The Paid:Content:UK blog has recently published an article which informs us that “A Fifth Of BBC Sites Are Already Dead“. The article begins by annocing that “Nearly half of the websites most likely to be closed as part of its big Strategic Review have already long been shut, some for as much as eight years“.

A list of a number of the sites which have been ‘mothballed’ is given in the article. Some of the sites are for programmes that have ceased broadcasting (eg. On The Record) and others are for events which are now over (e.g. Politics ‘97).

I was particularly interested to read about the BBC policies regarding the decommissioning of such Web sites. The article provide a link to the BBC’s policy which describes that inactive pages are left online for reference as “We don’t want to delete pages which users may have bookmarked or linked to in other ways. In general, our policy is only to remove pages where the information provided has become so outdated that it may lead to actual harm or damage.”

With the promises of large cuts for public sector organisations in the offing after the general election I suspect that we will find Web sites in many higher education origanisation being decommissioned.  But will  content be simply deleted, will the content be left ‘as is’ or will a more manged approach to such decommissioning take place? 

I feel there will be a renewed interest in the decommisioning of Web sites.  I hope the JISC PoWR’s Handbook on the Preservation of Web Resources will be of interest to organisations which find themselves  in this position.

An article published today on the Guardian Technology Web site entitled “Geocities: dead but not lost” describes how Geocities, which was founded in 1994 and was at one stage the third most-browsed site on the web, is now dead.

We discussed Yahoo’s announcement that the Geocities service was to be shut down some time ago in a post entitled ““Seething With Anger” at the Demise of Geocities“. What I find interesting in the article is the information that “… there’s the real effort, by the Archive Team, who have been trying to archive as many Geocities pages and sites as they could“.

I’d not come across the Archive Team wiki before. They describe themselves as a “project composed of volunteers, currently coordinated by Jason Scott” which invites.

• Writers, who can create clear essays and instructions for archivists and concerned parties.
• People with Lots of Hosted Disk Space who have a proper hosted webserver and fat pipe, who are willing (when asked) to consider hosting mirrored dead sites or archives.
• People who love setting up torrents who can do the same as the mirror folks, but do so hosting torrents.
• OCD-rich individuals who want to download things who will respond to our alerts and call outs and download entire sites or diagnose ways to get at obfuscated data.

The wiki home page informs us that “This website is intended to be an offloading point and information depot for a number of archiving projects, all related to saving websites or data that is in danger of being lost. Besides serving as a hub for team-based pulling down and mirroring of data, this site will provide advice on managing your own data and rescuing it from the brink of destruction.”

Hmm. I wonder how effective a volunteer organisation is likely to me? My initial thoughts were fairly sceptical, but other volunteer-led initiatives, such as Wikipedia, do seem to be successful. What are your thoughts?

A news item on The National Archives Web site has recently announced a new study on “Web Archives: Now and in the Future“. This study, which is funded by the JISC and will take place in collaboration with the UK Web Archiving Consortium, will look into how archived Web sites are collected and made available to users.

The study aims to:

• Investigate how UK Web archives are delivered to users now, and how they might be delivered in the future
• Define the long-term historical and research value of online content in the UK
• Look at different organisations that collect Web archives, and their interests

The study will run until late July 2009, and the results will be published on The National Archives and UK Web Archiving Consortium Web sites in August 2009.

We’ll published details on the availability of the study once it is published.

A blog post entitled “The Death and Life of Geocities” has been published recently on the Adactio blog by Jeremy Keith, a Web developer living and working in Brighton, England. In the post Jeremy describes how he is “seething with anger” but then goes on to add that “I hope I can tap into that anger to do something productive“. The reason for the anger is his concern that “Yahoo are planning to destroy their Geocities property. All those URLs, all that content, all those memories will be lost …like tears in the rain“.

Although in an update to his post Jeremy does admit that “no data has been destroyed yet; no links have rotted” and that his “toys-from-pram-throwage may yet prove to be completely unfounded” Jeremy is right to raise concerns regarding the recent announcement that “Yahoo [is] to shut down GeoCities“.

Some people, as illustrated by JR Raphael’s article in PC World entitled “So Long, GeoCities: We Forgot You Still Existed” are not losing any sleep over GeoCities demise whilst others, such as the Online Lunchpail blog feel that “the demise of GeoCities … proves my point that the U.S. government never should have approved the takeover of GeoCities by Yahoo!“.

