Yesterday evening I found a very interesting new research paper (via the Twitter search for ‘RT microblogging’). It was presented by Daniel R. Sandler at IPTPS09 on 21 April and deals with decentralized microblogging:

Daniel R. Sandler and Dan S. Wallach. Birds of a FETHR: Open, Decentralized Micropublishing. 8th International Workshop on Peer-to-Peer Systems (IPTPS ‘09) April 21, 2009, Boston, MA, 2009. [BibSonomy: microblogging p2p] URL

Wow. At first that was quite shocking as one of my current research projects deals with the same thing. On the other hand it is great to find researchers with the same interests and thoughts. And: a nearer look at their work shows that they have a different solution for the same problem.

The first part of the paper is a great motivation for decentralized microblogging. They show the disadvantages of Twitter’s monolithic architecture and I strongly agree with them. However, their solution is a new protocol, ‘FETHR’, which has to be spoken by all applications in their decentralized microblogging space. Further, via FETHR the microblogging postings are sent to the subscribers (rather than fetched by the subscribers).

Personally, I strongly believe that the big advantage of microblogging is its characteristics of blogging enhanced with a social network (following/followers, @-refers, replies) and combined with the publish-subscribe-mechanism. There currently exist wide-spread standards in the web which could help us implementing decentralized microblogging. In my opinion there is no need for a new protocol.

However, they wrote a great paper, they go in the right direction and they were the first to publish their approach. Kudos! I am looking forward to future discussions on the topic!

From time to time I am involved in discussions about the best reference management programme. Personally I use BibSonomy and I am  very happy with it. BibSonomy is a Web 2.0-like online tool with a powerful tagging functionality and an open API (see BibSonomy content at the sidebar here or the tag cloud on the my research page).

However, good friends of mine use other tools like Citavi. They say that Citavi has nice features for knowledge organisation. For example you can create the structure of a research paper in Citavi and add references to the sections. Finally it takes seconds to export the reference list for this paper.

As I know BibSonomy has only rudimental support for such use cases. You can “pick” references from your existing lists and collect them in a basket. The basket’s entries can be exported for a reference list. Unfortunately there is no support for paper creation.

One possible way would be to use BibSonomy for collaborative reference sharing/tagging and export/import all the references to Citavi once I want to create a paper. Ok, this is not the most elegant way but it would work. But imagine you would have a Citavi-like functionality for BibSonomy. You would not only be able to share your references but even your paper structures and drafts. This mashup would be able to suggest a further reference D for a paper’s section just because there is another author who used A, B, C and D in the same paragraph while you only use A, B and C.

Another lack of function of BibSonomy is the missing support of knowledge extraction from the sources. When saving a HTML page or a PDF document I probably have one or two sentences I am interested in. Maybe there are even different parts of the document and I want to give them different tags. It would be great to mark interesting parts of a reference and write comments/give tags.

Both extensions - support of paper creation and extended knowledge management - would perfectly fit together. Ingredients for the mashup would be the BibSonomy API, a powerful JavaScript-library (paper creation would be good with drag and drop) and maybe an online editor like Google Docs to finally write the paper. It would take much time to build a ready application. I do not have this time. But maybe I try to build a little prototype of the paper-creation-functionality and than we will see. Maybe someone wants to participate?