A paper presented at the
42nd Annual Meeting
Society for the History of Discoveries
Denver, Colorado
September 8th, 2001 (minor revisions, February, 2006)
The author's research on the human and cartographic history of the Galápagos Islands began in the traditional manner: on paper. But given the inevitable cross-referencing between maps and manuscripts, books and other sources, the HTML document format lent itself nicely to bringing some order to the resulting bibliography. In no time at all, the bibliography became the focus of the project, and a move to the web was the next logical step.
This paper reviews the rather short history of the author's discovery of the Internet as an ideal medium for the presentation of his topic at several levels, from a simple bibliographic reference work to an in-depth hyper-linked resource of all those maps and manuscripts. The same techniques can of course be applied to the production of a CD-ROM disc, and may be of interest to others doing research in the history of discoveries.
The conventional bibliography is of course a static device, tucked away in the back of a book to reveal the author's sources. At its simplest (and often, only) level, it gives the reader enough information to identify each source. An annotated bibliography may supply more information—a welcome addition for those who want it, but apt to “get in the way” of a simple perusal of the source list. By contrast, an electronic bibliography can present the same information at several levels. Unlike its predecessor, in which an underline does nothing more than add visual emphasis, here each underlined word or phrase indicates a link to another location. The target of that link may take the viewer to another location within the same document, to another document in the same collection, or to a website on the other side of the globe, as shown by this typical example:
Bowen, Emanuel | |
---|---|
1744 | The Gallapagos Islands Discovered and Described by Capt. Cowley in 1684. In John Harris' Navigantium atque Itinerantium Bibliotheca (Volume 1, p. 79, 2nd and 3rd editions). Bowen's map is based on a Herman Moll map in “Cowley's Voyage Round the Globe” but with Text Commentary about Captains Woodes Rogers and [Edward] Davis added at bottom. Compare Bowen and Moll maps. Compare five versions of this map. |
At first glance the entry looks like an onscreen version of the traditional paper document, and the viewer who seeks nothing more than a conventional bibliographic citation has it here, just as it might appear on paper. But a single click on one of the underlined segments presents more information; either the actual map cited in the entry, details about the edition in which it appears, or supplementary information about related documents and maps. Thus, this simple entry offers the viewer convenient—and instant—access to the following information:
Of course all of the above resources are available on paper, but are presumably not within easy reach of every viewer. Or even if they were, it would be awkward at best to find the desired information within each document in a reasonable amount of time. But thanks to HTML (Hyper-Text Markup Language) and modern browser technology, the resources of any researcher can be made readily available to a much larger audience than would otherwise be possible.
The specific examples given here are taken from the author's own website but of course can be modified and applied elsewhere as appropriate. The underlying code is not presented here, but is readily available to anyone who visits the site at http://www.galapagos.to.
On traditional printed media, a comparative study of any two maps or charts is subject to the limitations of the obvious side-by-side examination. Especially when comparing the old and the new, there is little likelihood that the two items will be drawn to the same scale, thus making it all the more difficult to compare physical relationships between places found on each. Computer technology makes it reasonably easy to conduct a detailed comparative analysis that would not be practical by any other means.
As a typical example, the map collection at the Library of Congress contains a ca. 1530 vellum chart of the Pacific coast of Central and northern South America. To compare its coastline with a modern map of the same area, the old chart and the new map were both scanned and imported into Adobe Photoshop. The distance and angle between two prominent points (Cabo Blanco, Costa Rica and Punta Mariato, Panama) were noted on the modern map. The same landmarks were found on the vellum chart and again the angle and distance were noted. This image was then resized and rotated to align the landmarks with their modern counterparts. Thus the old chart had now been “drawn” to the same scale and orientation as the new map.
The next step was to make a copy of the old chart and remove all landmarks, rhumb lines and other extraneous details from it, leaving nothing but the coastline on a transparent background. This image was saved as a separate file, and then placed directly on top of the other version. For the moment, the presence of this extra coastline layer has no visual effect, because it simply duplicates the original coastline.
But now, the image of the coastline layer may be dragged from the old chart to the new map, to see how it compares. If any one prominent point is aligned with its modern counterpart, the relationships between other points along the coastline are immediately apparent, and we have a good idea of the accuracy (or, lack of it) of the old chart.
Compare coastline with modern map, by dragging coastline image.
Alternate step-by-step comparison
As a variation on the above technique, it may be convenient to make a simpler comparison of two charts. The obvious procedure would be to place them side-by-side within a browser window. Of course a single comparison might show either the relative size of one chart with respect to the other, or the two charts displayed to the same scale, but not both views. The solution here would be to present one of these views, with a link to display the other. Thus the viewer could make whichever comparison was required, and have the option to conveniently view the other one too, if desired.
Compare Bowen and Moll maps. [This example adds the resizing feature to the Bowen/Moll comparison in the “Hyper-Linked Bibliography” example given above.]
