Archive for November, 2002

New approaches to web services.

This article points to  acouple of good examples of leveraging Amazon and Google’s web srervces APIs….This other example – this time from Calin Uioreanu – shows various eCommerce offerings (in this case cameras) being aggregated together – and presented to on-line shoppers in a comprehensive fashion. This aggregated camera-shop – was created entirely based upon Amazon APIs.

Though I realize allot of this was done during the dot com hayday – that aggregation was all done – by hand (i.e. manually gluing together systems – with no underyling industry standard to leverage or utilize.)

Paul Bosch (Meg’s buddy) has also created some cool ‘aggregated’ interfaces – combining – guess what? Blogging and web services!

That’s the genius behind XML-RPC and Dave’s original reason for creating it.  It frees the UI and app developer from all the ‘misha-gash’ of the backend.  And for us – it’s a zig while others are zagging.

[Marc’s Voice]

This post jarred an idea loose.

Ultimately, the promise of web services, multimedia conversations, smart aggregation, metadata, and king client apps, is to put the potential power of web scripting into the hands of the non-technical users.

Users should be able to assemble complicated networked ‘apps’, with easy to use (visual?) tools, for their own consumption, and for publishing to others. Today this is called rss aggregation and weblogging.

The standards are emerging for gathering data – the web services and sematic web standards (and heck throw in screen scraping) – and publishing data, in rss, opml, html, etc. More standards need to develop to stitch this together. A layer above scripting, encompassing all the gathering and publishing standards, in XML presumably, which can support a tool layer above that.

ebXML and all the WS-* are somewhat in this direction, but completely tied to SOAP, and years from emerging. I like Apache Cocoon, for the idea of stitching together different functionalities in a simple XML document, to create portals. We need this sort of XML instruction language for rich clients, to be used locally, for personal aggregation, and shared, for publishing.

Developers need to build so the unexpected can happen. I’ll think up some future examples. In the present, myRadio nuggets are knocking.

Comments

Comments

Comments

Comments

Comments

Matt Mower’s Knowledge Log – ( liveTopics, k-log, radio, blogging, RSS ): "This will enable a smart aggregator to use the topic’s for filtering & combining feeds together."

+ filtering is in the pipeline for myRadio, on dates, keywords, and now topics. will be tricky to devise a UI.
+ filter a single feed, or multiple feeds. multiple feeds would require agreement on a common pool of topics, i think.
+ Syndication, with meta-data, gathered
by smart aggregators, has a lot of possibilities. It would be cool to
hear more about usage scenarios.

Comments

Using inline frames (iframe elements) to embed documents into HTML documents

just a reference. myRadio’s micro-content blogging uses iframes in RSS (discussed in this thread). the layout looks horrible, but it works as a stop gap solution.

Comments

Comments

Syndication is not Publication

Absolutely, on the Mark.
Isn’t there enough to do out there, without fretting to death the syndication format? Geeking out of control.

The size thing is number 1.

What’s the big deal about publishing in two formats? Or more. Isn’t there a need for mobile phone, pda, etc formats.

RSS has momentum, beyond the ubergeekdom, and even beyond Dave Winer. Switching formats leaves behind all the people and groups who have just started to get it.

Comments

‘Intelligent’ design personalization at bbc.co.uk | Holovaty.com

Nice to see more behavioral personalization. The best thing about it is that it’s completely passive for the user. Though I think small color changes may be a bit too subtle to be useful.

From the sbc-yahoo commercials, I gather that My Yahoo has finally tapped into this sort of personalization. Yahoo has a huge amount of data on users. Tracks every page view visited on the y network, every search, etc. This is datamined into a very detailed profile, which was generally used for targetted advertising. I put together a prototype tool, that suggested My Yahoo content based on the datamined profile. Nice to see it finally released, 1.5 years later 😉

Comments

« Previous entries Next Page » Next Page »