John Haltiwanger is a New Media MA graduate and an autodidactic programmer with a strong interest in typesetting and open source software. Haltiwanger collaborates with the Open Source Publishing platform and Universiteit van Amsterdam. The main focus of his presentation is generative typesetting, with his MA thesis used as an illustration. Haltiwanger argues for liberating humanities from proprietary control of tools such as Microsoft Office or Adobe Suite by implementing open source tools within academia. A man standing behind his beliefs, for his presentation he uses an open source version of Prezi (an alternative to PowerPoint) – Sozi.
John Haltiwanger @ The Unbound Book Conference photo cc by-sa Sebastiaan ter Burg
“It’s not who or what you are, it’s where you’re at” (reference to Rakim’s “It’s not where you’re from, it’s where you’re at”) opens the third presentation in the Open Publishing Tools panel on Day 1 of the Unbound Book Conference. Haltiwanger starts by mentioning LaTeX and LyX, common libre tools which can be successfully used for typesetting documents such as theses and argued for their superior typographical and referencing management advantages. However, he also mentions that extensive stylistic customization in these tools can pose major problems and that such realization lead him to exploring other options and discovering ConTeXt.
Haltiwanger exemplifies the possibilities enabled by tools such as ConTeXt with his own Master thesis whose case study was its own typesetting. What follows is a discussion of the technicalities of producing the thesis using generative typesetting, such as the necessity of setting it in both HTML and PDF and being dependent on automation. Later he explains how people began to start applying the visually semantic developments found in email communications (such as ALL CAPS to indicate shouting or underscores for _emphasis_) to enable a precursor format for generating HTML (an example being Markdown) and concludes that in terms of informational impact and widespread use, MediaWiki has been the most successful visually semantic format. However, he doesn’t see wikis as particularly fruitful in producing essays because of their fragility and not fully flowing visual semanticization. On the other hand, the relative popularity of wikis within the humanities proves that it is not so difficult for people to comprehend and work with visually semantic textuality.
The core of Haltiwanger’s discourse on generative typesetting is unraveled within the introduction of Subtext, a tool he is designing. Its most distinguishing characteristic is transformability of both the semantics and procedures of dealing with them. In result, the same semantics can be interpreted in multiple ways and a file can be easily made into a PDF for screen or for print; an HTML version or ePub can also be generated. Thus, he believes that the Next Great Format does not pose threat to Subtext. While Microsoft Word privileges the human and HTML privileges the computer, Haltiwanger envisions Subtext as introducing a productive balance of agency between these two, while at the same time bringing out the best in the text itself. An effect of this balance is that tools for distributed source code development could be applied in a generative typesetting.
Some controversies during the Q&A session are driven by Haltiwanger’s suggestion that the contribution of these developer tools could possibly revolutionize the class room in academic humanities’ workflows, collaborative homework and peer review situations. While the server knows who each individual contributor is, it does not need to give this information to others and therefore enables for more just grading or collaborative work. While Haltiwanger imagines the tool to allow teachers to have new ways of having their stylistic wishes respected and for new ways of grading and reviewing, some of the audience members voice their concerns that he suggests machines (the server) grade human contributions based on the quantity and not the quality of input. Haltiwanger acknowledges those doubts with a clarification that this was not his suggestion and that by keeping the interface of the tools flexible, anything can be imagined: live anonymous peer review, conversations occurring without the power dynamics of names and granular grading of group writing are just the tip of the iceberg.
The conclusion of Haltiwanger’s presentation is that while current generative typesetting workflows are still too complex for a widespread implementation, Subtext as a F/LOSS tool is capable of reflecting the relative simplicity of humanities’ workflow. People need to care about open source tools in academia and give up the embodied comforts of the current proprietary workflow. Humanities writing can be successfully liberated from proprietary control through merging the toolsets of distributed programming and reconfiguring them for one’s own specific needs. While rather technical, Haltiwanger’s presentation is inspiring: although still a distant vision, a widespread implementation of open source tools within academia would no doubt enable many new possibilities.
View Presentation here: http://drippingdigital.com/conf/unbound-book/textual-liberation.svg
Text document of notes here