CD ROM and the laser disc out of the water. Again, that's getting rolled out in the third quarter of this year. Let's talk about it a minute. In its original incarnation in the literature, you would see it as the digital video disk, but they can put a lot more stuff on there. There was an article in the New York Times on Tuesday about how manufacturers are now looking at a different audio, putting a lot more audio data on a single disk. Finally, the manufacturers are owning up to the fact that maybe CD quality sound does sound a little too tinny, lacks subtlety. So they're thinking if they increase the sample rate, increase the bits, they can get a more subtle, much more lush sound, and using the DVD as an audio medium - so anyway, there's a lot of other things you can do with this standard besides video. But let's look at it strictly from the perspective of video. A lot of major - the major manufacturers of media got together and came up with this standard, and was sort of rancorous, but they figured it out, kind of gathering under the banner of "remember the Betamax." They didn't want two competing standards out there, confusing consumers and slowing down the acceptance of this format. So, a lot of cooperation. Extraordinary, I would say.
At its initial inception, it's going to be about $500 for the deck, and the new decks supposedly will play CD ROMs, but your CD ROM player, of course, will not play DVD disks. It's backwards compatible, in a sense. Major motion picture studios are gearing up to release titles for DVD, sort of bodes well for the format, for the medium and format, I should say. And from a preservation standpoint, there is a ROWM version due for 1998, ROWM read once, write many, so a recordable DVD format. I'd be surprised if it actually took two years for them to get that to market. And the reason I say that laser disc is probably going to go by the boards, because of DVD, it's going to be cheaper, and the image quality should be superior. Should be. Who knows? They're really hyping this a lot in the different video media. So, depending on how you define standards, I am cautiously optimistic that if you applied this standard to digital video, that you could perhaps come up with format, medium and equipment that meet national and international standards. Certainly not ISO or ANSI standards, but maybe joint industry standards, maybe de facto industry standards. Next, make the format accessible on standard equipment at various levels of access. Well, again, how we define standard becomes key. Right now, if you consider your television set standard equipment, we can get digital video, digital broadcast satellite with a converter. You can get digital video delivered to your home via satellite. Optical disc players, laser discs, certainly, I would consider standard in some sense. DVD out by fall - there are a lot of manufacturers putting a lot of R&D into making sure that we as consumers consider that as a standard. That's coming. If you consider your desktop computer standard, you can definitely get digital video via the Internet, and other computer networks. There are many, many ways to get video on the Internet, Library of Congress' American Memory Project being one. Just to go back and talk about Direct PC. Fairly soon, there's going to be a $600 converter device, a receiver and converter device that you can get. And for a subscription of approximately $15 a month, they promise, you will have a satellite connection to computer networks, including the Internet. And the bandwidth will be such that you can download almost immediately, and get MPEG 2 compressed video.
HDTV - when that does come, there's going to be a lot of effort to make high definition television standard, so that's access. Multiple levels for access. This is one of the things that digital video really does very well. In terms of digital broadcast, broadcasters will probably - definitely have the capacity - whether it gets used or how it shakes out, who knows, but they'll definitely have the capacity to broadcast one high definition program, for instance, be it a motion picture or a - like the Olympics, major sporting events, something like that - in HDTV. Extremely high resolution. But they'll also, if they wanted to, send you three standard TV programs in the same bandwidth, so you can definitely get multiple levels of image quality based on content. And that way they can sell you for the same channel. They can sell a lot more advertising if they send three programs simultaneously on the same channel. You can also, as a user, you can modulate - supposedly - the version of the movie you want to see. You can see the G rated, the PG rated, or the R rated version of the movie quite easily. DVD supposedly will incorporate that as well. DVD supposedly, if you want to see the editor's cut, the final edit cut, versus the release cut, versus outtake cut, whatever cut. It's all going to be multiple levels and channels that will be on that disk.
