RECMGMT-L Archives

Records Management

RECMGMT-L@LISTSERV.IGGURU.US

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jesse Wilkins <[log in to unmask]>
Reply To:
Date:
Fri, 30 Mar 2007 10:50:05 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (80 lines)
My responses to Larry in roughly the order he made them: 

- I agree that policy is important. So are procedures, training, etc. But
bottom line is that users ain't records managers - and you can't fire all
20,000 of them for not doing proper information management. Gateways and
other technology solutions can help, to be sure...but reality has to kick in
at some point. 

- I'd be surprised to see most classical records programs' classification
structures with less than several hundred buckets. And users simply cannot
understand that many choices. For that matter, many RIM professionals can't
understand that many choices or cannot understand the logic used to create
it in the first place. Cf. also organizational attempts to create and
enforce electronic file naming conventions - and how frequently *they*
change. I have seen some success with using technology to hide all but the
most applicable buckets, so when a user logs into Outlook they only see
maybe the 10 most applicable buckets plus an "Other" bucket that someone
else (here, RM) cleans out periodically. 

- Automatic classification relies on AI to work (or not, as in my previous
thread). Full-text indexing is not intelligent at all: create a list of
words in the corpus of messages, create a list of which words appear in
which messages, even create a list of how many times a term appears in a
message (inverse weighting). Indexing typically takes up 10-40% of the
original information; this percentage is actually going down for email
(primarily because the email message size is going up due to attachments and
increased use of verbose HTML messages rather than plaintext). So there is a
bit more storage required - but contrast $1000 for 2 terabytes of storage
(yep, at your local consumer electronics store - check it out) vs. the ease
of retrieval. 

- 36 CFR 1234.10 is a non sequitur. It doesn't say anything about whether
indexing can be done of email or any other type of electronic information.
(http://frwebgate.access.gpo.gov/cgi-bin/get-cfr.cgi?TITLE=36&PART=1234&SECT
ION=10&YEAR=1999&TYPE=TEXT). 36 CF 1234.24 is more reasonable, and arguably
provides that leaving email in a messaging app is not allowed because emails
are not available to OTHER users who have a business need for it. 

- Would you consider that same "three bucket practice" for paper records? If
not, why would you treat email differently? Email is or isn't a record
according to content, not format...just like any other record. For those
records that are longer-term (in the three-bucket mode), how is this
different from the problem I described earlier? In other words, yes, someone
has to do this work, but with thousands of emails a day, it's either a)
users - and you can't fire them all when they can't, don't or won't do it;
b) records managers - and the volume will be overwhelming for them; c)
technology, and it ain't there yet; or d) streamline even the long-term
bucket - which brings us full circle. 

- I would never assert that anyone save everything, but I think
organizations have to temper the RM ideal of managing email the same way we
manage all other information with the reality that the volume of email and
its intrinsic limitations regarding file format and media obsolescence will,
IMHO, start to force organizations to change the way they address
information management generally and with electronic information in
particular. 

And to paraphrase Larry's final point, I submit that it's a better business
practice to have it all *AND* be able to find it using full-text indexing as
well as classification tools than to keep it in the messaging app, or print
it, or rely on users to do the right thing given the overwhelming evidence
to the contrary. Again, I don't want folks to think that I've gone to the
"dark side" and am telling you to keep everything. I'm not. But you have to
be realistic, and sometimes the crusty heel of a loaf (save it all in the
not-correctly-named email archival app, full-text index it) is better than
nothing at all. 

My tuppence on a clear & chilly Colorado morning, 

Jesse Wilkins
CDIA+, edp, LIT, ICP, ermm, ecmm
J Wilkins & Associates
[log in to unmask]
blog: http://informata.blogspot.com
(303) 574-1455 office
(303) 484-4142 fax

List archives at http://lists.ufl.edu/archives/recmgmt-l.html
Contact [log in to unmask] for assistance

ATOM RSS1 RSS2