
	SDAST Weekly Chat Meeting #57: Thursday 14th March 2002
	
Participants: Peter Kretschmar, Carol Anne Oxborrow, Niel Joergen Westergaard,
	Stefan Larsson
	
Subjects discussed: ISDC's Festive Event
		Science Verification work
		Installing ISDC scripts at DSRI (problems)
		Science Verification Document
		SCREW00481 and updating ADD6.3 to ADD7.0
		Agenda items for SDAST meeting at ISDC 26,27/03/02
		AI list

Action Items;

AI020314_1 ALL	21/3/02 All corrections to ADD6.3 as listed in the change  
OPEN		document to be committed to CVS latest next Thursday 

AI020314_2 CAO  26/3/02 Produce a draft version of the JEM-X ISSW Science
OPEN		verification document according to RW's template, to be
		discussed at next SDAST meeting




**** Logging Started : Thu Mar 14 09:54:17 MET 2002
> Hello Peter - it's good to see you here already. END
<peter> Hi Carol Anne, sorry I was just distracted. END
*** larsson (~larsson@tenma.dsri.dk) has joined channel #jemxadr
<peter> Good Morning Stefan - thanks for the latest explanations. I'm still 
tuning my talk. END
<larsson> Good. Nice to "see" you around. END
> I just got that nice invitation to the  `Festive Event' Thierry's always
<larsson> I almost forgot our meeting. We just got 10 milj kr on an application
> promising us - it does look like it's going to be really festive. END
<larsson> and I am busy spending it! END
<peter> Sami won't come - what about Niels Jrgen? END
> I'll go and rustle up NJW. END
<peter> Stefan - if you need any more ideas, I'd be glad to make some 
propositions :-) END
<larsson> Thanks Peter.....(I have a few of my own though.... END
> Peter, about this invitation, it asks `Nous assisterons au repas, oui/non'
> Does that mean we participate at the meal as diners or as waiters? END
<peter> As diners - even ISDC would not make its guests serve the officials 
:-) END
> I'd like to eat and make merry, but I have no intention of serve food. END
*** njw (~njw@uhuru.dsri.dk) has joined channel #jemxadr
> Great! I'm all for food and drink! END
<njw> Hi there END
<peter> But french can be funny. I recently received a formal letter (Swiss tax
 office) telling me that they "accuse" the reception of my letter to them, which
  obsviously means acknowledge in this context. END
<peter> OK, let's begin - first of all thanks to all of you for the input - the 
viewgraphs are almost finished. END
> So since Sami can't make it, let's get on with the agenda. So
> when do you give this presentation Peter? END
<peter> Tomorrow at 16:50. But I'll be there from 09:00 or so onward, meaning 
I get up quite early here. END
> Go sock it to them, on our behalf! END
> First hot topic on the agenda: science verification - how far have
> we come on that? END
> So, the test sky has been specified by NJW and NL; have all the
> simulations been run, does the test sky now exist in some file(s)END
<njw> I have not advanced much since the draft account of the DSRI simulations.
<njw> On the other hand I have made a lot of simulations for the SPR 
(Science Performance
<njw> Report) END
<larsson> I like to ask Peter. Is there an available XSPEC-version which 
understand
<larsson> our ARF format. END
<peter> And I have not at all progressed. After some discussions hear at ISDC, 
the formal part has been pushed back somewhat to be finished within April.
<peter> Yes, our XSPEC here has no problems. Xspec 11.0.1ag  10:20:18 
14-Mar-2002  END
> I haven't installed any scripts here, or the complete JEM-X ISSW
> components because I was waiting for Silvia to do this. But she's 
> had so many problems (and so have the people in Valencia) that she's
> more or less halted all work on the scripts until she can see Peter
> again. So where do we go from here? END
<peter> It seems Bruce has found the source of the problem - a library issue 
and I'd be interested to hear if Silvia can work with his fixed version now. END
> Could Bruce please contact Silvia and tell her how to proceed? END
<peter> In any case, these things just have be resolved and will be - it 
sometimes takes a few weeks as Bruce is quite loaded with other stuff as well 
and it was not fully trivial for him either. END
<peter> He has sent a mail to her, as far as I remember. END
> Okay, I'll talk to Silvia and see if she can get the scripts running
> now. Should I use her installation, or must we install the officially
<peter> Mail from yesterday, 14:10. END
> delivered versions that you've delivered to the ISDC library, Peter. END
<peter> That would be better, but wait a few more days. Bruce is finally 
getting to the point of setting up a general "analysis" account with a common 
support software installation for all instruments. Assuming he manages to get 
that ...
<peter> to work, you should be able to download and install the same selection 
he has.
<peter> The difference to my integration setup is that the latter sometimes has
 patched versions (the ROOT of the problem Silvia had actually - pun intended). 
 END
> I might have guessed ROOT was the problem? Isn't it more efficient
<peter> Ideally you would probably log in to ISDC and use the envrionment of 
that new account, so there is full control over the system. END
> to do these Science Verification runs on the ISDC system, instead of
>  installing everything here? END
> Great minds think alike. So who does the processing of the test
> sky SWs once it's simulated, and who goes through the results and
> determines their correctness? END
<peter> You all - especially those people calling themselves astronomers - and 
yes, I have to it too. I think we can do the full homework attribution at our 
meeting. END
<njw> I can go through the results and compare with my own results. END
<larsson> I'll do my share. END
> Okay, so basically we'll make AIs to get this done at the meeting,
> so that's one agenda point for the SDAST meeting - is there anything
> else we need to clear up NOW, or can it all wait till we meet at
> ISDC? END
<njw> Are you opening the discussion on Roland's request for a status document
 ? END
> That's next on the agenda, and a separate subject. I've just sent
