Abstract

An exploratory sample of 114 personal home pages, drawn from a home page directory available on the World Wide Web (People Page Directory, http://www.peoplepage.com), was analyzed to detect patterns and trends in home page content and design. Covered in the analysis were types of informational content included in the home pages; internal organization and structure of the content, including type and number of hypertext links; miscellaneous content elements, such as "sign guestbook" and number of hits to the page; and physical design features such as photos, motion, and audio elements. Metaphors used in the design of the pages, and degree of self-revelation were also considered.

The home pages displayed a great variety of content and of specific types of formatting within broader formatting approaches. While some content elements were quite popular, none of them--not even name--was found on all home pages. Nor did the pages evidence reliance on any single dominant metaphor, such as home page as "home" in the sense of domicile. It appears that though certain features may be frequently found in it, the personal home page as a social institution is still very much under development.

Use of the Internet and the publishing and navigation tool known as the World Wide Web (WWW) has grown explosively in recent years. Many companies, organizations, and individuals have created a virtual presence on the Web by mounting a website, which anyone else with Web access can visit, read, and explore. The primary entry and access point for a website, generally called the "home page," is the focus of this article.

Navigation tools, such as "Netscape Navigator," enable people to access websites by entering the site’s "URL," or Universal Resource Locator; by searching on vocabulary terms in websites with the help of "search engines," such as "Infoseek;" or by searching in directories of names of people and companies.

Vast amounts of attention in the form of news reporting and discussion have been devoted to the Web in newspapers and journals; however, empirical research on this dramatic new communication and information medium is still in its early stages. In particular, while numerous articles have advised people how to create home pages (December & Randall, 1994; Falcigno & Green, 1995; Spear, 1995; van Brakel et al., 1995) no research analyzing the character and content of personal home pages or websites was found in preparation of this article. Just one article analyzing the content of corporate or organization websites was found--a study of library websites (Clyde, 1996).

Home pages are commonly developed and made available on the Web by companies, organizations, and individuals. It is the purpose of this article to describe a study of a sampling of personal home pages; business and other organizational home pages were excluded.

The authors became interested in the question of the character of personal home pages because it appeared to us that the home page represents a new social form or device. People are representing themselves to the world through a new medium. How do they present themselves there? What kind of information do their home pages contain? Does this new social device of the home page draw on old social and document forms or create entirely new ones?

The objective here is to develop a preliminary profile of some personal home pages, and thus provide an initial characterization of this new social form.

Questions the authors asked of the data included:

(1) What was the purpose of the home pages, specifically, were there common patterns of social types of home pages?

(2) Did the home pages draw on any pre-existing dominant metaphors, or were they developing their own metaphors?

(3) What kinds of information were most commonly included?

(4) How was the information organized within the home page?

(5) Were there any regularities in the hypertext links to other websites within the home pages?

To answer these questions, the home pages were analyzed for the general character of information provided, metaphorical references, specific elements of information, the organizational structure of the home page, and the number and type of hypertext links within the pages. In addition to the quantitative data, our qualitative assessment of the character of the pages will be presented in ensuing sections.

To examine personal home pages, it would clearly be desirable to select a random sample of such pages from across the World Wide Web. That, in turn, however, is a very difficult challenge. With the number of available websites growing explosively, and available directories always incomplete and overlapping, selecting a true random sample may be next to impossible. We experimented with another alternative--drawing names randomly from a telephone book and then searching them on the Web. This, too, proved impractical, and almost certainly not, in fact, representative. It was impractical because most of the names selected did not show up on the Web, and unrepresentative because there is no guarantee--in fact, it is quite unlikely--that the mix of people appearing in telephone books (or other name listings) parallels the mix of people appearing on the Web, which still represents a minority of the population.

However, since we did not know if studying home pages would prove to be either interesting or feasible in any case, we decided in the end to select a systematic sample from one of the available Web directories of home pages. Such a sample would obviously represent a self-selected group--those who submitted their sites for inclusion in that particular directory--but the sample would have the advantage of not being consciously or unconsciously biased in selection on our part. Using such a sample, we could then learn of the characteristics of personal home pages in at least an exploratory manner.

