[SATLUG] OT: a typical grad student question

Robert Pearson e2eiod at gmail.com
Fri Mar 28 22:49:00 CDT 2008

On Fri, Mar 28, 2008 at 9:23 AM, Jeremy Mann <jeremymann at gmail.com> wrote:
> On Thu, Mar 27, 2008 at 10:22 PM, toddwbucy <toddwbucy at grandecom.net> wrote:
>  > sometimes I come up with these crazy anthropological questions...I think
>  >  that this one might be of some interest to you guys...
>  >  recently I had to go through the hassle of learning how to update my
>  >  bios without the aid of a 3.5 floppy drive and it occurred to me that
>  >  there might one day be the need for a digital archaeologist.  So I
>  >  googled it and came up with this old (circa 1993 dawn of the
>  >  Internet...damm I feel old) wired
>  >  article...http://www.wired.com/wired/archive/1.05/1.5_archaeology.html ... the article basically notes that the abundance of obsoleted archival media and that in the future this data may be of interest to social scientists.  My question is this, has there been any study in which various forms of archival media life spans were tested under variable environmental conditions?  Basically a future archaeologist who comes across say a DVD or 8mm video tape at a particular archaeological site.  For purposes of the argument lets assume that it resides in a sealed time capsule and that its been there for say 300 years.  What is the likelihood that the data on the DVD or 8mm tape would survive?
>  >  one final question, in part this question touches on another...given
>  >  that google is said to keep everything, Is this true? it seems to me
>  >  that this is true and it is also true that the amount of data on the
>  >  Internet is growing exponentially the it stands to reason that they will
>  >  eventually run out of archival space. (Finite resources and all) How is
>  >  google addressing this issue? Do they eventually destroy old data to
>  >  make room for new or are they hoping that eventually technology will be
>  >  able to address the problem though more efficient use of resources (i.e.
>  >  building better archival storage which allows for greater data density)?
>  >  I suspect that the answer is the latter?
>  Its pretty doubtful the data would be usable. I'm just now seeing my
>  data DVDs I've burned a few years ago starting to have read errors
>  when I try and access them. I've taken the usual practice of storing
>  them in a dry, cool dark place, and always in cases, and I try and use
>  the best media I can buy.
>  --
>  Jeremy Mann
>  jeremy at biochem.uthscsa.edu
>  University of Texas Health Science Center
>  Bioinformatics Core Facility
>  http://www.bioinformatics.uthscsa.edu
>  Phone: (210) 567-2672
>  --

CD's and DVD's burned on any PC burner have a shelf life of about 2-5
years according to this article and its cited source---
A second opinion on Network World of the same source article---

Here  are a couple of more good ones on the Adaptec Storage Advisors Blog---

A thing called a "media server" can solve these problems. A few people
have built them and have them running. Most have not and will not. For
a variety of reasons mostly having to do with Cost (initial +
maintaining) and Management. The "media server" concept occupies about
the same budget position as Human Resources. Great to have when times
are good and the first thing cut when times are hard.
The main thing the "media server" concept does is roll the media
forward based on archive life. Typically the media archive life is
longer than the required Information retention period or the useful
lifespan of the application to read that format with.

[The Timeline Possibilities - some of this is dated]
On Fri, 2006-09-22 at 02:41 -0500, wrote:
> But there are no magnetic discs of the capacity of DVDs,
> where the drive mechanism is not bundled with the media.

Not true!  There are several Removable Rigid Disc (RRD) with capacities
over 100GB/platter!  Now if you mean "standard," no, there doesn't seem
to be an open standard.

Matsushita/Panasonic DVD-RAM (based on their prior CD-PD approach with a
pre-formatted media, verify-after-write error correction and optional
cartridge) is a 30 year, 100,000 re-write media (quite unlike
"commodity" DVD-RW/+RW which has high error rates, typically 1 in 10^9,
on even the first write -- 1 byte in ever 1GB byte!).  It's not
commodity, but many vendors support reading the disks, even though the
format has been around 10 years.

That's why DVD-RAM was first (before the non-standard/Japan-only DVD-R
+W, DVD Consortium DVD-RW and finally, as a follow-up to DVD-R+W, DVD
+RW).  It was designed by and for optical _archivers_ and replaced a
number of proprietary formats.

> Hard drives aren't long-term devices, as in 10-20 years.

Correct, especially not "commodity" drives.

> And don't tell me tape is reliable long-term, in that it gets brittle,
> begins to adhere to itself and the loader mechanism can become
> unreliable.

Depends on the technology used.

IBM offers a proprietary linear tape (not LTO) that has a lifespan of 30
years.  It's very, very expensive, but it's used in the engineering and
financial industries.

Otherwise, DLT and multi-vendor standard LTO typically have lifespans of
about 10+ years.  LTO is known for reading at least 3 generations back
(if not more), so you should still be able to find a drive 10+ years in
the future.

But I agree, don't rely on tape for more than 10 years.

[There is a real IT Position for this called "Legacy systems archaeologist"]
[Slashdot article]
IT: The Dirty Jobs of IT
Posted by Zonk on Monday March 10, @06:06PM
from the we-care-a-lot dept.
dantwood writes "In an Infoworld article, Dan Tynan writes about the
'7 Dirtiest Jobs' in IT. Number three? Enterprise espionage engineer
(black ops). 'Seeking slippery individuals comfortable with lying,
cheating, stealing, breaking, and entering for penetration testing of
enterprise networks. Requirements include familiarity with hacking,
malware, and forgery; must be able to plausibly impersonate a pest
control specialist or a fire marshal. Please submit rap sheet along
with resume.'"
Paging Mike Rowe, Mike Rowe to the IT desk

[InfoWorld article lead in]
Working in IT isn't always pretty. After all, we can't all work on the
cutting-edge technologies all the time. Some of us have to get dirty
-- in some cases, literally.
Unfortunately, dirty jobs -- whether you're being chained to a help
desk, hacking 30-year-old code, finding yourself wedged between
warring factions in the conference room, or mucking about in human
effluvia -- are necessary to make nearly every organization tick.
(Well, maybe not the human effluvia part.)
The good news? Master at least one of them, and you're pretty much
guaranteed a job with somebody. We don't guarantee you'll like it,
Here are seven of the dirtiest jobs in IT, and why your organization needs them.

[The Jobs]
Dirty IT job No. 7: Legacy systems archaeologist
minimum is either humor or wishful thinking).

[rdpnote - I actually tried for a couple of these type jobs. Got to
the 3rd round of one. I agree fully with this comment below]
cpg343 2008-03-10 10:38:42
Dirty IT job No. 7: Legacy systems archaeologist
That's a great job! That app may not even have source code, or the
source code is in someone's "special" dialect of JOVIAL, with binary
Or it came on unlabeled 1/2" tape from some unknown system. Data
formats unknown, program operation unknown...
You get to work by your lonesome without being bothered--nobody really
wants to know what you're doing.
It's a grand combination of haruspice, crossword puzzle-solving and
looking for the Rosetta Stone.
You learn about long-forgotten systems and languages. And any progress
confers the aura of "genius" to the worker.
I'd take the job in a flash over grinding out volumes of Java for
throwaway applications. Dirtiest? Hardly.

More information about the SATLUG mailing list