LSA Annual Meeting Albuquerque 2006

[Jeff Good]

> For the tutorial, what about making it similar in length and format,
> but really focussing tightly on the nitty-gritty how-to level. Like
> specific advice about recording equipment and techniques and formats,
> for both audio and video. (If we go this route, we have to say something
> about video. People are filming - they need advice.)
>
> And/or, we could easily spend the whole three hours on text production:
> transcription, alignment, annotation, markup, formats, Unicode. We could
> include databases in here, because people tend to lump them together,
> probably because so many people use Shoebox for interlinearizing. A
> Shoebox 5.0 / Toolbox tutorial would be well attended, I guarantee!!

These are both two good ideas--and, they probably mean we have good topics
for 2006 and 2008. (Heidi and I have been discussing the idea of
alternating LSA Meeting attendance with LSA Institute attendence--the
latter only takes place on odd years).

Though it's further from my heart, I tend to think an audio/video tutorial
may be more needed than a text tutorial. On a completely impressionistic
level, I think the average documentary linguist's awareness of issues in
managing texts is much higher than managing audio and video. Even when I
was doing recordings recently, I realized that, while I knew a lot about
recording formats and media, I didn't know much about physically setting
up the recording space. Nor did I know if there were recommendations for
recording oral metadata with the recording. (I can imagine this could be
as valuable as good labelling in the long run.)

And, for video, I was in a similar situation: Are there recommendations
for physically setting up the recording space? Should we _always_ get
video, even for vocabulary elicitation? Are there any video formats whose
accompanying audio formats are best practice or should there always be a
separate DAT (or whatever) audio recording?

I guess all this is to say that, if I'm confused about all this, probably
I'm not alone. Texts, on the other hand, are relatively well understood
and there's already a long tradition about how to collect them--and,
crucially, while Ph.D. advisors generally know nothing about audio and
video, they know something about texts.

So, I guess, what I suggest, as an initial proposal is a tutorial on audio
and video covering topics like:

1. Audio and video formats (pros/cons of format X over format Y, being
open-minded to the fact that some people love their minidiscs and won't
part with them gladly)

2. Technical recording techniques (what microphones to use, where
to position them, minimizing background noise, unplugging refrigerators
before recording [a problem I had recently], when to have separate audio
and video records, etc.)

3. Archival recording techniques (labelling media, recording
metadata with the recording, informed consent and recording)

4. What do I do with all these tapes? (this one is crucial--we need
someone to say, (i) give them to an archive and (ii) if you're working on
language X, go to archive Y, and (iii) if you don't have an archive for
your part of the world do Z--this has to be ultra-concrete with no
possibility for someone to leave thinking: I'll just leave the tapes at
home for now)

Please anyone feel free to criticize this--or suggest something completely
different. Assuming that at least the first three topics could easily be
multiple presentations I think this could be a full schedule. We'll
have to think sooner rather than later about who can talk on these
topics--especially video.

Also, in writing this, I realize that we might want to think "big" and try
to have an information "packet" accompanying the tutorial summarizing all
this information. (Perhaps just as a .pdf on a web site.)

> I'm really coming around to Tony Woodbury's view that most field linguists
> want to be told what to do and how to do it, in the tech domain at least.
> And it's our job to tell them. Last year's attendance supports that. We
> could be much more explicit and instructional and it would just make our
> audience all the happier!

I agree with this entirely. As long as we can give them "actionable"
advice, I think they'll just listen. Especially since most of them are
desperately interested in their documentation not being lost. The problem,
as I see it, is too often we given non-actionable advice like: Use XML or
use OLAC metadata. The typical response being something like: What's an
olack?

Jeff