More Unreasonable Demands on IM

The other day in  Unreasonable demands on IM Tools I talked about automated linking tools that I want built in to my blogger. But today I came across another idea that has made me even more demanding.

Over at Liquid Information, Frode Hegland, Doug Engelbart and Vint Cerf are playing with an idea called hyperwords; a tool that enables us to view the net, especially the web, through a tool that gives you more control over the presentation and view of the text on the page. here''s the vision:

Imagine if every word on the screen was truly connected to this 'cyberspace' we heard so much about during the late 1990's? Imagine if you could flip around on any word - rearrange your views as if you were a pole-vaulter, around any word - go anywhere based on any word. That would be cool indeed. And probably very productive

Now, I'm not so sure about it being more productive. I think words are less important than utterances and the whole lot mean very little without context; but then I';m married to a systemic functional linguist so I would say that, wouldn't I?

More to the point, the interface looks to me like my favourite nemesis, more work

But it does give me an idea that started brewing with the earlier post.

Lets say that I run an indexer like Copernic that looks after my own files, Eudora emails, browser favourites and history (not Opera yet dammit) and generally knows what words and phrases, also names, I find important. Lets say it indexes my blog as well, plus any blog pages that I open in Awasu, and its database.

Now lets have it talk to a tool such as LI. Now let me tell it to

• highlight in red any names of people in my PIM or any other file that it indexes,
• lets put in blue words or phrases that are the titles or in the first paragraph of any document I have on my hard drive or the blog
• How about green for text that I have used as an anchor for a hyperlink etc etc

Now let me tell it that if a page rates higher than, say 60% on some personal, arbitrary sliding scale of importance or interest, just automatically archive it on Furl.

And so on. What I'm saying is that tools like this should rely on the emergent properties of what I do, based on the text that I use, or link to or in some other way, value more highly than "lorem ipsum". Once it can do that it will be able to tell me, or remind me, that this page is "interesting", not because I say so, but because it intersects with a high percentage of what I do.

And even if it doesn't reach that exalted state, the tool will be able to give me a very fast scan of the page to see whether it contains any references top things in which I might have an interest.

What I don't want is to have to do all that manually. But hey, its early days, stick Liquid Information on your watch list.

Comments

[Mark]

You bring up some good points, Additionally; liquid information sounds very much like Xanadu's "Survivable deep linkage". Ted Nelson would probably have something to say about this project (even if Xanadu is technically "open source")

Mark

[Frode Hegland]

Yes, Ted Nelson does have a lot to say about this project. In fact, I am very honored to say he's on the project Advisory Board.