Various web service APIs have been maturing, other technologies have become available, and my own personal thoughts have congealed such that I can create a useful personal knowledge management platform.
I started thinking about PKM (related to some of DavePollard'srecentblog entries) in my May 13, 2004 entry, Tip: Enabling personal knowledge management. I'm not using most of those tools at the moment, mostly because it takes too much time to manage the software. I also have taken a strong liking to layered web applications that build on smaller-scope web services. I have been a fan of combining blogs, wikis, and mindmaps - part of the reason that I like SnipSnap (bliki software that has a mindmap plugin). I've done some considerable thinking about using mindmaps and am not sure it's the best solution - other visualization techniques (like hypergraphs of tagged items) may be more appropriate. I can't find the blog reference that got me thinking about it in the first place, but the gist of the entry was that it's easier to tag a link rather than to categorize it. Hypergraphing (tags) vs. mindmap (categorizations) seems to be a very similar issue.
I think that now I'm to a point where I can start working on a concrete product offering. The features of the PKM platform would be:
accessible online with future offline access
primary focus is grouping and searching
tags provide context for the grouping and searching (I have some cool "magic" in mind that would differentiate from other providers)
visualization tool(s) would assist navigation
utilize the most popular web applications such as Furl, del.icio.us, gmail/yahoo/etc., and google
would be hosted and integrated with a bliki
would be accessed using an Ajax interface
would provide multiple layers of privacy so that some content could be shared with friends, etc.
would value-add existing web service apis
each web service api used would be openly accessible including the apis with my special "magic"
some set of basic features would be free, but the aggregation of all features would require a monetary subscription
> Looks like you call feature the way "how" you are going to > make it not "why". I wonder which problems this tool > solves?
For me, it scratches an itch. I have a lot of stuff floating around amorphously in my head with a lot of links, hard references, and quotes. This tool would help me find what I'm looking for (and possibly spur new ideas) while browsing my personal knowledge base. Also, an executive that I've been discussing this with has some tools that help him manage his personal knowledge base, but they suck at integration and context-related searching. Finally, I've seen lots of rumblings from various A- and B- list bloggers that are requesting some of these features.
I'm thinking about developing quite a different application currently and my thoughts led me to at least two points that are on your list as well:
> <li>tags provide context for the grouping and searching (I > have some cool "magic" in mind that would > differentiate from other providers)</li> > <li>utilize the most popular web applications such as > Furl, del.icio.us, gmail/yahoo/etc., and google</li>
I think that these two are really important. Any application should be as open as possible to other applications, and with the possibility to access blogs, del.icio.us, a whole lot of google services etc with simple links, the need to do everything within one application is overcome. A PKM shouldn't try to change the way you use email but should support your usage of your favorite email service. And tags (or lables), well all has been said by gmail...