Meeting notes from Seattle meet-up

Are now available here, thanks to Paul Furio.

It was a long discussion (three hours!), but there was particular interest in setting up events for Seattle PAX 2010 and possibly for SF/reading-related conventions in the area as well.

Another area of interest was interpreter possibilities for browsers and mobile devices (including the idea, which got a lot of play at PAX East, of having browser interpreters capable of preserving transcripts from every player, so that the author could analyze the results and tweak the game accordingly).

We also talked a fair amount about choice-based (rather than setting-based) approaches to IF design. We were focusing in particular on a draft tutorial that Ron Newcomb is writing that introduces I7 starting with the concept of beginning and ending scenes, and building up a plot. The conversation kind of spread outward from there, but I thought Ron’s ideas were pretty interesting, especially in light of some recent blog posts (1, 2, 3, 4, 5, 6, 7, 8, 9, 10) I’ve been reading from Inform 7-using students who were frustrated not to be able to start implementing story immediately.

I had a great time meeting everyone who was there.

7 thoughts on “Meeting notes from Seattle meet-up

  1. Emily,

    I’ve gone through the blog posts you pointed to and I must say they look like amazingly rich feedback on the Inform7 software. As a software developer, I wish I could mandate that my users write detailed accounts of their experiences with my work ! (something to think about, actually…)

    This particular group of users is merciless in their critique of the authoring system. What’s funny is that my own experience with Inform7, as an accomplished programmer, in many ways mirrors that of those non-programmers : at first, marvel at the elegance of describing a world in plain English; but soon, frustration at having to bend the language into such broken statements as “The ocean currents are a direction that varies.”, to achieve the (seemingly) simplest things (of course, at that point, opinions on what would be the most logical way to express a given thing differ between programmers and non-programmers).

    I guess what I would like to know is, how does the Inform7 team analyze such feedback ? In short, how well does the idea of natural language IF authoring stand up to actual use by authors ? Have alternative authoring experiences been studied ?

    I realize that such questions may well have been rehashed among IF systems authors – if that is the case I would appreciate being pointed to such discussions so as to learn more about the current trends in that fascinating area.

    Cheers,
    –Jonathan

    • What’s funny is that my own experience with Inform7, as an accomplished programmer, in many ways mirrors that of those non-programmers

      That’s certainly a common kind of feedback, and something that has been discussed a good deal. (In such a variety of places that I don’t have one single place to point you to in order to summarize it, really.)

      I guess what I would like to know is, how does the Inform7 team analyze such feedback ? In short, how well does the idea of natural language IF authoring stand up to actual use by authors ? Have alternative authoring experiences been studied ?

      Well, a couple of points here. One: there is only one person on the Inform7 team who is in a position to make decisions about this kind of thing, and that’s Graham Nelson; this project has always been intended to work on the particular vision he had in mind. It may work or not work, but he wanted, and wants, to carry out the experiment as well and thoroughly as possible. He does accept and make use of lots of other suggestions about how the system might become more useful and accessible; however, the natural language component was a founding feature of his vision, so there was never a point at which I7 might have been developed along different lines.

      I think, personally, that that is the only way this experiment could have been conducted meaningfully; it would have been easy to stop at many different points along the way and go a more conventional direction, but the question of how well this kind of thing works is only really explored if we are committed to doing as good a job of it as we can.

      In short, how well does the idea of natural language IF authoring stand up to actual use by authors ?

      There’s some information about that we do have — anecdotes from people who have used I7 successfully when they tried and failed to develop IF in any other language, and then the statistics about I7 releases relative to other language releases each year. (It would take a little digging, but one could liberate some numbers from, e.g., the games-by-VM categories at the ifwiki).

      There is also some anecdotal counter-evidence — e.g., in Jeff Nyman’s explorations working with storytellers from other media, he has at least sometimes reported that I7 was considered to be inadequate vs. TADS 3. But since I have no direct access to the data from those groups it’s a lot harder for me to know what to make of that feedback.

      Nonetheless, my impression is that I7 was the most-used language for IF development last year, despite the existence of many languages with more conventional syntax (TADS 2 and 3, Inform 6, Hugo, et al). So my own reading of the situation is that there are some people who find it off-putting and inaccessible, and who are better served by other languages — which already exist to serve them — but that there are also other people who prefer I7 to more standard-syntax approaches, and it is the only tool available to those people. So it seems to me to be worthwhile to continue supporting the development of it, aside from my own investment in using the language.

      I suppose there are some ways in which that data could still be misleading: for instance, if there is a very large body of potential authors interested in IF who do not produce games because they dislike Inform 7’s syntax but for some reason never discover the existence of alternative languages that might be more suited to their needs. But I don’t know how to collect data from people I never hear about and whose existence I’m not sure of; the only thing I can really do about it is try to leave around some pointers to the other languages, especially TADS 3, in the hope that those pointers will be useful to people who find I7 unsuitable to their mental paradigm.

      Now, after that long digression: it’s not at all clear to me from reading these that the students were primarily running into the same problem that you describe. Some were to an extent, but several were frustrated by things that would have been an issue in any programming language, such as the need to punctuate accurately, while others were annoyed that they weren’t able to get as much done as they hoped or because the process of thinking about storytelling in such an environment/object-oriented way did not jive with what they were expecting. Which I think are issues that we’d have to approach in terms of design education, rather than syntax redesign.

      • Thank you for taking the time to write such a detailed reply.

        I had begun writing a long response but then realized I should first read the vast volume of discussion and commentary already available “out there”, rather than waste your time with more uninformed comments.
        For a start, I will continue to enjoy your blog and your works.

        Cheers,
        –Jonathan

  2. Pingback: The week in IF #3 « IF URLs

  3. Pingback: So, Do We Need This Parser Thing Anyway? « Emily Short's Interactive Storytelling

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s