HTML Tables and the Data Web

Some time ago now, I wrote a post about progressive enhancement of HTML web pages, including some examples of how HTML data tables could be enhanced to provide graphical views of the data contained in them.

I’m not sure if anyone is actively maintaining progressive enhancement browser extensions (I haven’t checked) but here are a couple more possible enhancements released as part of the Google visualisation API, as described in Table Formatters make Visualization tables even nicer:

A couple of other options allow you to colour a table cell according to value (an implementation of the ‘format cell on value’ function you find in many spreadsheets), and a formatter that will “format numeric columns by defining the number of decimal digits, how negative values are displayed and more”, such as adding a prefix or suffix to each number.

I’m not sure if these features are included in the QGoogleVisualizationAPI Google visualisation API PHP wrapper yet, though?

Also in my feed reader recently was this post on Search Engines Extracting Table Data on the Web, which asks:

what if Google focused upon taking information from tables that contain meaningful data (as opposed to tables that might be used on a web page to control the formatting of part or all of a page)?

What if it took all those data filled tables, and created a separate database just for them, and tried to understand which of those tables might be related to each other? What if it then allowed for people to search through that data, or combine the data in those tables with other data that those people own, or that they found elsewhere on the Web?

and then links to a couple of recent papers on the topic.

It strikes me that Javascript/CSS libraries could really help out here – for example structures like Google’s Visualisation API Table component and Yahoo’s UI Library DataTable (which makes it trivial to create sortable tables in your web page, as this example demonstrates: YUI Sortable Data Tables).

Both of these provide a programmatic way (that is, a Javascript way) of representing tabular data and then displaying it in a table in a well defined way.

So I wonder, will the regular, formalised display of tabular data make it easier to scrape the data back out of the table? That is, could we define GRDDL like transformations that ‘undo’ the datatable-to-HTML-table conversions, and map back from HTML tables to e.g. a JSON, XML or javascript datatable representations of the data?

Once we get the data out of the HTML table and into a more abstract datatable representation, might we then be able to use the Javascript data representation as a ‘database table’ and run queries on it? That is, if we have data described using one of these datatable representations, could we run SQL like queries on it in the page, for example by using TrimQuery, which provides a SQL-like query language that can be run against javascript objects?

Alternatively, could we map the data contained in a “regular” Google or Yahoo UI table to a Google spreadsheets like format – in which case, we might be able to use the Google Visualisation API Query Language? (I’m not sure if the Query Language can be applied directly to Google datatable objects?)

It’s not too hard then to imagine a browser extension that can be used to overlay a SQL-like query engine on top of pages containing Yahoo or Google datatables, essentially turning the page into a queryable database? Maybe even Ubiquity could be used to support this?