[dmis-bof] Comments on the charter?

Tue Dec 13 16:20:55 CST 2005

OK.  I have attached drafts of the 7 questions and the charter in GGF
format.  The more I think about it, the more I think Allen is right and that
this schedule is too aggressive....  there is a lot of stuff that needs to
be done that is not on there.  Here are some of the issues I would like to
get comments back on:

 - I changed the name to OGSA-DMI
 - we need to address naming.  What will we accept as source and destination
names? URLs? URIs? any string? EPRs?
 - There is a general issue which will affect a lot of this and that is just
extensible WSDL.  How do we allow parameters to change when options in the
WSDL change.
 - I think we all agree that this needs to be transport agnostic, but we
need to figure out how best to implement that (related to the WSDL question)
 - what statement do we want to make (and does the OGSA data architecture
need) about delivery semantics
 - what about scheduling / planning aspects?  Do we want to include elements
in this WSDL that specify rate (bandwidth), quantity (file size), and timing
(START BY, FINISH BY, etc)
 - everybody's favorite: security.  I hope we can basically punt on this and
say we will do whatever OGSA does
 - A sort of pet project of mine is monitoring / troubleshooting.  I would
like to potentially include elements in the WSDL or state that is exposed
that would enable better/easier monitoring and troubleshooting.  For
instance, some sort of unique job identifier that can be passed down to
children, so that you can trace the chain back when you have a failure.
 - groups that we need to be aware of to one extent or another include OGSA,
OGSA-D, info-d, gsm, byte-io, naming, grid file systems, authz (other
security groups), ws-agreement (GRAAP?).  are there others? how should we
liaise with those groups?  some will require more work than others.

Let me know what you think.