The Internet and other computer networks, in terms of multiple levels for access, already, there are multiple resolutions, frame rates, screen sizes, that kind of thing that are possible, which allow you to research at higher volume. You can look at a lot greater volumes of material at lower resolution, in the context of research, glean out what you want to see and maybe take a look at that in higher resolution. That's something that digital does very well. Also, in terms of access, there are new tools for access possible. IBM has something called Query By Image Content, which they're applying to digital video. The first step in QBIC is to process the video and extract features by color, texture, shape, camera motion and object motion, and then you can search that database by user-defined queries, which is absolutely unique in terms of video, as far as I know. Are there any automated search tools for video? Upper left is a texture query, and on the upper right, there are video stills that were returned from that texture query. More impressive to my mind is the lower example - another user query on the left, you put in a profile of a fish and all kinds of different fish from different videos are gleaned and brought forward. Upper left, another user-based query, just a very simple linear-based drawing. What's returned are a box, a bridge, there are grids... Color-based query on the bottom, juxtaposition of red and green returns a pepper and a tomato next to each other. So this is potentially a very useful tool for research access of digital video. Definite thumbs up. That's something that digital does very, very well - the multiple levels of access.
The last one is where we get bogged down, I think. It's to insure a data migration path as a hedge against format and machine obsolescence. This is why we get bogged down. The challenge of constant change. Hardware and file format obsolescence, I would argue, is inevitable. Period. Take a look at the graph on the right. In 20 years from 1980 to 2000 - this is aerial density growth - which is just storage capacity. In this case, we're talking about gigabytes per square inch, and notice the Y access is exponential - it's an exponential leap. So we have three exponential leaps in about 20 years. And you look at the rate. This is the situation - I could have put up processor speed, I could have put up a number of different ways to measure the progress of computing, and so, in the face of this progress, we're just going to have hardware and file format obsolescence. Period.
So the rate of technological change is ever-increasing. I don't want to berate archives, but I do think there's going to be a real need to adapt to this landscape. The role of the archive is redefined. I don't know if I really stand by this - I thought I put some provoking stuff at the end. Let me just sum up my feelings about this, and you can tell me I'm out of line. I would argue that before digital information is created, it would help a great deal if there was consideration as to how that information will be preserved. Without such consideration, there is a great risk of losing that information after a very short period of time as a consequence of file, format and machine obsolescence. I would argue, in terms of for artists - those who create video, especially digital video - that those who rely upon digital technology to create art and information may want to consider preservation issues before they embark on a technology-dependent creative process. This may sound limiting to some artists. In terms of archives, when it comes to electronic information, no longer do archives have the luxury of moving in long after an event, gathering, identifying, cataloguing and preserving information relating to that event. Amid quantum leaps in technology, such business as usual, will probably lead to great amount of digital information becoming lost and irrecoverable. And having said that, really both archivists and artists need answers when they pose preservation questions. That's a trick right there, as we're finding out. So, within this context, those that create and attempt to preserve digital information may have to accept the responsibility to acquire the knowledge and dedicate the resources - which I think is key - necessary to make sound decisions regarding information technology. I'm less optimistic about the last one- cautious to almost a pessimistic - cautiously pessimistic, because ensuring migration paths has nothing to do with technology. Information technology will continue to develop at ever-increasing rates, and that's fact, that's law. Porting electronic material from one format to the next is really a cultural issue or cultural mandate. And I want to conclude with a question. Certainly digital video has the potential to be a preservation medium. But the real question is, are we as a culture willing to dedicate the resources required to manage the preservation problems brought on by an environment of rapid, constant change? Thank you for your attention, and I'll be happy to try to field any questions. Yes.
MAN: On the D formats, I would turn D1 and D2 around. D1, there's only been a couple of hundred sold...
PAUL: So, if you're talking about - I don't want to get too much into composite/component issues, but if you're talking about a digital component video...
MAN: No, I'm talking about D formats. Don't get me started on composite formats. The reason we don't have D4 is that it's a bad number in Japan. We don't have D13, either.
PAUL: So, D2, you would say, in terms of the digital formats, you would say that would be the quality standard bearer?
