[DAS2] Notes from DAS/2 code sprint #3, day three, 16 Aug 2006

[Steve Chervitz]

Notes from DAS/2 code sprint #3, day three, 16 Aug 2006

$Id: das2-teleconf-2006-08-16.txt,v 1.1 2006/08/16 17:05:24 sac Exp $

Note taker: Steve Chervitz

Attendees: 
  Affy: Steve Chervitz, Ed E., Gregg Helt
  Dalke Scientific: Andrew Dalke
  UCLA: Allen Day, Brian O'Connor
        
Action items are flagged with '[A]'.

These notes are checked into the biodas.org CVS repository at
das/das2/notes/2006. Instructions on how to access this
repository are at http://biodas.org

DISCLAIMER: 
The note taker aims for completeness and accuracy, but these goals are
not always achievable, given the desire to get the notes out with a
rapid turnaround. So don't consider these notes as complete minutes
from the meeting, but rather abbreviated, summarized versions of what
was discussed. There may be errors of commission and omission.
Participants are welcome to post comments and/or corrections to these
as they see fit. 


Topic: Spec Q&A
---------------

bo: perusing spec, saw mention of XID as a filter. can I get more
explanation?
ad: can't remember without looking at docs, but think I was not sure
what XID was supposed to be, lincoln sent email to clarify.
aday: an external db id trying to resolve into local space, eg., for gene
das.
ad: don't think there was enough info there to be useful.
gh: just uri and nothing else?
ad: looking at steve's notes from 16 march. looks like we deferred it.

gh: input was minimal. I have no particular use for it.
bo: need to know what support to provide for the biopackages server.
in the read spec, says "it's not well though-out. should have
authority, type, id, description."

bo: type vs. exact type
gh: did we get rid of exact type?
ad: see gregg's email from 16 march:
http://lists.open-bio.org/pipermail/das2/2006-March/000655.html

The assumption was, there's no type inferencing done on the
server. it's just done on the client. we were to rename 'exacttype' to
'type' and use exacttype semantics for it.
gh: there is no parent-child structure to types. there is to ontology
though. 
ad: type records in das aren't parent-child relations because they
combine other info about type, e.g., ways to depict it.

bo: looking for places where our server disagreed with spec. segments
feature filter is not supported on our end. overlaps segments. but
this is just work we need to do, not a spec issue.

gh: allen and lincoln were struggling with xml:base resolution yesterday,
looking through the xml:base spec, dealing with edges. are you satisfied?
aday: yes
gh: for implementes that don't already deal with xml:base resoultion,
it may take a day or so to deal with it. nomi and I struggled as
well. I was suprised it is not so supported in xml libraries.
ad: just a matter of walking up the xml tree.
gh: recursively had to verify that the resolve stuff in the java
networking libraries actually worked according to the xml:base spec.
but we've moved through this.

bo: url example, uses 'segment' and 'sequence'. not so consistent.
gh: pros and cons to this. it shows that das/2 links can be built
using different uris.
ad: used different url structures to show that this was possible.
bo: confusing when you only see a snippet and don't see where the uri
was coming from. showing variety is useful though.

gh: are both specs frozen now?
ad: yes.


Topic: Status Reports
-----------------------

bo: went through spec. updated our bug queue. added bug re: passing in
id filters vs. uris. working on this today.

aday: need to resolve type ids, need to deal with relative ids given
in the document. now can go back to working on writeback.

gh speaking for lincoln: perl stuff for gbrowser to connect to das
servers. went through xml:base abyss.
updated uris for sequence and genome version ids for human and mouse
on the wiki page: http://open-bio.org/wiki/DAS:GlobalSeqIDs

sc: should we allow anyone to edit this, of just lincoln?
gh: would like to restrict it. worried about wiki graffiti.
ad: you have to register. we can always back things out.
sc: lincoln will get notification upon any edits.
gh: ok. 

gh: working on igb release. adding parsing abilities. can now focus on
das/2, mostly writeback stuff, refining that in igb client.

ee: finishing up bugfixes before igb release. will start on gff3
parser today.

ad: looked into content negotiation stuff. why validator server on
open-bio site isn't working: I updated underlying webserver
framework. working on that.

sc: worked on creating new data files used by the affy das server for
exon arrays using gregg's new parser.
gh: this is generating more efficient versions of probe sets for exon
arrays. important since the affy das server is in-memory.
sc: this will help us support more arrays in the das server and also
move away from having to maintain two different das servers, so we can
focus on just the das/2 server.

sc: also working on final touches on web page describing available data on
our das servers.
gh: we can modify xml from the server to point at that page as an info
url. sources element has info url, and sub elements as well, but we
can just put the info page at the top level.

sc: also was working on ann's fly data project, where she needs to
pull genomic regions relative to probe sets. we need to update our
das alignment file (link.psl) to be based on dm2.
gh: we don't provide residues. she'll have to do a das/1 query at ucsc
to get residues.