
	Weekly chat #40: Tuesday December 5th 2000
	
Participants: Peter Kretschmar, Stefan Larsson, Niels Joergen Westergaard,
	Sami Maisala, Carol Anne Oxborrow
	
Topics discussed: Consulting with GRANAT people about backgrounds
		Defining Interactive Background Analysis
		ADD updates deadline - pushed back yet again!
		j_sci_format - replaced by ISDC generic tools
		j_ksc - to be discussed at SDAST meeting in January
		SDAST meeting in January 9-10/01/01
		Background data structure DOL parameter names - to be updated
		Next chat meeting: December 20th 2000

Action Items:

AI40.1 PK	10/12/00 Organise informal meeting between SDAST and GRANAT
OPEN		visitors at ISDC consortium meeting to discuss backgrounds and
		background handling.
		
AI40.2 CAO	5/12/00 Ensure that background data structure DOL parameter
OPEN		names are all in the parameter name list. Parameter names to
		use, consistently, dbg and bkg instead of ibg.
		
AI40.3	ALL	22/12/00 All developers with S/W that uses background data
OPEN		structure parameter names to correct them using the new
		consistent parameter names and redeliver the affected
		executables.


**** Logging Started : Tue Dec 05 09:28:19 MET 2000
> Me neither! END
<Peter> But it's easy to call up the chat window while still editing away somewhere else. END
> How true ! END
<njw> Hello Peter and Carol Anne. I'll try to be a little less punctual next time END
<Peter> Why that? END
*** larsson (~larsson@tenma.dsri.dk) has joined channel #jemxadr
<larsson> Morning down there. END
<njw> You shouldn't think I'm bored END
> Hi Stefan: we're all here except Sami: I'll try to get him on the phone. END
<Peter> So by CA's qualification, Sami is usually the least bored :-) END
*** smaisala (~smaisala@taurus.astro.helsinki.fi) has joined channel #jemxadr
<Peter> Hi Sami!
> sami's coming - slowly
<smaisala> Hi all
> Hello everyone, we're all here almost on time too - this must be some kind
> of record. Let's get stuck into the AI list (unless you've got better
> suggestions) END
<Peter> Not really ... END
> AI13.17 NJW	11/00 	Ensure that the calibration document will contain 
OPEN			descriptions of data-formats, products etc. which will 
			be in telemetry format to be provided to ISDC

> So how's that calibration document coming along NJW? END
<njw> Unfortunately it is still on my list of to-do's END
> So a new due date (REALISTIC please) ? END
<njw> 5/1 2001 END
> Okay. Next one:
> AI14.3	SM,AH	10/00 	Contact Russian GRANAT specialists to get background
OPEN			information from mission
> Have you contacted these people yet Sami END
<smaisala> Yes. I have sent email to Rodion Burenin but no response yet. ENd
> Good stuff - I'll close that AI. Next one:
<Peter> Sidenote: Alexander Lutinov and Sergei Mol'kov who have GRANAT are here at the ISDC this week and the two next weeks.
<Peter> (have GRANAT experience)
> Could you talk to them about this Peter - I think it's quite important
> expecially while we're defining interactive BKG handling. END
<Peter> It may be interesting for all of us available to discuss with them. Who from DSRI is finally coming to the Consortium Meeting? And with what schedules? END
<Peter> I will try to discuss matters with them but am extremely loaded these days. END
> Me Allan and NJW, the new SDAST head. END
<njw> and Soren END
> We'll all becoming monday morning and leaving tuesday evening. END
> yes, and soeren END
<Peter> OK, I'll try to arrange a little chat around the Consortium Meeting. Maybe Monday after the V2 demo. END
> That sounds good: a new AI? END
<Peter> Not really, either I do it or you won't knoe before it is to late to pressure me. END
> Too late: it's already written! Don't worry I won't hassle you THAT much
> over it! END
> Next on the AI list: ADD updates - they're all due 11/12/00, that's to
> say next wednesday. We'll discuss the CVS repository a bit later.
> Next AI is:
> AI16.4	SM	10/00	To define interactive background handling
OPEN

