On 10-03-13 11:01 , Neil Gould wrote:
"Alan wrote:
On 10-03-12 17:57 , Neil Gould wrote:
(snip)
No, IMO that's overall, and non-linear, since that figure includes
media that won't make it 20 years, and so on.

I'd bet that a lot of images are going to be copied forward (hard
disk
to hard disk) even if they are ignored, they will survive. Further,
many copies will branch into parallel paths.

It's getting more and more popular - as well as cheap - to keep data
on
hard disks and just keep moving it forward as larger disks come out
at
ever decreasing prices.

I'll hold my number.

To retain your own images for the remainder your lifetime, this
practice is reasonable, but not infallible. To maintain those images
across two generations? Three? Is the light coming on yet, Alan?

Light was never off from the beginning of this issue, I've never stated
it in terms of a specific image or collection surviving, but that given
the huge number of photos taken (what did you WAG?) that there will be
long term survivors.

Further of course, those that are dead in 10 years are dead. Those that
are prepared to go 100 years are more likely to go 500 than those that
aren't prepared at all.

In other parts of this discussion I've raised social reasons that will
result in surviving images as well as commercial entities like Corbis
and Google playing a role in long term preservation as they move their
own data set forward in time (Picasa, Panoramio). While they may not
survive as entities, the data likely will under some other organization
or trust. Even the Mormons might take on a lot of it as part of their
genealogy database.

I seem to have to repeat this time and again: the probability of a given
image surviving is pretty low; the probability of various images
surviving is a certainty.

And so it is for film images, however with a low volume of film
shooting, today, film images from _today_ are becoming lower probability
survivors.



In that 1% is an even smaller number who take very good care of
their
image data and a portion of those who ask themselves, "how do I
make
these REALLY last a long time."

And out of that last bunch is the statistical likelihood of some
images
surviving.

That's all. No guarantee at all that a _specific_ image will
survive.

But the probability can always be improved.

Not without extraordinary effort.

I think we're at odds over the following:

You (and perhaps stephe) are looking at it from the POV of given,
specific images surviving for a long time.

I'm looking at it from the POV that some images, though I can't
say
which, will survive by chance.

Yes, we are at odds over the idea that any digital images will
survive
*by chance* for that length of time. I've seen nor experienced
anythiing that supports the idea that this will be the case. If
you
have some information to the contrary, perhaps you should present
it
at this point.

I've stated it several times and repeated below.


The cheapest method that requires no long term plan is to use
archival
CD/DVD (BluRay?) and to store them benignly.

I suggest you do some research on "archival" digital storage
media.
The writable materials will not survive for even a small
fraction
of
500 years.

Not so. The archival CD/DVD media (usually gold based) have 100
year to
200 year lives when stored in benign conditions. This is
predicted
based on accelerated life cycle tests of such media v. the common
CD/DVD's which use silver or aluminum and which suffer
oxidization
over
the long term - even in well sealed media.

Without going into the failures of such tests to predict longevity
of
materials (I heard many claims since purchasing my first CD
recorder

Quote source.

The source for what I heard? Me.

before 1990, and have experienced many failurs of "archival" media

What was the archival media?

Various removable writable media, including "Gold" CDs, etc. Since my
business is affected by the cost of maintaining client data, I always
use the best available media.

since that time), it appears that your above commentary is
conflating
replicated media with writable media. Replication and the
preservation
of replicated materials would be included in my idea of
"extraordinary
effort", and would not include any of the typical every-day images
nor
most of the professionally taken still images.

I'm solely speaking of the use of archival (gold based) CD/DVD's
from
reputable disk makers. You have to do your homework to make sure
even
those are the best ones (sealing of the disk is very important).
Only
buy disks made in Japan and Taiwan. They quote lives of 100, 200
and
300 years (the more conservative, Verbatim and Memorex quote 100+).

This is pure nonsense, and only confirms that you are a victim of the
marketing drivel that drives such notions. First of all, there are not
many primary manufacturers of writable media. Secondly, you can't
determine quality by region, except coincidentally. Thirdly, Verbatim
and Memorex are not primary manufacturers of discs, their products are
sourced from low cost suppliers, as a result are inconsistent and in
my experience are typically unreliable for even a couple of years.

There is at least one site that talks about sources of the discs as
being important. It is certainly not "nonsense".

When you talk about sourced manufacturing I agree.

But I'm not talking about pricing the lowest cost CD/DVD's at Wal*Mart
either. The reputable gold archive disks (made in Japan or Taiwan) need
to be ordered online. At least here.


Now that the basis for your opinion is established by comments that
reflect a lack of experience and technical knowledge of the issues
that affect the longevity of writable materials, I will bow out of
this discussion.

My technical knowledge is better than most. I have experience with
accelerated life cycle testing (for other purposes). The key points for
image preservation on CD/DVD (and BluRay at some point) remain
non-oxidization, sealing and benign storage. These are the things that
the archive disk manufacturers address.

I've never come close to claiming that all images can be conserved with
certainty, only that a number will certainly survive.


--
gmail originated posts are filtered due to spam.