The People Page Directory (http://www.peoplepage.com), found through Netscape Navigator (2.0) Net Directory, contained "about 5000" entries, according to its founder and Webmaster, Rhese Hoylman, in June 1996, when the sample for this study was drawn. Entries available at that time in the directory had been submitted since a year earlier, June 1995. The first five entries under each letter were taken for the sample. As the letters D, K, Q, T, U, X, Y and Z each contained fewer than five entries, the size of the total sample was 114 home pages.

The directory was arranged by earliest submitted entries first, so this sampling method tends to draw on entries that were submitted toward the beginning date of the directory, that is, mid-1995, while randomizing the entries alphabetically within each letter. As home page owners can update their home pages at any time, however, the sampled home pages represented their current state as of June 1996. The sample was limited to the United States context; any site submitted from outside the country was passed over for the next listing.

Submission to the People Page Directory is voluntary and free. The following types of information are asked of people when they submit their listing: "real last name," "real first name," "alias/handle," email address, home page address, password (so the submitter can revise entry if desired), city, state/province, country, URL (Universal Resource Locator address) of favorite website, favorite newsgroup, and favorite IRC (Internet Relay Chat) channel.

Drawing on the work of others, van Brakel et al. (1995) characterize home pages variously as:

•"a designed and designated entry-point for access to a local Website" (December & Randall, 1994).

•"...a graphical door and screen that are full of information, in which links to related information are included" (Dougherty & Koman, 1994).

Operationally, in this study, a home page was defined as the first screen of information that appears upon entering the URL address (drawn from the People Page Directory) of a website in a search on the World Wide Web. The page may be of varying lengths, compared to paper pages, but reaches its end when the page cannot be scrolled down any further. The only exception to the "first screen" rule occurred occasionally when the first page was a referral from a older to a newer location, or was in some way quite evidently a sort of "see reference" page to what was clearly the main home page. There were only a handful of such instances, and there was no ambiguity in interpreting them.

The home pages were then analyzed for their characteristics. A trial sample of home pages was first examined, and a wide range of features identified as being of possible interest. These traits were then grouped into several classes:

•Information about the website as a whole

•Purposes of home page

•Personal information content of pages

•Home page internal structure, including links

•Miscellaneous content elements/features

•Home page physical features

Categories within each of these classes of information that had been discovered in the pre-sample were then tallied in the sample proper, and are presented in the results section.

Among the 114 home pages, 40 percent (45 pages) are located on commercial servers (i.e., final characters of Web address are "com"), 36 percent on educational servers ("edu"), 19 percent on network servers ("net"), four percent on organizational servers ("org"), and one on an unknown server. The text of 105, or 92 percent, of the home pages was written by the owners (the individuals submitting the home pages to the directory), with the remainder having been written by someone else.

In reviewing the home pages, several functions of the pages were identified. One page could possess more than one of these functions, and were so tallied, but few did. These functions were identified by the contents of the page and the "feel" we perceived compared to paper documents. For instance, if the information largely overlapped with typical contents of resumes, even if the word "resume" was not used, it was assumed that a principal purpose of the home page was to present the author’s professional capability and experience.

Fully 45 percent (51 pages) presented professional capabilities or background. Among these, six pages made explicit mention of looking for work. For example, Carolyn Gibbs states: "Take a look at my resume...I’m available for hire June 1996!!" (http:www.ced.berkeley.edu/~cgibbs). Another 20 percent (23 pages) announce products or upcoming performances. So nearly two-thirds of the home pages present--and implicitly or explicitly advertise--the home page creator’s professional capabilities or business.

Such information can be interpreted in a couple of ways. On the one hand, announcing one’s work experience is a way of attracting possible interest on the part of employers, even if one is gainfully employed currently. And certainly, in those cases where people are explicitly looking for work, or explicitly advertising a product or upcoming performance, there is no question that the home page is directly serving a work-related purpose.

However, the extent of professional information provided varies across the sample home pages. In some cases, it appears that the author of the page is not interested in employment so much as in offering information for exchange with others as a part of a conventional get-acquainted ritual. Exchange of information about one’s work is a very common and important component of such conversational interchanges for Americans--in fact, that piece of information is often the second proffered, after one’s name. In these cases there is generally less detailed information and the organization of the information on the page has less resemblance to a conventional resume format.

In fact, even in those pages with the most resume-type organization, there are often substantial deviations from the standard format. The creators of the home pages appear to enjoy experimenting with the more flexible possibilities of web documents, and include photographs, icons, and other format deviations from conventional resumes (e.g., Andy Dong’s page, http://hart.ME.Berkeley.EDU/~adong/adong.html)

On the other hand, two areas generally considered delicate in American social life, politics and religion, are correspondingly scarce in the content of the home page sample. Just three pages mention religious beliefs, and three discuss politics. One home page brings both of these hot topics together. Its title: "Jewish Dude With a Conservative Attitude: Alan’s New Conservative Home Page" (http://www.cris.com/~Alcanh/).

The 33 home pages (29 percent) categorized as playing with system capabilities appeared to be presenting their home page largely for the fun of it, to share in the new phenomenon, and present themselves to the virtual world. For example, one home page is titled: "Jeffrey and Patricia Gyurke present: Informative and fun things to explore!" (http://www.pennet.com/jgyurke/)

Another states: "Yes, It’s...Meredith’s Homepage." It continues: "Hello. This, as you might have already gathered, is my home page. I have actually only had my computer for two months and sixteen days, so you may have to bear with me as I struggle to figure out what all these funny little buttons do." (http://www.geocities.com/Sunset Strip/1764/) Still another states: "I have no idea why I’m creating this web page except for the sole purpose of doing it. Oh well, I hope it works." (http://studentweb.tulane.edu/~ahernan/)

Miscellaneous purposes were to look for a missing person, and create a focus for a family name and people having that name (one each).

We also examined the pages to determine whether any identifiable metaphors or pre-existing social or document forms seemed to guide the wording and organization of the pages. As noted in the previous section, the most commonly used basis for the design of the page appeared to be the resume, sometimes explicitly labeled, other times seeming to influence the presentation of the page without the use of the term. Here we examine whether explicit or implicit metaphors and social forms guided the home page creators, at least partially, in their development and presentation of the pages.

The phrase "home page" probably originated from the sense of "home" found in the phrases "home base" in climbing or "home plate" in baseball. The home page in a website, like the home base or home plate, is the starting point, and the point to come back to in order to get reoriented or start again. Once "home" was used in the name, however, some interpreted the word in the sense of primary domicile--which, after all, is closely related to, and probably the origin of, the meanings in the sports usages.

We found six home pages that used explicit or implicit references to home as domicile in describing the home page. "Aaron’s Home Sweet Home Page" (http://www.lookup.com/Homepages/70500/main.htm), "Spartan’s Haven. You are guest number _____ to visit my house" (http://www.clark.net/pub/spartan/), and "Welcome to my humble abode" (http://www.colum.edu/~phillips/), are examples. One of the six presented a graphic that looked like a room, with a sign on the room’s wall: "Welcome to my home" (http://www.enter.net/~gquier/).

Other terms implying location that were found one or more times among the home pages were the following: quadrant, world (as in "maggy’s world" (http://pinky.interaccess.com/maggy/), space, spot, domain, and inferno. On the whole, however, there were far fewer references to home pages in this sense of a physical home than there were home pages built on the model of the resume.

There were hints of other pre-existing social forms in the language and format of the pages as well. In the physical world, American homes are sometimes labeled with small signs, often attached above the front doorway, or hanging down from the porch roof, stating, "The Smiths," or "Welcome to the Smith Family Home," or the like. Because these signs usually have so little text, it is difficult to determine with confidence whether any of the sample home pages were developed with a similar function in mind. However, two or three of the home pages seemed to these authors to be evoking that same spirit. For example, one page has a title, "Welcome to Mike & Nancy Nalbone’s Home Page!", and is accompanied by a photo of the couple smiling in a warm and inviting way (http://pluto.njcc.com/~nalbone/). Such "welcome" signs may be seen not only as physical signs, but also as declarations of the family home as proud social unit, declaring itself to the world, in a modern, democratized (and therefore much simplified) version of the heraldic shield.

Another form the home page might be seen to resemble is the business card or calling card. The small standard size of the card today limits the amount of text that can be placed on it. However, in the nineteenth century, calling cards were often more elaborate. When photography was a new innovation in the last century, one type of calling card, known by the French name, "carte de visite," included photographs as a special, impressive feature.⁺ (See Mathews, 1974; Morgan, 1942, pp. 665-6.) These cards measured from 5 X 8 centimeters up to 6 X 10 centimeters. Lena Marianne Arvola’s home page evokes such a photographic calling card (http://pearl.mhtc.net1-chase/index.html). With a frame measuring approximately 11 X 14 centimeters, in turn containing several smaller frames each outlining a class of information, it contains the usual business card information, a photograph, and an image of her personalized automobile license plate. (Personalized plates may be seen as another form of modern heraldry!)

After a while the cartes-de-visite consisted primarily of the photos, and less the usual calling card information. People would exchange the photos and then put them in an album. The twentieth century Web parallel to creating an album might be "bookmarking" a site, so it can be revisited easily again.

Since preparing home pages involves writing, it might be expected that metaphors of document forms might drive the development of home pages. Elements of the resume and calling card format have already been discussed. If there is a standardized format for the home page as document, however, we were unable to identify it yet. Other than occasional mentions made of the home page creator as its "author," no clear, dominating sense of home page as type of document were recognized, other than ones already mentioned.

Since the creators of these home pages develop them themselves, for the most part, another possible way the creators might see their home pages is as an explicit "creation," a work of art. They may have done so privately, but other than one person who titled her home page "Mary’s Fantabulous Web Oeuvre" (http://grove.ufl.edu:80/~mebbahl1/index.html), we did not detect explicit evidence within the pages that their creators viewed the website in this manner.

On the whole, while home page creators did draw partially on metaphorical senses for how they conceptualized their home pages, it would appear that for the most part they are remaining open to the development of the home page as its own new social form. Resume-like forms are popular but by no means overwhelmingly dominant, and considerable freedom was exercised in the internal structure and content of home pages. Remaining sections of this paper will detail other aspects of home page design.

One of the ways in which the authors of the pages differed considerably was in the degree to which they revealed themselves and took social risks. Making a home page available to the virtual community can both reveal and hide the person. The truth or completeness of the statements could not be determined from a reading of them, of course, but it was possible to note how extensive and self-revealing the provided information appeared to be.

Two of the sample proved to be Trojan horses in this directory of personal sites; they were company home pages--one of them selling website development. Another home page was submitted by an individual but contained nothing but the specifications of that individual’s house for sale.

Three others were curious sorts of borderline cases as personal home pages. One was called the "Shoe Salesman’s Home Page" (http://funnelweb.utcc.utk.edu/~lyle/shoe.html) and was dedicated to "my fellow shoe clerks around the world." It was clearly a site that was personally very meaningful and well thought through by the author, containing, among other things, frequently asked questions about shoes, links to shoe companies, and even humor. However, the home page contained no personal information, not even the name, of the author, though a hypertext link was made available at the end to "me."

Another home page called "Yenzo’s Climbing Corner" (http://www2.coastalnet.com/~cn3111/index.html) was, similarly, "dedicated to North Carolina Rock Climbing" and contained hypertext links to a variety of information on that subject, but no personal information. Those links, as with the shoe salesman home page, contained one that led to the page author, however, titled: "Read article from Who is Yenzo?" Finally, one other site gave the author’s name, but no other conventional personal information--only links to over 40 poems written by the page author. Poetry, can, of course, be very revealing too, but generally in a different way.

Only 90 (79 percent) of the home pages provided the author’s full name; 17 (15 percent) gave nicknames or aliases. One hundred five (92 percent) gave electronic mail addresses, and 18 (16 percent) provided telephone numbers.

Another important form of self-revelation and presentation of self is the photograph. Forty (35 percent) of the home pages contained photographs. Of these, 22 had photo(s) of the author, self only, while four had photos of self with other people, and six had photos of self in a scene (some with pets or objects) more extensive than just the self alone. Thirteen home pages had other photos. (Any given home page could have more than one type of photo.)

Typically, the home page authors provide a range of information that, however extensive, is moderately revealing--listing work, favorite websites, hobbies, etc. Table 2 lists the various kinds of information tallied, arranged by frequency of appearance of the type of information in the home pages.

An average of 5.9 items appear on the home pages out of a total of 32 types of items we tallied. Thus, in this sample the top six items listed in Table 2 can be seen as the most likely to appear on personal home pages, though the frequency of the last of these is already down to less than one third of the sampled pages.

We identified three gross types of internal structure within home pages. The first consisted of strictly a contents menu of links to other pages, called here "menu home page." In these cases there would be no or virtually no text on the page other than the menu listing of links. The second type of home page consisted entirely or almost entirely of straight running text, with no or few links. The third type of internal structure had many links and often contained a menu or contents list at the beginning of the page. Recall that in all these cases, the stated structure refers only to the first, or "home" page at the website, scrollable from top to bottom.

The "menu" type of home page was found in ten, or nine percent, of the home pages. There were 16, or 14 percent, of the second type of home page, the straight text. Finally, the great majority of the home pages, 88, or 77 percent, were of the third type.

Links were defined in Netscape 2.0 as the blue underlined elements of home pages. No distinction was made in the general count between links within a page or website, and links to other websites. All but five of the home pages, or 96 percent, contained one or more such links. Altogether, there were 2184 links found in the 114 home pages, for a mean of 19.2 links per home page. The median number of links per home page was 14, and the mode was 11. The highest number of links found on a single home page was 215. Table 3, "Distribution of Home Page Links," displays the link frequencies.

Content types of links were examined as well. Nearly all home pages--107, or 94 percent--contained links to other home pages. In addition, 41, or 36 percent, contained links to search engines, and 42 (37 percent) contained links to shareware.

In the pilot test we identified a number of other miscellaneous features of the home pages that may be a part of the developing culture and character of home pages. These were then counted in the sample proper. Table 4, "Miscellaneous Elements/Features," lists the features in order of frequency.

Several of the most popular features draw upon characteristics specific to the interactive and ever-changing environment of the Web. Number of hits to the page, latest update date, clock or calendar, and "under construction" all reflect the distinctive character of the network environment. These four would be meaningless, a demonstration of inadequacy, or impossible to do if they were to appear in a book or business card.

Here they are a potential source of pride for the home page creator. Number of hits to the page may show the popularity of a page, but it also shows that the creator of the page knows how to add such a feature to the page. Latest update date, clock or calendar, and "under construction" also highlight the mutability of Web records, and imply a home page owner actively engaged in the Web through an ever-changing website. Though "time of creation" appears in the paper world as publication date, it is understood by visitors to Web home pages that such a time can change frequently without requiring the issuance of a whole new page or site, so in this way reinforces the sense of rapid changeability associated with the Web.

The phrase "under construction," which appears in no fewer than 41 home pages (36 percent), may serve another function as well, given how new the Web is, with the social norms and expectations still under formation. If one puts up a home page and is a little uncertain about how it will be received, adding the "under construction" rubric can imply that an actually-finished home page it is not yet in its finished state, while one tests the waters to see if the home page gains acceptance. If one concludes that one’s page is not fashionable or has some other flaw, one can change it with no loss of face--after all, it was still "under construction" up to the time it was changed. The 15 cases where the icon "New" was used presumably represent instances where the home page creator wants to signal changes that have been made recently in the page.

