DailyDeli for 2006-11-30

This could be an interesting way to distribute e-paper like products too:

  • PrestoSM is a combination of the Presto Service and the HP Printing Mailbox It allows you to send email, photos, and content to people who don’t have a computer or Internet connection. You can send email just like you do today from your regular email ac

Interesting stuff from James Tauber:

  • Many people, when first finding out about Google Web Toolkit, wonder “why can’t I use Python instead of Java?”. pyjamas is designed to make that possible. And we’re drawing heavily from Google’s work.
  • Leonardo is an extensible content management system written in Python, architected in a REST-like style and initially focused on providing for personal websites with a password-protected wiki and blog (including Atom feed).
  • Demokritos is a Python library and content repository implementing the Atom Syndication Format (RFC4287) and Atom Publishing Protocol (draft-ietf-atompub-protocol-08). Persistence is via a subversion repository. You’ll need Subversion 1.3 with the SWIG Py

Google Co-op FAQ:

  • A Custom Search Engine is a tailored search experience, built using Google’s core search technology, which prioritizes or restricts search results based on websites and pages that you specify, and which can be tailored to reflect your point of view or are

Impressive flash-based data-visualization from gapminder:

BBC Touch – something for every newsroom

bbctouch.pngI already credited the BBC a couple of times for opening up their content vie RSS feeds and APIs on Backstage.

This encourages and enables users to come up with innovative ideas how to use that content.

One idea i especially like is BBCTouch that uses the BBC Homepage RSS feed and information about the most read stories (also provided by the BBC) to determine the degree of synchronisation of both collections in near time.

I think that this kind of tool should be available in every newsroom (maybe with an even better visualisation like treemaps ;-)
Don’t get me wrong. It is not about following the masses. It is about making informed decisions. Having the basis for defining something that goes into a balanced scorecard for news operations.

Via BBCTouch i also discovered the Yahoo term extraction API one of the lesser known APIs of Yahoo. At least i didn’t know that it was there although i signed up as a yahoo developer quite some time ago. I couldn’t withstand to testdrive it immediately yesterday evening:

  • It took me 2 minutes to get a new API key
  • Thanks to the availability of an JSON output format and the included batteries of Python, only a few minutes later I was test driving it with sample content from the english and german newswires of dpa. (available to me because of my current employment situation :-)

I was sceptical escpecially wrt. the term extraction for german texts but was more or less pleasantly surprised. On first look it looks quite promising at least as an input feed for specialized classifiers (e.g. for persons, places, organizations etc.).

I definitely will test it on a large scale but first have to come up with an appropriate data model etc. that wil be useful for subsequent steps like cterm classification, term clustering and the likes.

If i find time during my daytime job i’ll try to compare this with dpa’s inhouse term extraction and classification solution.

DailyDeli for 2006-11-29

Afterd finding WebCollage and Driftnet/EtherPEG (very interesting ;-) via jwz.org i got interested in Xscreensaver for MacOSXand Ubuntu:

Something like this should be in every newsroom (in order to make informed decisions, not in order to follow the masses):

Very nice API (Initial tests show that it’s even not too bad with german texts):