Obelus intends to be a tool to help modelize complex ideas and arguments, representing them through semantic networks. Tools and examples must be developed. Just the same way as written word has been a breakthrough to express and propagate non-trivial ideas in an unaltered way. Representing knowledge in a network of simple concept and annotation nodes (not constrained by a linear layout as in long texts, etc) might have advantages over ordinary (even hyperlinked) text (specially when extra comprehensiveness and rigour is sought, or to be handled by automatic tools) Obelus intends to be an semantic and open web annotation service, which allows to reference, quote, comment and attach information to any referenciable resource.
For example: In a discussion in the internet, over a certain topic one user might say a few things. It would be nice (and maybe faster) to make some custom annotations over specific items instead of answering the whole post. Many of these anotations would be relevant to few people but nonetheless useful ("i read this", "i am writing about this item" would be useful for the "duelers", but a random reader might easily skip them), and the actual comments would be more focused on a single concept (avoiding cumbersome hydra-like [monster with multiple heads] replies talking about too many things, some of them not closely related). These concerns might be useful as well when composing a compilation of ideas (either a single or multiple authors) on one project/research line.
However our mind is quite bound to articles and graphical representations, and is only able to work with a few concepts at a time. Therefore a bare knowledge ontology might be utterly unwhieldy. We need technologies to map concepts present in each existing resource.
The project was initiated as a web annotation system (similar to extint SideWiki and Third Voice). Tough some useful uses might be provided (linking to relevant resources), a lot of more controversial use (annotations that web owners would tag as vandalism, bigotry or worse) might be done. Such systems received a relevant ammount of complaints and where eventually discontinued. Same concerns (cost of operation, moderation, legal obligations)of an open annotation system would apply to an open ontology. Some close internet service sites (Wikipedia, reddit, etc) are self-sustaining even with same problems.
Also, the users should be able to provide new annotations on a page/resource, even annotating on annotations, so the annotation service must provide for it.
Some sites (Such as blogs, forums, newspapers) have their own comment systems. Some other internet services (such as Google search, or Delicious) allow as well to link and annotate on pages. Why then yet another system?
Obelus Wiki(Compilation of web annotation concepts, known developments, etc)
Obelus02 is a prototype Java Swing client application for modelling ontologies, and hyerarchical representation of these.
This sample is about some questionable work (A rulebook about how Java programs should be writen somewhere). As the work is not suitable to be modified, a reference to it is made, and also a facsimile (the work copy that is actually annotated). We also show a composite comment (references several anchors).