The "sign guestbook" feature is a particularly interesting one for our exploration of the social character of home pages. On the Web it represents another "cool" feature that a home page maker can add to his or her site. It also gives the page creator an opportunity to receive feedback from people visiting the site.

Socially, outside of the Web, signing a guestbook is something one does both in homes as well as in public places, such as churches, museums, restaurants, and funeral homes. The custom, generally, is to sign one’s name and address, with brief comments appropriate to the location. Guestbooks are very easily provided in the real, as opposed to virtual, world. A guestbook is simply placed in a prominent place, and visitors have the option to sign and make comments or not. While this may be technically somewhat harder to achieve in the world of the Web, the social arrangements are remarkably similar. The visitor has the option to sign or not, and the opportunity to provide a little feedback. Twenty of the home pages in this sample, or 17.5 percent, had this option. As only some people provide guestbooks in their real homes too, it would be interesting to see how this rate in the virtual world corresponds with the availability of guestbooks in real homes in our society.

While the above features highlight what is new about the Web as a communication medium, the remaining features in this section are evidence of social forms and issues being transferred from the paper to the electronic environment. ("Sign guestbook," along with being a technical feature that could be added to home pages, was also a social device carried over from the paper world.) The large number of home pages with the "free speech" blue ribbon symbol present reflect the controversies currently brewing in our society about how to handle intellectual freedom issues in the context of a new communication medium with its own distinctive parameters of access and availability.

Copyright notices and listings of directories where the home page is indexed are features carried over from the paper world. While much of the talk in the early days of electronic access centered around the freedom from constraints that network access was thought to have made possible on the Internet, it is not surprising that people want to maintain ownership of their intellectual property--at least, that is, until and unless our society undergoes a sea change in its attitudes toward intellectual products. Likewise, the model of professional associations in the print world is carried over in the "HTML Writers’ Guild" and the award for the website continues the pattern of awards for the development of intellectual products in the paper world.

Home pages may have textual, visual (including both still and motion), and audio components. Table 5, "Visual and Audio Features of Home Pages," lists the frequencies of various features in the study sample.

Among 114 home pages, 44, (39 percent) have a standard white or gray background; 18 (16 percent) have a plain colored background other than white or gray, and 52, (46 percent) have some kind of patterned background, such as flowers, logos, etc. (regardless of color). There are also 90 home pages (79 percent) that use icons and/or artwork to represent or illustrate the contents of the pages. Also, as mentioned earlier, 40, (35 percent) of the 114 home pages contain photographs.

Some motion feature or features appear in 38, or 33 percent of all the home pages. Of these, in 33 the motion appears directly on the screen, while five must be clicked on to pull up the motion feature. (Motion includes scrolling text, blinking text, as well as full-fledged moving images.)

Sound features are relatively new on the Web, and correspondingly, showed up less frequently in the sample. Six home pages, (5 percent) contained an audio feature; in all cases, one had to click to summon the sound.

Personal home pages in this sample of 114 pages appeared to serve a number of functions for their creators. Fifty-one pages (45 percent) displayed professional capabilities and experience (with or without explicit mention of "resume"), and another 23 pages (20 percent) announced products or upcoming performances. Thus, though these were all personal home pages, they were also personal business home pages too.

In American society, however, presentation of one’s business background serves an important social function as well. When two people first meet, they generally give each other their names, then their occupations. So publishing this kind of information on a personal home page functions to announce oneself to potential social and business contacts both.

Another large group (33, or 29 percent) appeared to be simply enjoying the capability of presenting a home page to the world, and displayed no specific purpose other than this. A few home pages were serving miscellaneous other functions: Six of those presenting resume information explicitly stated they were looking for work. A smattering of others were seeking dates, engaged in political activities, looking for an adopted person’s birth parents, etc.

