Digital BW, The Print

On Nov 23, 2004, at 2:26 PM, Peter Nelson wrote:

>
>
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com, Christer
> Rosewelll <christerart@m...> wrote:
>>  you gotta leave something behind for those who's
>> supposed to dissect you and your work when you're
>> not around anymore - if nothing else - this will
>> make it a lot easier for 'em...=*^)
>
> WRONGO!
>
> It will make it a lot HARDER for 'em!
>
> Digital technology changes so fast that even 15 years from now there
> will be no easy way to read those DVD's.
>
> Assuming that the shelf life of the media itself isn't an issue, in
> 15 or 20 years we'll be - what - maybe 3 generations of storage
> technology past DVD's.    Traditionally PC's overlap one generation
> prior to what's current.   When 3.5" floppies came out most PC's also
> came with 5.25" floppies, too.   When CD's came out they still
> shipped with 3.5" floppies but 5.25's were gone.  Now that DVD's are
> standard, 3.5" floppies are going away (NONE of my last 3 computers
> came with one) but PC's can still read CD's.
>
> So say XYZ replaces DVD in 5 years - DVD's will still be around.
> But 5 years later, when ABC replaces XYZ, DVD's will go and computers
> will still have old XYZ drives.

First, CD's have been around a lot longer than 5 years; 3.5" floppies 
have been around about 20 and DVDs will have a retail lifetime for 
consumer media of well over 10 years. That said, when you discuss 
digital preservation, the physical medium issues must be separated from 
the data format issues; they are completely different facets of the 
large digital preservation question. The assumption today *must* be 
that the physical media has to be migrated to avoid erosion 
(delaminating CD-Rs, fading signals on magnetic media, etc); there are 
varying recommendations depending on the media in use, but it's quite 
reasonable to expect even without a major shift in the market that 
today you at least need to assess your physical media every few years, 
and depending on storage conditions will likely have to migrate every 
3-10 years.

> And nevermind the drives - you would also need DRIVERS

This should be moot; no one should be storing physical media beyond 
it's market lifetime. Media must be migrated for two reasons - erosion 
and obsolescence.

> and something
> that can read the file formats, etc!!    How many PC's today have
> drivers and display software for hardware and image formats that were
> in use 20 years ago?

Again, we must separate hardware from software issues. Actually, I 
guarantee you that a CD-ROM drive or even floppy drive I used over 10 
years ago can be used on a modern PC; but let's assume that hardware 
does become obsolete (it will), but that the data is migrated regularly 
to avoid the two key physical issues with digital preservation.

Then we come to the data; I have zero doubt that a well formed JPEG or 
a basic TIFF will be completely readable in 20 years. This is without a 
doubt in my mind. Comparing the insanity of zero standards and no 
market of the earliest days of the PC era to what are two of the most 
common, documented, and well supported file formats on the planet today 
is nonsense. JPEGs will probably still be the bulk of online images in 
5 years, maybe even 10, and will be extremely well supported. The 
formats are open, documented, and understood - unlike that WriteNow 1.0 
word processor file, which is understandable long dead because it was 
never documented, had a market in the few thousands, and had 100 other 
competitors.

Granted, proprietary formats have much less a chance for easy survival, 
though there will likely always be an entrepreneurial market for file 
format conversions. This is where standards are critical; your simple 
ASCII .txt file will be easily supported in the future, due to its 
simple structure. Your Unicode XML file will be tolerable because of 
its broad documentation and consistent, cross-industry, 
cross-application, cross-platform support. A proprietary camera RAW 
file from a small camera manufacturer that went out of business in 2007 
will probably be on very few radars in 20 years, and will require some 
extraordinary efforts to decode.

> So your plan forces your heirs and descendants to keep copying your
> images to whatever is current every few years!   That's a lotta
> trouble for them to go through!   What makes you think they'll be
> willing to do that?

No, his plan requires him to maintain his collection, just as you do 
with analog materials. You don't leave your materials in the sun; you 
put them in quality binder sleeves. You keep the climate reasonable and 
know how to handle them. As the digital preservation issues become more 
pressing, so will the wider world's understanding and awareness of the 
issues; right now it's a few people preaching to huge markets that 
simply don't want to hear it. Backup software, migration processes, 
none of this is rocket science; but so far there have been no market 
forces demanding it be made easier and maintainable by mere mortals. In 
20 years, I expect enough people to have been burned HARD by loss of 
every photo, document, and email they created during this era (~1996 
and on) that preservation technologies and practices will finally take 
a position as a desirable feature of both hard- and software products.

> Now *I*, on the other hand, have Kodachrome slides and BW negatives
> my father took over 60 years ago!!    They have been sitting around
> in attics and drawers all this time and NOBODY had to do ANYTHING to
> update them.   Yet I can pop them into my Nikon Coolscan and read
> them like they were taken yesterday.

And since I don't shoot slides or any kind of film, I'm not sure the 
relevance of this. Do you expect us to all go back to film, or are you 
suggesting we simply archive to slides from our digital files? If the 
former, I have one word for you: no. If the latter, yes, this is a much 
discussed option. I work at a major cultural heritage institution, and 
we do this reverse migration (digital to analog) for many critical 
materials, just like your lawyers likely keep signed paper copies of 
documents that originate electronically.

> If you want archival images for your descendents, you need to pick a
> storage medium that does not force them to do constant maintenance.
> I suggest BW silver-emulsion film.   Even if there are no scanners,
> because no ones uses film, in the future, there will still be
> cameras.   And those cameras will be FAR higher-res and wider dynamic
> range than today.   So your descendants can just "scan" the old film
> by taking a picture of it, like people do today to copy slides with
> their cameras.

I'm not planning to die any time soon; if so, my current will asks that 
all my data be destroyed. However, my career in particular is oriented 
around solving these problems; while there may be a semi-viable analog 
backup medium for digital photographs, most digital data is not 
practically stored in an analog form that would be any less susceptible 
to the tides of time than is digital (in fact, the opposite). 
Photographs and other static (ie, non-interactive) content typically is 
tolerant of some loss - some noise introduced by writing it to analog 
media, or to simple erosion over time - a faded photograph is still 
recognizable as a photograph. Many other data and forms of human 
productivity and creativity - whether interactive pieces, raw numerical 
data, databases, and so on - are not tolerant of digital to analog 
conversions or the introduction of noise. So while it's easy for us in 
a photographer's forum to just say "bah, just print to an archival 
medium and be done with it", this transformation in fact loses so much 
data that it's virtually inapplicable to many other forms of critical 
digital content - so, in other words, the same digital preservation 
challenges must be solved for a variety of other applications, and will 
be.

This is only the highest level summary... I'm pragmatic, and realize (I 
deal with it daily) the huge challenges of preservation. But it's not 
impossible, nor unnecessary (by a long shot), and ultimately will 
benefit from many of the same reasons we all use digital technology for 
everything already - it's inherently easier to automate interactions 
with bits than it is with molecules. In the long run, I'd rather manage 
an archive of digital materials than analog.

-R