NINCH guide home interview table of contents previous interview next interview
Vikki Fenner, the Project Manager for Images of England, and Phil Turton, the Technical Manager, were interviewed by HATII on February 16 2001. As part of the National Monuments Record, the Images of England project aims to encourage knowledge and enjoyment of English Heritage sites. The project was therefore undertaken in order to provide public access to the archives as well as documenting and preserving sites of historical and cultural value. Through this project, the NMR is building a digital library of photographs of 370 000 listed buildings in England.
Images of England is managed by the National Monuments Record (NMR), the public archive of English Heritage. The project has mobilized volunteer photographers, the majority of whom are members of the Royal Photographic Society, to capture an image of every listed building in England. An initial pilot was carried out in 1996. They received the first phase grant from the Heritage Lottery Fund (HLF) in August 1999 and have a completion date for October 2002. The practice of listing buildings to preserve and protect them for future generations began in 1947 in Britain when the importance of preservation was realized after years of wartime destruction. Listed sites include palaces, castles, churches, country houses, as well as garden sheds, fishponds, pigeon lofts, mile posts, army camps, telephone boxes, pigsties, post boxes, and lamp posts. A prototype is now available on the website and the whole database will be available on the Internet by 2002. The project will generate around 370,000 new images. The aim is to create a ‘point in time’ record of English heritage, which will include buildings from every town and village.
They did not carry out a collection survey, as they already had a complete database with information about all the listed buildings in England from the NMR and so knew what they had to digitize. Setting priorities for digitizing also works quite differently in their case, as the staff are all volunteers. It is therefore the responsibility of the volunteer photographers to select the ‘defining’ view of the listed buildings in their area, following the training and guidelines set by the project team. Due to the voluntary nature of their work, the project team no longer gives them targets on how quickly they should process material. They tried using strict deadlines in the first year, but this proved unsuccessful, so they changed the approach in response to feedback from the volunteers themselves. There are about 400 volunteers who receive a list of ‘target’ buildings in their area, and the time commitment to the project and schedule planning is up to them, although they are generally encouraged to commit at least two days a month to the project. They used the pilot project loosely for planning purposes, to estimate for example the rate at which material was being photographed.
One of the obstacles to both planning the development and building the digital deliverables, was that they had not anticipated that they would need a database in place to manage and monitor the survey, so they had to design it themselves after the start of the project. There was a time lag between putting in the HLF application and receiving the money. In addition, the grant was received in two stages, with a delay between them. The organisations did not have experience working with so many widely distributed volunteers in such a concentrated way and faced initital challenges because of this lack of experience. In practice it also proved quite resource-intensive, as they had more volunteers than they originally anticipated. Technology was not a problem in their case. In fact, they benefited from starting a little later, when technology was established in this field.
The project will create a ‘defining image’ of the exterior of each listed building in England. This needs to capture the essence of the architectural character of the building and something of its function and context. Photography is only taking place from publicly accessible land, unless the owners have given permission to photograph from privately owned land. The remaining selection and prioritization questions in the instrument do not apply in their case.
There was wide-ranging regional and national co-operation with other organizations including archives (internally, as part of English Heritage), libraries, academic institutions, and corporations (as part of the consultation process and at the road shows organized across the country), foundations and charities, and government agencies. This experience showed that in any collaborative project involving volunteers, the organization should consider the population to draw upon very carefully.
The aim is to create a ‘point in time’ record of English heritage, which will include buildings from every town and village. Between 1999 and 2002 volunteer photographers, the majority of whom are members of the Royal Photographic Society (RPS), will capture a single defining image of the exterior of every listed building in England. These 370,000 images will then be linked to existing listing information about the buildings, creating a digital library of England’s architectural heritage.
In order of priority, the project manager would say that the project’s purpose focuses on the following areas:
A statement of intent was produced with the application to the HLF. They also state the aim of the project on the website. This was explicit about the project’s rationale, the scope, the significance, the primary audience, the long-term sustainability, and the suitability for different target audiences.
The program digitizes only 35 mm color photographic negatives of all the listed buildings (so the entire body of the material in that sense). The digital deliverables are intended to be re-used. After the whole library is created, there will be spin-offs; for example, they might produce educational packs, both hard copy and digital, as well as sell copies of the images commercially.
