[Sigcr-l] Sending message again

[Barbara Kwasnik]

Hi everyone, apparently, ASIST listservs don't allow attachments, so here's my message again, cut and pasted. 

Hello fellow SIG members,

The deadline for submitting our SIG-session proposals is fast approaching (jan. 21). I'm leaving for a month's stay in Canada next Friday (19th), and Kathryn has agreed to help out with the submission details. So, we have a few days to get our act together, so to speak. Please, see the attached summary of what's flowed in so far. If I've missed someone, sorry, and please just pipe up, but at this point, so close to the deadline, please make your idea a fairly firm one that could eventually be acted on as a submission with SIG/CR sponsorship.

So far we have one proposal that is well-worked out. It's being spearheaded by Mark Rosso (details in attachment) and is a panel whose aim is to discuss the current state of genre research, including the creation of a classification of digital genres, and other issues near and dear to SIG/CR hearts. In addtion the notion of genre fits this year's theme. So, if nobody objects I will work with Mark on getting htis submission ready and we'll send it with SIG/CR sponsorship, OK?

The other ideas are summarized as they came in but would need to be formally proposed and someone will need to take "ownership". If you're interested in submitting another idea besides the ones I've included in the attachment, please check the ASIST website to see the panel submission forms so you know what you have to include. Then post your idea to this forum and we'll take it from there. 

Anyone can propose a panel. Basically, you need a short description, who will organize and moderate it, the format, and some commitments from speakers with their topics (one or two will do). This is not a hard-and-fast commitment on the speakers' part, since these proposals first have to get accepted, but your speakers or participants shoudl be willing to be involved "in principle."

Barbara

And here's what was in the attachment:

IDEAS FOR PANELS SPONSORED OR CO-SPONSORED BY SIG-CR
ASIST- 2007
DEADLINE, Jan. 21, 2007

I can submit these panel proposals if someone is willing to take "ownership" for writing them up in the format needed. Here's what we have so far. 

GOOD to GO (almost):
This panel was submitted by Mark Rosso previously to another conference but did not get accepted because of its interactive panel format (which I think ASIST welcomes). He is in the process of reworking the wording slightly to make it fit the conference theme. 

Towards the Use of Genre to Improve Search in Digital Libraries:
Where Do We Go from Here?
Organizer: Mark A. Rosso 
Moderator: Stephanie W. Haas
Panelists: Andrew Dillon, Barbara H. Kwasnik, Mark A. Rosso
Marina Santini, Elaine Toms
As the collection sizes of digital libraries grow, access to materials through keyword-based searches becomes more problematic. Research over the past ten years has investigated several aspects of using digital genre for improving information
retrieval. Work has included proposals for what constitutes a digital genre, the automatic and manual classification of documents by genre, users' ability to recognize the shape of
digital documents, the solicitation of users' genre terms for digital documents (e.g., personal homepages or blogs), and users' ability to recognize and agree on the genre of digital documents. Despite the efforts put forth in this area, no one has been able to show that retrieval by genre can be effectively implemented. Why not? What directions should research take to bring this seemingly intuitive concept to a working reality?
The panel brings together some of the researchers who are investigating the use of digital genre in retrieval. The goals of the session are first to establish an understanding of what we know about genre, and then to discuss what the remaining questions are and how we can find answers. Between the panelists' ideas and contributions from the audience, we hope to develop a research agenda for the future.
Part one of the session will review the "state of the art" in digital genre research: what do we know so far, what can we and others in the field agree on? Questions in this segment will include:
·	How do users perceive the genre of digital documents?
·	How do users think about genre in the search process?
·	What are the known constraints and obstacles specific to automatic classification by genre?
·	What inferences can we draw from the answers to these questions regarding how a viable genre classification should be developed?
·	What research methods have been informative?
Part two will focus on "Where do we go from here?" We will discuss the open questions, areas of uncertainty, and topics on which we disagree. Issues will include:
·	Is genre too subjective a concept to be useful at all?
·	Can we really ever separate genre from topic? Does it matter if we can't?
·	How should document classification be done: automatically? By end-users?  By catalogers? By authors?
·	How should we approach the problems of multiple genres per document, multiple documents per genre occurrence?
·	Where in the search process should genre be introduced, and how should people use it?
·	What types of interfaces might best leverage this genre metadata?
·	Can genre be effective for the general webpage search problem, or is it only feasible for certain collections, user communities, and/or tasks?
·	Will adding genre to search ever be cost-effective?
·	How can we identify useful and usable genres?
Through discussion among the panelists and members of the audience, we hope to cast some light on the dimensions of these questions, and establish the focus of genre research in the future. The end result will ideally be a clearer sense of what the next important steps in genre research for digital libraries should be.



NOT FULLY WORKED OUT IDEAS:

1.	From Kathryn La Barre, who is also working with SIG/HFIS:

HFIS interest is quite keen in another reprise of the FRBR panel, as is interest in full engagement in the current debates that were set off by the recent changes at Library of Congress, the planned bibliographic summit at ALA this summer and the possibilities currently being discussed of several pre-summit gatherings in the coming months, that are being planned in order to formulate responses to these changes and their broader implications within the LIS community.

I would very much like to see CR and HFIS play a significant role in the discussions at these upcoming gatherings. In the context of panels for the upcoming annual meeting of ASIST, we might also want to consider planning a panel or two in order to address these developments. The January deadline comes long before any of the planned meetings, but I think we could draft an abstract or two that would be sufficiently general to allow the incorporation of the developments that occur post-January to engage the interest of ASIST.


2.	From Andrew Grove: Since I will edit the conference proceedings again for '07, I haven't time to do much more than suggest an idea or two: 
a.	1. folksonomies as seeds for indexing languages, categories and categorization, classification, thesauri, and other taxonomic structures.  I have some ideas but no time to really develop them.  Of course, at the speed things happen in "cyberspace", all will be old news by October '07.
b.	2.  walking the fine line between order and chaos - springboarding off the excellent panel on uncertainty at this year's conference.  What is the role of classification at the edges of knowledge?  How do we, or do we, classify unknown or emerging knowledge?  How about multiple "private" classifications?  How about the intersection of new and private - speaking of chaos...?  This is something I deal with everyday, it's a very real (and sometimes painful) issue.  Again, many ideas but little time to formally develop them. 
c.	Corinne Jorgensen follows up: I like Andrew's two ideas as they seem to follow on the social tagging aspects. It seems that many commercial ontologies are closer to classification in everyday life, or to the classification emerging from the data (something we are seeing happening in the Morphbank project - a large biological image database, even though the organizational schema follows traditional taxonomy, participants create their own groups, and it is interesting to see what emerges as organizing elements from those).
d.	 (Note from Barbara: Corinne and Andrew, do you want to take the lead on either of these two ideas?)
3.	Joseph Busch had mentioned a reprise of last year's session on "classification in everyday work life" which was very successful. Joseph, will you submitting a proposal for that session again? I'd be happy to participate again.