From my perspective I feel that the concerns raised by Jeremy Keith (who, it should be pointed out, is a professional Web developers) will become more widely appreciated as ordinary Web users, who might have used the first generation of public-facing Web-hosting services such as GeoCities for their initial simple Web development activities, realise that their may be sentimental attachments to one’s early work – just as I regret having lost my scrap book from primary school (I remember writing “When I grow up I want to be a Beatle, sing ‘She loves you, yer, yer, yer’ and earn £100 a week“). And what of the social historians – have we lost our cultural memories of the initial take-up of the Web outside of the universities and business sector?

In a blog post by Jason Scott on the ASCII  “weblog of computer history, punditry and trivia” Jason describes the efforts being made to preserve content published on GeoCities. But Jason admits that

“I can’t do this alone. I’m going to be pulling data from these twitching, blood-in-mouth websites for weeks, in the background. I could use help, even if we end up being redundant. More is better. We’re in #archiveteam on EFnet. Stop by. Bring bandwidth and disks. Help me save Geocities. Not because we love it. We hate it. But if you only save the things you love, your archive is a very poor reflection indeed.”

What is to be done? Should the digital preservation for the general public’s digital heritage (as opposed to an institutional digital heritage) be left to volunteers? Or will future generations regard us as having failed in our responsibilities as previous generations failed to preserve the built environment and left us with the soulless shopping centres and high-rise building which were developed during the 1960s?

The dangers of reliance of externally-hosted Web 2.0 services has been mentioned previously. And there have been recent incidents in which companies have given a short period of notice of impending closure of services, with users having little time to migrate their data to alternative providers. A recent article in The Guardian (Thursday 2 April 2009)  entitled “Can I assume that my online data is safe for ever?” addressed such concerns in an article on the closure of the Filefront.com service, who gave their users just 5 days to migrate their data.

Coincidentally I recently received the following email from a service I subscribe to:

Our previous request to you to provide a new owner for the  list has not produced a response.  Therefore, we assume the list is no longer useful and aim to close it in one week’s time.
We would be happy to provide a zipped copy of the archives and any files on deletion of the list, should they be required.

In this case it appears that the service has been little used for over a year. And yet what if useful information is still available on the service? Is a week’s notice enough for users of the service to consider the implications of this decision, identify appropriate solutions and then implement them? And let’s not forget that this email was sent outside of term time when researchers could be away.

The email did not make it clear if data was to be deleted, the service was to continue to be made available in a read-only mode or the interface to the data hidden – all possible solutions if it is felt necessary for a little-used service to be withdrawn.

There’s still a need to establish the best practices when Web-based interfaces to services are to be removed, I feel. And such issues do not just affect the third party services outside of our community.

The JISC-funded TASI (Technical Advisory Service for Images) is no more. This service, which is based at ILRT, University of Bristol has been reborn as JISC Digital Media, with an expanded remit for supporting digital media in general and not just images, which was the focus of the TASI service. Further information is available on the JISC Web site.

This change has been accompanied by a new domain name – http://www.jiscdigitalmedia.ac.uk/ rather than http://www.tasi.ac.uk/.

Now the TASI service provided many useful resources on best practices for digitisation.  But what has happened to links to these resources? Will we get a 404 error message? Or, even worse, will we get a message saying the domain no longer exists?

The QA Focus briefing document on “Improving The Quality Of Digitised Images” contains a reference to a Digital Imaging Basics resource which was available at the URL <http://www.tasi.ac.uk/advice/using/basics.html>. Following the link takes you to the resource, which is now available at <http://www.jiscdigitalmedia.ac.uk/advice/using/basics.html>.

There seems to have been a simple mapping of resources from the TASI domain to the new JISC Digital Media domain. And as the original resource has ‘cool URIs’ (i.e. they had no dependencies on a specific technology (such as a CMS, Java server pages, etc.) it was technically not a difficult task to migrate the links to the new domain.

Well done TASI / JISC Digital Media. The challenge now is to see how long such redirects will continue to function.

