The Library: Building a Resource Collection for the

A few efforts, such as the ACIA's IA Guide and Jesse James Garrett's IA resources, have attempted to collect resources of relevance to IA theory and practice.

These efforts are often doomed due to issues of scalability. Could the IA community be enlisted to identify, evaluate, and classify IA resources? The solution might lie in using an active "economic" model, like Slashdot's, or a passive one, similar to Amazon.com's reader-reviewers, to channel volunteer effort into maintaining such an ambitious collection.

This might also be an opportunity to create a model content development policy for use by a distributed community. Which would be cool.

So, the $64,000 question: is there a scalable model for a distributed library of IA resources?

Posted by Louis Rosenfeld at November 03, 2001 04:02 PM
Comments

i've been doing A LOT of thinking about this for the last year or so. in fact, i went so far as to start designing the service a little. to my mind, the right model is neither slashdot or amazon. it's epinions.com.

epinions.com's basic model is a combination of the others. it gets the content creation and management of slashdot without all the overwhelming feature creep and the collaborative filtering power of amazon. it's actually more complicated than that but, trust me, it works. especially, if you aren't tied to making a profit. :)

actually, the idea i had was a simplified verison of epinions because i think we could remove the $$-earning aspect and a good deal of all the trust stuff. (the ia community is small, active, and trustworthy enough to make them unnecessary.)

i've written more about this at MoreSmarter.

Posted by tpodd at November 12, 2001 11:39 AM

okay, i've been doing some more thinking. this time more tactically. there really seem to be 2 models for the community built/maintained resource repository: 1) IAwiki or 2) epinions. (i know there are probably others but i think these represent the two basic camps pretty well. please offer alternatives, correct me if i'm wrong, etc.)

IAwiki:
- simple, simple, simple
- easy to understand and implement
- basically static
- not very scalable, not that easy to manage (especially once it gets BIG)
- not very usable from an info-seeking perspective (yes, there's a search. but i'm talking about some functionality that could really take advantage of the social strength of the IA community)
- i mean this only as constructive criticism. :)

epinions:
- complex
- could be easy to understand but definitely NOT easy to implement
- very dynamic
- scalable and easy to manage if designed/implemented properly
- very info-seeking oriented; especially due to the filtering and rating focus...not to mention possibilities for favorite people, resource, etc lists.

it seems like what we really want is a mixture of the two or, more likely, a starting point in between the two that will allow us to move toward the epinions model in stages. so, i'll propose some possible stages:

1) build the team (we need to establish responsibility and control to avoid a design-by-committee situation. though, i'm definitely considering this will be a participatory design process.)
2) needs/use analysis (this is actually a BIG stage but not the focus of my thoughts right now. as a community we're damn good at this so we can work out the details later.)
3) specification
4) basic DB-driven site (think IAwiki++ ; input/output to support faceted classification, filtering by basic categories, etc.)
5) some basic rating and recommendation features
6) more complex collaborative filtering
7) ???

just some thoughts. i have a bunch of thoughts about specifics of the functionality too but i figure those should wait until we get some bigger picture things figured out. thoughts?

Posted by tpodd at November 12, 2001 4:40 PM

not very usable from an info-seeking perspective (yes, there's a search. but i'm talking about some functionality that could really take advantage of the social strength of the IA community)

It is true there is usually very little functionality hardwired into wiki sites. No rating, no karma, no meta-tagging...

Ironically, wiki's with their collaborative editing base, really does take advantage of the social strength of communities ... as people find a need it gets implemented just by creating pages and writing links. It is through this process than an emergent architecture arises, one which includes such info-seeking usefulness like category pages, roadmaps, intro guides, and reviews.

One very useful wiki-tool is the use of two-way-links .. from any page you can pull up a list of all pages that reference that page. This is particularly useful if you start from a category page, for which every relevent page at some time created a link to.

On hardwired ratings functionaly you want to be very careful, as these can generate feedback loops that will adversely affect your site. Not to say they aren't useful, just that they are difficult to get right.

Posted by Eric Scheid at November 16, 2001 3:26 PM

true. there are definite advantages to the ultra-flexible nature of a wiki. that flexibility is its strength and weakness. while it can be molded to fit any need the community has, there is very little control or structure to it, and that makes it very difficult to provide tools for navigating, filtering, or searching it.

i do like the two-way links, though. that's a powerful idea.

also, you're right about rating systems being hard to get right. but a lot of the difficulty comes from having an irresponsible user base. i feel like the users of the IA library might be a little more trustworthy in their behavior. though, i could just be fooling myself. :)

Posted by tpodd at November 19, 2001 2:08 PM

true. there are definite advantages to the ultra-flexible nature of a wiki. that flexibility is its strength and weakness. while it can be molded to fit any need the community has, there is very little control or structure to it, and that makes it very difficult to provide tools for navigating, filtering, or searching it.

i do like the two-way links, though. that's a powerful idea.

also, you're right about rating systems being hard to get right. but a lot of the difficulty comes from having an irresponsible user base. i feel like the users of the IA library might be a little more trustworthy in their behavior. though, i could just be fooling myself. :)

Posted by tpodd at November 19, 2001 2:10 PM

Other people who might be good to bring into this conversation include Kat Hagedorn, Vivian Bliss and Chiara Fox; I'll email them and point them to this conversation.

sb

Posted by Samantha Bailey at November 19, 2001 3:59 PM

I've been looking into Hebbian learning just lately and that has sparked some ideas.

