[Sigia-l] Open-source IA tool in the making

[Alexander Johannesen]

Hi All,

I'll try to answer most stuff in this one email, but so far I'm really
happy with what's being said. From the top ;

Some have asked about the financial aspect of this. I have thought
about that in the past, to create a cool IA tool, charge money and
laugh all the way to the bank. But, hey, who am I kidding? I don't
think the market is big enough, the customers are scary (!!), and
there are already lots of similar tools out there. I just want to
create an IA tool that I want to use. The advantage I'm bringing to
the table is that we all can contribute instead of relying on a vendor
to have the foresight and hindsight and strata to make the right tool
for you, and that it is free; you've got nothing to lose if it sounds
interesting.

Stewart ; good thoughts, although I don't want to make it various
applications, but more like one web application with a Wiki style for
most things and especially for content (I don't want to fiddle with
content *management*). Just click a field, type some stuff, and Voila!
the site is different.

Listera wants a demo, and I'm sure many more would like that. Um,
sure, of course, whenever I have some time, I'll do just that. I want
to make at least wo nice looking default styles first, otherwise it
will go from "features" to "visual impression" in two seconds flat. :)

Betsy wants technical info ; I've used BSD in the past, but there is
no reason not to go with either GPL 2 (not sure about 3) or Apache
License. Others that suits should be considered.

Still Betsy ; As to other technical tidbits, I use PHP classes that I
think myself are pretty good, and I will indeed put AJAX into this. If
people would point me to nifty and *small* widgets that could be used
(like a small tree-editor with XML support, for example), please do.

Jacco has gone the Visio way ; I'm going to stay right clear of that,
I think. It brings too much baggage, IMNSHO, besides IA is more about
structure and content than about how much like Visio it is. :) He asks
"I'm curious about the model you are planning to use to create
screenflows from XML" and the answer is, I don't; I want the thing to
be a site, almost like a prototype tool, only closer to reality. Once
you've specified how a page works, what links its got and some
validation info, you can generate flowcharts based on it if you need
that extra info. I don't believe in making a lot of work to prove a
little point, nor do I think workflow diagrams add much value to the
development cycle, except perhaps to the people who are creating it. I
think that's what makes Visio popular; you can think while drawing,
but after that fact, I shun it.

And about the summit ; if someone is willing to sponsor me, I'm happy
to go. :) *hehe*

Jonathan ; "Do you see this [the IA IDE] as being something this tool
could incorporate naturally, or is it essentially page-based?" It is
in essense a site / project tool that can view the whole shebang,
parts of it, a page in it, or a widget within a page. I don't want to
make it *too* IDE as that would add complexity that I don't feel is
needed. We need to *quickly* put up prototypes, sitemaps and skins,
overviews and such, but an IDE (would) offer more features than the
above needs to be. I want to design it so that you define and focus
from one angle, and other angles are created automagically based on
that.

About Axure ; looks good. Not sure I can nor want to compete with
that. :) Where Axure seems to be more like an IDE for IA/UX/design
people, I aim more to have a real-time website you extend and fiddle
with using AJAX until you're happy, flipping between different views
of the site as you go. (For example, Axure *generates* sites and
reports, while my tool would be more WYSIWYG in a browser)

Peter (the Topic Maps abandonista :) asks "You might wanna study some
of the other attempts that've been made at this and why they haven't
really taken off." I have, and especially my own efforts. :)
Satisfying the IA is a hard one; they are scary people *because* you
can't give them crap, they will see through your bluff, and generally
poke at the inabilities of any given thing. I hope the open-source
part of the project would take away most of that sting.

Further ; no it isn't just a prototyping tool. In fact, I can run
production websites with it, no problem. It is a fully functional
website in itself, depending on how much data you feed into it. The
framework is fast enough to run heavy-traffic websites, but it hasn't
been designed specifically for that, no. Also, I don't (at least, not
at this stage) have any management of content and publishing, nor do I
think I want to do that. (I don't want to create a CMS)



Ok, I think that's where were up to, and please, keep it coming. For
example, how would people like the interface to be? Application? Web
browser based? Do you *want* external people to participate with the
tool (so you can hire a greasemonkey to add content, for example)?


Regards,

Alex
--
"Ultimately, all things are known because you want to believe you know."
                                                         - Frank Herbert
__ http://shelter.nu/ __________________________________________________