A good data browser allows you to navigate the knowledge space by car
Or so I would like to paraphrase David Huynh’s words that I read today on the W3C’s Semantic Web mailing list, where he wrote in response to Michiel Hildebrand:
It’s very perceptive of you to ask about the tasks that Parallax is presumed to address, and who the users are. I don’t have a specific answer beside “browsing graph of data more efficiently”.
I tend to think that contemporary graph-based data browsers either fly the user at 50,000 feet and show her the whole world in one window below (render a huge data graph as a huge visual graph), or leave her at the street level to wander around on foot (single resource view). I’m just wishing to provide her a car. Perhaps the good thing is that the car doesn’t come with a destination built in. (It’d be quite bad in real life if you need different cars to go grocery shopping and to go to work, for example.)
I quite like this metaphor he uses to describe the motivation behind Parallax, the UI prototype David designed as a novel way to browse Freebase data. It also ties in nicely with a wish made by Richard Cyganiak in an interview with him we published yesterday:
On the top of my wish list would be a really good data browser. The current crop of data browsers for RDF, such as Tabulator, Disco and the OpenLink browser, are still very basic and geeky. I hope for some sort of “Excel for Web dataâ€, an application that allows me to browse through different datasets, find the bits that are relevant to my problem, and lets me slice and dice and correlate the data in different ways. I think such an app would be key to the kind of serendipitous reuse I mentioned earlier.
In the mailing list post cited above, David pointed to the Spellbound blog where Jeanne Kramer-Smyth published a showcase of faceted browsing across Olympics games facts using Freebase Parallax and suggested that Parallax would be particularly useful for exploring connected information:
Now take this idea to the world of archives and libraries, OPACs and finding aids and imagine the sorts of questions you can start asking. Yes – it does depend on the data being connected, but that is happening more and more all the time. The promise of the semantic web is structured data everywhere we turn.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=1974531b-b7ee-44d5-b916-d3f07515428e)
August 20th, 2008 at 3:26 pm
The OpenLink Data Explorer (ODE) is not “Geeky”. It is actually a very modular delivery of Linked Data interaction that “Geeks” don’t seem to understand.
Please read:
1. http://ode.openlinksw.com
ODE is a modular Linked Data Exploration shell comprised of the following (unbeknown to most):
- Cartridges for Display (Fresnel) the View component
- Cartridges for Data Extraction (RDFizers) the Model component
- Cartridges for Actions; the Controller component
ODE at this juncture is about a modular foundation for unobtrusive exposure and interaction with Linked Data.
ODE is built with knowledge from the pre Web world of Report Writing Tools, Executive Information Systems, and other Query Tools.
Ironically, when David states that he is focusing on UI rather that “Web Essence” (Links) everyone seems to accept that. But when those of us working from the bottom up deliver minimal UI becuase we are trying to show what we are up to (at the core), the rules of task focus cease to apply
ODE is not for “Geeks” and most “Geeks” will more than likely not understand it for a while to come.
The issue with Linked Data has little to do with Data Visualization (graphs have existed for eons in computing time), but everything to do with understanding the Web and delivering options that provide the user new ways of exploiting the Web.
Kingsley
August 20th, 2008 at 3:56 pm
I think in this day and age, being ‘geeky’ does not necessarily have negative connotations. I would not see this as a harsh criticism.
August 20th, 2008 at 4:06 pm
Kingsley, so whom is the OpenLink Data Explorer for? Okay, it can show the data “behind†a web page. That’s a powerful idea, but not very useful in itself. The usefulness comes from being able to do things with the data that were not easily possible with the data embedded in the layout of a web page. Filtering, pivoting, cross-correlating, summarizing. All the data browsers out there still have to improve a lot before they can consistently add value to the data using features like this.
Of course, getting the data out of the web pages is a necessary first step, and the current crop of browsers, and especially ODE with its Sponger cartridges, have made a lot of progress towards this goal.
August 25th, 2008 at 2:03 pm
[...] the Semantic Web interfaces are only about to become more accessible to web 2.0 people (once more, hooray for Parallax), I think a VITAL next step in promoting the Semantic Web is to find human-readable explanations of [...]
September 17th, 2008 at 3:29 pm
[...] takes us right into the interface debate and, e.g., the question whether data browsers are or are not too geeky. What do you think – what would be strategies to deliver the best to be achieved with a data model [...]