Some days ago a new service launched, called Zemanta, which integrates with the most popular blogging platforms (Wordpress, blogger, typepad) to suggest tags, links and images for your post.

After installing the firefox plugin, zemanta runs next to your content editor. When you have written a post of at least 300 characters, zemanta provides a list of resources.

Although, the service is still in an early version and there are some slight problems, it is nonetheless a very promising application.

Andy Beard’s review of zemanta is very thorough, if you want more information about it.

While reading about the service, it made me wonder how semantic is actually Zemanta?

The delay of the Semantic Web

What takes so long for the further development of the semantic web? In his interview on the semantic web, Tim Berners Lee explained that the technologies are already there; what is missing the implementation of these technologies in the current systems. The way I understand the complex term “semantic web”, what is required for complete semantic functionality is:

• semantic applications
• implementation of semantic technologies in the current systems

This implementation is actually a fundamental aspect of the semantic web, since it will certify the correct interaction between machines, databases and humans.

How semantic is Zemanta?

Zemanta comes in the blogosphere as one of the first semantic applications, open for the wider public (twine is still in ‘invite Beta’ version). Zemanta uses a semantic algorithm, which - according to the ReadWriteWeb - compares

the words in a blog post to their pre-indexed database of other content in order to suggest related items which will display next to your blog post.

This is of course great news for the further development of the web. We are witnessing and participating in the first semantic application. But although zemanta, works on a semantic algorithm, it cannot take full advantage of it. The reason is, that the current systems haven’t

yet implemented semantic technologies.

For the take off of the semantic web, linked data and killer applications are required. With Zemanta, we have a killer application, which cannot show its potential.

Until this point of writing the current post, zemanta suggests 9 images:

• 2 of Tim Berners Lee
• 2 of the wordpress administration interface
• 1 screenshot of OpenOffice
• 1 screenshot of Chatzilla
• 1 Data set in the linking Open Data Project (image on the right)
• 2 screenshots of two blogs.

It seems to me, that these suggestions do not take full advantage of semantics, but rather function “web 2.0-style”, based on keywords.

Conclusion

I really like Zemanta. It integrates in the new wordpess 2.5 very well and it makes blogging significantly easier.

Unfortunately, it is quite far from being a semantic application; not because it doesn’t have the correct background, but because the rest of the web is not ready yet.

But my hope is, that if we use the application more it will help the development team to expand the service and the whole web experience.