Feed on
Posts
Comments

[The exciting sequel to "The Cuneiform Code"]

Having established the elements, theories, and principles, what I really wanted was pretty simple. I know what I wanted to keep (element one); I had a place to keep it (element two); and what I thought was a simple way to find it all again (element three).

Element One — Know what you want to keep:

What I wanted to keep were all the bits and pieces of information that are crucial to a sane IT operation. Here’s the dirty secret. There is a vast amount of stuff — facts, figures, incantations, mystical folklore, secret handshakes, twiddles and tweaks — that IT folks have to remember to keep tens or hundreds or thousands of computers happy and healthy. There’s even more to remember if you want to keep a vast army of squeaky users happy and healthy too. To most folks IT stuff is voodoo. I needed a simple system to remember all the voodoo, Papa Legba be dammed. What I wanted was a simple system to track all these assorted permutations, combinations, and incantations.

[My secret goal was to simply avoid the 11th circle of IT hell — the hell where all bad programmers go, along with whoever invented the concept of "opt-out" email. (It should be "opt-in," you bastard — no fruit baskets for you! Oops that slipped out.)

It's a hell composed of forever clicking "Next" again and again and again, while simultaneously explaining how to print an envelope with Word. It's a hell where all three heads of Cerberus nip gently at your heels while you un-jam a printer. It's a hell where you are forced, again and again, to diagnose why "X" program doesn't play nice-nice with "Y" program.]

So, I needed a knowledgebase. My knowledgebase will serve as a repository for a host of esoteric knowledge, making it easy find again, right at my proverbial fingertips, so it’s there, when memory fails — a shared repository of the critical yet esoteric. By the way, the sharing aspect is as crucial as the “keeping it simple” part. It does no good if I know it and others don’t. Those “others” must be able to find it too.

Element Two — Have an organized place to keep it:

Abraham Maslow once observed “When the only tool you own is a hammer, every problem resembles a nail.

[My variant on Maslow is: "When all you have is a hammer, you're bound to hit your thumb. So swing lightly." My father's variant on this was: "Nothing screws things up faster than a power tool." With a power tool, you're likely not just to hit your thumb, but cut it, and a portion of your arm, clean off. Beware the power tool. ]

It’s only natural I suppose, that when I wanted to build a knowledgebase, I turned to my preferred hammer, SharePoint. I’m sure there are other options. Maslow aside, SharePoint is well suited to the task. It’s perfect — really. It hits all the criteria. In fact, a couple of features in MOSS make this a cakewalk — specifically these new things called “email-enabled document libraries.” This feature, new to MOSS, eliminates a dozen headaches. Moreover, they make it easy to feed the beast. Email-enabled document libraries are also kind of smart. MOSS is a power tool, though, swing lightly.

First, using email-enabled document libraries, a knowledgebase is easy to feed. People already spend their days inside of Outlook; I figured I might as well capitalize on that. Second, MOSS’s document libraries are flexible. They’re flexible enough that they handle just about everything I’ve thrown — or emailed — at them, including multiple attachments, mixed attachments, weird attachments, and Buddhist email (no attachments).

Element Three: Have a Way to Find It Again

Finally, the “easy retrieval” requirement is well met by SharePoint, almost out of the box. Once you twiddle with it a bit, SharePoint’s indexing is terrific. It handles all “Office” documents, including (important for me) Visio drawings. I tend to think in pictures and flow charts drawn on napkins and, subsequently, translated to Visio. Other file types are a piece of cake too, as the search engine is extensible via add-ins called IFilters. There are IFilters available for PDFs, and JPGs, and MP3s, and all sorts of other stuff. Find a list here. There’s a filter for just about everything, from GIF’s, to TIF’s, through ZIP’s.

Creating and Feeding the Beast:

With SharePoint, feeding the beast is easy. You can send it an email message; you can send it attachments via email. SharePoint will magically convert the message to a file and put it in the document library. Along the way, it takes all the information it can from the email system and adds it as metadata, automatically. Everything — metadata, message and attachment(s) — are completely searchable.

