UK Web Focus

Innovation and best practices for the Web

Archive for July 25th, 2011

Do We Want Technical Diversity or Harmonisation?

Posted by Brian Kelly (UK Web Focus) on 25 July 2011

Current Diversity of Approaches to the Mobile Access to Institutional Services

Do we want diversity in the technologies used to provide various institutional Web services or should we be seeking to gain benefits which may be provided by adoption of a small range of technological approaches?

The instinctive answer for some is likely to be a desire to embrace diversity and to encourage ‘a thousand flowers to bloom‘. Others, however, will be concerned that such approaches will be costly and lead to confusions for user communities and will make it difficult to provide a harmonised services once the best approaches become accepted.

Such issues are likely to be revisited in the context of approaches and technologies which will be used to deliver mobile web services.  UKOLN and CETIS recently carried out a survey on Institutional Use of the Mobile Web. Although we are still working on our report on the survey the  findings for the initial question “How is your institutional website(s) delivered to mobile devices?” appear interesting. It seems that the most widely used approach to the provision of access to mobile devices is no different to that taken to provide access to conventional devices.  A significant number are providing a separate site for mobile users whilst similar number have developed stylesheets for their Web styles which are specifically designed for mobile devices.  One institution is providing access to mobile devices using a mobile plugin which is provided by their institutional CMS.

Whilst CETIS’s Mobile Web Apps briefing paper (PDF format) (which is described on the CETIS blog) provides the advice that:

There is no such thing as the Mobile Web. Design for the usual Internet and then make your site adaptable for mobile devices for example decreasing the screen size using CSS media queries and then scaling up for larger devices like tablets and PCs by progressively enhancing access for larger audiences.

in reality it appears that such advice is not currently being widely implemented. There will be understandable reasons why such advice cannot necessarily be easily implemented (there will, for example, be existing technologies in place which cannot easily be updated or replaced and there will be the need to find resources to carry our usability testing on a variety of devices). In addition, as discussed on this blog in the context of the Shhmooze app for Apple’s mobile devices,  there may be business reasons for developing an app for a popular mobile device in order to validate the potential demand for a new service by providing a tool which maximises the usability provided on a specific device before developing device-independent solutions once the demand has been established.

But whilst one can appreciate the current diversity, there will be a need to understand how the landscape may develop in the future.  The comments in the survey describe how, in addition to existing implementation challenges, staff in institutions are still debated longer-term strategic policies.

Revisiting Decisions on Institutional Web Site Search Facilities

Might there be understandable reasons for diversities in the technical directions which institutions across the sector take?  Is there value in welcoming a thousand mobile flowers blooming?  In order to provide a historical context to such discussions I thought I would revisit the ideas which were being discussed regarding the provision of search engine technologies on institutions’ Web sites over ten years ago and look at how institutions are currently providing such services.

In a short paper on Approaches to Indexing in the UK which was delivered at conference on Managing the Digital Future of Libraries hosted in Moscow in 2000 I presented the results of a survey of software used to provide search facilities on institutional Web sites in UK Universities.  As shown in the accompanying table the most widely used indexing tool was the open source ht://Dig solution. The survey shown a wide range of applications were used, with 13 institutions using software which was used by only one or two institutions.   It was also noticeable that no few than fifty higher education institutions in the UK were failing to provide a search facility on their institutional Web site back in July/August 1999.

My recollection of the discussions on the mailing lists back then tended to focus on a variety of factors: ht://Dig was preferred by many as it was an open source solutions, whereas others were happy to use the service provided by the Web server software provided.  I can recall the Ultraseek’s management capabilities where appreciated by institutions hosted multiple Web servers who were  willing to pay the licence fee for this commercial product, whereas Harvest’s distributed indexing was felt to provide a scalable solution which can be used to provide a national indexing across UK University Web sites, known as AC/DC. Only three institutions, however, made use of an externally hosted solution (two used Freefind and one used the public Alta Vista search facility.

A survey of institutional search engines was carried out for the twenty Russell Group Universities in December 2010. As described in the post on Trends For University Web Site Search Engines we found that “15 Russell Groups institutions (75%) use Google to provide their main institutional Web site search facility, with no other search engine being used more than once“.

In this case we can clearly see that arguments for a diversity of solutions based on preferences for open source or bundled solutions, ease of management or the distributed architecture seem no longer to be relevant, with a Google solution now being the preferred option.

Will we see simple arguments for diversity in the ways in which institutions provide support for the Mobile Web until we eventually arrive an approach which is used by most institutions? And whilst it may be dangerous to mandate a preferred solution too soon (after all, the majority of search engines used in 1999 are probably no longer in existence) might there not also be risks in failing to engage with mainstream approaches?

Posted in Web2.0 | 1 Comment »