[Bioperl-l] EUtilities policy change

Chris Fields cjfields at illinois.edu
Thu Jan 28 14:18:22 EST 2010


I think warning is fine for now.  I've reimplemented that so it occurs
lazily (warns only when a request is actually made).

Will also change the tool to 'BioPerl' (currently 'bioperl', all lc).
We'll obviously have to address this in the test suite as well in some
way, maybe ask for an email if network tests are requested.

chris 

On Thu, 2010-01-28 at 14:05 -0500, Mark A. Jensen wrote:
> Thanks Chris-- 
> The soap modules currently set tool to "SoapEUtilities(BioPerl)". 
> I agree that a default email is a bad idea (tm) (unless maybe it's 
> hilmar's...?). I'd say a warning on unset email parameters is a responsible
> "there be dragons" sort of treatment.
> MAJ
> ----- Original Message ----- 
> From: "Chris Fields" <cjfields at illinois.edu>
> To: "BioPerl-l" <bioperl-l at lists.open-bio.org>
> Cc: "Mark A. Jensen" <maj at fortinbras.us>
> Sent: Thursday, January 28, 2010 2:00 PM
> Subject: EUtilities policy change
> 
> 
> > All,
> > 
> > Per NCBI's recent change in eutils user policy (effective June 1):
> > 
> > http://bioperl.org/pipermail/bioperl-l/2009-December/031698.html
> > 
> > Both the tool and email parameters ('-tool', '-email') are now required
> > when making requests.  Note this will significantly break all modules
> > requiring remote access to eutils (Bio::DB::GenBank/GenPept, some Biblio
> > and Taxonomy stuff as well, IIRC).  This also applies to web services
> > (SOAP-based access).  Mark, not sure how this affects your SOAP-based
> > modules.
> > 
> > I have reconfigured Bio::DB::EUtilities to follow this policy; the
> > default tool setting has been 'bioperl' and will remain that way.
> > However, there has been no default email, therefore setting this is now
> > required for future requests unless we (the bioperl devs) decide there
> > is a safe default email to utilize.  My gut tells me, however, that
> > falling back to a default email opens up a can of worms for the devs and
> > is very likely a 'BAD IDEA'(TM).  
> > 
> > Regardless, be aware that, after June 1, NCBI will very likely exclude
> > requests with no email and will notify users who are considered to be
> > violating their policies.
> > 
> > I will likely make further changes to Bio::DB::EUtilities in the
> > meantime to ensure that using the tools by default will not violate
> > NCBI's policy (e.g. override this at your own risk).  
> > 
> > chris
> > 
> > 
> >




More information about the Bioperl-l mailing list