We had expected that the popular usage of the phrase "home page" might elicit a number of references to the personal home page as a "home" in the conventional sense of domicile ("home sweet home," "my humble abode," and the like), but only a half dozen such cases were found.

On the whole, the creators of the home pages seem to be open to it as a new social form, and the pages were relatively free of efforts to carry over old metaphors or forms.

A wide range of types of information was provided on the home pages. Of the content types that were tallied, an average of 5.9 items appeared in each home page. The ten most popular items, in descending order, from 92 percent to 19 percent, were email address, name, favorite websites, gender, photo, current work, educational background, hobbies or interests, mailing address, and previous work experience.

The great majority of home pages, 88 (77 percent), had an internal structure of text, many links, and often contained a menu or contents list at the head of the page. Of the remaining, 10 (9 percent) of the home pages consisted of menus solely, leading to other pages with content, and the final 16 (14 percent) consisted entirely or almost entirely of straight running text, with no or few links. Ninety-six percent of the sample contained links of some sort--to other locations on the page, to other pages, or to other websites. The median number of links per page was 14, with the highest number being 215. Forty-one, or 36 percent, contained links to search engines, and 42 (37 percent) contained links to shareware.

The home pages were observed to contain several features that would not exist or would be different in character in a non-electronic environment. Number of hits to the home page, latest update date, and the rubric "under construction," or variants thereof, all appeared on more than a third of the home pages. Intellectual freedom issues from the paper world were addressed here, too, with 25 instances of the free speech blue ribbon, and 15 copyright notices. In 20 cases, a "sign guestbook" feature enabled visitors to the site to respond to the site in a manner very similar to signing a guestbook and commenting when visiting a real home or museum.

Regarding the home pages’ physical appearance, 70 pages (61 percent) of the pages used a colored or patterned background (other than white or gray), and 90 pages (79 percent) used icons or artwork to represent or illustrate the contents of the pages. One or more motion features appeared in a third of the home pages; just five percent contained audio elements.

Overall, the home pages reviewed displayed a great variety of content and of specific types of formatting within broader formatting approaches. While some content elements were quite popular, none of them, not even name (90 percent), was found on all home pages. Evidently, the creators of these pages felt free to structure them as they wanted, and no standardized, or required fashionable approach was identified in the sample.

Nor did the pages evidence reliance on any single standard metaphor, such as home page as "home" in the sense of domicile. The closest pattern in the paper world that a substantial minority of the pages resembled was that of the resume, but even those home pages were quite freely and variously structured. In sum, based on this small exploratory sample, it appears that the form and content of personal home pages on the World Wide Web is still quite open and various. Though certain popular features may be found in it, the public social form known as the "personal home page" has not yet developed a fully standardized character and social role, recognized by all. The home page as a social institution is still very much under development.

Clyde, L.A. (1996) The library as information provider: The home page, The Electronic Library, 14 (6), 549-558.

December, J. and Randall, N. (1994) The World Wide Web unleashed. SAMS Publishing, Indianapolis.

Dougherty, D. and Koman, R. (1994) The basic handbook for Microsoft Windows. O’Reilly & Associates, Sebastopol, Calif.

Falcigno, K. and Green, T. (1995) Home page, sweet home page: creating a Web presence, Database, 18 (2), 20-28.

Mathews, O. (1974) The album of carte-de-visite and cabinet portrait photographs 1854-1914. Reedminster Publications, London.

Morgan, W.D. ed. (1942) The complete photographer: A complete guide to amateur and professional photography. vol. 2. National Educational Alliance, New York.

Spear, B.J. (1995) Preparing pages for the World Wide Web, Online & CDROM Review, 19 (6), 325-327.

van Brakel, P.A., Roeloffze, C. and van Heerden, A. (1995) Some guidelines for creating World Wide Web home page files, The Electronic Library, 13 (4), 1995, 383-387.