1. Background
A number of respondents who left their details with the survey were contacted about further participation and invited to attend a focus group in their local area. A pilot focus group was conducted with History PhD students and postdoctoral researchers at the University of Sheffield to test the suitability of the topic guide (Appendix 2). This was followed by three focus groups held in London (Courtauld Institute), Sheffield (Humanities Research Institute, University of Sheffield) and York (Humanities Research Centre, University of York). Three to four participants attended each session; a small group allowed for each person to develop their opinions and ideas; therefore discussion was easier to foster. Each focus group was led by two members of the project team acting as moderators; this was audio recorded and subsequently transcribed. Consent was obtained from the participants to record their interactions.
The key themes taken from the focus group analysis were similar to those discovered through the survey and the issues identified in the studies discussed. These are transparency, access and research community. This section will present some of the most interesting observations made by the participants and discuss how they relate back to search.
2. Key Themes
Three key themes were distilled from analysis of the focus group transcripts. Direct quotations will be used to illustrate these, which are as follows:
- Transparency: content and extent of resource; resource definition and contextualisation; expectations and limitations.
- Access: advantages of online research; pathways and layering; browsing; access issues.
- Research community: own knowledge; sharing knowledge; leaving traces; ideals.
3. Key Theme: Transparency
3.1. Content and Extent of Resource
Providing information about the content and extent of the resource is extremely useful as it gives researchers a starting point and it is also important for issues of trust if a site makes explicit what is included or whether something is a digitised form of the original source or not.
“It is important to be aware of what is and isn’t there and what you’re missing and how representative things are that you’re looking at which is sometimes difficult to tell from just the online site even if it’s a very reputable one you can’t quite be sure.”
“Sometimes it matters sometimes it doesn’t but often you may assume that was when it was first published when in fact that’s not the case and you may not need to see the original but it would be important to know that it was not the original.”
This relates in a way to the contextualisation of the source material. If you know absolutely nothing about what you see in front of you and it is difficult to infer anything from the interface then it would be difficult to know how to start engaging with it. Preserving as many of the physical attributes of an artefact as possible was therefore considered a great advantage because they might be part of the questions you are asking about it.
“It’s somehow providing information about what can be searched for or what sort, sometimes it might go back to descriptions of this is what kind of stuff we have in the collections.”
“It’s fine to have the text of this Anglo Saxon bit of writing but for me to get the most out of it I really do need to know which words are at the end of the line because that explains what’s going on in the text … the whole positioning of everything is very useful and important.”
“Out of context the stories are no good at all particularly in a newspaper or magazine it’s got to be contextualised; an article for instance about potato rationing needs to be yeah well because there’s a war in Europe and those sort of things.”
“Just the physical form of a document sometimes is part of what you’re asking about it I mean I look at guide books to zoos and maybe you want to know how big it was how it would feel if you were carrying it round how would you have physically used it.”
3.2. Resource Definition and Contextualisation
Defining a resource and providing enhanced contextualisation can also be achieved by ensuring that what is going on behind the scenes is more transparent, for example, in terms of metadata. If researchers want to investigate how entries have been tagged it might help them to figure out how best to conduct their searches and what permutation of search will be likely to supply the data they are looking for.
“Even though I might not use that all the time I think having access to metadata and what’s behind the scenes is always a good thing.”
“Certainly the value of metadata and more expansive data is highly valuable cos you can never fully kind of pre-empt or envisage what kinds of searches people will want to construct.”
However, it is a fine balance as many people would equally prefer not to have any contact with the inner workings of a website.
3.3. Expectations and Limitations
Scholars have high expectations of what search should be able to achieve but at the same time they would rather know what the limitations of a particular search function areand have that clearly defined from the outset. It seems important to provide hints as simple as what kinds of queries you can type into the search box and whether the search function takes a recognisable form.
“It’s almost like it will be limited. You want to see what to see what the limits are but you don’t want it overly limited.”
“There’s nothing worse than going to a website and thinking I need to look so and so up and you try it and you think I’ve obviously looked for the wrong thing cos I’ve got nothing back.”
“I’ve had cases where I’m not sure what I was supposed to type whether it’s a name of a person or a keyword.”