The digitized images will complement the existing database of listed buildings. This uses various standards for cataloging: e.g., BS76666, NGDF (spatial data standard), and ANSI Z39.50. The information that the project is adding includes fields such as the name of the photographer, the date photograph was taken and the direction the camera was facing. They have standardized the way these should be recorded.
They did not look at existing guidelines for digitizing particular document and object types at the initiation stage, but did so later. For example, they looked at TASI, the Joint Information Systems Committee’s (JISC) Technical Advisory Service for Images, and at the Visual Arts Data Service (VADS), which is part of the Arts & Humanities Data Service and is funded by JISC and the Arts and Humanities Research Board in the UK. They also talked to several people with imaging expertise. They did not identify a particular standard or a specific way of organizing things that would apply to the whole project. There were no guidelines that they rejected.
The primary intended audience for the digital deliverables was the general public. The secondary audience was the whole spectrum of teaching and learning, and third the subject specialists, such as archive users and conservation professionals. At the moment the website is used more by the general public.
Extensive evaluation of the targeted audience was carried out between the pilot and the main application. This was both qualitative and quantitative and included the educational users (for example, sessions with teachers about how the material from the project could be used in conjunction with National Curriculum areas) and the general public. Market research was carried out by a professional company.
They are currently looking into ways to cater for those with disabilities.
Where there are limitations on the use of the digital deliverables, these are stated clearly by disclaimers on the website (e.g. use of text from the Stationary Office).
They did not use external advice or consultancy on managing the digitization project, as there is sufficient expertise in-house. The National Monuments Record and English Heritage have considerable project management expertise and a long track record of managing digital records.
Eighteen new members of staff were hired for the project, and one project manager was re-deployed. They have adopted a very flexible approach based on team work and have cut a lot of red tape. The project has had an influence on the wider organisation in various ways. For example, the project has put more material on the web than any other part of the organization. Considerable knowledge has been gained on digital imaging and on web-enabling datasets, which has proved useful in other areas of the organization. Volunteer management was also observed with great interest by the rest of the organization, since it involves a different way of working. On a very simple practical level this meant, for example, that they had to design a different, much simpler expense form than their usual one, for the volunteers to fill in.
Project management procedures include: a Project Board, which meets at least quarterly; a Project Assurance Team and IT Assurance Team; and a User Group. The latter includes carefully selected members of staff to represent different parts of their audience, such as experts from architectural history or from communications media. This has proved very useful, even though its management has not always been easy. They started public access trials in June of 2001. They also have regular meetings with two HLF monitors, a lead monitor and an associate IT monitor. They use PRINCE II as a project management tool – this is probably one of the most PRINCE-managed projects in the organization. Although PRINCE can be overly involved in documentation, it is generally excellent for planning. Both the project manager and the IT manager are PRINCE-trained.
Digitization is outsourced. They have contracted Ewan Associates Limited, which they selected after an 18-month long, European-wide procurement process. The company sub-contracts different aspects of IT tasks related to the project, such as data capture and web design. The reason for selecting the option to outsource was that there were no resources in-house and they did not want to invest in the type of equipment and specialized staff that the medium and volume of this project required.
Each volunteer photographer receives an addressed pre-paid envelope and a form to fill in for every film they are given from the project. After they finish a film, they post it directly to the company doing the data capture who print the photographs and send them to the volunteers. The contractor scans the negatives, puts them on CDs and sends the digitized images to the project offices. All digitized images are checked for both photographic quality, metadata, and scan quality (pass/fail). If they fail the scan quality, they have to be re-scanned. After digitization and quality assurance of the images is complete, the negatives are also returned to the photographers. Because of the type of contract that they have, there is a set deadline, set budget, and set quality requirements.
They carried out a pilot study to provide them with information to assist project management and planning. This provided information for scheduling (for example, giving them a better idea of the different progress and cost of photography of urban, suburban, and rural areas), training needs, technical feasibility, technology forecasting (e.g. the web was very new when they started) and to some extent also for workflow analysis and piloting. User needs’ analysis was carried out after the pilot and prior to the application being made to the HLF. After the pilot study, the design of the project was scaled up. Time and motion studies were part of the initial pilot, which provided them with information about the rate at which the material could be produced and the number of volunteers needed. The project team itself took photographs of buildings as a small pilot, which informed the process of providing guidance to the volunteers and was very useful. They use Gantt charts, staff charts in Excel, flow diagrams, resource leveling. They produce job descriptions (where there was some flexibility in adjusting job descriptions from the original HLF application) and an annual list of targets to be achieved by each member of the team as part of performance-related pay.
