Fwd: treebase.treebase3 (harry.shyket@yale.edu)
Karen Cranston
From: Hilmar Lapp <hl...@nescent.org>
Date: Wed, Jun 15, 2011 at 10:31 AM
Subject: Re: treebase.treebase3 (harry....@yale.edu)
To: Rutger Vos <rutge...@gmail.com>
Cc: "Shyket, Harry" <harry....@yale.edu>, Karen Cranston <karen.c...@nescent.org>, "pi...@treebase.org" <pi...@treebase.org>
Yeah, something like OAuth would be much preferable to us hacking our own non-secure system together, though I'm not sure how easy it would be to build client-side uploaders (mesquite? command line scripts?) that can authenticate through that.
On Wed, Jun 15, 2011 at 3:17 PM, Shyket, Harry <harry....@yale.edu> wrote:
Has there been any thought of using a possible single sign in across the multiple sites? If implement an authentication method across all sites, ideally one person can sign in to multiple sites with one login. OAuth has a specification for doing this. This is what Facebook, Twitter, MySpace etc. use for allowing other sites to implement their login. This could open up to the possibilities of sharing across the sites include a possibility of the curation features. With that said, everyone would still manage their own user lists for their given sites.
From: karen.c...@gmail.com [mailto:karen.c...@gmail.com] On Behalf Of Karen Cranston
Sent: Wednesday, June 15, 2011 10:13 AM
To: Rutger Vos
Cc: Shyket, Harry; hl...@nescent.org; pi...@treebase.org
Subject: Re: treebase.treebase3 (harry....@yale.edu)Based on our conversations with Matt (and others at the recent cyberinfrastructure summit at NESCent), the social networking / crowd sourcing aspect of this is definitely something that we want to do very carefully. There is plenty of active research in this area, and we need to be aware of that in the proposal. Given that ToLWeb already has curation features and a community (albeit limited), this piece seems to fit best within the ToLWeb interface. One of the challenges with this proposal is going to be deciding how to structure the web interface components (what goes on treebase.org vs tolweb.org vs new_site.org?) so that we don't have parallel / redundant development across resources.
On Wed, Jun 15, 2011 at 9:40 AM, Rutger Vos <rutge...@gmail.com> wrote:
Hi Harry,
- Looking at the first documenting for Lookup by identifier, it seems that you are suggesting skipping the codebase all together and delivering directly through mod_rewrite. This will certainly speed up the download process especially with larger files. but will we ever have a need to the user hit code (for validation or anything else)? Seems we may have to also have a tracking mechanism for downloads directly off of the server.
Given a directory structure that can be reconstructed with simple regexes from request URLs in principle there is no need to have any code intervention. I also can't think off the top of my head why we would want the code to intervene: access could be managed using mod_access/.htaccess, downloads are automatically recorded in apache's access_log. I realize I take an extremist, spartan position here, though and I'm happy to stand corrected - just figured I might approach this initially from the KISS position :)
- The submission process looks great! I would recommend that we utilize social networking to enable crowdsourcing of the curation.
Absolutely!
The question is which social media are best suited to go easily from suggested curations to them actually finding their way into the database. Rod Page is running this Mendeley group (http://www.mendeley.com/groups/734351/treebase/) but none of that has found its way in the db yet.
Perhaps a solution where suggested curations are in a more structured format would make it easier to ingest them automatically in a soft-typed part of the schema. For example using something like Freebase (http://www.freebase.com/).
On the topic of social media - it would be very easy, even now, to add a comment feature to selected pages using a Disqus widget (http://disqus.com/)
- Do you think it would be worth it to run Apache Solr (rather than Lucene) to harvest the data (or push to)? It would have to be another tool that has to be managed.
I don't know enough about how these tools compare to say one way or another. In any case, the plug in idea is partly to give treebase the opportunity to push the responsibility for additional functionality onto the complainers themselves, so that management of additional indexes and indexing tools is distributed. This is also somewhat inspired by the distributed annotation system that the genomics community uses (http://www.biodas.org/wiki/Main_Page)
Rutger
From: rutge...@gmail.com [mailto:rutge...@gmail.com]
Sent: Tuesday, June 14, 2011 8:17 AM
To: Shyket, Harry
Cc: karen.c...@nescent.org; hl...@nescent.org; pi...@treebase.org
Subject: treebase.treebase3 (harry....@yale.edu)
<image001.jpg>
I've shared treebase.treebase3
Message from rutge...@gmail.com:
Hi guys,I sketched out some ideas for how various things might work on TreeBASE3 (for ABI). The sketches describe how you might retrieve something based on its accession/PhyloWS URL; how you might submit batches asynchronously; how searching would work in general; how specific search plugins work and how they would be implemented. Curious to hear your comments!Rutger
Click to open:
Google Docs makes it easy to create, store and share online documents, spreadsheets and presentations.
--
Dr. Rutger A. Vos
School of Biological Sciences
Philip Lyle Building, Level 4
University of Reading
Reading
RG6 6BX
United Kingdom
Tel: +44 (0) 118 378 7535
http://www.nexml.org
http://rutgervos.blogspot.com
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karen Cranston, PhD
Training Coordinator and Informatics Project Manager
nescent.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
--
Dr. Rutger A. Vos
School of Biological Sciences
Philip Lyle Building, Level 4
University of Reading
Reading
RG6 6BX
United Kingdom
Tel: +44 (0) 118 378 7535
http://www.nexml.org
http://rutgervos.blogspot.com
--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Karen Cranston, PhD
Training Coordinator and Informatics Project Manager
nescent.org
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