“[I use] PASE, Prosopography of Anglo Saxon England, and that’s odd because it took me ages to work out how to search for something cos you’ve got buttons and it turns out that if you press them a little drop down thing does come in and you can enter what you want to look for, but it’s not obvious it’s a lot easier if you’ve got a text box and a go button.”
4. Key Theme: Access
4.1. Advantages of Online Research
On a basic level, access could be applied simply to the advantages that online research brings, such as access from remote locations to resources that would have previously been out of reach; this is as a result of being able to spend an hour here and there on research when you do not always have time to visit a library or archive.
“I had a meeting with a Chinese academic and a PhD student from New Zealand and in the area that we all work in which is English medieval history it would have been absolutely impossible for them to have engaged in the kind of work that we all do 20 years ago or even 10 years ago so digitisation has actually transformed things not only for me in terms of the way in which I do my work it’s also made my subject global.”
“Sometimes practicalities maybe dictate what you do as much as the ideal would be to go to the archive but certainly myself I find I use quite a lot of digitised stuff because I can it’s something you can do in an hour wherever you are whereas I can’t always get to the British Library or the National Archive.”
4.2. Pathways and Layering
Pathways through a resource and layering came up again and again. There are certain schools of thought around creating digital resources that argue it is important simply to get as much material digitised as possible; whereas others maintain that it is absolutely crucial to provide apparatus alongside it to enable different ways of engaging with the source material so that its use is not restricted to a very small group of specialist scholars. Instead it is opened up to the wider research community allowing for new research questions and approaches to the data to be developed. What this apparatus should be and how it should function and be presented was something taken forward to the Design Group phase of the project (PHASE TWO). The fact that people think and work in different ways was also raised and highlights why it is important to offer varying methods of navigation.
“It’s not just about my own access to that material it’s about how that’s then going to be used by wider communities of scholarship… the projects that I have been involved with and led I’ve always tried hard to create multiple forms of access to the data so you can have the digitised document but you can also have an English summary of it and you can have keywords for searching or that kind of thing… a catalogue entry, an abstract searching facility, it’s layering… people have different logics so you have to cater for that, more than one process of navigation, and it’s not to say one is right or wrong but they bring different experiences from their background… somehow you’ve got to be able to get at what you want in more than one way not just different words you know.”
4.3. Browsing
Browsing is something many researchers value but feel can be lost in web searching, especially if a resource is vast, because it is not as easy as flicking through a stack of journals to quickly assess whether the content is useful to you or scanning library shelves to see what other books are around the one you came to find.
“You can’t kind of flick through and find things as easily you have to know where you’re getting to I think.”
“If the collection, the dataset, is somehow manageable in size simply having the browser function where you can one by one click through every record might sometimes be useful too.”
4.4. Access Issues
Some associated with ease of access were identified. A lack of context was the main worry, that sometimes it is difficult to reconcile the original artefact with what you see on the screen.
“I think I remember things by the position of the page and sort of quite physically and that’s where I have actually a lot of difficulty with accessing. well not accessing but managing with things online because you don’t have that physical [context].”
“A lot of people working in all kinds of areas are finding more material but there are dangers obviously associated with that and you can lose context but it certainly has made a lot of things possible.”
“I suppose the danger with the ease of access is the fact that it’s just on your screen in front of you, you stop asking some of the questions you would ask if you went to an archive about a primary source.”
5. Key Theme: Research Community
5.1. Own Knowledge
As a scholar, bringing your own knowledge to bear on your research is important. You never really start from scratch. This allows you to find ways around the resources you need to use even if they are not particularly user friendly or the material itself makes it tricky to search.
“I take for granted that I tend to know where to look for things… it’s something that’s accumulated over the years.”
“Huffington Post uses search engine optimisation where they basically trick things into giving them higher ranking you have to find ways to discount that so if you don’t know how a tagging ranking works behind the scenes or if you don’t have confidence in it your research will be flawed and if you ignore it someone else will point out how flawed your research is.”
“Old English isn’t standardised language it doesn’t have standardised spelling or anything like that so you have to be quite ingenious with the spellings that you think will come up again to be sure that you’re going to hit everything.”
5.2. Sharing Knowledge
It would seem that where possible researchers are trying to pass techniques and tips on to their research network, whether that is students they are supervising or colleagues. A common practice is simply asking people you know to answer specific queries as well as placing reference lists and personal bibliographies in the public domain. Social media was also mentioned for keeping up to date with research developments.
“We all know the resources that we work with; we have developed the shortcuts we all know the secret ways of getting to the bits of material that we really want and … I do try to share that experience with my students or whoever, there are all kinds of little techniques and hints you can give people.”
“I just try and arrange a meeting with the people I know in the same way as dissertation students of MA or PhD actually come to me because I may be able to make links they wouldn’t think of by searching.”
“I would have to keep these references myself anyway and it’s only a little bit more work to put them up.”
“Other topics which I’m interested in which is the digital and digital humanities and that mainly happens in blogs and Twitter and you get a lot of information from that.”
5.3. Leaving Traces
It was also discussed how this sharing of knowledge could perhaps be incorporated into digital resources, leaving traces for subsequent users. Various methods were suggested for achieving this, such as keyword tagging and tips on how to use the search function to find very specific things.
“User input is quite important in that users can leave a trace of what they were doing whether that’s by keyword tagging and have a cloud like a keyword cloud or commenting or other types of interaction and contribution for me that’s an important aspect.”
“You could have a Wikipedia sort of set up where you find something that doesn’t have a tag that you think it should have and you add it.”
A good example of this came out of the pilot focus group. One participant was searching for references to levellers in Lancashire. Going through the newsbooks on Early English Books Online would take days and days and so would not be worthwhile. However, just from speaking to a colleague about the dilemma the participant was put in touch with someone who had created a database of newsbook references who could immediately say, “there are 4 references to levellers in Lancashire and this is where they are”. In such a vast body of material this kind of thing is invaluable, especially if someone is prepared to share their work. Once the newsbooks are full-text searchable a keyword search should pick up these references, however, there could still be a place for reference collections and similar shared research findings.
6. Ideals
Participants were also asked about search ideals which ranged from: shortlisting items of interest and providing breadcrumbs to orientate yourself within a resource; saving preferences even if only temporarily; being able to add tags and being provided with something as specific as a good zoom function.
“If you can click on the results you want and then be able to see them perhaps on a shortlist or something cos there’s nothing worse than having to go into the thing you’re interested in then finishing with that coming back out and then you have to reorientate yourself thinking where was I on this page.”
“Remember either your preferences or your searches… some way of sort of saving even if it’s just temporarily… if you could export that just click a button and it would send it to you.”
“I suppose if you could select a tag set that’d be really useful you know a linguistic tag set and then a different tag set for something else.”
Simplicity of resources was also frequently mentioned. There seems to be some conflict here because a lot of functionality is required and expected but if a resource is cluttered and challenging to navigate and use, then it could be off-putting. We have seen in the survey and focus groups that scholars will find ‘work-arounds’ to deal with online resources that they find difficult to use but that they need for their research; however, our goal is to create something intuitive and user-friendly that means users do not need to develop work-arounds in order to obtain the information they are searching for.
7. Summary
The findings from the landscape survey and focus groups have echoed one another and shown that researchers are engaging with search in different ways. The ways in which they relate to search have been identified and from here we could start to think about how these observations might be incorporated into improving search design.
- Transparency of a resource, in terms of contextual information and definition of the dataset, is important in terms of both search and overall site functionality. Researchers will struggle along with difficult resources but that is not an ideal situation. Many participants expressed how important it is to preserve as much of the integrity of original artefacts as possible and include some notion of the extent of the resource. In terms of developing some kind of coherent web resource, this means finding ways to contextualise so that people can search intelligently.
- Access to different types of search is important, i.e. from basic keyword to semantic or linguistic search, as is the ability to search in different ways, i.e. from entering keyword(s) into a search bar to browsing categorised lists. This is not to make search necessarily easy but to make it more accessible to more people at different levels of interest, with different kinds of questions.
- Engaging with the research community through a digital resource was found to be an interesting way of sharing search methodologies, approaches and findings. It could also be a method to make a resource more sustainable by continuing to add value through expert contribution.