> So far I've had an email from SL after the dedicated IBKG chat, but noe
> from Sami - how's it going Sami? END
<smaisala> It's still under work. The IBKG chat wasn't so productive as I assumed. I think we still have to discuss more concerning IBKG. END
<smaisala> There is still open questions what we really want. 
<smaisala> I will send as soon as possible my proposal what you want and after that we can circulate that proposal and decide the final solution END
<Peter> Thanks, I was just about to propose this. END
> That's exactly what we want Sami. END
<Peter> But I'm quite sure the soultion will not strictly speaking be final. EMD
> Let's give your IBKG AI's a new due date, say in a week? END
<smaisala> Maybe week and a half. We have that ADD updating. END
> I'll put 15/12/00 END
<smaisala> OK. END
<Peter> About ADD update: I really can not guarantee that I make that deadline. Currently it's crazy here and after the Consortium Meeting I need to prepare a nice talk for the MPE in Garching so they hire me after my presentation. ...
<Peter> END
> Is that pretty much the same for all of you? END
<larsson> I am in favour of Peters presentation! END
> Okay, let's make the ADD updates due date one week before the SDAST
> meeting, and then I can go over any problems at the meeting - or at
> least provide each of you with a list of corrections that need
> correcting. Is that a good idea? END
<larsson> Yes. END
<smaisala> Yes END
> Due date is Tuesday 2nd of January 2001 - you should have recovered from
>  your hangovers by then! END
<njw> OK END
> So your IBH deadline is back to 10/12/00 Sami - we've got to get going
<Peter> I can just see the SDAST writing ADD updates with firecrackers flying and champagne glasses :-) END
> on one front or another sooner or later. END
> Next on AI list:
> IBH1.2	ALL	15/12/00Read and comment SL and SM's emails as 
OPEN			preparation for more detailed AD figures

