The results page has several components, the most important of which is the list of textual parallels. Other components include a set of options for sorting and exporting results, a pager, and session details.
The table of results has four columns.
The line number and text of the target (alluding) phrase. The matching words are highlighted. Click on the line number to display the target phrase in context in a new window.
The text of the source (alluded-to) text. Again, clicking on the locus should display the phrase in context.
This gives the features shared by the two phrases. The matching features may not be the same as the highlighted words—for example, if the feature set searched on is stems, the default, then this column will list the dictionary headwords matched, not the specific inflected forms. In cases where an inflected form is ambiguous, it may cause several possible stems to appear here.
This score is a guess as to the significance of the result. Score is based on the frequency of the matching words and on the distance between them within each phrase. For more information on how these are calculated, please look at the “Advanced Search” help page.
Initial testing has shown that Tesserae's score tends to correlate roughly with human-perceived literary significance, but it remains a work in progress. Scores may change as we continue to update the program.
Adjust the options at the top of the page and click “Change Display” to re-sort your results without performing the search again.
By default, results are sorted according to location in the target text. You can also sort by location in the source, or by score.
Change the sort order to “decreasing” when sorting by score to put the most significant results first.
By default, the table is split into “pages” of 100 results. Choose more or fewer from the drop-down list and click “Change Display” to change. An abbreviated page index appears above the table. Selecting “all” will turn this feature off, displaying all results on the same page.
By default, results are displayed as an html table. If you change the “format as” drop-down, clicking the “Change Display” button will cause the results to be downloaded to your computer as a separate file.
This format is appropriate for import into a spreadsheet such as Microsoft Excel or OpenOffice Calc. Each result appears on a separate line. Columns are separated by commas. Text appears within double quotation marks. Newlines are Unix-style.
CSV results will be sorted according to the options set when you click the button.
Results are marked up in a custom XML format. This is probably only useful to advanced users who want to do their own automated processing of the results file. For more information please email us.
This section gives a summary of the search settings which produced the results. The search itself is done only once, when you click “Compare Texts”; these settings are not affected by subsequent changes to the sort order or other aspects of results display.
This information is for the most part probably not helpful to casual users. It can be useful if you are saving and comparing results from multiple searches. Note that the complete stop list is recorded here.