Semantic Web Application Ideas
Yesterday I attended a very interesting Tech Talk at Mailtrust about semantic web. Manu Sporny introduced the idea behind semantic web and compared between Microformats and RDFa. Here are some thoughts and ideas inspired by the talk.
Manu demonstrated the importance of implementing semantic web from several perspectives. Surprisingly, he stressed the lack of interest from the community. Specifically, there isn’t enough participation to fully enable semantic web. I think that there is pretty good explanation to why this is the case.
Researchers try to find the technique to be used in order to describe the data on the web. Users, on the other hand, apply this technique and start implementing data descriptions. However, users face two obvious problems:
- Describing data is a tedious and time consuming task.
- Users face a lot of arguments about the accuracy of the description they provide.
Manu gave an interesting example of real-life scenario. In his business, he wanted to describe a song title using semantic web. He mentioned that it took them five and half months to solve the argument about tagging the title of the song using the word “title” or “audio-title”.
The good news is that there is a pretty interesting infrastructure that web developers and authors can start using. With that said, I think that the community needs to see killer applications that motivate them to do the hardwork of data description. The following ideas are pretty simple and intuitive, but they might lead to a combination of ideas to be the killer application.
- I believe that authors might be interested in tools to help them create their pages (editors, validation tools, plug-ins to existing IDEs, …etc.)
- Web browsers need more powerful ways to make use of these semantics. For example, fuzzbot does a good job on parsing the documents and so forth. However, more interesting tools can be driven from the use of the semantic information. Maybe, automatic synchronization of objects (event’s information online and your copy in your calendar).
- For hardcore programmers and/or researchers, can we automate or at least simplify the process of describing meanings? Although that there are several research questions in the area of web semantics, this question is considered crucial to make substantial steps towards practicality.
I am pretty sure that almost all of these ideas are already implemented, but improving is the key to progress. If you are interested to discuss any of these ideas, similar ones, or related ones, please don’t hesitate to contact me. I always like to work on cool stuff with cool people.
Share your thoughts.
