PureGin.org

Sixteen years ago, I started translating "FLP": A. Fathi, F. Laudenbach and V. Poénaru's notes (in French) from a Seminar on the work of Bill Thurston on surface diffeomorphisms, measured foliations, and hyperbolic geometry. At the time I thought it might take a couple of summers (this being a fundamental division of time in the life of a graduate student). Time passed, and many things got in the way, including a thesis, work, offers to star in Hollywood movies... well, OK, not the last. The project had been sitting on a back-burner since 1996, when a lovely summer in Montreal provided the perfect setting for a bout of translation work. I faithfully copied my files from an old Compaq 8086 portable, to a Sparc Classic LX, to a Sparc 20, to a Pentium II Linux machine, and finally to my current Apple Powerbook, never fully expecting that they would ever see the light of day again.

Thought people might be interested in some download stats:

These are from my server logs, since I started automatically generating Handbook PDFs on a weekly schedule on Nov. 1st:

(URLs relative to http://www.puregin.org/)

I set up a cron job to automatically generate the Drupal handbook PDF files (using the old HTML2PS based scripts). The job runs weekly, early on Monday morning Pacific time.

I attended a presentation tonight on DITA, hosted by the Society for Technical Communication, Canada West Coast Chapter.

Paul Prescod and Su-Laine Yeo of BlastRadius gave an overview of this XML-based standard/architecture for creating, using, and re-using content, and illustrated how some of the key features, such as topic focus, information typing, composition of documents using maps, customization of document specifications via specialization, and definition of re-usable content/document components can be used within BlastRadius's XMetaL DITA edition software.

Just as object-oriented software design can be regarded as a collection of conventions and tools to support re-use of software components and larger structures, the design of DITA can be regarded as an attempt to bring a 'topic-oriented' approach to the design, implementation, and maintenance of large documents.

1. Morkes, J., and Nielsen, J. (1998). Applying Writing Guidelines to Web Pages. http://www.useit.com/papers/webwriting/rewriting.html

Our study suggests that scannable, concise, and objective writing styles each make a positive difference in Web users' performance and subjective satisfaction. Promotional writing, which is the style most commonly found on the Web today, had much lower scores on virtually all usability measures.

Much work remains to fully understand the best way of writing for the Web. In our Study 3, we treated the three variations (concise, scannable, and objective) as dichotomies, whereas to a writer, they are continuous dimensions. A simple follow-on study would vary each of our dimensions in a number of steps to assess the shape of the usability curve. For example, conciseness could be tested with word counts varying in 10% increments from 10% to 90% of the control condition. Similarly, scannability could be tested with fewer or more headings, with no bulleted lists, and with more or less use of highlighted words, background coloring, pullquotes, and other typographic tricks. We welcome any such additional studies, and to facilitate them we have made our original test pages and the complete set of test questions available for download at this URL:

To determine how much better or worse in percentage terms each site version was relative to the control, we normalized all participant groups' mean scores for the 5 major measures. For each measure, the control condition's mean score was set to equal 100, and the other conditions' mean scores were transformed (by division) relative to the control (see Table 2). Scores above 100 are "better" than the control, and those below 100 are "worse."

Next, we calculated an Overall Usability score for each version of the site, by taking the geometric mean of the normalized scores for the 5 measures (the geometric, rather than arithmetic, mean was used because we compared ratios). Again, the control version's score was 100.

Version	Task Time	Task Errors	Memory	Sitemap Time	Subjective Satisfaction	Overall Usability
Promotional (control)	100	100	100	100	100	100
Concise	172	205	142	124	156	158
Scannable	157	273	94	130	133	147
Objective	128	164	116	121	112	127
Combined	242	818	162	142	122	224

Table 2. Normalized mean scores for five major measures and Overall Usability. Scores above 100 (the control score) are "better." For example, the scannable version is 57% better than the control for Task Time.

Main measurements are presented in Table 1.

Condition	Task Time	Task Errors	Memory	Sitemap Time	Subjective Satisfaction
Promotional (control)	359	0.82	0.41	185	5.7
	(194)	(0.60)	(0.14)	(43)	(1.5)
Concise	209*	0.40+	0.65**	130***	7.1*
	(88)	(0.70)	(0.21)	(41)	(1.9)
Scannable	229*	0.30*	0.55*	198	7.4*
	(86)	(0.48)	(0.19)	(93)	(1.8)
Objective	280	0.50	0.47	159	6.9*
	(163)	(0.53)	(0.13)	(69)	(1.7)
Combined	149**	0.10**	0.67***	130**	7.0*
	(57)	(0.32)	(0.10)	(25)	(1.6)

Table 1. Mean scores for five major measures. (Standard deviations appear in parentheses.) Time measures are in seconds, Task Errors and Memory are percentage scores, and Subjective Satisfaction is on a scale from 1 to 10.

+ p < .10 * p < .05 ** p < .01 *** p < .001 (test for significant difference from control condition)

Time to recall site structure was the number of seconds it took users to draw a sitemap.

A related measure, sitemap accuracy, was a percentage score based on the number of pages (maximum 7) and connections between pages (maximum 9) correctly identified, minus the number of pages and connections incorrectly identified.

Subjective satisfaction was determined from participants' answers to a paper-and-pencil questionnaire. Some questions asked about specific aspects of working with the site, and other questions asked for an assessment of how well certain adjectives described the site (anchored by "Describes the site very poorly" to "Describes the site very well"). All questions used 10-point Likert scales.

The subjective satisfaction index was the mean score of the following four indices:

• Quality of the site comprised four items: accurate, helpful, useful, and the question "How satisfied are you with the site's quality of language?"
• Ease of use of the site comprised five items: easy to use, and the questions "How easy is it to work with the text in this website?", "How easy is it to find specific information in this website?", "Compared to what you expected, how quickly did the tasks go?", and "How hard was it to concentrate on searching for information (because of distractions)?" (this item was reverse-coded).
• Likability of the site comprised six items: entertaining, interesting, likable, engaging, fun to use, and boring (reverse-coded).
• User affect comprised three items (all reverse-coded): "How tired do you feel right now?", "How confused did you feel while working in this site?", and "How frustrated did you feel while working in this site?"

For each index, the items were averaged so that the possible range was from 1 to 10.