Anything mailed to the library gets the following information automatically added as related metadata:

MOSS (SharePoint) Mail-Enabled Document Library Metadata

Email Fields

Document Fields

SharePoint Fields

To:(contents)
From: (contents)
CC: (contents)

Sender: (contents)

Subject: (contents)

Title: (Office Title)
File Size (bytes)
FileType (extension)
Created (Date)
Created by (AD)
Modified (Date)

Modified By (AD)

Version (for versioning)

I added one additional field — a calculated field to store the concatenated value of the “Title” field and the “Subject” field. This is just so I have one field that contains either the “Title” or the “Subject,” instead of having to look at two separate fields. As such, if one or the other was blank, it didn’t matter.

Helpdesk Site Showing IS Knowledgebase View

When you email an item, SharePoint automatically detaches any attachments and saves them with the same metadata as the message (From, Subject, etc). As I mentioned, you can also feed the document library directly, dropping files directly into it, uploading, or by saving files to it from any Office application. You can even set it up as a mapped drive, should you want to. True to my needs, it handles just about anything, including images and Visio drawings, as well as Word documents, Excel spreadsheets, PDF documents, and the body of an email message.

Knowledgebase — View Showing Items Grouped in Folders by “From” Address

There is one nice option called “Group Attachments in Folder.” This option lets you decide to have MOSS automatically store things in folders based on the “Subject” or “From” field of the email message. I show a view of this above. I chose to group by “From” — neatly organizing the stuff by each contributor to the knowledgebase.

(Note: As you’ll notice in the above screen capture, I’ve once again mucked up the names and such. This is to protect the appropriately paranoid. I’ve left most of my info stand as is — it’s too late for me to be paranoid.)

If you turn on the “From” grouping option, items sent to the Knowledgebase are automatically sorted and filed into folders named for the originator’s email address. It creates the folder automagically. All the stuff I send goes into a folder named for me. Stuff from other people is automatically sorted into their own folders.

I should note that getting the mail-enabled features to work took some effort. I had to tweak it. In the end, I followed a guide from a site called “Combined Knowledge” verbatim. Here’s a link to the PDF.

(I emailed a copy to our Knowledgebase, by the way — illustrating the rule that you must use it to make it worthwhile. See it works!)

Knowledgebase — Standard View Showing Basic and Custom Metadata

In addition to the regular metadata fields, you can “roll your own,” adding as many fields and types and other crap as you might need. (So far we haven’t needed to add any. )The sky (and your imagination) is the limit. Beware, however, if you create the beast, you must feed the beast. Keep it simple.

In the end, I ended up with a simple, inexpensive system for managing information, something that met all my particular, curmudgeonly needs. A system that:

  • Accepts any sort of information I can throw at it, including documents, pictures, diagrams, web pages, and just about anything else that can be digitized.
  • Accepts that sort of stuff via all imaginable methods, including “drag and drop”, email, scanners, upload, and regular old file/folder access
  • Automatically tracks the “who, what and when” in metadata (who put it there or who changed it, when they did it, and what it is).
  • Is customizable and flexible (because, well, I’m going to customize it, like it or not);
  • Is easy to search — a “Google-like” interface, except none of those annoying ads about linoleum, free cell phones, or timeshares in Essakane.

Nevertheless, the most important lesson in all this is clear: Never , unless, of course, you the and first. Everything else pales in comparison. Oh, and watch your thumbs.

2 Responses to “Cracking the Cuneiform Code — The KM Supremacy (2 of 2)”

  1. [...] miss the exciting sequel: “Cracking the Cuneiform Code — The KM Supremacy” – coming soon to a blog near [...]

  2. on 07 Jan 2008 at 5:42 am Andrew Denby

    Hi Gavin, I’ve just finished reading through your “Cracking the Cuneiform Code — The KM Supremacy” article and was hoping to talk to you about the same. I run a small company (based in the UK) that has developed a solution in this area that i was hoping to ask your advice about. I’ve not been able to find a contact email address for you as yet please could you drop me a quick email so that i can get in contact?

    Best Regards,

    Andrew Denby
    ColourBlack Ltd

Trackback URI | Comments RSS

Leave a Reply