Our next step is to put these discussions into practise. This may get slightly technical but don’t be put off, it’s all interesting really.

Outline

The system is made up of two basic parts:

  • the client, or browser, which applies the styles and provides means for the user to provide feedback about a particular source
  • the server, which stores referencing data, links & any other important information

The Client

In simple terms, this element (most likely as a Firefox plugin) will scan through each site looking for referenced information in the form of <cite> tags. The plugin will take this data, and lookup that reference on the central server which should respond with an appropriate confidence factor based on the rules describe previously. The client will then apply the dotted-underlines with shading to convey the level of confidence given.

User interface ability will be restricted originally to clicking to rate a particular page (reference) on a percentage-confidence scale. Over time this could be extended to applying opinion to specific blocks of text within a page & querying information thought to be incorrect/providing counter-argument.

The Server

For the most part the server software will simply receive requests, access the database and return the appropriate confidence factor levels. However, as mentioned previously there will be times when no confidence information is known and in these circumstances server software can use a number of means to assess the likelihood of accuracty.

For example, simply checking the next “parent” level of information – i.e. if no information exists about a particular BBC article, we could check the confidence factor of the BBC region, subject, or indeed the whole BBC site. Other more intricate methods of filling in the gaps could of course be used as developed.

The server could also take responsibility for checking the validiting of referencing claims – i.e. ensuring that referenced documents really do support what is suggested. If this is beyond coding capabilities, simply providing a means for human-input to report such inaccuracies. Documents found to be misrepresenting their references could be marked automatically as inaccurate/fraudulent.

Advertisements

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: