Here is a modern data management challenge that Harvey solved manually many years ago. Technical journals were a very important way of keeping up to date. Harvey read or scanned many journals. We had a library of journals that was the envy of most. When Harvey read a good article he copied it put the date on the top and filed it according to the subject in his "Subject Files". The neatest things was when he read what he considered a classic, he would put a "C" in a box on the top. If any of us wanted to research a subject, we always went to the subject files first. Managing the volume of data that would accumulate over the years was a challange which Harvey solved very simply. Annually he would purge or have purged, everything that was older than three years except fo those marked "classics".
I would love to find a way of managing any of my data that I store electronically that way. I have so many old emails and I have to read them all before I delete them in fear that I will delete the classics. Any ideas? I am sure many others remember these simple systems that Harvey created for complex things.
Hi Jim,
I avoid FOD (Fear Of Deleting) by pre-empting the requirement. 16 months ago, I created a ‘New Folder’ called ‘Keep’.
Whenever I send or receive an e-mail that I think I should keep for a long time, I move it from either ‘Sent’ or ‘In-Box’ into the ‘Keep’ folder at that time.
Then, when I periodically empty my ‘Sent’ & ‘In-Box’ folders, I don’t experience FOD.
After 16 months, I have only 129 e-mails in ‘Keep’, which is not a lot to search by name or subject.
Warren
This was very timely, Jim, as I was out with a client today and we ended up talking about enterprise information management specific to data warehousing.
We talked about the Google/BearingPoint relationship and then the solution we built that funnily enough the client is also building.
As these tools become more and more robust, we are going to see a blur. Google just announced being able to google databases. To what end we discussed today, to do business intelligence or to help with the Metadata issue?
For emails I use gmail, which for many reasons I consider to be the best implementation of email around. Given the powerful search you need not throw anything away, so you can always find something you were looking for. More than that though, you can tag emails with labels. Emails can be given more than one label as well. What you could do Jim is tag emails that you find paritcularily valuable or insightful with a label called Classic. Then in the future you could at any time ask to view only emails with that label.
For files I do a similar thing but use Google Desktop search to achieve it. I have a folder on all my PCs that is kept in sync across all my PCs using a tool called Groove. This folder is called mip’s reading and I keep all the real gems in there. If I’m at work and I drop a file into that folder, it replicates out to all my other computers. So if I’m at home and want to reference that document, it’s there. But I use Google Desktop search to quickly find the item I’m looking for inside this folder which now has just over a 1000 documents inside it. With Google Desktop search I just type in a search term and it comes up. What you could do Jim is simply add the word Classic to the documents or files you consider to be classics and then you could do searches in the future with that keyword added (i.e., data warehousing classic).