The Future of Open Source Shakespeare
Open Source Shakespeare has fulfilled its initial goals and in several respects gone beyond them. All but the most complex searches are completed in ten seconds or less, meaning it is quick. “Quick” is admittedly a relative term, and reflects my personal judgment that most users will be content to wait a few moments for accurate results. But simple keyword searches are typically returned in two seconds or less, and often take a mere fraction of a second. Right now, OSS is hosted on a shared Web server, but if it had a dedicated server, it would be blazingly fast. The big functions – advanced search, concordance, and statistics page – are all there, with the capabilities listed at the beginning of this paper. Of course, the site includes Shakespeare’s complete works, too.
Where will OSS go from here? Dozens of people have downloaded the OSS source code and database. A few people have inquired about its use in their own literary projects. Although OSS is designed with freely available tools and can be easily replicated elsewhere, modifying it to do something else would take a decent amount of work. This is not because it would be difficult, from a programming perspective – there are no arcane programming techniques, and any intermediate-level programmer could modify the code if he wished. The problem is the time commitment. A person would have to learn how to mark up the texts, modify the parser to accommodate them, set up some data in the database, and modify the view pages to display the new texts. Again, none of that is difficult, but it would take a while to execute.
On the other hand, that effort would pay off handsomely. The developer who modifies OSS would not have to design a database or think through all of the ramifications of storing a collection of texts and displaying them. The collection would have a ready-made concordance, a search function, and the statistics page could be adjusted for the new texts, too. OSS could process non-English texts, even with non-Western character sets, as all of the technologies used to build the site can handle UTF-8 characters, which display any language included in that standard.
What about the future of OSS itself? It is not in its terminal form – I hope to continue extending and refining it long after this paper is completed. I see three main possibilities for improvement:
1. Include multiple versions of the texts. The Internet Shakespeare Editions has already transcribed the folio and quarto versions of each text, with the original spelling. Having an editorial edition (Moby) alongside the early texts would be ideal: readers could use Moby for everyday use, and scholars could compare the early texts onscreen. There are some technical challenges to be overcome – namely, how does one collate, or “map,” the passages in one text to the passages in another? What about passages that are in one text, but not in another text – how will they be stored or displayed? I have no doubt that these issues are soluble, but they require careful thought.
2. Include folio and quarto images, audio clips, and video clips. There are sites such as the Electronic Text Library that will let you look up a passage, then display an image of a First Folio page onscreen, where you can see the passage yourself (Electronic). This strikes me as an extremely useful tool for scholars. Keeping track of which passage is on what page is a monumental task, so OSS would have to use texts that were already mapped to the pages. Such texts exist; whether or not they can be used legally is a different matter.
Considering the inclusion of audio and video clips may be a flight of fancy. It would involve taking very large computer files and breaking them up into smaller files, then mapping them to each passage. Yet would it not be wonderful to read a soliloquy, and then hear it read out loud – or, when you are trying to understand a passage of dialogue, to see actors interpret it on your computer screen?
I do not underestimate the amount of work involved with this. Completing all of the works would take years of full-time effort. But in the short term, I would like to take a single scene – most likely Act I, Scene 1 of “Romeo and Juliet” – and add multiple text versions, folio and quarto facsimiles, audio clips, and video clips. I have that particular scene in mind because the folio and first quarto versions differ significantly, so it would show the value in comparing variant texts side-by-side. Also, the scene has a lot of action, and it is universally well-known, even to high school students who started to read the play and then decided to fake it for the test.
3. Build another site, with another text collection. I have thought of the Gospels or Chaucer’s works as possible candidates for a new collection, to demonstrate that OSS’s parser, database, and display code could potentially ingest and display any kind of literary work. That may happen eventually, but the thought of embarking on another project like Open Source Shakespeare, even one requiring far less effort, makes me want to lie down for a while.
If I had thought about it, I would have recorded the amount of time I spent developing OSS from its inception. Since I started it on a whim in the Kuwaiti desert, I have spent at least 500 hours on it, and probably significantly more. Using a relatively low billing rate of $100 an hour, that would make OSS’s theoretical value something like $50,000.
That does not mean it could be sold for that much. If it were used commercially, it would have to use a modern editorial edition as its texts, which would have to be licensed from its publisher. Then the texts would have to be converted to the OSS format. Still, with a month of steady, full-time work, it could be done.
Ultimately, I would consider donating OSS to a foundation or an educational institution. I could make some changes so the whole thing could work on a single server, or a group of servers, and after that it would pretty much run itself. I would only do this if the recipient wanted to continue the project as a going concern; I would not want to give it away, only to watch it die from neglect as other sites arise to surpass it.
It is also satisfying to know that OSS is gaining public attention. I have received unsolicited positive messages from every part of the world, including professors from the U.S., Canada, the U.K., and Argentina. Dozens of other Web sites have linked to it, many of them singling it out for praise. About twenty sites have it listed on their “permanent” links, with blogs making up most of the total, but some institutional sites link to it as well, including the Cleveland Public Library and the Shakespeare Theatre of Washington, D.C.
According to Awstats, a program that generates site usage reports, OSS had about 7,000 unique visitors in April 2005, a respectable total for its seventeenth month of release. To give an idea of the site’s global appeal, users in each of the following non-English-speaking countries downloaded more than a hundred pages from the site: Germany, Japan, the Netherlands, Hungary, Hong Kong, China, and Singapore.
If nothing else, I hope Open Source Shakespeare demonstrates that you can build a useful literary site using off-the-shelf technologies, public-domain texts, and Web development skills. There are many other Web-based projects that use the same elements, but I believe my site is unique in that it is free, and that you can download it for non-commercial use. I hope that other people will use the code and database as examples for their own work, and I hope that Shakespeare lovers and scholars everywhere continue to embrace it.