Posted by Eric Scheid at November 19, 2001 4:22 PM

I would love to help out with this! Thanks for the heads-up Samantha. :)

My time is a bit limited at the moment, but it should get better come the first of the year.

Let me know when we are going to get a meeting together to discuss things. I'll go back through our brainstorms for the ACIA IA Guide - might get some ideas there.

Posted by Chiara Fox at November 28, 2001 7:55 PM

Samantha did indeed drop an email my way. And darn it - this looks too interesting to not be involved.

At the risk of sounding obtuse, let me ask a few questions to make sure I understand the idea of the IA Library.
1. The resources themselves are not being collected but rather surrogates are being created with information on how to attain the resources, right? (Can you say metadata?)
2. Desire interested persons located anywhere to add to the collection. This could happen in several ways. Making the initial contribution of the surrogate or adding to the surrogate with editorial comment or such?
3. Any thought on overarching editorial control affecting:
-collection development policy - Lou mentioned this one.
-metadata standard
-controlled terminology
-search
-browse
-classification, not necessarily in the traditional library sense. (It is the most fun to figure this out in a digital collection w/o the need for physical storage and going beyond resources that are traditionally library.)

How serious is everyone? And keep in mind that 'resources' could be many, many things. Ahh, once again we are back at the idea of collection development policy. :-)

VB

Posted by Vivian Bliss at November 28, 2001 8:11 PM

The IA Library group is posting its thoughts and ideas here.

Posted by tpodd at November 28, 2001 10:42 PM

sorry, the IA Library group is posting SOME of its INITIAL ideas there.

We'll be posting progress reports here.

Posted by tpodd at December 2, 2001 1:41 PM

Hi all. Apologies for the late response. I'm still unpacking my life...

Chiara's right. There are many lessons that we could take from the ACIA IA Guide re-design. I also am limited in my time until the beginning of next year, but would be more than amenable to starting a discussion then.

Posted by Kat Hagedorn at December 6, 2001 9:41 PM

Hi folks; I'm (re)posting Todd Wilkens' 12/2 project plan here for reactions:

--------------------------------------------
1) Generate rough list of goals/requirements (I'll send out my take on this to start the discussion.)
- As a group, we know a lot about this problem already. Let's start from there.

2) Research
a) Competitive analysis (both user and system approaches)
b) Survey the community to test assumptions and get a little more info.

3) Iterate the goals/requirements list

4) Brainstorming/Innovation/Etc. (This is where we create, synthesize, and brainstorm about how the system could/should work based on the goals/requirements and research.)

5) Decisions, decisions, decisions...
a) Make decisions about the system model (e.g. content entities, relationships, etc.)
b) Make some decisions about policy (i.e. how can/should the users of the system behave?, what can/should we do to control/not control this?)
c) What technologies will we use?

6) Specification and Detailed project plan
a) Specify the thing down to the nitty gritty details
b) Assign roles/responsibilities
c) Work out the the practicalities (where will it be hosted?, who will maintain it?, etc.)

7) Build it (this will be fleshed out in more detail as we complete steps 1-6)

Please, comment, suggest, critique or praise as you see fit.

-tpodd

Posted by Lou Rosenfeld at December 11, 2001 11:43 AM

...and here are some goals that Todd Wilkens posted for comments:

--------------------------------------
These are the broad goals/requirements. They will need to be made more explicit eventually. I just want to make sure that we're all in agreement about the big picture first.

The IA Library should...
***Primary***
- Collect links (surrogates) to IA/UX resources.
- Allow these links to be annotated.
- Allow users to retrieve relevant resources.
- Allow users to discover relevant resources. (different from retrieve.)

***Secondary***
- Allow annotations to be updated/edited.
- Allow classifications of links to be updated/edited.
- Allow users to learn about special topics in IA. (special collections)

***Misc***
- Allow users to find people in the IA community.
- Allow users to find examples of IA deliverables.

The way I see it, the first 4 are primary/immediate needs.

You could argue that Discovery might not be an immediate need, but I'm sure it's a primary requirement for the Library. There's a huge soup of IA resources out there. It's a big problem to find what you're looking for. It's an even bigger problem to find what you don't even know you should be looking for.

Also, I'm not tied to any of the "Misc." items. I listed them because they've been discussed on the list.

Comments, comments, comments, please.

-tpodd

Posted by Lou Rosenfeld at December 11, 2001 1:56 PM

This tool (the first and second priorities) sounds very close to something I built for the IRS in 1997/98. The tool was built in ASP, but I do not have a platform to resurect the tool in that format. I am beginning to rebuild the tool for my own site in the next few weeks. I can easily hand off the tool for use elsewhere after that.

Posted by vanderwal at March 18, 2002 4:18 AM

buy viagra online_
viagra_
anal sex_
big tits_
black sex_
blowjobs_
boobs_
girls_
hardcore sex_
hentai_
lesbians_
milf_
nude_
porn_
pussies_
shemale_
teens_
tits abc

Posted by Klaus at October 12, 2003 4:06 PM

CaptainStabbin InTheVip BoysFirstTime 8thStreetLatinas TrannySurprise Cumfiesta MILFHunter BigNaturals MikesApartament TopShelfPussy StreetBlowjobs WeLiveTogether

Posted by as at December 26, 2003 4:53 AM

Test

Posted by levitra at December 27, 2003 8:31 PM

Hi

Posted by phentermine at December 27, 2003 11:55 PM


Comments closed
This site exists for archive purposes only. Please check Asilomar Institute for Information Architecture for the latest on the community effort.
 
  Home

Keeping Up
Progress Report
Credits
Powered by Movable Type