[SAGA-RG] Call Notes

[Andre Merzky]

Below are the meeting notes for todays call.  We'd be happy to see
feedback, in particular to the python discussion.

Cheers, Andre.


  - When: Wed Oct 07, 2009, 22:00 CEST
  - Who : Steve, Hartmut, Ole, Shantenu, Andre
  - What: 

    - update from OGF27
      - all participants have been at OGF, or have been updated<F2>

    - info about DRMAA/SAGA collaboration
      - no DRMAA callin - postponed

    - python bindings
      - no VU calling, thus only partially discusses
      - see Steve's mail: file io more native.  
        - CCT groups agrees
        - VU answer awaited
      - attributes
        - have been discussed at OGF
        - CCT proposal:
          - continue to use dict interface, but do not expose as
            class attribs
          - saga objects which implement attrib interface would
            thus implement the python dictionary interface, but
            could have a remote implementation behind it
          - set_attribute()/get_attribute() is required for async
            ops though

    - advert extension
      - flags are missing on short C'tor
      - TTL was discussed, and agreed that it MUST NOT impose
        anything on the implementation. Default ttl should be
        infinite (-1).
      - it has been discussed if TTL should be exposed as attribute.
        After some discussion, it seems that set_ttl and get_ttl
        seems fine after all.
      - the advert class should not have a second c'tor, in sync
        with other SAGA packages.

      - the doc will be revised and put into last call on mailing
        list.



Quoting [Andre Merzky] (Oct 21 2009):
> Date: Wed, 21 Oct 2009 10:03:00 -0500
> From: Andre Merzky <andre at merzky.net>
> To: SAGA RG <saga-rg at ogf.org>
> Cc: Andre Merzky <andre at merzky.net>, Steve Fisher <dr.s.m.fisher at gmail.com>
> Subject: Re: [SAGA-RG] phone call cadence
> 
> In respect to the python bindings, please all have a look at the
> proposed bindings at http://www.cs.vu.nl/~kielmann/pysaga/ , if
> possible *before* the call ;-)
> 
> Issues raised at OGF27 are:
> 
>   - file read returns int *or* string??  Uhm ... 
>     possibly split into read_buffer or read_string?
>   - app managed buffers are supposed to make zero-copy-read
>     possible.  If a copy happens anyway, one does not nead
>     application managed buffers!  Clarify!
>   - buffer CAN be deleted if bindings can do same semantics 
>     natively
>   - possibly two flavours for 2.x and 3.x: signed char array vs
>     byte buffer
>   - NORMAL -> SYNC
>   - attribute interface: possibly to ONLY
>     set_attribute/get_attribute, even if that is not 100%
>     Pythonesc.  SAGA attribs *are* *not* class member attribs,
>     but conveye a different semantics.  Also, ther are sometimes
>     extensible, sometimes not, sometimes readwrite, sometimes
>     readonly - all difficult to express in the 'native' way it
>     seems.  Not to speak of async...
>   - shouldn't exceptions be nested?
>   - What is an SagaIOError?  Does it make sense?  Use Case?
>   - no set_defaults on context!
>   - enums in UPPER_CASE_WITH_UNDERSCORES (object_type)
>   - bindings seem to be designed before ERRATA - please recheck
>     EVERYTHING!
> 
> Thilo, do you have an answer on those issues by now, by any
> chance?
> 
> Cheers, Andre.
> 
> 
> 
> Quoting [Steve Fisher] (Oct 21 2009):
> > Date:    Wed, 21 Oct 2009 09:49:32 +0100
> > Subject: Re: [SAGA-RG] phone call cadence
> > From:    Steve Fisher        <dr.s.m.fisher at gmail.com>
> > To:      Andre Merzky        <andre at merzky.net>
> > Cc:      SAGA RG             <saga-rg at ogf.org>, Steven Newhouse <s.newhouse at omii.ac.uk>, 
> >            Thijs Metsch <Thijs.Metsch at sun.com>,
> >            Daniel Templeton <Dan.Templeton at sun.com>
> > 
> > I will try to join today because I am interested in the Python
> > bindings. However I will not generally participate as it is too
> > inconvenient at this time.
> > 
> > Steve
> > 
> > 2009/10/20 Andre Merzky <andre at merzky.net>:
> > > Mathijs was the only one who cannot make the proposed time.
> > > So, sorry to him, but we'll go ahead with that timeslot then.
> > >
> > > Please mark your calendars then for Wednesday, Oct. 21, at
> > >
> > >   3pm Central Time Zone (Summer Time)
> > >  10pm Central European  (Summer Time)
> > >
> > > The call will repeat every second week.
> > >
> > > Agenda for our first call:
> > >
> > >  - update from OGF27
> > >  - info about DRMAA/SAGA collaboration
> > >  - possibly discussion of python bindings (depending on
> > >    participation)
> > >  - possibly discussion of advert extension (depending on
> > >    participation)
> > >
> > > For this week I could get us the LSU telephone bridge.  Dial-in
> > > details are:
> > >
> > >  +1-225-578-4942 (international)
> > >  866-573-0359    (US, toll free)
> > >
> > >  conference-id: 118682
> > >
> > > Using skype to dial in to that bridge works of course, and is cheap
> > > from Europe.  The following calls will either use the same line, or
> > > skype conferences - we'll sync about that before the calls.
> > >
> > > Cheers, Andre.
> > >
> > >
> > > Quoting [Andre Merzky] (Oct 14 2009):
> > >>
> > >> Hi all,
> > >>
> > >> the topic of reviving our WG phone calls has come up at OGF27 (and
> > >> before actually).  We would like to propose a bi-weekly schedule, at
> > >> Wednesdays 10pm CET (3pm Central).  Note that the DRMAA WG will
> > >> hopefully pick the same slot, also bi-weekly, so that the calls for
> > >> both groups will alternate.
> > >>
> > >> Is there anybody on the list who would like to participate, but
> > >> can't make this time slot?
> > >>
> > >> Cheers, Andre.
> > > --
> > > Nothing is ever easy.
> > > --
> > >  saga-rg mailing list
> > >  saga-rg at ogf.org
> > >  http://www.ogf.org/mailman/listinfo/saga-rg
-- 
Nothing is ever easy.