> This one still stands with it's due date: I really do want to see some
> comments on the SL/SM emails so that we have a thorough discussion of
<Peter> OK. END
> the subject before any design decisions are made. END
<njw> We'll put best efforts into this END
> That's it for AI's due: there are however lots due in December, so
> let's keep an eye on the A-eye list and get our stuff done! END
> Next on the agenda: what's happened to Stefan's V3 components - there is
> a nasty gaping hole in the ADD and I've not idea what to put there.
> Are j_sci_format and J_KSC completely scrapped or what? Which tools,
> ISDC-generic, ISSW, other are going to perform the previously imagined
> tasks? At the very least this calls for a major AI on Stefan to let
> me know what goes here. END
<larsson> j_sci_format is ISDC task, right Peterr?
<Peter> IMHO they are all replaced by not-yet-written generic ISDC tools. END
> ALL? END
> Does anyone at ISDC know about these tools other than you? END
<njw> Do you have any idea on the time schedule for these ISDC tools ? Now that we
<larsson> J_KSC (known source calibration??
<njw> can extract spectra and lightcurves and that OSIM will include time variability
<njw> we will soon need to look more closely at the simulaition results END
<Peter> The tools have been foressen for a long time.
> That's a relief! END
<Peter> (spec_pick, lc_pick, image_pic)
> So I just rip out all of j_sci_format and all of `j_ksc' and put in
<Peter> They have been assigned in general to James Peachey (Bruce's replacement) but their realization is still pending.
> a little not about ISDC generic tools. Talking with Soeren, it seems 
> likely that some continuing detector spatial calibration will be
<Peter> I'll try to increase the pressure to have first versions sooner. END
> performed here at DSRI using various spurious lines that arise from
> the gas mixture - but this will be done way-offline and will result
> in new JMX-SPAG-MOD files being delivered once ina while. As I see it,
> this ad-hoc calibration will replace the j_ksc component, and maybe
> be more reliable than a  pre-ordained ISDC executable in as much as we
> can constantly tailor the process to whatever data we have available
> at the time. eND
<Peter> I still think we should have the tools to build up shadowgrams from large sets of data, also uncorrected.
<Peter> And to fit shadow structures (collimator) to that. END
> So j_ical isn't dead, just sleeping? END
<Peter> In my opinion we should try to have a generic description in the updated ADD. END
<larsson> I agree with Peter regarding such tools. They will be needed anyway. END
> Generic description of what: j_ical_collimator? END
> j_ical_fit_collimator replaces all the position parts of j_ksc,
> and off-offline analysis by DSRI to produce new -SPAG-MOD replaces
> the energy parts of j_ksc? I agree the tools are still needed - but
> I'm not clear on the ISDC/SDAST division of labour. END
> Hello? Anyone? END
<larsson> I think we should discuss this at our january meeting, when we
<larsson> might know a little better what is needed. END
<njw> It is hard for me to imagine that ISDC will make the fit_collimator executable END
<Peter> For me this part is largely ISSW.
<Peter> ISDC will have generic tools to build up an OG of suitable data, but then it's ISSW. END
<njw> Sounds reasonable END
> So this ISN'T going to be covered by ISDC generic tools - I think this
> is the exact opposite of what you said 10 minutes ago. I am all in
> favour of us using part of ISDC generic tools and  putting ISSW on
> the top to meet our needs - but I must know how things are going to
> be split to be able to put it in the ADD. END
<Peter> Wait, wait ...
<Peter> I said that the sci_format tools will be done with generic ISDC tools.
<Peter> Known source calibration, or rather it's succesor, is ISSW again. END
<njw> Stefan is right, we should discuss this at the january meeting - it is a bit
> Okay, that's a little clearer. j_sci_format is all done by ISDC generic
<njw> too much for this chat. END
> tools but we're still 100% responsible for whatever is the latest
> version of j_ksc.  Finally, I'm  with you END
> Yes, let's put this on the January agenda - I had no idea it was such
> a quagmire! Will we need one or two days for the SDAST meeting do you
> think? END
<Peter> Maybe two would be better.
> I think so two because we have two BIG subjects: IBGH and j_ksc. Does
<njw> I agree that two days would be better but some of you had some constraints ? END
> everyone else agree with 2 days? END
<Peter> Now, I can do this again. Before we had an important date on Jan 10 (Bonnet visiting) but this has moved to Jan 24. END
<larsson> I have no problem with that... END
<smaisala> No problem END
<Peter> OK, let's settle for two days. Can we still start inthe morning on Tuesday? END
<njw> January 9 and 10 then ? END
> Yes sirree! END
<Peter> And 8 is QLA, right?
<larsson> Hmm, maybe slightly later if thats OK with you. END
<larsson> (an hour or so) END
<Peter> That should be possible. END
<njw> We can adjust the agenda so that Stefan need not be there to begin with END
