Tag: semweb

Automated metadata

it is amazing how much semantics can be had for free these days. while i’m keenly interested in metadata and the semantic web, live is probably too short to bother with extensive annotation. the way out? maybe it can all be automated to produce low fidelity, but constant metadata. google doesn’t do too badly on just links, although they may move beyond that soon.
so i’m willing to play with new tools, or teach old tools new tricks. i signed up for the internet topic exchange, which collects posts, sorted by category, via trackback. simple concept with potentially serious ramifications. weblogs.com has long scaled out of being useful. to really scale, it needs to support some form of faceted metadata, such as XFML. of course people wont agree on taxonomies, but since every taxonomy is arbitrary why not just pick one and be done with it.

Making emails public

Imagine the field day that Google could have if 1) all email files had access controls removed, and 2) people started surfing each others’ email messages. Unrealistic, right? Well, think again. Why have we grown so accustomed to the social norm that email should be private?

this hits home for me. in 1999), i had similar ideas, and decided to give some of my friends access to my private email archives. i had an interesting discussion about the implications back then:

(me) my private section contains all my mails (or at least all i could save) from 1993 to present. its like a diary, only more frank and complete. so far, only one person besides me has access. i may give you access some day, but not yet.

(her) somehow this scares me. i fear that i may could do something that could destroy that fine construct of our friendship and would tear it down. this went on too fast. give me access to your personnel diary? you did even think about it a few minutes? i AM overwhelmed. in my diary are my deepest thoughts written down. not all off them, because some things are not to be held anywhere – some things will always stay in your mind. but there are things that nobody knows.

in retrospect, the time was not yet ripe for that experiment back then. neither were there tools to discover interesting content (no google), nor was the narrative form of the blog widely known. i believe that people have become more accepting of trading in some of their privacy in exchange for other benefits. it has become acceptable to share with the world, and this process will only continue from its modest roots (“i had cereals for breakfast today” no shit!) towards more meaningful exchanges. i have been archiving my email in IMAP for years (100 MB and growing), and it would be relatively easy to make some IMAP folders browsable and searchable from within my blog. this would be one more facet of the personal CMS, a concept that has been taking shape in my thoughts recently.

A personal CMS gives a unified interface to a users’ thoughts (weblog), his emails (IMAP web mail), his contacts, his schedules (Web PIM) and also his files. The personal CMS supports the discovery of information within the personal data of a user by offering pervasive rss feeds, deep searches and extensive hyperlinking.

RDF in my neighborhood

the new professor at my institute, abraham bernstein, has some very interesting research going on. i shall definitely try to attend his seminar social and economic foundations of computer science. i know i pledged never to consider a phd at university of zurich, but i could imagine a external phd in the field of semantic web. as long as i don’t have to sit at ifi all day and play underpaid secretary and handle unmotivated and lazy students, plus i don’t want to deal with office politics there. who knows, so many things to do.

Personal CMS

Mitch Kapor (ex Lotus) is building a personal CMS with Andy Hertzfeld (ex Apple). Very interesting architecture, and with these people behind it has a high chance of seeing the light of day.
2004-10-15: Google Desktop Search brings my vision of a personal cms (for lack of a better term at the time) a step closer. As I am writing this, outlook express is synchronizing my 2 IMAP stores to the local disk so that the indexer may pick them up. this gives me at least access to my existing emails while the wait for thunderbird support continues. I’ll start using slogger to save all my Firefox sessions permanently to disk and see how it goes. (Don’t forget to filter out 127.0.0.1 or you’ll have a nice little feedback loop with slogger picking up your desktop search pages, storing them, desktop search indexing them, etc)
Using adblock aggressively should help to keep the signal to noise ratio of those saved pages as high as possible.
I wonder where SharpReader keeps it’s local copy (currently 23841 posts) and if this facility gives me a way to search through posts that have expired.
I will try to get a good-sized gmane nntp feed in through outlook express to see if it gets picked up as well.
I also noticed that my Trillian chat logs are not being picked up even though they are text files. maybe it is just a file locking issue, but it still makes me wonder why AOL chat logs are singled out in the preferences.
Of course, once you have full-text search over most of your digital footprints (which now seems within reach), you begin to wonder what else you could do. correlating information (what sites was i visiting while I had that IRC conversation?), visualizing connections (show me other mentions of the term “projectx” over time), bayesian techniques (show me sites I might find interesting based on my accumulated data). Eventually we will all be using MyLifeBits.
2005-05-17: For those who already freaked out over the minor changes the google toolbar makes on their site (only if you specifically trigger it, a fact that was conveniently swept under the rug), what will they make of this? personal content management? the writable web? another step towards Xanadu?

