Spotlight + Data Detectors + AdSense: Champagne PIM on an OS Budget

As I continue working with Mac OS X I'm continually impressed by Spotlight, Apple's included text indexing feature. It is seamless, fast, effective, and has significantly changed and improved how I find information, as compared to my Windows XP days.
Having a solid idea file [1] is crucial to Knowledge Work [2], an on-going interest of mine (see Where's The IDE (Integrated Development Environment) For Personal Information?)
Given that, I was struck by an idea to integrate Spotlight with all apps to support an 80-20-level information support system. I suppose this is essentially a work-around to connect all data in the OS, i.e., to remove data islands [3]. Lest you think I know what I'm talking about, see my disclaimer below [4].
Tell me what you think!
Elements
As I see it the elements are:
- Spotlight, Data Detectors [5]: Text -> Objects (converts text to OS data)
- Google AdSense (retrieval of hits plus display)
- Firefox's Smart Location Bar (learns hit priority)
In other words, Spotlight on steroids.
I envision these interconnecting thus [6]: Pull the text from the front most app (option: just selected text), look up all the phrases in Spotlight's database, pull out most relevant ones, order by learned patterns, then display summaries and links to the side. Suuuure.
Applications
How would this work in practice? In a word, PKM (Personal knowledge management) [7]. Suppose you're reading an email from a prospective client (or anyone of interest, really) and you want to remind yourself of who she is and your history of interactions with/notes about her. Boom! As if by magic (and I may not be exaggerating :-) a selection of relevant hits show to the right, perhaps categorized ala Spotlight's results bar.
What would show up?
- Contact information
- Times, Dates, Names, Addresses/Places
- Calendar evens
- Email messages
- Projects
- Tasks
- Files
Another application: A visual indicator of which information sources are most valuable (see Information Provenance - The Missing Link Between Attention, RSS Feeds, And Value-based Filtering). As you're browsing your news feeds, an icon (?) shows on the right how many idea file entries the current feed has. Use this when it's time for your regular purge to make an informed decision.
I'm really curious...
- Whaddya' think?
- Reality check: What's are the major arguments against this? They center, no doubt, around reasonable retrieval given all text in an app, and possible hits.
- How would you use such a beast?
- What tools do you use to accomplish PKM?
References
- [1] Mine's a Big-Arse Text File, but alternatives abound including (ick) paper-based systems. (I have nothing against paper - it has compelling affordances (see The Myth of the Paperless Office).
- [2] I'm not fond of the term, but it's the defacto standard (buzzword alert!) describing modern white-collar work. By far the best source for information on the topic is the wonderfully-named Thinking for a Living: How to Get Better Performances And Results from Knowledge Workers, mentioned in Custom Workflows For Knowledge Workers.
- [3] Developers: check out Soup (Apple Newton)
- [4] I'm not a researcher in this field, so please be forgiving. I'm shooting for an idea that's "good enough to criticize" (supposedly Alan Kay's comment about the Dynabook). I'm positive this approach has been outlined before, but I wanted to capture it Matt-style.
- [5] Mac OS X Leopard - Features - Mail, Apple Data Detectors Are So Useful.
- [6] I know, pompous as hell :-)
- [7] AKA PKM (Personal knowledge management). Gotta' good ref? I found: