The HOTBED acronym stands for handing on tradition by electronic dissemination. As this suggests, the ethos of the project is about handing on a tradition, in this case a musical one, by promoting the active use of networked sound materials. The project is funded under the JISC’s resources for learning and teaching, with three main aspects to our work:

• the creation of a body of digitised recorded sound resources, mostly archival field recordings of Scottish music from the School of Scottish Studies sound archive
• investigating and developing useful software tools for manipulating these audio materials, therefore extending their mode of use beyond simple playback
• evaluating their implementation within a very specific learning and teaching context, namely the RSAMD’s BA in Scottish Music and courses in Scottish ethnology at our partner institution, the University of Edinburgh’s Department of Celtic and Scottish Studies.

The BA (Scottish Music) course offers unusual opportunities to study the use of networked sound resources in the curriculum. Students on this course not only need access to existing archival recordings in the course of their performance and academic studies, but they also make their own field recordings and document them as part of their studies. As students both use and create primary resources, the project is not only evaluating access and use, but will also introduce digital resource creation and archiving into the curriculum and report on the results. This provides a rich agenda to examine the issues of implementation and engagement with digital resources.

The main purpose of the project is not to build collections as such (there are other projects and programmes doing that) but to look at the particular needs of a group of users and how best to serve those needs. This general area of what users want and need to do with networked digital resources is still relatively under-researched. Despite the spreading use of digitisation and networked resources in the cultural sector, we still have a rather limited knowledge of their effects on user groups and the way digital collections affect learning, teaching and research. There is surprisingly little attention paid to how users interact with materials, even though there are considerable amounts of money poured into resource creation. There are of course honourable exceptions, for example the Library of Congress, which did a serious piece of work as early as 1993 as an evaluation survey of its pilot American Memory collections and a current example in the work that SCRAN is undertaking.

However if there is a lack of research and knowledge about how users generally interact with and learn from digital resources, there is even less known about specialised groups such as performance students. The HOTBED project is seeking to learn more about the specific needs of performance students in their interactions with digital materials. This covers such practical but vitally important matters as the location of workstations. HOTBED machines are located in practice rooms and teaching studios, i.e. where performance students work and can make music (they need to sing!), rather than in the library or "computing labs". In the future it is possible that materials will even be located in students’ pockets using portable computing devices. Other differences involve the sort of criteria these students want to search for, and the manipulation of audio materials in a manner useful to them. A corollary of that is discovering what is perhaps not at all useful, such as staff notation (the musical "text").

At this point in time we have now completed the first part of the project by building a system, which is part resource-based and part Virtual Learning Environment, digitising materials and preparing for implementation. We are now ready to go into first major phase of evaluation. Development of the audio manipulation tools is proceeding, but is scheduled for a later version of the system.

Now I shall hand over to my colleague Karen Marshalsay who is the project’s Learning and Teaching Officer and subject matter expert (in addition to various other roles in the Academy’s Scottish Music department). Karen will talk more about the particular needs and characteristics of our primary target audience and how they are accounted for in HOTBED. After this, I’ll end the paper with some of our experiences so far and lessons learnt.

[Karen Marshalsay continues.]

I am going to talk at some length about the users of the HOTBED system – what their background is and what kind of learning environment they have come from. This is important as the system has been designed to meet the needs of 2 very specific groups of users, firstly performance students of traditional Scottish music and secondly students at the Department of Celtic and Scottish Studies here at The University of Edinburgh, along with staff of both institutions.

The learning processes which the students have undergone prior to their arrival at the RSAMD are no different to the way traditional musicians have been learning for many years. The BA (Scottish Music) course is fairly new, as indeed is the concept of traditional music as part of the educational establishment rather than something which exists outside of it, and as yet there is no evidence to suggest that young people are learning or being taught any differently because they aim to study for a degree in it.

Basically traditional musicians learn by LISTENING, WATCHING and IMITATING.

LISTENING, because you need to have the sounds in your head before they are anywhere near your fingers, and for that to happen you need to be exposed to the tradition (although it would be false to think of any such homogenous thing as ‘the tradition’ – Scotland being what it is there are in fact many traditions).

WATCHING, because although we talk about ‘the oral tradition’ and ‘learning by ear’ we actually expect the student to watch the teacher closely and observe their technique. Indeed a good reason for not working from sheet music is that people immediately look at the piece of paper instead, a piece of paper which can never hold all the information needed to play the tune properly and well.

And IMITATING, because, of course, once you have heard it and seen it played, then you try to do it yourself.

At this point let me show you something of the system.

Having logged in with my own unique user name and password, I shall initiate a search for items by our Head of Scottish Music, Brian McNeill :-

PLAY - TO ANSWER THE PEACOCK

The method of learning I have discussed is not confined to Scotland. It is the way in which

traditional music is passed on in most cultures. To quote the ethnomusicologist Judith Becker:

"The term oral tradition has a much wider meaning than simply the absence of an elaborate notation system, just as the term written tradition implies more than just the use of notation. Too often, oral tradition is equated with oral performance, that is playing without written music, or with aural learning of a composition, that is memorization through repeated hearings … or with improvisation on a theme or motif … none of these interpretations describe the practice of the musician in an oral tradition who has mastered a technique of composition which allows him to perform and compose at the same moment … acquiring this technique is a slow process beginning in childhood, unconsciously internalising aspects of the tradition, later beginning to learn a specific instrument, learning the formulas which he will imitate, vary, expand and rearrange for the rest of his life … The final stage of development of a musician comes when he has internalised the underlying forms of his formulas and their proper places within a composition. From then on he is free to use them creatively within a performance. The best musicians rarely repeat themselves precisely …"

That quote, which aptly describes the learning process I began with, could so easily be about Scotland, but is in fact describing Javanese gamelan musicians. Nor is this methodology confined to traditional music, there are many similarities to what Shinichi Suzuki called ‘the mother tongue’ approach and developed into the Suzuki method of teaching classical music.

So, prior to their arrival at the Academy the students will have had one to one tuition in their main instrument and possibly in their second study as well. This is likely though by no means certain to have been regular. It has probably happened outside the school system with the exception of some pipers who may have been lucky enough to live in an area with a schools Piping Instructor, and some fiddlers who may have also had some classical violin lessons in school. Of course as the educational establishment develops for traditional music this will change, and indeed is already doing so, particularly with such ventures as the National Centre of Excellence in Traditional Music based in Plockton High School.

So, what happens in one-to-one lessons? There is of course an emphasis on good technique, tone and musicality, just as there would be in classical lessons. The main differences would be the stronger emphasis on individuality and the student’s own decision making abilities. A clarsach (Scottish harp) player for example needs to decide how they are going to finger a piece, what decorations they are going to add and when they would be appropriate, what chords and rhythmic emphasis they are going to add with the left hand. All of this is part of ‘arranging’ the piece. They may even have had to initially decide what to play. The passing on of repertoire and the ability to search out tunes to add to your own repertoire is an important skill. Players are always looking out for a good tune, whether they get it directly from someone else, or indeed from what is now a very firmly established method, learning it from someone else’s CD.

Environment is a huge factor in the creation of a good traditional player and cannot be over emphasised. One of the African musicians I have worked with used to point out that an African child is not born with rhythm – but is born into it – they are brought up in a culture and tradition. It is the same thing with our Scottish music students – they weren’t born knowing the difference between a strathspey and a reel – the knowledge is gained through exposure – having heard it again and again and again, having asked questions and thought about the answers, having listened while others, especially elder and respected musicians are playing and talking. Constant exposure to a tradition is what can make you a part of it.

HOTBED will become part of this environment at the RSAMD in the coming academic year. One of the things we must consider is why these students choose to come and study with us. They are all already players of a high standard, and indeed many of them are out there gigging on a professional basis even before they arrive at the RSAMD. Obviously there is the opportunity of 3 or 4 years instrumental and vocal tuition of the highest standard. But the main thing that the Academy course can offer is a chance for students to expand out from their already strong base to learn more about other instruments and styles, from the many Scottish traditions. To expose them to the Scots language if they are from the Highlands and Islands, or to improve their Gaelic, to let a Borders fiddler hear and learn something of Shetland style, or introduce a North East ballad singer to port a beul. There are also the historical and analytical aspects of the course, but above all students are encouraged to keep a healthy respect for what has gone before. I remember being told that it is like using your rear-view driving mirror – it’s not safe to keep going forward if you don’t keep looking at what is behind you. In this case the interviews with older players and the chance to learn directly from original field recordings are incredibly valuable. They are the next best thing to actually being in the room with that person, and as some of these people have now passed on, these field recordings are even more valuable.

HOTBED is providing access – easy, convenient, quick and infinitely repeatable. Digital media and databases mean you can search for what you want and and listen to it as many times as you like. As a traditional musician involved, as most are, in both performing and teaching, this kind of digital archive is the most exciting thing to happen since the first fieldworkers went out with the heavy Uher and Nagra reel-to-reel tape recorders. When you think of the time it would take to just to locate the item you want on a cassette tape, never mind having to keep rewinding it to hear it over again, instant access at the click of a mouse seems almost magical by comparison, and that’s not even mentioning the absence of geographical boundaries.

Let me show you some of the searches our students are most likely to do.

You can currently, and this is only version 1, search on title, tune type, performer, composer, instrument and language.

Search under language - SCOTS – play Jock Duncan Cutting Peats – open up transcription window.

Search in title for BROSE – play both versions: J Scott Skinner’s 78rpm recorded in 1910, then fiddle tutor Iain Fraser’s version.

The possibilities lead us toward a learning environment. Users can group items of their choice into favourites lists, they can email those lists to other users. Staff can make their lists public – in effect creating the equivalent of an online assignment, grouping together tunes that the students have to listen to and learn for their next class.

HOTBED makes comparative listening very easy.

Search in title for YARROW - various versions of the ballad The Dowie Dens o’ Yarrow.

