The American Verse Project represents a major and successful effort to put full‐text poetry on the Internet. It focusses on American poetry published before 1920 and, while it is not at all comprehensive, the depth it has can be quite breathtaking. At first glance, it is neither an exciting site nor a dull one. The initial page sports one main attractive graphic along with links to various types of searches on a neutral background. The initial page also makes very clear from where the page came, the conditions of its use for institutions as well as people, and the means by which one may access the poetry. I considered it a very solid first impression; and I looked forward to exploring the page.
There are two main ways to search the collection. One can search what is called “The book‐bagged amverse”. Using this page, one can limit a search to selected works within the collection. One can also search the entire Amverse collection. In both cases, the user may use “simple searches”, “Boolean searches”, and “proximity searches”. None of the searches requires (or allows) very advanced search strategies. With the “simple search”, the user may only enter one search word. The only way to manipulate the results is if the search provides more than 100 hits. In that event, the user may choose to view the first 100, a sample 100, or all of the records. It is not a very precise way to search; but it does put the user in the thick of full‐text poetry rather quickly.
“Boolean searches” provides a bit more searching ability in that the user may enter three search terms instead of just the one allowed with the “simple search”. These three terms may be searched using the AND/OR/NOT operators. What’s more, the user is allowed to look for the search term in various fields within the records (i.e. poem, entire work, line of verse, etc.), and that can be quite handy. As with the “simple search”, the user may manipulate the results if the search provides more than 100 hits.
“Proximity searches” allows the user to look for given search terms following or preceding another search term by 40, 80, or 120 characters. Again, the user may manipulate the results if the search provides more than 100 hits.
There are two other ways to search the American Verse Project and these involve browsing. The first browsing link from the main American Verse Project page is for the American Verse Project Texts. On the other side of this link is a list of authors in alphabetical order by last name. From here, all the user needs to do to view works by a given author is click on the last name of the author. The user is then taken to a short introductory page on the author with links to the author’s work in HTML or SGML ‐ I found this quite impressive.
The second link that makes use of browsing is the link to the “HTI Bibliography of American Poetry”. This is a Hyperbibliography ‐ it links to full‐text when it can; but it lists other items in the University of Michigan libraries as well. The Hyperbibliography also links to other online full‐text poetry collections, most notably the Chadwick‐Healey, American Poetry Database and Making of America, as well as the American Verse Project. It all adds up to a veritable wealth of full‐text poetry; and the user can browse the Hyperbibliography in full or by the letter.
Getting around the site is very easy. Links are provided so the user can venture all over the site and still know how to get back to the main page. The only instance this is not the case is when one jumps from the main page to the book‐bagged amverse. On that particular page, there is no link to get back to the main page. It is an easy matter to accomplish getting back to the main page. I would consider that a trivial inconvenience in comparison with the rest of the site. One of the more astounding aspects of the site is the depth to which it goes in covering the works of given authors. For example, I chose to look at the works of Emily Dickinson. In the bibliography (found in the “Browse the American Verse Project Texts”), there are three links to her poems, series 1, 2 and 3. Each one of these series is 200 pages long or more. Granted, the pages are short; but they are still 600 pages devoted to the work of a single author; and this is typical of the whole Web site. While not every author is covered in such depth, this indicates that the creators of this page have no fear of covering the larger works. Too, they did the job very intelligently in that only small chunks of the works are covered at a time making them easy to digest for the user.
At the bottom of the main page, one can find links to other humanities text initiative projects and resources. While some of these resources are for the use of University of Michigan students, there remain a large number of other sources (i.e., The Book of Mormon, The King James Bible, Samuel Johnson’s A Dictionary of English, etc.) to which the public has free access.
There is one thing that troubles me about the site: it was last updated in October 1996. While it is true that poetry published before 1920 is not changing all that much these days, one wonders if the items listed on the page found through the “Volumes of poetry in‐progress” link (which is on the “About the HTI American Verse Project” page) will ever be made available.
One rather fun part of the site was the random poetry generator. This link may be found on the “About the HTI American Verse Project” page. All the user needs to do to create an amusing poem using lines from the complete collection of the American Verse Project is click on the link. The rest is automatic; and the results can be quite humorous. To conclude, this site is marvelous. The creators of the site created a very user‐friendly Web site that offers the full‐text poetry online. They also limited their scope to American poetry published before 1920 so that, while it is not a comprehensive site, there is a great amount of depth to what is present. Overall, I found the site a magnificent example of the potential offered by the Internet; and I recommend a visit if not a bookmark.