> RW's email to Sami and Stefan. His wishes for this document are 
> pretty specific, and what I suggest is that I make a draft in time
> for the SDAST meeting, and we all comment on it then. What do you
> guys think? END
<njw> Good idea and I hope that you'll be the editor for it, Carol Anne END
> I don't mind taking responsiblity for this one, if you'll handle
> the document for the instrument science verification. END
<peter> I like the idea. END
<njw> You got a deal END
> Stefan do you want to be editor on this one? END
<larsson> Me 2 END
> Okay, I'll try to get the draft ready so that you can all read it
> before the meeting, so that there's lots of constructive comments.
> END
> Any more comments on Science Verification and documentation of the
> same? END
<njw> N
<peter> N
> Okay the next hot topic: SCREW00481. This basically is a SCREW that
> says we should do all the changes listed in my document, unless these
> prove contrary to ISDC practice, as in the case of the naming of
> Peter's pipeline script parameters. I've already started on all the
> general overview sections, parameter name lists, data structure
> dictionary etc. etc., but on top of the component descriptions which
> each developper is solely responsible for, there are other things
> you each have contributed: background handling descriptions etc. 
> which should also be updated, self-consistentized, etc. and NJW
> has to write a piece on data structure formats and XSPEC 
> compatiblity. When would be a reasonable deadline to get all these
> changes made? END
> You have all read the changes document haven't you? END
<njw> Yes, and I have made the first draft for the data structure formats. I'd
<njw> like to have Stefan's comments as well. END
> Good stuff NJW! END
> Okay, when can you finish your updates - by the SDAST meeting:
> we could discuss any lingering problems then END
<njw> Next week, say tuesday END
> Okay, next week, say Tuesday, for all your inputs sounds great to
> me, what about the rest of you? END
<peter> Wednesday or Thursday. END
> Okay how about Thursday evening, and I'll assemble the whole thing
> on Friday. I can raise any outstanding problems I find at the
> meeting, and we can publish ADD7.0 on the net by the end of March. END
> I'd prefer not to drag this out too long: I want to CODE!!! END
> No comments?
<peter> No comments - I think we all agree. END
> Okay, next agenda item: agenda items for the SDAST meeting at ISDC:
> I suggest the following:
> Science Verification what's still to do and who should do it?
> Science Verification Document: what's still needed and how can the
> draft be improved/corrected
> ADD: any lingering problems, otherwise, open the champagne
> V1 status: integration, testing, verification, outstanding changes
> known to be needed (hopefully not too many of these); outstanding
> SPRS/SCREWs
> V2 status: as above
> V3 status: have we got anything delivered here?
> Instrument model status: how many of our tables actually contain
> realistic numbers?
> Operational activities: who's doing what, and where?
> ISDC news views and status
> Other suggestions welcome. END
<peter> After some discussion with Sami about his software I have realized we 
should also spend some time again to discuss background model status and 
selction schemes.
<peter> While we all know there is a lot to learn after launch, we must verify 
that his software and the scheme he uses to populate the models is adequate 
for launch and allows updates in reasonable time without huge changes to the ...
<peter> structure. END
<peter> This is easier to explain face to face ... END
> On a purely technical side, I'm slowly (very slowly) getting to
> grips with C++ and ROOT, and I think I'll be able to help Sami by
<peter> We also should have an item on IC files: which need updating and 
when/how will this be done before launch. END
> providing so code for JEM-X related objects and methods he can use.
> I thought IC files were included in Instrument model END
> But, yes, background handling should be on the agenda. END
> Any more suggestions? END
<peter> Oh sorry, read to fast - that's just another way of saying the same. END
> Sami, NJW have you anything particular to contribute: source separation
> problems, ideas, verification? END
> Should we have an agenda item on refined image reconstruction: it's
> strengths, weakensses and algorithms, for example. END
<njw> We could put that up as one of the later points since we (I) have no solid
<njw> contribution yet. END
> Okay, we'll add a  little imaging seminar by NJW at the end.
> If there are no more suggestions, lets get on with the AI list:
> Can I assume that all the IAs having to do with verification of the
> ROT data have been done: how did that whole thing go Peter, did we
> find anything interestingly anomalous in the data? END
<peter> Well, there were a lot of small details.
> So it was a productive exercise? END
<peter> Yes, surely. The mazing thing is how much you still find, even though
 we've had similar exercises before.
<peter> This being said, for JEM-X it was mostly on the trivial side, except 
for Nicolas Produit calculating the OBTime wrongly. END
<peter> Which was actually caused by a parser problem, if I remember correctly.
 END
> It's good to find this stuff now......
> Okay, can I cancel the following AI, since we'll be doing the SV
<peter> And then we found this whole issue of no FRSS data in various modes, 
the description of this behaviour was hidden away inthe detailed HK table of 
teh User Manual (the new version is much clearer). END
> processing at ISDC? :
> AI20.16 CAO	Jan 2002 Install RP, scripts, software etc. needed for SV tests
OPEN
<peter> If you agree to work like this, then this is fine. END
> Basically, I don't think I've got time to do this now. END
<peter> If at all, Silvia should do it. Let's just have that development natural
ly and remove the AI on you. END
> On the still open side:
> AI19_17 NJW March 2002 Provide SM with diffuse sky maps from which to make
OPEN		diffuse background models