Lorcan Dempsey picked up on the work of the JISC PoWR project in a blog post entitled The institutional record and web archiving. Lorcan described the presentation given at the first JISC PoWR workshop by Alison Wildish and Lizzie Richmond in which they described the changes to the University of Bath printed prospectus over the lifetime of the University of Bath.  Lorcan drew parallels between this print publication and the digital environment:

“The University would always have kept the print manifestation; what now to do with the web manifestation? One of the interesting changes they note over this time is the ‘rise of the logo’, and tracing changes in how the institution presents itself over time is also interesting.“

In a response to Lorcan’s post Tony Hirst referenced a blog post by Michael Nolan on the Edge Hill Web Services team blog in which Michael pointed out “one [example of interesting use of RSS] that caught my eye was the University of Warwick’s recent changes feed which allows you to subscribe to find out when the homepage changes. Better still, they have this for every page in their CMS.”

An example of this can be seen for the Research page on the University of Warck Web site. Although not nornmally visibile to most end users who visit this page, there is a link to an RSS feed of recent changes to the page. Using tools such as the Greasemonkey RSS Panel (available for Firefox) you can view the changes, as shown below.

In his comment on Lorcan’s blog Tony Hirst went on to suggest that “A change feed, like on a wiki, could be one way (maybe) of facilitating 1st, 2nd or 3rd party web page archiving?“. I think Tony might be right. And maybe we are seeing the University of Warwick pioneering this approach, as the feed of recent changes seems to be provided by their in-house Sitebuilder 2 software, “the University’s web publishing tool“.

Perhaps when institutions are next procuring a CMS system they should be asking if vendors provide RSS feeds of changes to pages.  

A recent blog post by Lorcan Dempsey on “The institutional web presence again” provided a link to a page on “the history of U.Va. on the web” which provides details of 14 year’s history for the University of Virginia’s Web site from 1994-2008.

The page provides details of the Web usage statistics in the early years, with screen images shown of major changes to the home page from 1997 (unfortunately no screen images are available for the first three years of the service).

Information is provided on the people and groups responsible for the design, the changes which were made as new technologies became available, significant additional content that was added and details of awards which the site won.

This is an approach which I feel all institutions should consider taking.  And let’s start recording the history of those early years quickly, before the first generation of institutional Web managers start to retire, leave or forget the details of the institution’s Web history.

I recently posted an article about the history of the University of Bath home page which included a link to a display of versions of the home page, based on data taken from the Internet Archive from 1997-2007.

Andy Powell, a former colleague of mine who used to work at the University of Bath, posted a Twitter message in response to my post in which he said:

@briankelly all pages prior to http://tinyurl.com/47pydq were mine – that’s what web design was like back then! – but all records now lost

03:56 PM June 18, 2008 from twhirl in reply to briankelly

But although formal records of the decisions made related to the home page (its design, the content, the links and the technologies used) may have been lost (or perhaps not even kept) I do wonder whether it may be possible to document such history based on anecdotal evidence from those who were either direectly involved with the decision-making process or perhaps who observed the results of the decisions.

From the museum’s sector and the experiences of The National Archive (with the public Wiki service) we know that the general public does seem willing to provide anecdotal information on resources such as old photographs.

This approach seems to reflect some of the discussions held at the first JISC PoWR workshop. As described in Ed Pinsent’s summary of the event “there was a lot of ‘folk memory’ and anecdotal evidence, also sometimes called ‘tacit knowledge’“.

Would it be possible, I wonder, to provide access to images of an institution’s old Web pages and, though use of social networking technologies, encourage members of the institution (and perhaps the wider community) to document their recollections of the Web site?

The JISC PoWR is concerned with ensuring that Web sites and their content don’t disappear. Right? Actually this would be to misunderstand what Web site preservation is about. Sometimes there may be a need for Web sites to be deleted. Indeed there may be dangers (both in terms of brand management and legal issues) if the content of Web sites outlasts its welcome.

Take, for example, the Web site for the National Open Centre, which is illustrated.

If you visit the Web site you will find a nicely designed and easy to use Web site for the National Open Centre (NOC) which is a:

“national policy institute, a think tank to understand and articulate strategies to make effective use of Open Source Software and Open Standards (OS&S) for the benefit of all. It will focus on nationally relevant issues leading to proactive strategies to ensure that the UK effectively exploits the opportunities that arise with OS&S. The NOC will be independent, strategic and proactive and seeks the participation of interested and informed people.”