The external contractor uses Sony film scanners for digitizing the negatives. The project established a set of specifications for the quality of images that is acceptable and guidelines for scanning. Although they use grayscales and color charts, these are more suitable for controlled indoor conditions. It is very difficult to apply them in practice, because of the huge volume and variability in lighting conditions and exposure when the photographs were taken. They try to come up with a compromise of color balance correction applied to the images for acceptable quality, as they cannot color-balance each image individually.
Apart from the volunteers, external contractors, and collaborates, there are 19 members of staff who work full-time on the project. These are:
|
Type of Staff |
Number |
% of time on the project |
|
Director |
1 |
2.5 |
|
Project Manager |
1 |
100 |
|
Technical Manager |
1 |
100 |
|
Image Selectors |
8 |
100 |
|
Finance Officer |
1 |
100 |
|
Office Manager |
1 |
100 |
|
Office Support Staff (necessary for all HLF and internal paperwork as well as supporting the survey and communications work) |
1 |
100 |
|
IT contract administrator |
1 |
100 |
|
Communications Officer |
1 |
100 |
|
Survey Coordinators |
3 |
100 |
Only the project manager was re-deployed from another area of the organization. The background and profile of most staff is professional (financial, press, architectural, project management, IT and GIS, architectural history and archaeology). They design a training plan for each member of staff each year. The NMR is part of the Investor in People scheme (the UK initiative which encourages use of standards in training and other areas of personnel development and regularly monitors organizations to see whether they still qualify to be part of the scheme).
Areas where training needs were identified included: managing and appraising staff; presentation skills; interviewing; architectural interpretation; use of post digitization and other software (e.g. Photoshop, DTP packages, MS project); handling of the media and dealing with customer complaints. They have a system of shared learning regarding architectural history, whereby members of staff give presentations to colleagues about their specialized area. All members of the project team are eligible for training. This was organized in a variety of ways: in-house, using project staff, external consultants, by attending external courses, with independent study, and by learning on the job. The training met the needs of the organization.
All copyright of the photographs rests with the volunteer photographers. The project signs a licensing agreement with each photographer to use their images in digital form. They will also have to draw up licensing agreements for third parties who want to take a copy of the digital images and use them in various ways. Users can download thumbnails and medium quality images and associated documents and use them for personal and research purposes, as long as it is not for commercial purposes. No electronic management systems are used to control copying.
This does not apply, as no analog material is kept.
They use the cataloging and reference systems used by NMR for listed buildings (a custom-built Oracle database). They used a subset of the original information stored in the database for the digitization process. The contractors also record some quality data associated with the negatives (e.g. whether they are scratched or otherwise physically damaged). Very little material was rejected before scanning, only when the film was physically damaged or not exposed.
They use the English Heritage & Royal Commission for Historical Monuments England (RCHME) Thesaurus of Monument Types, controlled vocabulary, and classification system. The metadata records information about the original object, the digital object, and technical and staffing details (e.g. who carried out the QA). Volunteers supply descriptive information, then some information is added by the image selector about the quality of the scan and of the photograph that is classified, using four grades. This means that the people who record metadata include volunteer photographers, the images selectors who are specialists in architectural history, and for some of the metadata information, the contracted digitizers. The catalog for digital deliverables is in electronic form on an intranet server. A section of it is available on the web. The original digital deliverables and the associated records are related on a single database.
The file formats used for images were JPEG for delivering and TIFF for capturing and preserving. They initially used Photo-CD before they realized its limitations for preservation. The resolution used is 2200dpi for capturing and preserving and 120*80; 360*240; 3072*2048 for delivering images. The internal bit depth is 12 bits per band and the output color depth is 8 bits per band. LZW (Lempel-Zif-Welch) compression is used by the contractor when copying the images on the CD. The original scans are retained in uncompressed form for preservation. JPEG is used for delivering. The aim of the compression is to improve access and to enhance usability.
