I have this interesting problem with regard to material that I employ from e-mail notes, news groups, and web sites.  Even when I have a way of citing the material and its source, I don't have any way to assure access to the original.  Yet the original is an ephemeral thing that may be altered, moved, or destroyed.  How do I work something like a chain of custody and evidence for material that I rely on, want to credit and cite, and that is not mine to publish?

In the context of preserving the work so that someone could verify it, review it, and perpetuate it in my absence -- maybe long after my absence, there is the question of where do I put my confirming materials, and how to I assert their provenance?

My intention is to use web sites.  That certainly makes sense for nfoWare, where confirmable experience is a theme.  On the other hand, I can't have copies of source materials just lying around.  They need to be in a form where someone could access them and confirm my use of them and also have it as a source without my basically publishing the material on the web.

I ran into the down side of this when Karen Anderson's mother wanted to know what the mention of her on my web site was all about.  That was because I had filed an e-mail of hers as a backup to a discussion, and search engines found it.

I want one that remains a good backup, but that search engines won't find, and the merely curious will not access meaningfully. 

I have a similar problem on the ODMA site, because I keep e-mail notes that back up bug reports, discussions for improvements, and so on.  I want to preserve those, and it works best to do so on the site itself, linked from the material that relies on those contributions.  But in some cases, it might be awkward to leave the note in plain site.  

I have thought of a solution and a practice.  It's structure is as follows:

• The material is encrypted using a private key of mine.  This makes it unlikely to be useful to a search engine, but it is clear that I did it and that anyone can confirm my use of it and what I say about it by decrypting it with my public key.

• Because I may need to assert some things about the provenance of the material, I will also digitally sign, with my certificate attached (or easily locatable) my identification of the material, its source, and how it was encrypted.

• This effectively creates a separate signature of the material, in a signed object.  I need to think this through, but that's the basic idea.

• [dh:2004-02-20] This use of signed detached signatures could be tied into my bibliographic entries.  That has me thinking about my bibliographic database versus bibliographic pages.  There should be a way to mark availability of archival materials in my collection of source content, either way.

I will try this out with some specifications that I have obtained from the web.  There is no need to encrypt the material when there is permission to use it, but in cases when there is not, I will apply this device.  

[dh:2004-02-20] This thinking first came up when I had found the Microsoft-IBM Technical Note on WS-Inspection, and that I noticed it was becoming difficult to find on msdn.microsoft.com, there being no links to it on the XML and Web Services pages there and I could only find it through search by knowing exactly what to ask for.  It may be permitted to repost this particular document -- I will have to see what it says, but the thinking not wanting to lose it inspired in me brought me to these ideas that I can apply elsewhere.  This doesn't deal with having my sites and materials outlive me, so that is something to take up separately.

[dh:2004-02-23] I found another reason for scooping material.  I began working with network capture utilities and drivers that allow me to monitor LAN and Internet activities of my computer.  This is partly in response to work in my Computer Communications course, and also because I do want to be able to monitor activities and understand how my computers (and the software that I use) are playing on the network.  However, the sites that I downloaded my software from disappeared over the past weekend.  They are back now, but it reminded me that there are other reasons for capturing what I did, what I found, and being able to duplicate it even if the source disappears.  So I need to work on this more, quickly.

-- Dennis E. Hamilton
Seattle, Washington
February 19, 2004