Subscribe: Comments on: Getting Machine-readable Spreadsheets
http://blogs.ecs.soton.ac.uk/webteam/2010/12/17/600/feed/
Added By: Feedage Forager Feedage Grade B rated
Language:
Tags:
csv  data  don  excel  foo  machine readable  machine  office staff  problem  readable  spreadsheets  staff  tabular data  tabular  web 
Rate this Feed
Rate this feedRate this feedRate this feedRate this feedRate this feed
Rate this feed 1 starRate this feed 2 starRate this feed 3 starRate this feed 4 starRate this feed 5 star

Comments (0)

Feed Details and Statistics Feed Statistics
Preview: Comments on: Getting Machine-readable Spreadsheets

Comments on: Getting Machine-readable Spreadsheets



Ideas and Tips from the Team



Last Build Date: Fri, 17 Nov 2017 14:30:33 +0000

 



By: Christopher Gutteridge

Sun, 26 Dec 2010 11:04:01 +0000

What I'm trying to do is find something which can be realistically maintained by office staff, rather than IT staff. In an ideal world we'd build a custom database & web interface for everything but I'm looking at something slightly less IT-staff expensive to get running. My point is that CSV/excel/tabular data isn't automatically a useful database without some effort being made to either post-process it or restrict the way it is used. My idea is to produce some spreadsheets *designed* to be converted into RDF, but easily maintainable in the office by staff who don't know or care what a URI is.



By: Jakob

Thu, 23 Dec 2010 13:50:14 +0000

Sorry, but I just don't get it. What's wrong with CSV? Sure "there’s lots of Excel & CSV files in it which are not noticeably more machine readable than an HTML document". That's not a problem of CSV, but a problem with the particular data. You can create CSV, XML, JSON, and even RDF data, that is far from machine readable. Converting CSV to RDF should be similar to converting relational data to RDF. I'd keep the tabular data as it is and introduce a mapping from rows to graph patterns in turtle syntax with '$foo' as replacement for the table cell in column $foo. For more complex mappings (conditionals, splitting field values etc.) you need a full programming language and a customized script for each tabular format anyway.



By: Neil Crookes

Sun, 19 Dec 2010 16:31:50 +0000

There's also Google Spreadsheets, in front of which you can build a pretty web form, and there's an API for querying it. But it is proprietary of course :-(