A very worthy organisation, it would seem (and I should add that I was a member of the NOC’s Advisory Group and attended the first meeting). Sadly, despite having a launch event at the Houses of Parliament, the NOC was unsuccessful in its attempts to gain funding, despite having a launch event at the Houses of Parliament. To paraphrase the Monty Python sketch “the NOC is not resting. The NOC is no more! It is bereft of life. It is an ex-NOC!“.

But this isn’t what you’d think if you explored the Web site. The home page urges visitors to “Get Involved!” and describes how it has “established the first set of subject panels. The topics being researched/discussed are: Public procurement, Open Standards and Open Source/Open Standard for SMEs. The Get Involved page then encourages visitors to participate with the NOC in a number of ways, including joining the Advisory Board, Subject Panels or the NOC Community. The only subtle indications that the NOC is no longer operational are the dates on various pages (206 or 2007) and the broken link to the NOC’s wiki from the Get Involved page.

The failure to provide any indication that the NOC failed to receive funding may be embarrassing to the partners of the service, which are list on the home page. But as well as such possible embarrassment, what would happen if visitors arrive at the Events page and read details of the one-day event on Document Standards planned for 4 July, which is illustrated below.

There is no indication that this refers to an event which was planned for 4 July 2007.
And there are no details about registration, although a location for the event is given (NCC offices, London). What might happen if someone travels to London to attend the workshop (which covers interesting aspects related to open document formats, with apparent participation from companies such as Microsoft). If this happened, I’m sure the potential participants would be pretty upset to discover that the NOC folded last year.

This is, I would agree, unlikely to happen. But what if the information about the event had been held on one of the NOC’s partner organisations, such as Birmingham City Council?

This example is taken from the wider public sector. But within the higher and further education sector, with short term project funding provided for much development work, institutions may find themselves in a simple situation, with the intentions of a project team failing to be realised due to a failure to win funding, and perhaps a loss of project staff.

How should this possible scenario be addressed? This is something to be addressed in future posts, but for now your comments and suggestions would be welcomed.

In response to a post on ULCC’s DA Blog Chris Rushbridge, director of the DCC (and contributor to the Digital Curation Blog) commented:

The enthusiastic way in which web-site owners “re-brand” or “re-launch” their web-sites suggests that they are not particularly interested, long-term, in the details of the experience; continuous improvement means continuous discarding. One hopes that they are more interested in the information content, in some more abstract sense. Maybe we could measure this by tracking older pages across re-launches?

Perhaps a measure of commitment to the “look and feel” might be the lifetime since last reorganised?

Is this right? Don’t Web site owners care about preservation, preferring instead to continually add new features to their services?

I have to say that I disagree. Rather than continual changes to Web sites due to the Web site owners’ enthusiasms, I would argue that such changes usually occur in response to user needs and expectations, the growing importance of Web services (which mean that institutions have greater expectations of the services which will be provided) and an increasing understanding of the limitations of approaches taken to Web site development in the past.

One example of this has been the obligation (for legal and moral reasons) to enhance the accessibility of Web resources. Initially HTML authoring tools and Content Management Systems (CMSs) provided little support to enhance accessibility – indeed many CMSs generated low quality HTML which could not be processed by assistive technologies. Continue reading →

Exploit Interactive was an e-journal which was funded by the EU’s Telematics for Libraries programme. Nine issues of the journal were published between May 1999 and October 2000.

After the project funding had ceased, additional funding from the EU was obtained to publish a new e-journal known as Cultivate Interactive which was launched in July 2000. However there was a need to define a policy and accompanying procedures for the preservation of the Exploit Interactive Web site and content.

As described in a case study document on Providing Access to an EU-funded Project Web Site after Completion of Funding the following policy decisions were taken:

• The Web site’s domain name will be kept for at least 3 years after the end of funding.
• We will seek to ensure the Web site continues for at least 10 years after
  the end of funding.
• We will seek to ensure that the Web site continues to function, although we
  cannot give an absolute commitment to this.
• We will not commit to fixing broken links to external resources.
• We will not commit to fixing non-compliant HTML resources.

The case study went on to measure the disk storage used by the Web site and to quantify the costs. The storage requirements of less than 500 Mg of disk space were not significant so it was agreed to continue to pay for the www.exploit-lib.org domain name until at least October 2008.

Periodic automated links checks were carried out on the Web site to ensure that the internal links on the Web site continued to work.

A similar process was established for the Cultivate Interactive e-journal, when its funding ceased in February 2003. Continue reading →