Some post-scanning processing is carried out on the images. They use Photoshop for color corrections, removal of visible defects (e.g. scratches and dust). The original scans are not cropped but this may be done in the future for web delivery (e.g. to remove irrelevant features) using bespoke software. The average file sizes created are 18 MB at capturing and preserving stage (12-12 MB with LZW), and 10K (thumbnails) and 6MB (medium quality) at delivery. The dynamic range of the contractor’s equipment is specified at a minimum of 3.0 : 1.0.
Their experience showed that the digitization contractors were not accustomed to dealing with large volumes of material. In their case it was difficult to find someone who could apply consistently, the level of correction and guaranteed quality of output with the kind of varied input that the project produces. A large number of initially supplied images were rejected because of the presence of defects. The initial contractor may have assumed that some corrections and editing would be carried out by the individuals within the project. The current contractors are contracted to do 100% correction of images.
In terms of Quality Assurance, the way they worked (according to PRINCE guidelines) involved deciding what they wanted to produce at each stage, such as a handbook for the volunteer photographers, using a three month planning horizon, before defining what the acceptable quality for that would be and how they would check it. As quality was very important for the project, they spend a lot of time with the volunteers explaining about the expected quality, also including in the manual information about security, health and safety, example lists of entries which the photographers would have to produce, what constitutes a defining image and how to choose their views of the buildings, lenses. (Quality control procedures are also described above in the Project Management and Planning section.)
Users have free open access to both the catalog and the digital deliverables for browsing. Some of the material is available on the web. In October 2000 they put onto the prototype website the first 15,000 images which can be searched by type of building, date, county, historical figure or individual associated with the building, and by photographer. There will be terminals at the Swindon and London offices for the general public to use, with access to all the material, once the project has been completed. The web interface is a front-end to the database. They have taken a complex database and simplified it. All the fields that are indexed are metadata. They have no special hardware or software needs at the moment. They will be asking users to register if they want to receive added functionality in the future, and will also be developing e-commerce software. They monitor use via automatic data capture and web logs. The level of usage of the web pages is high. There were 480,000 images viewed in January 2001, a sharp increase from previous months related to press work on the project that was carried out in the US and Canada (on TV and radio). Most usage is from the English-speaking countries, but there have also been some surprises, such as Lithuania. Generally they find much useful information in the analysis of the statistics and the feedback for the users.
In addition to announcements on the English Heritage website, press releases, a national press campaign when the project started, staggered local publicity during the survey periods, print and broadcast media coverage (such as the January 2001 one from the US), the project produces a free newsletter which comes out three times a year. They also participate at conferences and events (e.g. a demonstration at the Historic Buildings, Parks & Gardens Event 2000).
The project is funded at £3.9 million from the Heritage Lottery Fund, through the Millennium Festival Fund between 1999 and 2002. This was the single largest grant awarded through the Millennium Festival Fund. Additional support has come from a development grant from English Heritage itself. The total cost of the project is £4.4 million. There was also some private sector sponsorship with Kodak’s donation of the photographic film used for the 1999 survey season.
The HLF have asked the project to provide a conservation plan for digital archiving, a risk assessment plan, a front-end evaluation report, a project management plan, cost models, workflow reports, a marketing policy, an education strategy, a post-project management plan, as well as plans for every stage of the funding and monthly highlight reports.
Within the remit of the project, no elements will be changed (apart from the file formats that had to be changed from Photo-CD to TIFF and updating the user interface) unless there is a specific need for them to do so. The NMR has a preservation strategy for both digital and paper records. The strategy is based on migration of data. Quality control procedures in the life-cycle management are based on the adoption of open or de facto standards and other procedures carried out by the NMR and not by members of the project’s team. The HLF contract specifies that they would need to keep the digital deliverables available for 25 years (for monitoring purposes). They plan to have them available in perpetuity.
As general comments, they emphasized the need for flexibility in digitization projects because the problems often do not emerge until the project is underway, but that planning sensibly can accommodate many of the problems. Projects must also allow sufficient time to plan effectively.
Particularly interesting aspects of this project include: the management of volunteers which seems to work very effectively and has the secondary output of raising awareness of the project and involving the wider community from the outset; the use of the PRINCE model for project management which demands considerable time from the project staff, but which has helped them in planning, coordinating, and carrying out effectively all the steps of the digitization process; the use of quality assurance procedures to monitor the quality of photography, scanning, linking with and entry of metadata and the use of the resource.