Archives & Museum Informatics
2008 Murray Ave.,
Suite D
Pittsburgh, PA
15217 USA

info@archimuse.com
www.archimuse.com

Join our Mailing List.

Time for Renovations: A Survey of Museum Web Sites

Table of Contents

• Introduction
• Museum Surveys
  • Virtual visitor survey
  • Museum site survey
• Conclusion
• Acknowledgements
• References

Introduction

If an army of monkeys were strumming on typewriters they might write all the books in the British Museum.

- Arthur Eddington (1882-1944) The Nature of the Physical World (1928)

More recently the following retort was made in the "Quotes of the Week" section of a national newspaper in the United Kingdom:

We've all heard that a million monkeys banging on a million typewriters will eventually reproduce the entire works of Shakespeare. Now, thanks to the Internet, we know this is not true.

- Robert Wilensky (born 1951) Mail on Sunday (16 February 1997)

The Internet certainly is full of a lot of rubbish, but the same can be said of the physical world. However, museums act as repositories of objects and associated information (sometimes grandly known as "metadata" for virtual objects in the world of information technology) which are worth saving for posterity. The role of the curator is to judge which artefacts are worth saving and then to preserve and, if possible, to display them in some form. Thus most museums are in the position of having a collection of interesting objects and information (both written and mental) worthy of presentation on-line for educational, research, and even entertainment and money-making purposes. Many organizations have no such resources. The problem of course is overcoming the barriers of the presentation technology, which may be problematic due to lack of expertise or lack of funds.

Even though the age of the "wired museum" is upon us (Jones-Garmil, 1997), there is still a significant way to go before the majority of the world's museums are on-line. Many (especially smaller and non-American) museums have yet to establish an Internet presence themselves because of the technological and financial barriers. However, a significant number have established a Web presence for several years and are now considering how this should be developed for the future. Certainly museums should avoid adding to the "rubbish" of the Internet and ensure that their Web sites are up to date and lively. Maintenance is a significant problem for any organization with a Web site and a budget for its continued development needs to be made on an annual basis.

The session associated with this paper considers the problem of museums which have established a Web presence, but now wish to develop this further. What facilities and information are worthwhile and cost-effective for museums to make available on-line? Of course the answer will vary for different institutions, depending on their size, resources, physical location, collections, technological expertise, etc. In this session, four case studies are considered by the people directly involved.

Eija Liukkonen and Riikka Haapalainen have been involved with the revision of the Finnish National Gallery's Web pages (Liukkonen & Haapalainen, 1999). After three years since their establishment, the pages have been completely overhauled, using the latest technical Web facilities such as Cascading Style Sheets (which allow greater control of fonts, etc., on recent Web browsers) and HTML Version 4.0 Web page markup. A user survey was undertaken to gain feedback before the changes were made. The pages are stylish and modern in design which matches the image of both Finland in general and the gallery in particular. Special attention has been given to navigation around the site. A good rule of thumb is that users should be able to move on from a given navigation page within 5 seconds of viewing it. Most Web users are relatively impatient given the amount of resources available and will go elsewhere if a site is confusing to traverse or slow to load.

The Museum of Science in Boston, USA originally produced a Web site in 1995 since demonstrating new technologies is part of its remit. Four years later, without a great sense of direction for the Web site during this time, the museum is now considering the purpose of the Web site. A committee within the museum has been considering this for almost a year. Edward Rodley, an exhibit planner at the museum, discusses the current situation at the museum and its plans for the future (Rodley, 1999).

Sarah Kenderdine of the Powerhouse Museum, Australia discusses the Australian Museums On-Line (AMOL) resource (Kenderdine, 1999). This provides a gateway (or "metacenter") for searching national distributed museum collections which has recently undergone a major redesign. 350,000+ museum records are accessible for an impressively fast keyword (and field-selectable) search via the Internet. A useful feature is the restriction of the search to records which include an image. A researcher will normally be interested in all records available, whereas a typical museum visitor (and especially children) may only wish to access records where the object can actually be seen.

