Suggestion from Chris Gutteridge, to be worked into usable text.
-- use your own data.
If you've provided open data, especially linked data, it is ideal for building applications and websites on top of. Your staff already know that the data is cleared for publication so don't have to job through hoops. Making even a simple interface on top makes the data far more valuable to your own staff/users.
I've been working on a PHP library to make it easy to turn data documents into friendly, maintainable HTML pages. http://graphite.ecs.soton.ac.uk/
Basically, after publishing data, the next step is to link to or provide a simple interface for non-programmers to gain value from it. Unless you're a government or have an amazingly important dataset, it's a bit optimistic to just wait for a coder to write a free app for you.
Such dogfood-eating pages should always give a clear indication of how to correct the data. We've found people notice errors and want them fixed but don't know how. People are far more likely to spot errors in an HTML document than in a data file.
Suggestion from Chris Gutteridge, to be worked into usable text.
-- use your own data.
If you've provided open data, especially linked data, it is ideal for building applications and websites on top of. Your staff already know that the data is cleared for publication so don't have to job through hoops. Making even a simple interface on top makes the data far more valuable to your own staff/users.
I've been working on a PHP library to make it easy to turn data documents into friendly, maintainable HTML pages. http://graphite.ecs.soton.ac.uk/
Basically, after publishing data, the next step is to link to or provide a simple interface for non-programmers to gain value from it. Unless you're a government or have an amazingly important dataset, it's a bit optimistic to just wait for a coder to write a free app for you.
Such dogfood-eating pages should always give a clear indication of how to correct the data. We've found people notice errors and want them fixed but don't know how. People are far more likely to spot errors in an HTML document than in a data file.