But, and there are buts, all of this is still within the computer. The user needs to be at a terminal, and preferably with a fast connection. I mentioned earlier about the importance of getting a tune in your head as the first stage of learning. Alison McMorland, one of our Scots song tutors, makes a tape which she gives out to each student of the songs they are to learn. She encourages them to listen to it constantly, but especially last thing at night and first thing in the morning. This advice is just as pertinent for instrumentalists. I have learned and been taught many tunes simply through singing them over and over again. Pipers’ canntaireachd is a highly developed form of this. Indeed as one of the interviews on HOTBED tells us many of the older pipers would still hear the canntaireachd being sung in their head even while actually playing the pibroch on the bagpipes.

You can’t have a computer with you all the time, the way the students always seem to have walkmans or more recently minidisk players with them. But you could have an mp3 player for example, and that, together with developing the use of video, would be a logical step for HOTBED to pursue – to allow students to download the pieces they want to learn and access them whenever and wherever they want. And on that logistical note I’ll hand you back to Celia.

Let us look at some of the practical realities and the difficulties we have faced, many of which are not unique to this project. I’ll highlight in particular metadata creation and when - in an institutional sense, small isn’t beautiful.

Small is not ideal … We not only work in a very small institution, but in a very small team, with only one full-time staff member (neither of us!). One problem is keeping our focus firmly in mind. The digitisation and building of a collection to work with is, of course necessary, but where this project’s focus should be is in facilitating usage and supplying useful audio manipulation tools. Ideally it would have been better to work with a ready-digitised collection of audio materials. Our focus is on usage, on pedagogy, and on users’ needs, but it has been hard to keep that focus sharp in the early stages when we had to build a system and populate it before doing anything else.

In framing projects it is all too easy, especially for those of us not from an information management background, to minimise the difficulties encountered in digitising from scratch when the original materials are not held within the institution, and when we have an imperfect knowledge of the source cataloguing with no automatic transfer of catalogue information from that source. All data has to be re-keyed and examined critically to see if it is what we need for our specific aims, and we underestimated the resource needed to accomplish that.

The nature of the sound materials in HOTBED (especially the archive fieldwork recordings) and the requirements of the users have made metadata creation particularly problematic. This partly stems from the nature of the first tranche of materials to be digitised which are field recordings held in the sound archive of the School of Scottish Studies. Rather than discrete musical tracks, much of the material mixes interviews recorded with fieldworkers and performances (e.g. when informants break into song to illustrate their points). Ideally, the name of the song or tune embedded in the interview should be recorded in the metadata to aid discovery for our users, but this is extremely resource-intensive.

Deciding upon the degree of necessary compromise on metadata description will be a problem familiar to many other digitisation initiatives and will be part of the scope of the evaluation on HOTBED. Other familiar challenges are working with a standard scheme like Dublin Core alongside non-standardised indexing schemes that have evolved for fieldwork material. We don’t need to rehearse the "linguo-centric" difficulties of fitting sound and moving image materials to Dublin Core, but it is perhaps worth noting the specialist nature of the cataloguing from the School of Scottish Studies sound archive when, again, our musical orientation may well require a different focus.

One of the evaluation objectives will be to look at how much metadata is a) necessary and b) desirable, to add value for this audience. Here is an example of the type of searching our students might want to do. Let’s suppose we have a fiddle student who has just been taught a stunning strathspey in B minor. They might want to search the system to find a suitable B minor reel to go into after the strathspey and then finish off a set (say for a gig or a performance exam) with a reel in D. To enable them to do that we would have to provide a very rich descriptive metadata set. Our initial speccing of metadata requirement to include keys and metre has proved impossible to carry out for all items, but on the advice of the project’s Advisory Group we will compare very rich description with rather more workaday levels of description and report on the outcomes.

Small institutions also have their own culture and infrastructural realities and although we have had very strong support and encouragement from the most senior levels, there are some difficulties. The staffing profile of a monotechnic (and this also holds good for art schools) consists mostly of part-time practitioners, whereas for a project like this a base of full-time teachers who are on hand all the time and able to commit fully is the ideal. We also work with the infrastructural realities of a very small in-house computer support team which is not geared to research projects and the sort of demands that HOTBED makes. For example, we have no web email yet, and it was only last year that computer access was made available to part-time staff. Indeed it was only the year before that a 20-seat computer lab was installed for students. Outside consultants do our networking and firewall issues impede smooth technical implementation, but even then we are still further ahead of most other UK conservatoires. This situation forms a stark contrast with most digital projects sited in universities.

However there are also positive lessons that may be learned from HOTBED, particularly for wider audiences. The project is generating data on user behaviour and the needs of users which might lead and inform decisions about digitisation and the presentation of digital sound materials that could be more widely applied. The creation of audio manipulation tools in a streamed networked environment is an important of the project. We know from previous work that users require more than just a simple playback facility, they need to manipulate their sound material in the way that digital text users can manipulate their materials. We are hoping to research the creation of such tools in a streamed environment and are working with software partners on this, which could help a potentially large market in distance learning. And finally, HOTBED has a contribution to make to passing on cultural heritage and to enhance our knowledge about the part digital materials can play in that process.