Finally the Technisches Museum Wien in Austria is considering more futuristic developments, based on ideas from the world-renowned MIT Media Lab. Otmar Moritsch from the museum and Harald Krämer, a museum consultant, discuss how future visitors could interact with an object in an immersive environment using artificial intelligence techniques (Moritsch & Krämer, 1999). This will be tested in the museum first, since state-of-the-art Silicon Graphics equipment is required, but could be made available on-line in due course. Increasing interaction with the visitor, perhaps improving the educational possibilities of museum Web sites, will be possible in the future (Bowen & Houghton, 1997).

The rest of this paper considers the results of two on-line museums surveys, especially with respect to their impact on how museums may wish to consider the updating of their Web sites.

On-line Museum Surveys

Virtual museums interact with virtual visitors, just as real museum interact with real visitors. This section considers the results of two surveys, one of virtual visitors and one of virtual museums, and tries to draw some conclusions on how the results should influence the development of museum Web sites.

* From March 1996 this has been collected using a graphical counter rather than "hits" to the main page in the Web site statistics log file since the resource is now mirrored at a number of sites around the world.

Virtual visitor statistics can be collected both automatically and through surveys via electronic forms and email. As an example of automatic collection of information, visitor statistics for the Virtual Library museums pages (Bowen, 1995, 1997a & b), a directory of on-line museums, have been collected from almost its inception. See Table 1 for the monthly visitor figures since August 1994. As well as numbers, other information can be gleaned automatically to some extent such as variation of activity through the day and the location of visitors by country.

The VLmp visitor figures have steadily increased year on year although the rate of increase is now reduced. Indeed towards the end of 1998 there is an apparent decrease. This could be for several reasons. The use of "proxy caches" by browsers is on the increase for security and charging reasons; thus these figures represent a lower bound since repeat visits by other visitors using the same proxy cache are not detected. A similar effect was apparent during 1995 when proxy caches were first introduced. In addition, it appears that December is normally quieter than most months, probably because many people reduce their use of the Internet during the holiday period and interact with real people instead.

Virtual visitor survey

Several surveys have been undertaken on-line using the Virtual Library museums pages as an aid. One has surveyed visitors to on-line museum sites using a questionnaire. This was undertaken as an M.Sc. project at the University of Leicester, UK as part of a Museum Studies course (Reynolds, 1997) and has been summarised in an article available on-line (Futers, 1997). Here we attempt to interpret some of the key results and, where appropriate, how these impact on the provision of museum Web sites:

¤   "88% of visitors to museum Web pages are based in North America."

Although around 59% of the world's Internet users are based in the US and Canada (Internet Industry Almanac, 1998), the high figure of 88% may be skewed by the fact that most North Americans are native English speakers (the questionnaire was only available in English). They may also be more willing to answer on-line questionnaires due to familiarity with the technology, the relative cheapness of telecommunications compared with the rest of the world, and availability of leisure time to spend surfing on-line. In any case, access to the Internet is becoming increasing pervasive around the world (Press et al., 1998).

¤   "women make up 46% of museum virtual visitors"

Overall, men still tend to outnumber women in the use of the Internet, especially those making high use of it. In the US, it is estimated that 78% of high-use Internet/Web users are male (Hoffman, Kalsbeek & Novak, 1996). Thus it appears that museum virtual visitors are much more balanced between the sexes than is the norm for Internet users. Museums could attempt to take this into account when design their Web sites.

¤   "the average age of people visiting museum Web pages is 40 - 64 years of age."

The average age of the typical Internet user (in the US at least) is 25 - 34 (Hoffman, Kalsbeek & Novak, 1996). This the virtual museum visitor age seems to be skewed to an older age group that is typical of Internet users in general. The questionnaire would naturally tend to be answered by adults rather than children. In the future, as schools go on-line, it is likely that an increasing number of children will visit museum Web sites under direction from school teachers, perhaps before or after an actual visit to a museum. This could well bring the average age of virtual museum visitors down in future.

¤   "74% of people expect to find on-line exhibitions when visiting museum Web pages."