> Let't start later on the Tuesday than originally planned and end maybe
> a little earlier on the wednesday and that should give us enough time
> for everything. END
<larsson> OK. END
<Peter> I wonder how Tuebingen reacts to a trip that ends on my last day hired by them ... :-) END
> Okay, we'll plan around that arrangement of things. Now, next on the
> agenda: any problems with CVS? END
<njw> None discovered yet END
> I hope you won't miss your own goodbye party! END
> Well, any way, if there's any problem checking things out or stuffing
> them back into the repository, just let me know. Like I said, I'm
> trying to find a way that we all can see the latest update of the
> document on a daily basis as a help to all you contributers END
> Next on the agenda: background data structure parameter names - can be
<njw> It will be best if people can retrieve their files - do the editing - and put
> get rid of the ibg pre/sufffix now that we no longer use it in the
> data structures themselves. END
* njw thme back. It is not good to wait several days before putting back. END
> Yes, that's a good point NJW - the whole system will work best if
> people are snappy about their editing, but so long as there's some
> file in the repository with the correct name then I can always make a
> new version of the document, it just won't have your very latest
> ideas in it . END
<Peter> Does anyone have an opinion on the background parameter name issue? END
<smaisala> Let's say good bye to 'ibg' END
> Yes - I want to use BKG and DBG prefixes only - no IBG. END
<Peter> It does mean redelivery of every executable that uses it and I need to change my scripts ... END
<Peter> Niels Jrgen, Stefan ??? END
<larsson> Sorry I was disurbed.
<larsson> (disturbed)
> If you and Sami are already living comfortably with the names as they
> stand then we'll leave them as they are - inconsistencies and all. END
<njw> I have not yet started to use these datastructures yet so I have no sharp
<njw> opinion on this END
> People will just have to be very careful to consult the list before
<Peter> Stefan, what do you think?
> doing anything with these parameters because the names follow now
> clear rules. END
> Sorry `no clear rules' END
<Peter> I do not mind too much making the changes, but if we change them it should be a coherent effort with a time window when
<Peter> new deliveries are made and my scripts get updated. Otherwise it's a mess for me to manage this. END
> I think it's up to the people who use these parameters to decide: what's
> easier: to change to a consistent naming system for parameters, or
> to stick with the current names and be very careful about looking up
> the names before using them? END
<smaisala> My opinion is that we should keep consictency in names. Peter are you using your suggested names or the deliveried ones in your scripts? END
<larsson> Sorry for interuption. Well, consistency is nice so lets go for that. END
<Peter> I am adapting to the executables. And I'm for consistency. So which flavor shall it be: ibg or bkg ? END
<Peter> Carol Anne and Sami like bkg, Niels Jrgen has no opinion. what about you Stefan? END
<larsson> bkg is ok. END
> ibg is obsolete END
<Peter> OK then let's change to bkg***.
> Also ibg is currently used in a non-consistent manner if Sami's email
> is to be believed END
<Peter> How about the following scenario:
<Peter> Everybody concerned, makes a new delivery of their affected executables in week 51 (Dec 18-22)
<smaisala> CAO. You have that complete list of all BKG names. Should we deliver that to everyone so we do not have to talk the names anymore. END
<Peter> I will change my scripts and integrate the software either within this week or on the first days of the new year. END
> AI on everyone to change their background data structure parameter
> names by Dec 22 and deliver affected executables.
> AI on CAO to ensure that the consistent background DOL parameter
> names are all in the parameter name list by this evening. Is that
> fair? END
<smaisala> Yes. END
<Peter> OK with me END
<njw> Y
<larsson> Yepp. END
> Tis done. END
> I think point 5 on the agenda's already been covered, so that leaves
> AOB: when should the next chat be? How about Tuesday 19/12/00? END
<larsson> 19th is fine with me. END
<Peter> I'll be in Munich presneting myself on that day, END
<njw> Wednesday then ?
> Sami: I'll send the list of new parameter names to you and Peter for
> final confirmation before updating the list END
<smaisala> OK. Sounds good. END
> Wednesday's good for me too. END
<Peter> Wednesday would be OK. END
<Peter> In the morning. END
<larsson> Wednesday is even better. END
> 9.30 Danish Galactic Time: also known as Paris time in the remoter
> corners of the galaxy! END
> Any other AOB? END
<larsson> No. END
<njw> No END
<smaisala> nope END
<njw> Peter ? END
<Peter> Sorry, an interruptin here
> See you on Peter: I'll `see' the rest of you on December 20th. Have
<Peter> No real AOB, but if Stefan could stay for a few minutes ...
> a good an productive week!Goodbye ! END
<njw> OK I'll leave in privacy - good bye END
*** njw has left channel #jemxadr : (njw)