Platypus is a tool for modifying web pages and then saving those changes so that they’ll be repeated the next time you visit the page. Changes are made by selecting an element on the page and then hitting a key to use one of the commands below. To save your changes so that they’ll be applied the next time you visit the same web page, hit Save (Ctl-S). This will bring up a window containing a GreaseMonkey script. Install this script and you’re done!

2018-08-12: Memory is central to problem solving and creativity.

In this essay we investigate personal memory systems, that is, systems designed to improve the long-term memory of a single person. In the first part of the essay I describe my personal experience using such a system, named Anki … The second part of the essay discusses personal memory systems in general. Many people treat memory ambivalently or even disparagingly as a cognitive skill: for instance, people often talk of “rote memory” as though it’s inferior to more advanced kinds of understanding. I’ll argue against this point of view, and make a case that memory is central to problem solving and creativity.

Webtop revitalized?

i found jon udell on zoe to be an inspiration for my thinking of what a web gui can do.

“Services could flow in the other direction, too. For example, ZOE spends a lot of time doing textual analysis of email. Most of the correlations I perform manually, using Outlook folders, could be inferred by a hypothetical version of ZOE that would group messages based on matching content in their bodies as well as in their headers, then generate titles for these groups by summarizing them. There should be no need for Outlook to duplicate these structures. ZOE could simply offer them as a metadata feed, just as it currently offers an RSS feed that
summarizes the current day’s messages.”

hmm.. xaraya running on localhost on users home computers with their idle 2 ghz cpus and connected 24 / 7 to the net. if i think about it, i am already trying to export a better view of my thoughts / files / email / contacts / calls to the web, accessible from everywhere (with the proper credentials).

  • thoughts -> blog
  • email -> imap web gui
  • contacts -> web gui to ldap
  • calls -> web gui for answering machine
  • files -> webDaV

unfortunately those apps are all separate at this point. there is no framework yet to embed them all.

disruptive plumbing

despite all the web services hype, there is steady progress on all fronts. while the e-commerce applications for web services have yet to emerge, there is another phenomenon brewing, almost unnoticed by the web services crowd.

besides connecting resources on the internet, web services are increasingly being found on the desktop itself, where they strive to fulfill the old promise of universal scripting. web services really could be the glue that holds it all together. perl, python, kde, visual basic, soon java, and now apple script have built-in support for web services. this creates interesting opportunities to leverage desktop automation (pioneered early on by apple with their apple script architecture, and later copied by microsoft with its windows scripting host.) apple demoed a few no-frills examples of what web services can do. very interesting stuff, especially in light of adobes recent announcement of their XMP infrastructure to provide standard metadata facilities across its product line. the semantic web is taking shape, driven by forces who would not necessarily be associated with the topic.

Agents and the Semantic Web

I hope that I have convinced the reader several strands of research in the fields of AI, World Wide Web languages, and multi-agent systems can be brought together in exciting and interesting ways. Many of the challenges inherent in bringing communicating multi-agent systems to the web require ontologies of the type being developed in DARPA’s DAML program and other efforts. More importantly, the integration of agent technology and ontologies may make significant impact on the use of web services and the ability to extend programs to more efficiently perform tasks for users with less human intervention. Unifying these research areas and bringing to fruition a web teeming with complex, “intelligent” agents is both possible and practical, although a number of research challenges still remain. The pieces are coming together, and thus, the semantic web of agents is no longer a science fiction future. It is a practical application on which to focus current efforts.