> Has this been done or does its due data need updating NJW? END
<njw> Not done yet - should probably be change to april
<njw> END
> And this one?
> AI18.16 NJW/NL15/01/2002 Make sure that IT provides a list of HK parameters
OPEN			to be monitored by operators in NRT

<njw> I'll discuss this with Niels when he returns END
> Due date: end March ?
<njw> Alright END
> Here's one that's REALLY been neglected, but should still be done:
> AI19_9	ALL 15/01/2002 Read, understand, comment and correct Peter's 
OPEN		most recently delivered scripts (download from ISDC config. 
		control page)

> Could we all take a look at these scripts and put a short discussion
> on the SDAST meeting agenda? ie Take a look at them before the meeting?
> END
<peter> Fine with me. END
> Okay, due date for that one is 26/3/02 now, so let's make and effort
> and get it out of the way this time. END
> AI19_11	CAO	Feb 2002 Talk with IT about what derived HK parameters 
OPEN		should be used for performance monitoring


<peter> Sorry - I have a seminar coming up - and I'm chairing. Can I leave you 
with the note that ALL our ISSW delivered up to now is now under Change Control
 - this means future updates have to be formally requested and agreeed by ...
<peter> the CCB. END
> Okay, have a good semiar Peter - we'll finish going over all the
> CAO/NJW/SL AIs. Have a good weekend! END
> Okay, since Sami isn't here, and the AI about performance really
> relates to him, I'll put that back to end of March. Next one for
> you chaps is:
> Acually, two for NJW:
> AI21.7	NJW	Feb 2002 Write architectural design for tool to do 
OPEN			source finding on merged/mosaiced images

AI21.8	NJW	Feb 2002 Write architectural design for tool to do robust
OPEN			spectral classification of found sources

> I think these are rather like `Dream Tools' and not a priority right
> now what with all the SV work. How about move this to July 2002, after
> we've all made last adjustments to V1/V2 ? END
<njw> Good idea END
> Well, that's it for AIs. Now there's just AOB - I thought peter 
> could tell us how his talk had gone, but clearly it hasn't yet, so
> any other business you two? END
<njw> No except for a question for Stefan: When wil you be 'up and running'
<njw> for SDAST work, I mean can we do some spectral studies before the
<njw> meeting at ISDC ?
> I think we're on our own NJW, so let's call it a day: we got some
> things decided and cleared off a few AIs. Have a good weekend. END
