TextBook Project

16 December 05

A little history about the TextBook, the community-produced documentation wiki for the Textpattern publishing system.

Background

TextBook is a wiki documentation project for the open-source publishing system called Textpattern. The TextBook project was marshaled by yours truly, largely out of personal frustration when first coming to know Textpattern, but encouraged by the fact many others in the Textpattern Support Forum community were clearly expressing themselves for authoritative documentation too.

Although each day that passes widens the gap, it wasn’t all that long ago that TextBook was launched, and since then, Textpattern itself has improved leaps and bounds and new sources of Textpattern documentation have sprouted up; some have perhaps even become a bit more official, such as the Textpattern FAQ. In any case, TextBook’s road hasn’t always been a clear one, and the verdict is still out for its future (though the future should be good), but one thing for sure, the project has been an interesting and valuable experience in both community dynamics and wiki management.

TextBook Milestones

As you can see, things happened pretty rapidly in the beginning.

  • 13 December 2004: I’m seeing red from clean URL troubles, or perhaps had too much coffee, and make a case for a centralized documentation resource for Textpattern — a Textpattern User Manual. (The manual is initially conceived as having Chapters and Appendices, just as an application manual would, and follows that organizational model for quite some time.)
  • 14 December 2004: After a bit of banter and product investigation, MediaWiki is selected as the best tool for the job and Remillard gives the manual it’s first home; thereby becoming the first backend administrator of the project.
  • 16 December 2004: With a unanimous vote of 2 to 1 (a remarkably huge community feedback response of three), the manual is dubbed TextBook, a label suggested by Alicson, coordinator of the Textpattern Resources site.
  • 16 December 2004: Dean Allen, creator of Textpattern and CEO of TextDrive, endorses the TextBook effort and offers it an official home under the Textpattern flag. Two days later, Remillard sets things up at http://www.textpattern.net/wiki/. All in all, a significant milestone indeed.
  • 19 December 2004: Joel Dueck offers up his extremely popular article, Textpattern Semantics. I personally respond to Joel with acceptance but never hear back so I presume he’s busy with other things, we probably both forget about it, and the article is never added until like 8 or 9 months later by a third party. But Joel’s a hip dude, has a groovy Web site (glad to see the chipmunk back, Joel), is a good writer, and his article has helped untold thousands of people so no mention was ever made…until now. Oh, and he has an interesting theory (though counter-productive) about structured documentation efforts in a community environment. Judging by how slow documentation has creeped along, I’d say his theory is yet to be disproved.
  • 21 December 2004: Perhaps riding a bit to high on the seeming success of TextBook to date, and like a parent eagle encouraging the chick to soar, I propose a “Top 10 Textpattern Sites” contest based on those in the TxP Sites collection. The idea being to use the cream of the crop as case studies in TextBook, giving potential TextBook authors something to actually write about. The idea failed to be very popular with the community and died away.
  • 26 December 2004: Alexandra, a prominant voice in the community and probably most known for her webzine, TxP Mag (of which TxP Sites is a component), and for Textpattern Resources [ge] (the German sister site of Textpattern Resources), gives TextBook its first, major (and needed) value challenge.
  • 27 December 2004: I respond in TextBook’s defense with perhaps a bit too much zeal. My response was really written to anyone else having similar views more than it was to Alexandra herself; her post simply opened the door.
  • 27 January 2005: TextBook, which up to this point had been running on a good faith sign-in basis, gets hit hard with spam, and I learn under fire what it means to roll back a diff in WikiMedia. From this point on, users must request a wiki account from administrators.
  • 31 March 2005: Dean Allen contacts the TextBook administrators (at the time, Remillard and myself) and informs us he likes what he sees, gives us full charge, and calls TextBook the “authoritative” documentation for Textpattern. Another huge milestone. Dean also introduces the idea of a documentation migration process, from wiki to Textpattern CMS, starting with an FAQ. (Side: For some reason I lost track of this FAQ biz, completely forgetting about this discussion with Dean, and go on to make an ass of myself in the community here, here, and here about why there’s two different documentation efforts, namely TextBook and the FAQ. The next milestone complicates things even more.)
  • 15 April 2005: TextBook begins supporting multiple languages, and the TextBook Language Support information is put into place. Only now, as I work on this history, do I realize this milestone likely sealed TextBook’s fate with respect to ever migrating to a CMS platform, which was the expectation for TextBook all along.
  • 13 May 2005: DavidM, one of the many outstandingly generous individuals in the community, becomes one of the first TextBook translation leads (French). Perhaps fueled by his newfound Sysop power in the wiki, or by painful awareness of TextBook’s idle state, David makes the call to arms. His valiant effort rifles a few burps and scratches, but the community quickly falls back into literary slumber.
  • 28 October 2005: After too long of period of relative inactivity, and a series of new threads in the Forum about Textpattern’s documenation still lacking (for example), I pull my head out and stick my foot in, again, by rousing a roundtable discussion about Textpattern’s documentation efforts overall. For better or worse, and I hope better, it resulted in the first TextBook redesign effort, and as a side, a new forum, Documentation Discussion, was created in the Textpattern Forum for which I become Moderator.

