What makes some data visualization tools like CKAN, Socrata, OpenDataSoft, and Junar specific to Open Data? Is there anything special about Open Data formats that make it easier to work with in one tool over another? Has anyone done any comparisons of Open Data Visualization Tools? I've searched, but haven't found anything extensive. I saw one that only compares CKAN and Socrata.
Indeed, current Open Data portals are definitely data catalog oriented. This is cool as a first approach because at least it gives people the opportunity to identify what kind of data exists and where they are located.
However, data catalogs are only a part of the answer to data opening needs. Data visualization is key to get adoption of Open Data policies.
If you provide citizens with the crimes bundled in a Shapefile archive, it's cool. If you have the ability to also automatically provide citizens with a dynamic and exhaustive map of crime data, displayed in their web browser or on their smartphone, it's even better. As anyone will be able to interoperate with the dataset.
The problem is that it requires substantial efforts to transform traditional Open Data formats into data visualization ready data flow. New generations of data platforms are entering the market space, which are promoting an API First approach, with data catalog AND data visualization APIs. See http://blog.balefireglobal.com/2014/12/api-is-key-element-to-foster-data-based.html for instance (disclaimer: I'm the guy quoted in this article).