Scratchpad
(cross-referencing, project, visualization — )
27 Aug. 2007
I got super excited a few days ago when I noticed that The Center for History and New Media has a Web Scrapbook. I mean, I look at their site all the time, but obviously I don't do it when I've had enough caffeine because I'd somehow overlooked this tool until now. Thinking to myself, "OMG! This is totally the tool I have been wanting to build, where I can organize 'trails' of the websites I visit and link them together just like my probably undeserving hero Ted Nelson suggested!" I jumped on and created an account right away.
Poo.
The thing that immediately turned me off was the way you organize your information. Look, there. On the left hand side. It's the damned ubiquituous folder hierarchy "tree" like you have in Windows and Linux and Mac and every other single computer file system visualization in the whole frigging world. You know the one. You see it in your sleep. There's a little folder, and then you click the plus sign, and then there's a line of folders underneath it.
I am sooooo sick of this system. It has outlived its usefulness. In fact, it just isn't useful, period. Here is what I want:
Okay, now, ignoring for a moment that the picture kind of looks like crap (I was on a computer without photoshop, so I whipped it together using DRAW - a simple and easy 2.0 drawing interface you can use to hack a diagram in a hurry), you can get an idea of what I want out of a file system viz. The large, dark circles are upper level directories. As you descend down the tree, folders get smaller and lighter colored. Shared folders are pink.
As far as cross-referencing goes - this already exists. In Linux and UNIX you can create symbolic links as a way to cross-reference files and folders, and in Windows you call it a "shortcut." You just can't look at them that way. I want a directory tree visualization that you can adjust to your own tastes. You control the colors, the relative sizes, and the positioning (the picture above is static, but I'd like if you could click and drag the circles out of the way, say, to bring focus on one folder). And, more to the point, it shows you how the folders are related - how much the topics share in common, etc.
Comments (0)
(aggregation, information overload, project — )
16 Aug. 2007
I've been a little overwhelmed lately with the reality of keeping up with all my friends online (not to mention keeping up with those who resolutely remain offline). RSS, aggregators, Zotero, blah blah blah - they're supposed to make it so you can squeeze everything into one box and manage your information from one location, but the reality is hardly that. Half my friends have blogs with feeds, half my friends use flickr - and I'm still daily finding new accounts on there I didn't even know about, half use del.icio.us, half use linkedin or friendster or myspace. I've got people on furl, google calendar, yahoo upcoming, and every day I discover a new service my friends use that I've never even heard of.
Goddamn, goddamn.
Perhaps, if I'm lucky, there's an existing entity out there that puts everything together nicely, but I'm seriously starting to consider that I need to build some sort of an interface for myself that compiles everything together. I've been knocking around the idea for ages to build a personal mini-memex (ala the bastard offspring of Vannevar Bush + Ted Nelson + Microsoft Surface) or somesuch, where I can visually and "physically" tie different pages together. Like streams. Yeah, I want a stream of information, not these goddamn islands. I have no doubts that it'd be biting off way more than I can chew, but I am seriously at the end of my rope trying to keep on top just of the people I know, much less keeping tabs on outside research interests or - gah! - finding new ones.
Comments (0)
(links, project, xml — )
5 Aug. 2007
Finally started delving into the XML book I borrowed from the PTP lending library sometime last year (sorry, PTP! Sorry!) and naturally looking for something that might lend itself to a schema. In preparation, I've started poring through existing DTDs to see what sort of things exist out there and where there might be a need. [cracks knuckles]
Here are some links and schema to check out. Er. Someday:
In compiling this stunning list, I've realized that another possibility would be for me to create a repository for this stuff myself. I mean, I know that my brain is only half-functioning today, but damn - is there seriously only like one list of public DTDs in the world? Apparently, there used to be three(!), but the one at OASIS seems to be gone, xmlns.com appears to be dead in the water (or maybe it's just brand spankin' new and hasn't kicked off yet? getting the HEAD indicates it was updated in the last year...), and then there's just crappy biztalk (wtf). Bleak. Can we really not just have a nice database with XML name, human name, url, description, and industry/audience/keywords for searching for existing schema?
For you 5 readers out there, please e-mail me if you know of any other existing resources. [ holla⊕redheadedstepchild⋅org ] (One day, I'll make a comments feature, I swear.)
Comments (0)
(project, reminder — )
4 Aug. 2007
- 2.0 public site for found history archives/digitization
- understanding how anyone actually believes that XML means a computer can understand text
(hint - <h1> [html] is as meaningful as <animal> [xml], as far as the computer is concerned...which means that someone could still screw the system with <animal>rocking chair</animal>...hello!?)
- devising a replacement system that would actually allow a search engine to understand text and recognize errors
- the tank
- DTD repository
Comments (0)