At this point, the redesign effort, particularly the wiki re-skinning process, is still in progress (aim is to get away from the typical MediaWiki mold, and produce something a bit more polished like the Mozilla Wiki skin, if only a little bit).

Reflections

Below are a few interesting observations from working on this project.

Project Leadership

Whether it was because I marshaled the TextBook effort, have been an active proponent for the project, developed the wiki’s initial content structure and organization, provided editing services, served as an administrator, communicated between Dean Allen and the community on TextBook issues, or all of the above (most likely), I seem to have become recognized by the community as the “Lead” for the TextBook effort.

I never considered the notion of project leadership when I first proposed the TextBook idea to the community. I simply saw a real community need, voiced a possible solution, and figured (if warranted) the community would follow up and shape the manual into existence; I was just a community participant who was willing to put my money where my mouth was, as the saying goes.

You begin to get a sense of things when other members are turning to you with project questions, or have interest in your opinions about project direction, which in itself could be nothing more than respect for ones voice, but it really sinks home when someone comes right out and uses the word “lead” (or a variation of the implication) in dialogue. When this first happened, which I believe was in communication with Wilshire (another outstandingly generous member of the community who has developed several of Textpattern’s most popular plugins), I found myself arguing against the idea. As more time has passed, I’ve become a bit more comfortable in the lead role, even accepting of it. However, I do find myself wondering more and more at what point does it end; do I hand the reigns to someone else, or is there another path for TextBook, one yet to be realized?

The “Official” Factor

The term “official” has been thrown around pretty loosely in the community with regard to TextBook, and I’m likely to blame for that; but I’m sure I initially used the word to give the impression that TextBook was meant to be a centralized documentation resource about Textpattern, and not some distributed Weblog affair. As I’ve always envisioned it, TextBook is a single Textpattern-sponsored source to help answer user questions about the system without the user having to search the Web high and low.

When Dean Allen dubbed TextBook as the “authoritative” resource for Textpattern (note), I further made the mistake of advertising TextBook as “officially adopted,” which for all intended purposes was true, but looking back, and for the sake of community good, I shouldn’t have expressed it as such.

The term has never really sat well with many popular and respected members in the community, which may have more to do with profound relationships between communities and open-source documentation efforts in general (a seemingly very amorphous thing) than being directly related to something more superficial, but I don’t know for sure. For newer community members, the term seems to only confuse, especially when they see TextBook growing so slowly and other things like the Textpattern FAQ come into existence.

Whatever the case, an interesting phenomenon is taking place: TextBook is becoming less official in nature, but more popular in standing. One can begin to see this when people like John Hicks talk of TextBook’s usefulness, as he does with respect to the tags index, see New Tag Goodness.

To Wiki or Not to Wiki

I am an ardent proponent of using a wiki for collaborative documentation.

(Ed. — I’ve talked of this and other related points in a more recent article: The Return to Textpattern, and TextBook.)

TextBook’s Future

I think it’s safe to say that TextBook has a long future ahead of it so long as there are people grooming it into maturity, but whether it exists as the sole source of authoritative documentation or becomes relegated as a first stage drafting area of articles bound for a CMS somewhere else remains to be seen. I have a hunch the latter is what we will eventually see, with both wiki and CMS having critical roles — the translation efforts of authoratative documentation being a good example, which at this point in time relies upon MediaWiki use.

Obviously there are pros and cons either way, but no matter what, the real issue is the content, the words that get written, the information that actually helps people use the Textpattern product. If you don’t have the words, it makes no difference what platform is being used.