Work Completed By: Richard Scannell
The corporate communications office currently has a process that involves manually searching local media for mentions of the authority.
Create an application that resides on the web and provides a single place to find mentions of Warwickshire County Council across as many news outlets and social networking platforms as possible. The application should provide summary info, then allow the user to visit the source of the information
DETAILS/LOCATION OF PRACTICAL DEMONSTRABLE OUTPUT
Details of where code, web pages etc can be found.
DESCRIPTION OF THE WORK CARRIED OUT
Describe how you approached this piece of work, any technologies, tools or techniques that you found useful or tried and discarded. Include any examples that you used for inspiration and any contacts you have made in carrying out the work.
Reputation monitoring dashboard based on Google http://www.google.com/ig
This was based on the steps suggested at
Followed the steps on how to monitor for keywords & common mispellings. Created a slimmed down version of the suggested model ( excluding the sweary searches.) Instead I created a search for ‘ Warwickshire’ key words. This approach lead to a lot of cricket related posts. The process to search news sites & RSS feeds was quite straightforward, but most of the Warwick shire mentions will be in papers without the necessary feeds.
What is needed is a way to build feeds from static sites.
Investigated Yahoo pipes. http://pipes.yahoo.com/pipes/
This was a way of building feeds from 3rd party websites. I built a search for Warwickshire County Council within the Stratford Herald & Leamington Courier sites & added them to google homepage. Date searches were not flexible. Its default search was the Yahoo search, which did not have a means to filter & sort result by date, also it was not as uptodate as google. Google search can do this, but cannot render the results as RSS. I couldn’t find a way to get a google search into pipes.
Feedify, can take a google search as input, & convert it to a an RSS feed I chose to make it with the job vacancies XMLfeed. This was straightforward to build & to imbed into google, but it imbedded adverts into the feed. These may be removed from the ‘professional’ version .
Tried feeding a dogpile search into pipes, butt we couldn’t sort by date or filter by date.
Netvibes. http://www.netvibes.com/ offered similar functionality to google, although it was less easy to add stuff to.
I looked into Twitter to create some content for the the rss feeds. I also looked into Yammer, which is similar to Twitter, but is more business oriented in that you are restricted to watching people with the same email domain & lets you build hierarchies. Integeration with external directories is possible, but it costs :
(table removed but available in original word doc)
This is a list of ‘local’ papers, which we’d need to include in a reputation monitor. I have not investigated which already provide RSS.
Kenilworth Weekly News
Leamington Spa Courier
Nuneaton Weekly Tribune
Royal Leamington Times
Rugby Evening Telegraph
Stratford Upon Avon Journal
Warwickshire Evening Telegraph
There is an issue with sorting by date, which will apply to any RSS feed which is not generated by the originating web site (EG our use of Google to create a list of entries). The date Google ( or any search engine) records is NOT the original creation date or modification date. Rather it is the date that Google indexed, or RE Indexed a page
Web 2.0 Seminar in New York
I also investigated the Web 2.0 seminar in New York. Items of interest include ManyEyes
http://manyeyes.alphaworks.ibm.com/manyeyes/ -. This is a flexible web based reporting tool, which as well as dealing with numerical reports, can deal analyse long streams of text . It analysed various political speeches & produced tag clouds to highlight major themes. It also allowed you to query certain phrases, Ie where does . Councillor ‘X’ appear near issue ‘y’…..
It was also interesting to note in these online key speeches that some of the companies are moving to a charging model similar to the one we are moving away from, whereby every requested new feature has to be costed & charged for so that cost benefit analysis can be done, so that products don’t get bloated.
Describe the degree to which the work was successful in addressing the project description. Include reasons why or why not.
The existing RSS feeds can easily be imported in to a custom Google home page. These have the advantage that they have an explicit date built into them.
There are a variety of tools, which can take a web page & convert it into an RSS feed. I used a Google advanced search, filtered by date ( the last week) & web domain to build a flexible dynamic search, which I then fed into feedify or Dapper to build an RSS feed. The drawback was that it was impossible to sort the results by date, as the ‘dates’ were when Google indexed a page, rather than when it was created. Additionally results found in the last 24 hours, are marked up as ‘n’ hours ago & so do not have a date on them. These could be addressed by setting the google search to look for stuff updated today, & recording the details in a custom database, which itself tags when the search was performed.
Yahoo search was not as uptodate as google.
Yammer provides a useful Twitter style function, which would allow mobile staff to keep in touch, but one which can be limited to @Warwickshire.gov.uk users. There may be issues, with publishing personal names , email addresses & reporting structures onto the web. The cost of the paid for options is a bit steep for WCC wide rollout.
Manyeyes has the potential to perform web based financial reporting & text searches
SHORT TERM BENEFITS
What immediate impact could the output of this R&D work have on the organisation – could it provide benefits without compromising our strategic approach?
A media watch search engine could built either based on the RSS feeds or a derived google search.
Yammer may also be useful for remote workers
Many eyes could allow web based reporting of financial systems, which might allow remote analysis & cut down costs of financial FOI requests. Its text based queries could allow reports & document to be summarised & interrogated more efficiently
How the work carried out fits with our strategic direction or how it should contribute to our strategic thinking.
Feasibility of using external data for WCC applications. But may be issue s with us deriving RSS feeds from data we don’t own using google, also we are vulnerable to changes in the Google API.