Fwd: draft of ABI proposal section
Karen Cranston
---------- Forwarded message ----------
From: Karen Cranston <karen.c...@nescent.org>
Date: Wed, Jun 15, 2011 at 10:43 AM
Subject: Re: draft of ABI proposal section
To: Matthew Bietz <mbi...@uci.edu>
Cc: Hilmar Lapp <hl...@nescent.org>
Hi Matt,
Thanks for the thoughtful comments. I too am very excited about
thinking about user interfaces and user engagement from a novel,
research-informed direction.
> I'm guessing that these systems don't have very robust (if any) user
> tracking.
I am not sure about TreeBASE, but ToLWeb does have a minimal amount of
user tracking for contributors - there are public pages that list the
contributors, the number of curated pages for each and links to those
pages. It used to have a contributors wiki, but was hit hard by spam
and removed. They do some tracking of users, but I don't know the
details (for example, they have data that shows that home schooling is
one key user community).
> your friends. In small-to-medium-sized communities the networks tend to be
> too dense. Everyone already knows everyone else, so friend finding may be
> less useful.
Does this suggest that we should think about how to potentially link
to other groups of scientific curators (EOL, Wikipedia, Wikispecies,
MyPlant, Morphbank, etc). Even if we don't specifically try and
connect X and Y, perhaps we can think about the required standards and
infrastructure that would be needed to have this happen in the future
for these and other not-yet-identified communities.
>
> This isn't meant to be discouraging, but instead to point out an
> opportunity. I think that there's a chance to do some interesting and
> innovative things using social media approaches. The novelty is to figure
> out what would be useful and what does and does not work for communities of
> this size and in scientific contexts.
So, in some sense, we can pitch this as a test of methodologies
without having to say "we are going to do X and we just hope it
works". It seems like a challenge to do this in a way that still has
real value for our curation / user community, but I am more than open
to the discussion.
Karen
>
> On 6/13/2011 12:20 PM, Karen Cranston wrote:
>>
>> I am thinking about the outline of "our" section of the ABI proposal.
>> Hilmar and I have a good grasp of the features& limitations of
>> TreeBASE and ToLWeb as they stand. I am not at all sure about the key
>> questions on the social science side.
>>
>> For Matt's sake, here is a very brief description of the current state
>> of affairs:
>>
>> * TreeBASE is a repository of phylogenetic trees and associated
>> sequence data from published papers
>> ** the TreeBASE interface is very poor, both for users adding data and
>> users searching for data
>> ** TreeBASE does not capture all of the raw data and metadata that
>> researchers generate
>> ** the underlying database architecture is dated and will not scale to
>> meet future needs (or current needs, actually)
>>
>> * ToLWeb is a curated resource where systematists sign up to curate
>> one or more clades of the overall Tree of Life
>> ** the ToLWeb database only has one tree (it does not deal with
>> uncertainty in the phylogeny)
>> ** coverage is very limited, and does not get updated with sufficient
>> frequency
>> ** there is no automated updating - everything is manual
>> ** there is a large user community of biologists that contribute
>>
>> There are no links between TreeBASE and ToLWeb. You cannot go from the
>> curated ToLWeb tree to the research trees in TreeBASE. The metadata
>> that users can attach to trees in ToLWeb does not get stored in
>> TreeBASE. The publication of new studies in TreeBASE does not trigger
>> any updates / flagging in ToLWeb.
>>
>> Matt - what are your thoughts about the best targets for novelty and
>> innovation here?
>>
>> Karen
>>
>
> --
> Matthew Bietz
> Research Associate
> Department of Informatics
> Donald Bren School of Information and Computer Sciences
> University of California, Irvine
> Phone/SMS: +1 (858)405-5227
> Fax: +1 (949)824-4056
> E-Mail: mbi...@uci.edu
> Web: http://www.matthewbietz.org/
>
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karen Cranston, PhD
Training Coordinator and Informatics Project Manager
nescent.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karen Cranston, PhD
Training Coordinator and Informatics Project Manager
nescent.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~