Another technique may be convenient in a comparison of the same area as seen on several maps, perhaps to show the influence of one cartographer on another. For example, Aaron Arrowsmith published four versions of his map of the Galápagos Islands, the first of which appeared in James Colnett's Voyage to the South Atlantic. Although all four maps bore the same date (1st January 1798), certain island details were changed between the first and last editions.
Compare same island detail on different Arrowsmith editions.
By comparing another detail seen on the charts of Arrowsmith and John Fyffe, it's reasonably easy to deduce that Arrowsmith modified his 1798 map based on details found on Fyffe's 1815 chart of the same area, and these modifications appeared on Arrowsmith's 1817 draft and subsequent 1820 reprint edition. The influence of Arrowsmith on Vandermaelen's 1827 map can also be shown by including the same area from the latter's map in the comparison.
Compare details on Arrowsmith, Fyffe, Vandermaelen charts.
The track of H. M. S. Beagle through Galápagos was reconstructed by consulting the bearings in the Captain's log. As a typical example, the following table lists three bearings taken at 8:00 am on 10th October, 1835.
Landfall Description in Log | Bearing in Log | Inverted to: |
---|---|---|
Rocky Islet | S 14°30' W | N 14°30' E |
Peak on Breakfast Isle | S 52° W | N 52° E |
West extreme of James Island | S 81°40' W | N 81°40' E |
If 19th-century observers at each landfall had taken simultaneous bearings on the Beagle, the intersection of these bearings would fix the ship's position at the time the entries were written into the log. A simple inversion of the Beagle log entries (shown in column 3 above) provides these land-based bearings, thus providing the data necessary to plot the ship's position. But in order to do so, we must first determine the modern counterparts of the ambiguous “Rocky Islet” and the obsolete “Breakfast Isle.” The ship's log does not provide the necessary information, but in his 1844 List of Documents …, Robert FitzRoy writes that “Four miles East of Jervis Islet [the modern Isla Rábida], are the little Breakfast Islets.” Although this identifies the log's “Breakfast Isle” as the modern Isla Beagle, it calls into question the log's “West extreme …” notation. But if this is re-written as “East” instead, then the conflicts are resolved: The “Rocky Islet” is either the modern Isla Eden or Guy Fawkes. The “West extreme …” entry should indeed have been written as “East” and―given the ship's approximate location―a bearing taken to the east extremity of James Island would seem more likely in any case. A detail view showing several possible interpretations of the bearings helps confirm the actual location of H. M. S. Beagle at the time the bearings were entered in the log.
During the preparation of a JPEG image of the Galápagos chart, numbers were inserted along the Beagle track to indicate the date at which each bearing set was recorded in the log. The HTML usemap attribute in the image-source tag for the chart is a convenient means to assign a specific link to each date seen on the chart. Thus, by clicking on any number along the ship's track, the Captain's log for that date is displayed in a new window. Or, Charles Darwin's diary entry for that date may be viewed instead by pressing any key prior to clicking on the chart. Links within the displayed log or diary permit the user to toggle between these documents, plus FitzRoy's 1839 Narrative of the Surveying Voyages … .
Display track of H. M. S. Beagle.
Given the physical size constraints of even a rather large computer monitor screen, there may be a problem displaying a map or chart with readable placename text. Or, the nature of the original document may not lend itself to printed labels, as in a satellite photo where such labels do not exist, and if added, might obscure important image details. In a case such as this, a small empty text box can be positioned at a convenient onscreen location. As the viewer moves a mouse pointer over the displayed image, the appropriate identification of the area under the pointer appears in the text box.
Display satellite photo.
This technique can also be applied when presenting an image of a complex area. If all place names were printed on the image, the image itself would disappear beneath the text. By displaying the image alone, the viewer has an unobstructed view of the landmass and can identify any point on it by simply “pointing” to it with the mouse.
Display chart with complex detail.
One welcome advantage of presenting maps and manuscripts on the Internet—or on a CD-ROM—is that a certain amount of virtual “restoration” may help produce a virtual document that is in better apparent condition than the real one. As one example, an old map that has separated into two or more segments over the years can be put back together again. And, within reason, missing fragments can be reconstructed by cloning adjacent intact sections or by pasting a copy of a visually similar area over the missing portion. Finally, colors can be enhanced, stains removed and text sharpened. These and other techniques are possible with most modern image-editing applications and, if not overdone, can enhance (without distorting) the appearance of a time-battered document.
Restoration Example 1
Restoration Example 2
NOTE: If such techniques are applied to enhance an old document, a prominent notice should be displayed so that the viewer understands that the image does not represent the actual state of the original.
The examples given above describe only a few of the many HTML techniques that can be used to present a bibliography of historical citations in a format that takes that bibliography far beyond the limitations of the traditional print medium. The same techniques can be applied to any website devoted to the history of discoveries, and with little or no further modification used for CD-ROM production of discs for use as an alternative to online access.