MAN: No, I didn't say that. D3 is better than D2, - well, D5 is better.
PAUL: D5 is better because it's component. OK, my fault. Yes?
MAN 2: We're dealing with media failure issues here. That's the problem. The problem isn't that you can't get an analog signal off the tape, the problem is you can't get anything off the tape until it's hooked up. Whether it's a digital tape or an analog tape, whether it's ones and zeros, an analog signal. Digital tape is really an analog signal, let's really get into it. We're talking about media failure. Digital tape is one third the thickness of analog tape. So, there's no reason to think that five, ten, fifteen years from now we will not have similar kinds of problems.
PAUL: That's clear. If it's tape-based media. If it's digital media on magnetic tape, everything that we talked about magnetic tape this morning applies, one hundred percent.
MAN 2: It's really important, because when you look at the D formats, they still use tape. And all the compressed formats still use tape. You think you're going to have problems with an uncompressed format reading back information with media failure, just try a compressed format with media failure. There's absolutely no documentation, there's no research to tell us how to play back compressed digital formats when you have media problems. None. And I challenge the manufacturers and say, O.K., show me how machines that don't have any guard bands, they're very critical machines, they have to be in perfect alignment, tell us how you're going to deal with media scratching...
Q: [Unintelligible]
PAUL: The model has not been created yet, but as you perceive, I think the building blocks for the model may be out there. The ideal would be for these bits to be medium-independent, and the only way to do that, I think, is to have an ongoing migration - a program of data migration. This is difficult to articulate, but I think it can be dealt with within an archival context, but it has to be a dedicated, ongoing program. The resources need to be there, for someone to be paying attention - but here's the pitfall. OMF is on Version 2. What does that mean? Well, there was Version 1, and I know they're working on Version 3, because OMF2 doesn't incorporate MPEG. We can get bogged down in details of the little things, which can be very frustrating and discouraging if we hacked at them all day. But ideally - correct. Make the information medium-independent, and that's a challenge.
Q: [Unintelligible]
MAN 2: Well, if you look at servers, for instance. A lot of people talk about having vast quantities of information and spreading it out over a network somewhere. The information can reside in several different places at once, and the information, if you will, is mirrored to different places. That's all fine, but it has to sit on something. Until we get to a point where media becomes silicon or becomes organic, and is so cheap that there is no limit to the cost. Basically, technology will eventually make storage cost nothing, right? If that were to occur, if storage doesn't cost anything, if bandwidth doesn't cost anything, then why do we need to compress? We don't. So, if you follow that model along, if you look at that chart, that's where it's going toward. Eventually, if you followed it up high enough, you can store an infinite amount of data in zero space at zero cost. And that is the model, if you will, that industry is following.
PAUL: And there are some very interesting things coming in, like the year 2000. It's predicted that - by the way, tape for digital storage is probably going to be around to the year 2000, based on a NASA study , but this NASA study also said that we have different - solid state memory coming out as well, meaning there's no reader and medium, it's almost like everything held in RAM, holographic memory. And that's supposed to be very cheap - well, no - when it's mature. But that's down the road.
WOMAN: Are you saying that there is, through this process - for those of us saddled with this responsibility, of preservation - that there will be some sort of equipment guide? That there will be certain levels of standards that will weigh out all others, and that we have to accept those as migration?
PAUL: Absolutely. If you want to preserve any sort of digital media, you have to get it right now, and you have to set up a program of migration, and make the best decisions that you can make in terms of what's available in terms of file format, meaning the arrangement of the data, and the medium. And it's going to change. If it took the industry to change every five years, pretty soon it's going to be two years, and the rate of change is going to be increasing.
WOMAN: How do you get a consensus, though?
PAUL: That's a very good question, I mean how do you? I don't know. Maybe you need people who are really paying attention to the industry, and you need good sources.
WOMAN: There's no watchdog...?
PAUL: There's no Consumer Reports for digital video, that's for sure. I wish there was. Yes?
