Are Russell Group Universities Ready for the Mobile Web?

Yesterday I attended Nominet’s launch event for the W3C UK and Ireland Office (and note that tweets containing the #w3cuki hashtag are available on TwapperKeeper). A number of talks covered the Mobile Web including “Mobile web: where diversity is opportunity” by Dr. Rotan Hanrahan, the Chief Innovations Architect of MobileAware.  Dr. Hahrahan informed the audience about that many assumptions about Web sites are based on desktop browser experiences and many of the assumptions are wrong in a mobile context.

This made me wonder whether the assumptions we have regarding the design and structure of institutional Web sites will be valid for mobile access.  The W3C have developed mobileOk which is “a free service by W3C that helps check the level of mobile-friendliness of Web documents, and in particular assert whether a Web document is mobileOK“.

Are the home pages of Russell Group Universities ‘mobileOK’, I wondered, or have they been designed and tested for desktop access only? Yesterday I used the mobileOK checker service to check the home page of the 20 Russell group Universities.  The results are given below.

Ref. No.	Institution	Check	Critical severity	Severe severity	Medium severity	Low severity
1	University of Birmingham	Check	2	3	0	4
2	University of Bristol	Check	1	0	1	3
3	University of Cambridge	Check	2	0	1	8
4	Cardiff University	Check	1	1	3	3
5	University of Edinburgh	Check	0	2	0	3
6	University of Glasgow	Check	1	1	2	5
7	Imperial College	Check	4	5	0	7
8	King’s College London	Check	1	1	1	2
9	University of Leeds	Check	1	0	0	5
10	University of Liverpool	Check	0	2	1	3
11	LSE	Check	2	2	2	4
12	University of Manchester	Check	1	3	1	6
13	Newcastle University	Check	1	1	2	5
14	University of Nottingham	Check	3	2	0	4
15	University of Oxford	Check	4	2	1	6
16	Queen’s University Belfast	Check	0	3	4	4
17	University of Sheffield	Check	1	0	0	5
18	University of Southampton	Check	2	2	1	4
19	University College London	Check	1	0	2	5
20	University of Warwick	Check	2	2	0	7
	TOTAL		30	32	22	93
	AVERAGE		1.5	1.6	1.1	4.65
	BEST FINDINGS		0	0	0	2
	WORST FINDINGS		4	5	5	8

Discussion

About The Findings

How do these findings compare with other Web sites?  A survey of the W3C home page gives a score of 0 critical, 0 severe, 1 medium and 2 low severity errors which suggests that it is possible to avoid critical and severe errors. However the findings for the Vodafone.com home page were 4, 2, 3 and 5 which suggests that a mobile phone company is not doing as well as typical University home page.

But how relevant are the tests which are being tested?  Looking at the critical severity problem for the University of Sheffield home page we find:

The total size of the page (192KB) exceeds 20 kilobytes (Primary document: 8.9KB, Images: 180.2KB, Style sheets: 2.9KB)

It seems that pages should be less than 20 Kb in order to avoid this error.  Is this an realistic goal, I wonder?

Other critical errors which were found for other institutional home pages include:

• There are more than 20 embedded external resources
  
• The image does not match its supposed format
  
• An input element with type attribute set to “image” is present

Severe error include:

• The size of the document’s markup (78.1KB) exceeds 10 kilobytes
  
• The CSS style sheet is not syntactically valid CSS
  
• A pop-up was detected
  
• There are nested tables

A document listing the Mobile Best Practices 1.0 guidelines is available which provides further information about the tests.

Next Steps

The summer vacation may provide an opportunity for institutions to revisit the design of the institutional home page. The mobileOK tool should be a useful tool for those working in institutional Web teams in helping to identify whether the home page (and, indeed, templates used across the Web site) are mobile-friendly. However there will be a need to recognise that mobileOK is a tool and should not be regarded as providing an infallible means of identifying whether appropriate best practices are being deployed.  But at least we now have a benchmark which will allow comparisons to be made with other institutional home pages and we will also be able to see how these findings change over time.