Book People Archive

Re: Paul Duguid article: "Limits of self-organization: Peer

Bowerbird@[redacted] wrote:
> any "system" that requires _anybody_ (average end-user or
> p.g. bureaucrat) to "download and unzip a big bunch of tiffs"
> is a massively bad failure of architecture from the word "go".
> why download a whole book of scans to look at one page?

I've done it before, when it's been the only way to look at that page.
It's not what I'd ideally want to do, but it's better than not being able
to look at the page at all, when I've really wanted to know what was
on it.

I'll take what I can get, if it's sufficiently manageable and comprehensive
to be worth my time to adapt to.  If someone can provide me
page-addressable interactive scans of a substantial part of the
Gutenberg collection (say, 1000 titles or more, on a reasonable
trajectory to growing with the PG collection), and gives me an easy way
to plug links to these into my Gutenberg database (either by providing a
way of computing those links from Gutenberg etext numbers, say, or
providing up-to-date, machine-processable link tables) I'd be happy
to look at including them in my Gutenberg cover pages.

If no one's stepping forward with that, though, but someone is stepping
forward with that many titles, and easily-loaded valid link metadata, for
big zips of page images, I'll settle for that for now.  It's at least
better than no scans at all, and their availability may make it possible
for someone else to feed them into a nicer page-addressable interface, which
I could then switch to once it became available and well-supported for
the same title range.  (Remember, since I'm planning to maintain these
links via automated techniques, it shouldn't be too hard for me to
switch from one set of machine-processable link metadata to another set.)

So that's my initial offer.  Now it's a matter of seeing if anyone
finds it worth their while to provide something that meets the needs
and requirements I've stated, or, if they're interested in a similar
outcome but can't quite provide exactly what I've asked for, talking
with me to negotiate some arrangement that is workable for both of us.

And it seess to me that you're in a similar situation.  You've got
a prototype of a fairly nice page-turning system.  The user interface
has a few rough edges, in my view, but they're relatively minor ones that
could probably be smoothed out with some small adjustments.  So rather than
pick nits about the interface, I think the more useful question to ask is:
How do you see this fitting into the online book service chain?

Say, for the sake of argument, you want to work with me on the problem
I'm trying to solve.  Then I'd be asking you questions like:

  -- Can you provide me what I'm seeking (1000 titles + link data),
     in the form I've said will work best for me?

  -- If you can't, but you think other people could using the tools you've
     developed, are you willing and able to provide these tools, and required
     support, to people who can do it themselves?  Who do you see would
     be taking you up on the offer and providing me what I'm looking for?

  -- If my process doesn't fit well with your process, what would I need
     to change to work with you, and why would it be worth my effort to
     do so?  How are you willing and able to adapt if I can't or won't
     meet you where you are now?

In this example, I'm taking the role of someone who might want to consume
what you're producing.  And you might have a similar set of questions you,
as someone who might want to produce what I'd consume, would want to ask me.

Similarly, working down the chain, people producing things you might want
to consume, such as DP or other etext preparers, might have similar questions
for you.

These are the sorts of issues I was hoping we could discuss when this thread
started.  It's one thing to build a cool tool or service.  It's another thing
to make it work in a productive chain of people and projects providing online
books and services around them.

How you make that integration happen, in a world where other folks in the
chain may have different ideas, priorities, and trust relationships,
is a tricky but important problem to sove.  (You can also try to build
up your own chain from scratch, of course, which is basically what
Google's doing, but that's a rather expensive process if you want to do
more than occupy a small, specialized niche.)

So, how can we work together better?  (I throw that out as both a specific
question to you in this example, and as a general question for everyone
else who's interested in these sorts of issues.)

John