WOMAN: I'm with the Commission on Preservation and Access, and we have a couple of groups that are evolving as you say the technology evolves. And I do agree with you based on what these groups are saying, the technology is not the question. The question is how organizations manage their interaction with other organizations, and how they prepare for the future. In particular, the Commission has one group called the Task Force on Archiving of Digital Information. This task force operates in a virtual environment, just the way you're talking about information here. I think there are twenty five people on it and they operate on the Web. They put out one draft report and they'll be bringing out a final one in April. And it is an exploration, an ongoing work-in-progress about how archives will be established and will continue to be operated in the digital environment. I encourage people to get on the Web, get onto the site, pull that off and look at it.
PAUL: Is it under the Commission on Preservation and Access? I would go to Yahooor Lycos, one of the search engines to find it.
MAN: When you deal with all this migration, an advanced approach is for a computer to keep track of when the last time something was backed up, to make sure that the system itself keeps track of the migration, so the system perpetuates itself. If you have a collection of a certain size, and you know that a certain piece of media was accessed 100 times, and your system says, well, at 120 times that piece of media is due for recycling, so what we're going to do is automatically make a copy of that, in off time. We'll take the new piece of media, whether it's solid state or whatever it is, we're going to copy that information, whether it's local or at some other location, another country, wherever it is. The system itself can do that migration. There's a SMPTE meeting which has been set up to set standards for that kind of an environment.
PAUL: That's great, thanks. Very important. Yes.
WOMAN: [Unintelligible]
JIM WHEELER: I come from the industry. I worked at Ampex for thirty two years, and I'll tell you what drove the market is, our marketing people get out there and talk to people and find out what they want. What kind of editing they want, size of the cassette, quality of the video and all that kind of stuff. Then they'd design a machine that would fit the marketplace. But then there's also SMPTE, the Society of Motion Picture and Television Engineers, sets up standards. But SMPTE does not drive the standards. The companies like SONY and Panasonic are really driving them. So, the competitive market, as I see it, is what's driving them.
PAUL: There's a book I'd recommend, "Being Digital" by Nicholas Negraponte. He talks about control of bits being the new commodity item, instead of getting away from selling molecules, atoms, things, tangible things, it's the bits now that are being traded. And when you look at control of content issues, it's happening now quite clearly, with the alliances between Disney on the cable and telephone companies, it's - Microsoft buying the Corbus subsidiary - oh, it's separate from Microsoft entirely. He purchased the Bettman Archives. The Bettman Archives is composed of photographs, and that is my area of expertise. I know about the stuff, what a photograph is, but that's not necessarily the concern of Corbus. It's to deliver it as bits. And that's an important archive, and that's a good thing to do, there's nothing wrong with that. In fact, it's increasing the access to the Archives incredibly. But I do think it raises the question in that, is the profit motive the only reason to have an archive? Or do archives that have the potential to be marketed, are those the only ones that get saved or preserved? I think there might be a danger there. I think the Microsoft - who knows what it's going to be, but I think it makes a lot of sense if you're accessing - if you're a major company and you want a photograph of Winston Churchill for a documentary that you're going to do; that you pay $2500 or God knows, who knows how much - whatever would be fair. And some of that is a subsidy, and the subsidy goes into letting any school kid access that information, so the companies - the commercial companies are actually subsidizing - this is a model that may need to be set up. But I don't think it's so dire as maybe I just presented it - large corporations controlling media. There probably is a way to open it more. One more question.
Q: [Unintelligible]
PAUL: Recording to disc? With the Write Once Read Many, that's exactly what it's supposedly going to be - you'll be able to record directly - as a post-production step, you'll be able to put that material - as a final post-production step, deliver that material, put it onto the DVD and deliver it on the DVD. It's a recordable medium, and that's scheduled for ‘98, two years. Well, I thank you very much. This has been a great discussion.
KAREN: I don't know about you, but I'm in information overload right now. It's been a great, full, full day. Thank you very much to Dr. Peter Adelstein, Connie Brooks and Paul Messier for their very stimulating presentations today.