If a museum is planning to produce a Web site that provides more than "brochure-ware" (the equivalent of the printed advertising brochure issued by many museums and providing brief information to attract visitors), it is well worth considering some form of on-line exhibition. This could be a presentation of the best objects available in the museum or perhaps a record of temporary exhibitions as they occur, that can be left as a permanent record long after the physical exhibition has been dispersed, as at the Museum of the History of Science in Oxford (Bowen, Bennett & Johnson, 1998). Certainly any second-generation museum Web site should include some form of on-line exhibition area.

¤   "87% expect images and - perhaps a museum's greatest nightmare when going on-line - 52% of users expect to download images from a museum's Web page."

A museum Web site with no images of objects is normally a disappointment to virtual visitors and most will move on quickly if there are no images on offer. Although it is possible to make extremely high-quality images available on-line like the Museum of the History of Science in Oxford (Sphæra, 1997), it is probably better in most cases to make lower quality images available on-line, both to decrease loading times and to make high-quality printing of the images impossible. A screen is typically an order of magnitude worse quality than what is expected on paper and the eye appears to adjust for this loss of quality on the screen, but not when the same image is printed.

Perhaps a legitimate use of images would be for school children to create electronic scrapbooks for project work collected from a number of on-line museum sources, for example. In any case, it is well worth including the museum's policy on images prominently on the Web site. The site could attract commercial interest from people such as publishers and authors willing to pay for use of high-quality images in books, etc. Thus the Web site images should be seen as a promotional vehicle for what the museum has to offer, with increasing potential to raise revenue in the future.

The top three benefits of accessing information about museums through the Internet were found to be as follows:

¤   "to explore a personal interest"

Special-interest museums (e.g., associated with particular hobbies, sports, etc.) may especially benefit from an on-line site.

¤   "to learn about a museum not in your locality"

A significant number of people (around half) are interested in finding out about non-local museums. Of course many of these people are unlikely to visit the physical museum, but some may be planning a trip to the locality of the museum, and the rest will have the knowledge of the existence of the museum reinforced in their memory. Virtual visitors are much more likely to make an actual visit to a museum at some point in the future when they happen to be in the vicinity (e.g., on holiday) if they have found an interesting Web site associated with the museum beforehand.

¤   "it's fun and interesting"

To attract on-line visitors, museums will increasingly have to provide interesting multimedia material within their Web sites, similar to "interactives" than can be found in many physical galleries. Indeed, in the future some parts of exhibitions could be accessible both within the galleries and on-line, using the "write once, run anywhere" Java programming language for example (Singhal & Nguyen, 1998).

What constitutes an "interesting" Web site changes with the availability of new facilities on Web browsers, and must always be balanced against the browsers being used by the majority of a museums visitors at any given time and the speed of loading of multimedia content. Most users would like to be able to traverse navigation pages in a few seconds and are likely to lose interest if a Web page takes more that half a minute to load. The "Virtual Endeavour" exhibit at the Natural History Museum in London (Bowen, Bennett & Johnson, 1998) would be impractical on-line in the form presented at the museum itself since the cost of the hardware required is too high for an individual to afford. However some simpler virtual reality models are available on-line for viewing with an appropriate "plug-in" program.

How do people hear about museum Internet sites? By far the most import two ways that virtual visitors find museum Web sites are as follows:

¤   "the majority of museum Web pages are found through links from other pages"

Thus museum Web sites should be linked both from widely-used on-line Web search engines (e.g., AltaVista under http://www.altavista.com/) and directories (e.g., Yahoo under http://www.yahoo.com/). To ensure this, museum should expect to spend some time registering their site with major search engines on-line and information related sites of their existence. It is a good idea to do this again if there is a major overhaul of a Web site, since then newly added Web pages will be included. The use of the HTML <META …> tag to include metadata such as relevant keywords and a short description of each Web page is good practice and normally helpful for better processing and categorization by many search engines.

A page of links to other related sites and museums is well worthwhile in a museum Web site, perhaps under a reciprocal arrangement. Most people visit more than one site when using the Web and thus providing many entry points and exit points is a good idea.

¤   "many museum web pages are found by chance"

Very few museum Web site are found through electronic mailing lists or newsgroups. This is probably as much to do with the fact that most museums do not greatly use or understand such facilities. A specialist museum would find it well worthwhile issuing period newsletter (e.g., press releases) on a relevant mailing list and/or newsgroup to raise its profile worldwide at very little cost. The " comp.infosystems.www.announce" newsgroup is a good place to announce new museum Web site; once issued on this or other newsgroups, a Web site is more likely to be indexed by search engines and other sites. For an excellent Web interface to newsgroups with good searching facilities, see Dejanews under: http://www.dejanews.com/

Museum site survey

Another survey has been undertaken more recently from Japan of museums sites themselves linked from the Virtual Library museums pages (Oono, 1998). 206 answers were obtained from 986 sites targeted (representing a 20.1% success rate) covering 26 countries. The following questions were asked:

28.8% have 20 pages or less (which may be deemed to be "brochure-ware"), but the mean number is 50 - 100 pages, which represents a fair sized Web site. 16.1% have 500 pages or more which represents a very large site. These sites have embraced the technology very seriously and most of them have probably been through at least one major reorganization of their site.

One point to remember when updating a Web site is that search engine "robots" will scan your site periodically and record the location of all pages that can be found via internal hyperlinks. If the internal structure of you Web site is radically changed and previously existing URLs (Universal Resource Locators) become invalid, many people who try to access your site via search engines may receive a not very helpful error message on their attempted entry into your site. If URLs must be removed from a museum Web site, it is best to ensure that accessible invalid pages gives a helpful response, perhaps guiding the visitor to a number of good starting points within your Web site. It is even better to plan the hierarchical structure and layout of a Web site with great care and with an eye to its future maintenance. Then existing URLs can remain in place and any radical restructuring can involve adding new URLs and updating existing URLs rather than deleting them. It is worth ensuring that important internal pages on a Web site have short and memorable URLs that are not changed without good reason. Then external Web sites can easily and reliably link to these locations.

27.3% of museums update their home page once a month. 9.0% update on a daily basis and 9.0% update annually. Web sites can become dated quickly, especially if ephemeral (but useful) information such as entrance times, prices, temporary exhibition details, etc., is included. Monthly is probably a good aim for most museums, but larger museums should consider performing updates more often than this.

57.1% of museums have only one person working on their home page and only 3.0% have more than 5 people involved. It is well worth having more than one person involved if possible so that no single person is critical to its operation and success. 6.3% have no-one involved and thus rarely if ever update their Web site.

Not including personnel expenses, 70.2% spend less than US $1,000 per year. Only 9.8% spend more than $5,000. Thus Web pages are still not a high priority for most museums. This is not unreasonable considering that the majority of museums are small in size, but any national or international museum should already be in the top 9.8% above.

53.7% have been open for 1 to 3 years. 20.4% are less than a year old and only 2.9% are 5 or more years old. The useful lifetime of much computer equipment before it requires significant upgrading is around three years and the same can probably be said for Web sites. Museums with Web sites around two years old should be think about the possibility of a major overhaul to their site.

24.9% of museums own there own server which is quite a high percentage. These museums are taking the Internet seriously. For University museums this is easy and cheap because the Internet connection is typically available 24 hours a day anyway and no modem is involved; for other museums the communications cost can prove prohibitive except perhaps in the US. Second-generation Web sites should certainly consider the possibility of a dedicated Web server if communication costs can be covered and suitable computer support staff are available on-site.

43.7% receive under 50 visits per day, but 10.7% receive over 1,000 each day, which is an excellent number comparable to a reasonable-sized museum. 8.3% gave no answer and presumably do not have the statistic available. Most museums would like to know their real and virtual visitor numbers if possible, if only for marketing reasons. Owning your own Web server makes these statistics easy to collect automatically. Some Internet Service Providers will make log files available to customers, but it is worth checking if this facility is available. Second-generation museum Web sites should certainly be collecting and using automatically collected statistics.

34.6% replied "yes" which is a high percentage. This may be because those taking the Web seriously enough to attempt selling on-line are more likely to respond to an email questionnaire. It is still not financially worthwhile in general for an individual museum to provide its own on-line shop with full on-line credit card transaction facilities. Most museums (in the UK at least) do not even have an associated trading company (Carrington, 1998; MGC, 1998). In any case, designing an on-line store requires special consideration of the customer interface; a good design can increase the amount of traffic and sales significantly (Lohse & Spiller, 1998). The Metropolitan Museum of Art in New York have a full on-line secure shopping Web server (http://metmuseum.netcart.com) but this is very much the exception rather than the rule. A sensible alternative is to use an existing on-line shopping facility. For example, MuseumShop@Home (http://www.museumshop.com/) specializes in selling museum merchandise. Alternatively, Yahoo provides more generic on-line shopping facilities that are relatively easy to set up even for an individual (see under: http://shopping.yahoo.com/).

79.5% receive less than ten emails a day which is not a high burden. Only 1.0% receive more than 100 emails each day. At this level the handling of email becomes a significant resource problem that must be addressed and budgeted appropriately. Second-generation Web sites should allow virtual visitors to communicate with the museum via email. It is worth having a prominent and easily-found contacts page with one or more email addresses. A problem of providing email addresses on a Web site is the increased amount of unwanted "spam" emails as the address is discovered by automatic "robots" traversing Web sites to create junk email lists (Cranor & LaMacchia, 1998). However this is a generic problem with no good general solution as yet, although newer email software increasingly includes facilities to help filter incoming email. One solution which helps is to periodically change the contact email address given on the Web site (say at each major renovation of the Web site every two or three years).

Receiving information from customers (1.5%) and selling museum goods (0.5%) were very low on the list of priorities, although these are likely to become more important is the number of people with on-line access increases and appropriate secure technology for on-line transactions become more widely available and accepted. High-value and reasonably small-size and low-weight items are the most likely to be most successful when selling on-line. (Witness the success of book and CD sales on-line.)

This survey should be seen in the context of the total number of museums in the world against the number than have their own Web site. The percentage is still low (perhaps a few percent) and the scope for expansion is great. Near all museums of national or international status have their own Web site, but there are many thousands of small museums worldwide where the cost of setting up and more importantly maintaining a Web site is hard to justify on financial grounds. Free and low cost Web site hosting is becoming increasingly available, but the major cost is the preparation of Web pages rather than their provision on-line. Professionally produced Web pages can cost up to US $1,000 per page if production and editorial standards are to be of similar quality to paper publications.

Conclusion

The "contacts" page will probably be of special interest to museum professionals, and includes links relating to the museum surveys covered in this paper. The author welcomes information about other on-line museum surveys. Please send brief details and a URL to J.P.Bowen@reading.ac.uk for possible inclusion.

Acknowledgements

Thank you to all those who have helped develop the Virtual Library museums pages (VLmp), especially Cary Karp of the International Council of Museums (ICOM). Thanks are due to Rachel Reynolds (University of Leicester, UK) and Katherine Futers (MDA, UK) for information on the virtual visitor survey. Special thanks are due to Shinjiro Oono (Internet Museum, Japan) for the results of the museum Web site survey undertaken in conjunction with the Virtual Library museums pages. In particular, the graphical results presented above were originally produced as part of this survey. Minor amendments have been made for inclusion in this paper.

References

Bowen, J.P. (1995). The World Wide Web Virtual Library of museums. Information Services & Use, 15(4), 317-324.

Bowen, J.P. (1997a). The World Wide Web and the Virtual Library Museums Pages. European Review: Interdisciplinary Journal of the Academia Europaea. 5(1), 89-104, January.

Bowen, J.P. (1997b). The Virtual Library museums pages, VLmp: Whence and whither? In D. Bearman & J. Trant (Eds.), 1997 (pp. 9-25).
http://www.museums.reading.ac.uk/mw97/paper/

Bowen, J.P., Bennett, J. & Johnson, J. (1998). Virtual visits to virtual museums. In J. Trant & D. Bearman (Eds.), 1998 (CD-ROM).
http://www.museums.reading.ac.uk/mw98/paper/

Bowen, J.P. & Houghton, M.K. (1997). Networked multi-user and multimedia environments for learning and collaboration. Department of Computer Science, The University of Reading, UK. Consulted January 1999.
http://www.cs.reading.ac.uk/people/jpb/publications/moo/

Carrington, L. (1998). Data delivery. Museums Journal, 98(12), 37, December.

Cranor, L.F. & LaMacchio, B.A. (1998). Spam! Communications of the ACM, 41(8), 74-83, August.

Futers, K. (1997). Tell me what you want, what you really, really want: A look at Internet user needs. Proc. Electronic Imaging and Visual Arts, EVA, Paris, France, September. Varsari Enterprises, Alexander House, 50 Station Road, Aldershot, Hampshire, GU11 1BG, UK. Consulted January 1999.
http://www.open.gov.uk/mdocassn/eva_kf.htm

Hoffman, D.L., Kalsbeek, W.D. & Novak, T.P. (1996). Internet and Web use in the U.S. Communications of the ACM, 39(12), 36-46, December.

Internet Industry Almanac (1998). Computer Industry Almanac Inc., USA.

Jones-Garmil, K., Ed. (1997). The Wired Museum. Washington DC: American Association of Museums.

Kenderine, S. (1999). Inside the Meta Center: A cabinet of wonder. Museums and the Web 1998 conference, New Orleans, USA, 11-14 March 1999. Consulted January 1999.
http://www.archimuse.com/mw99/abstracts/prg_1062.html

Liukkonen, E. & Haapalainen, R. (1999). Revising the Finnish National Gallery's Web pages. Museums and the Web 1998 conference, New Orleans, USA, 11-14 March 1999. Consulted January 1999.
http://www.archimuse.com/mw99/abstracts/prg_18.html

Lohse, G.L. & Spiller, P. (1998). Electronic shopping. Communications of the ACM, 41(7), 81-88, July.

MGC (1998). Museum Focus. London: Museums and Galleries Commission.

Moritsch, O. & Krämer, H. (1999). New strategies of digital corporate identity: Technisches Museum Wien. Museums and the Web 1998 conference, New Orleans, USA, 11-14 March 1999. Consulted January 1999.
http://www.archimuse.com/mw99/abstracts/prg_1107.html

Oono, S. (1998). The World Wide Museum Survey on the Web. Internet Museum, Japan. Last updated September 1998, consulted January 1999.
http://www.museum.or.jp/IM_english/f-survey.html

Press, L., Burkhart, G., Foster, W., Goodman, S., Wolcott P. & Woodard, J. (1998). An Internet diffusion framework. Communications of the ACM, 41(10), 21-26, October.

Reynolds, R. (1997). Museums and the Internet: What purpose should the information supplied by museums on the World Wide Web serve? M.Sc. thesis, Department of Museum Studies, University of Leicester, UK.

Rodley, E. (1999). So we've got a Web site, now what? Museums and the Web 1998 conference, New Orleans, USA, 11-14 March 1999. Consulted January 1999.
http://www.archimuse.com/mw99/abstracts/prg_54.html

Singhal, S. & Nguyen, B., Eds. (1998). The Java factor. Communications of the ACM, 41(6), 34-68, June.

Sphæra (1997). Website revamp. The Newsletter of the Museum of the History of Science, Oxford, UK, Issue No. 6 (p. 5), Autumn. Consulted January 1999.
http://www.mhs.ox.ac.uk/sphaera/index.htm?issue6/articl10

Trant, J. & Bearman, D., Eds. (1998). Museums and the Web: An International Conference, Toronto, Canada, April 22-25, 1998. CD-ROM, Pittsburgh: Archives & Museum Informatics.

http://www.archimuse.com/mw98/