Surfing with Ed on the Internet…

by Ed Swires-Hennessy, Local Government Data Unit - Wales

Ed continues his appraisals of different national web sites to stimulate use of the Internet, share best practice and encourage debate.

 

Looking up data is always a dangerous pastime for me – I often forget the purpose of the data and begin to examine the formatting: today, I did consider going to Malawi (http://www.nso.malawi.net/) but was put off by the slow download time and by the poor presentation in the first table I looked at.  Then I remembered another table I had looked at this morning – and decided to visit the Organisation for Economic Cooperation and Development (OECD) – statistics part. By far the easiest way in is to visit http://www.oecd.org and choose statistics from the left-hand navigation list. Otherwise one can get confused by the full entry path to their statistics portal - http://www.oecd.org/statsportal/0,2639,en_2825_293564_1_1_1_1_1,00.html . I did write a plea for someone to introduce a redirect from a simple URL, such as www.oecd.org/statistics, - and then found it already existed!

 

The home page of the portal is around 3 screens deep – but mainly because of the advertisements on the right-hand side of the page. I know I have often indicated that lists of anything on a site should be arranged alphabetically: the principle here is that the eye can quickly scan a column’s left hand character to find the right word. In the centre of the home page the ‘Data by Topic’ is alphabetical – across the rows of the three columns! And the organisation of the 25 topics with white space pushes over half of the topics out of view on my screen. Reducing the repetition in the right-hand column (two live ‘OECD in figures’ links! And the heading that is not hyperlinked). The left-hand navigation is common to the whole OECD site. Navigation is very confused: bold or regular type; underlined or not underlined text; three different types of roll-overs and three colours of links.

 

My first data stop was to the most frequently requested statistics (top of the right-hand navigation). Two lists at the top of the page (27 items) are only bookmarks to information further down the page. If these were necessary links to further options, this may be acceptable but, for some, there is but one choice at the next stage (e.g. population). And this is usually one of my interests – so I followed the link, further down the page to another link called the same and then to a PDF version of a table of data. The presentation is poor and, should one want to use this data, it would have to be printed and then retyped. I know something better is coming – but even a simple Excel table would be better than this.

 

Further down this interesting navigation list, I wandered into the OECD Observer Databank. The first chart was not legible – so I clicked ‘here’ as requested to find ‘here’ was not hyperlinked. You need to click on the chart! One click gets a larger version of the chart which is very poor quality. I hope Scalable Vector Graphics is quickly adopted by the OECD. No data – though an advertisement leading to a CD-Rom costing nearly 1,500 Euro for a single non-profit user.

 

Turning to the Population link in the ‘Data by Topic’ – there wasn’t one! Fortunately I spotted Population on the end of Demography (not the usual term for the non-statistician). Following this, I found a link to the population data – the same table as earlier – or at least I think it was as I didn’t stay to read it. Interestingly, the brown coloured ‘Don’t miss’ on the right hand side in a box only contained ‘Contact us’!

 

Still on the trail of data, I remembered where I had actually seen some this morning – through the newsletter (or Statistics Newsletter) in the right-hand navigation on the statistics home page. In the middle of page 10 of the Newsletter is a link to the OECD in figures with a difference: could this be the future of data access? The new Digital Object Identifier (DOI) - an emerging international standard for identifying published material on line - appears below a table to link to the underlying data via the Internet. This identifier will always get to the same information wherever it finishes up on the web site. I duly followed the link to view my table in Excel format (I’m sure it would be easier just to make the whole of the booklet available in a non-PDF and usable format!), and, on my office and home computer, could see that the typeface used had proportional spacing for the numbers – so that the German figure in the first column looked an order of magnitude smaller than the one immediately above (France) whereas, in fact, Germany’s GDP is larger than that of France. And the last column of the table needs to be aligned on the decimal point.

 

A very disappointing visit though the prospect of this new technology may assist (if you can accurately type the 18 digits after the base address).

 

This review was undertaken using Internet Explorer version 6.60 on 1 March at 16.00 hrs GMT using a 2 Mbit link to the Internet on a Pentium M 1.6 GHz machine.

 

Please send and comments and suggestions for sites to review to

Ed.Swires-Hennessy@dataunitwales.gov.uk