Bill

> -----Original Message-----
> From: owner-dmis-bof at ggf.org [mailto:owner-dmis-bof at ggf.org] 
> On Behalf Of Hiro Kishimoto
> Sent: Tuesday, December 13, 2005 1:36 PM
> To: allcock at mcs.anl.gov
> Cc: dmis-bof at ggf.org; 'Foster Ian'
> Subject: Re: [dmis-bof] Comments on the charter?
> 
> 
> Hi Bill,
> 
> I really appreciate your support for OGSA and WSRF.
> If you will follow "OGSA spec name guideline" and willing
> to keep consistent it with OGSA-data architecture, please
> consider to use OGSA prefixed name, e.g. OGSA-DM, for your
> WG.
> 
> GGF steering group (including myself) is strongly promoting
> OGSA branding (of WG name and specification name) if appropriate.
> 
> Thanks,
> ----
> Hiro Kishimoto
> 
> William E. Allcock wrote:
> > 
> >>-----Original Message-----
> >>From: Hiro Kishimoto [mailto:hiro.kishimoto at jp.fujitsu.com] 
> >>Sent: Sunday, October 30, 2005 2:31 AM
> >>To: allcock at mcs.anl.gov
> >>Cc: dmis-bof at ggf.org; Foster Ian
> >>Subject: Re: [dmis-bof] Comments on the charter?
> >>
> >>
> >>Hi Bill,
> >>
> >>Data transfer is one of key data services and I am very 
> >>appreciate your
> >>effort to standardize its interface. I've read your charter proposal
> >>and have several comments below;
> >>
> >>(1) SOAP/WSDL interface
> >>I take it you define interface on top of Web Service. Do you plan to
> >>define it on vanilla WS-I or OGSA WSRF Basic Profile? As an OGSA-WG
> >>co-chair, I hope the latter. The better approach is hybrid; abstract
> >>IDL semantic definition plus one or more concrete rendering 
> >>definitions.
> >>OGSA-Byte-IO's work gives you a good example.
> > 
> > 
> > Since *I* need a WSRF compliant service, I will push for 
> one of those.  I do
> > think this is going to be a point of contention though.  
> Your idea about the
> > IDL is a good one, but means significantly more work :-(.  
> It means 3
> > documents not one or two, but it makes sense, so I will 
> definitely propose
> > this option to the group and see what kind of response we get.
> > 
> > 
> >>https://forge.gridforum.org/docman2/ViewCategory.php?group_id=
> >>154&category_id=1058
> >>
> >>(2) Relation to OGSA
> >>I know you are active member of OGSA-D WG. Do you intend to 
> place this
> >>work in the OGSA data architecture? Or, you make it to be combinable
> >>with OGSA works? If you agree to go with OGSA, the 
> following guideline
> >>gives you more advice.
> >>
> >>https://forge.gridforum.org/projects/ggf-editor/document/OGSA_
> >>related_working_group_research_group_and_specification_naming_
> >>guideline/en/1
> > 
> > 
> > Yes, this will go into the OGSA data architecture.  The 
> OGSA data group
> > "commissioned this work" and that is why they are pushing 
> for a non-file
> > specific standard as that meets their needs better.  I 
> understand the need,
> > but as is so often the case in computer science, we may 
> need to "eat the
> > elephant one bite at a time".  I suspect that the general 
> solution is to
> > much to take on the first bite, so we will likely go with a 
> file specific
> > version and follow up with a more general version, though I could be
> > pleasantly surprised and discover there is a reasonable way 
> to get both in
> > one shot.
> > 
> > 
> >>(3) WG name
> >>Since all GGF WGs work for standardization, I think you 
> don't need to
> >>name new WG as "Data Movement Interface *Standardization* 
> WG." I think
> >>- Data Movement WG, or
> >>- Data Movement Interface WG
> >>is better.
> > 
> > 
> > I actually put the Standardization on the end so we would have a
> > pronounceable acronym :-).  However, I will bow to the 
> wishes of the group
> > and area directors.
> > 
> > 
> >>(4) Charter draft template
> >>GFSG needs more information for charter draft.
> >>Please use the OGSA-data WG's charter draft attached as an template
> >>(which includes 7 questions you should answer).
> > 
> > 
> > Will do.
> > 
> > Bill
> > 
> >>Thanks,
> >>----
> >>Hiro Kishimoto
> >>
> >>William E. Allcock wrote:
> >>
> >>>I haven't received any comments on the charter.  I would 
> >>
> >>love to believe
> >>
> >>>that is because you are all waiting to have your computers 
> >>
> >>repaired after
> >>
> >>>weeping uncontrollably on your keyboards due to the beauty 
> >>
> >>of the prose...
> >>
> >>>unfortunately, it is more likely that you have not taken 
> >>
> >>time out of your
> >>
> >>>insanely hectic schedule to review it :-).
> >>>
> >>>Please do so.  It wont take long.  I opted for very little 
> >>
> >>prose and then a
> >>
> >>>timeline.  If people could comment on if they think the 
> >>
> >>text is sufficient,
> >>
> >>>and more importantly on whether the timeline is reasonable 
> >>
> >>and has the right
> >>
> >>>things in it, that would be wonderful.
> >>>
> >>>Please remember, this can only be a success if we have 
> >>
> >>participation from a
> >>
> >>>variety of groups.  Particularly for this one since we have 
> >>
> >>competing
> >>
> >>>existing implementations, and we will need to get some 
> >>
> >>momentum behind this
> >>
> >>>to get people to change and conform.  They have to believe 
> >>
> >>it provides some
> >>
> >>>advantage to do so...
> >>>
> >>>So, now! quick! before another email distracts you! read 
> it! send me
> >>>comments! I attached it again to this email to make it 
> >>
> >>*REALLY* easy!
> >>
> >>>:-)
> >>>
> >>>I had the above as "shouting", but then wondered how many 
> >>
> >>spam filters would
> >>
> >>>tag it as spam...
> >>>
> >>>Bill
> >>>
> >>>---------------------------------------------------------------
> >>>William E. Allcock
> >>>Argonne National Laboratory
> >>>Bldg 221, Office C-115A
> >>>9700 South Cass Ave
> >>>Argonne, IL 60439-4844
> >>>Office Phone:  +1-630-252-7573
> >>>Office Fax:      +1-630-252-1997
> >>>Cell Phone:      +1-630-854-2842
> >>>
> >>
> > 
> > 
> 
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: DMIS GFSG Questions.doc
Type: application/msword
Size: 30720 bytes
Desc: not available
Url : http://www.ogf.org/pipermail/dmis-bof/attachments/20051213/a82a43df/attachment.doc 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: CharterGGFFormat.doc
Type: application/msword
Size: 54272 bytes
Desc: not available
Url : http://www.ogf.org/pipermail/dmis-bof/attachments/20051213/a82a43df/attachment-0001.doc 