
	Chat meeting #45: Thursday 17th May 2001

Participants: Peter Kretschmar, Niels Joergen Westergaard, Carol Anne Oxborrow,
	Sami Maisala, Stefan Larsson.
	
Agenda:	1) AI list
	2) SDAST meeting in June - day, time, agenda
	3) V2 consolidation - what's delivered, what's not
	4) ADD - new updates with every delivery .ADD 6.1 for Science meeting.
	5) Error code proposals
	5) AOB



**** Logging Started : Thu May 17 09:09:35 MEST 2001
<njw> Hello Carol Anne, let's see if this works END
> Hi, NJW, you're good and early - checking that it actually works today? END
<njw> right END
> I'll be back again a little closer to chat time END
*** smaisala (^smaisala@mizar.astro.helsinki.fi) has joined channel #jemxadr
<njw> Hello Sami, so you succeeded tuning in on the right channel END
*** larsson (~larsson@tenma.dsri.dk) has joined channel #jemxadr
<larsson> Morning!
<smaisala> HI all ! 
*** peter (~pkretsch@crab.unige.ch) has joined channel #jemxadr
<larsson> Hi Peter.
<peter> Hi everybody - did we say at 9:00 and I have been missing the first 30 minutes, or why is everyone connected already? END
<larsson> Well, I am home (with a cold) and have finnished my brekfast so
<larsson> what wlse should I do? END
<njw> Perhaps this is the result of troubles last time END
<peter> We could call it an SDAST motivation burst :-) END
<peter> Shall we begin?  END
> Hello there chaps: we connected early just to see if it was all working END
> Let's get started. Point one on agenda: AI list. How are we doing there? END
> Instead of going one AI at a time, how about we each give an update of our
<njw> No calibration doc yet;
> outstanding AI's? NJW you get the ball rolling. END
<peter> AI18.10 - DAL3JEMX has already useful functions to extract REST data, image by image. 
<njw> I've had discussions with Niels on the name of the ISDC representative but not
<njw> committing answer yet.
<njw> unfulfilled SRD still exist; no pixel map yet END
> I'll close AI18.10, Peter END
> And the two lists of HK parameters to be monitored? END
<njw> wait till june END
*** Signoff: peter (Ping timeout)
> I'll put June 2001 on all yours then? END
<njw> OK END
> Of my 3 outstanding AI's all are going to be tackled round about end of June,
> and no sooner given the pressure for V1 and V2 updates. END
> So that covers all the AIs. Any comments? END
*** peter (~pkretsch@crab.unige.ch) has joined channel #jemxadr
> Next on the agenda: SDAST meeting in June: what days, times and agenda
> items. NJW and I already suggest Thursday 7th June 13.00 to Friday 8th June
> 17.00 for the times. What do you guys think? END
<peter> Ahem, I _hope_ those days are fixed (June 7+8) as I already bough back-to-back tickets together
<peter> with the JEM-X Science Meeting next Monday. END
<peter> Time and date fine with me. END
> Good to see you're on the ball Peter! END
<larsson> Should be OK for me too. END
> What about S'n'S from Up North, any comments? END
<smaisala> It's OK for me too. END
> Okay, so agenda items: what do we all want. Here's my wish list:
> 15 mins on status of ADD
> Everyone to have 15 mins to describe their V1/V2 status
> Everyone with a V3 application to have 0.5 hour to describe this for each
> application
<peter> Which should be: redelivered in consolidated version!
<peter> (the V2 status) END
> Status, changes, schedule news from ISDC - by Peter
<peter> OK
> Instrument news from NL
> Calibration news from NJW
> Other suggestions? END
<peter> I'd like to have ~30 minutes to go through the scientific analysis and to note down issues and questions
<peter> with regard to offline use especially. END
<njw> V3 components. Future of V2 components - how ready are they? END
<larsson> Good points END
> Also: issues raised by End Users at JEMX science meeting END
> Okay, so that's about it? END
<larsson> Let say so for now. END
> Any more suggestions should be sent to me or NJW by email. We'll discuss
> the colours of the flowers and limosine during the next chat. Next on the
> agenda: V2 consolidation (groan!). What's delivered what's not, what's
> VERY close to delivery? END
<peter> But you admit that there will be flowers and a limousine, right :-) END
> Absolutely: we
> 're a civilised SW developement team. END
<peter> OK, V2, shall we go through step by step - then you would start. END
> Okay, I just delivered j_cor_gain1.0 which is the consolidated, merged 
> version of j_cor_gain_events and j_cor_gain-spectral - it's tested for just
> about everything except nuclear attack! J_cor_position could be delivered
> late next week, unless V1 takes precedence. END
<peter> V1 _does_ take precedence. END
<peter> In the worst case I can show JEM-X analysis without the correction step as the OSim produces corrected data anyway, but we do need those v1 updates for the upcoming end-to-end tests. END
> That means till mid-June I'll be redelivering v1 and writing first versions
> of the `new' v1 executables. END
<peter> Probably yes. END
<peter> The script j_correction has been tentatively updated, but can not be fully tested until both executables have been
<peter> updated.
<peter> END
<peter> Nest step: GTI
<peter> A script has been written and very loosely tested (lack of suitable test data). I've also found some problems
<peter> with the generic tools that are being corrected. END
> So ISDC makes mistakes too??? END
<peter> Many! Seriously, the partial integration of JEM-X and SPI ISSW has unearthed quite a bit of flaws in various libraries (DAL3GEN for example) and generic tools. But that's also the goal. END
> Hopefully the more bugs we shake out now.........END
<peter> Next step: DEAD - delayed because CAO must first do V1 ISSW, right? END
> Right. So many components, so little time! END
<peter> We should consider redistributong work, if someone else (who?, where?) had some extra time END
<peter> OK, let's move on: CAT_I (j_cat_extract):
> `extra time' what's that? END
<peter> The script has been rewritten but not yet tested again. Besides the general lack of time, I'm waiting for an update of cat_extract, promised for today. END
<peter> Nest: BKG - Sami!
<peter> By the way, most of what I'm telling you is noted on the JEM-X integration WWW pages. END
<smaisala> I have corrected all problems in j_bkg_spat_sel and I have also added new selection functions. I can deliver that end of this week. END
<peter> Which means today or tomorrow, riogh
<smaisala> j_bkg_spec_sel and j_bkg_scal_calc are under work and I think I can redeliver them end of next week. END
<peter> (should be right) :-)
<smaisala> to Peter: Yes, As soon as possible I'd like to know  the new error code numbering... END 
<peter> Here the rewritten script appears to work correctly. END
<peter> For the error codes, I hope we decide today, and - as I said - at the moment there is no problem if you return 'nonstandard' error codes. It will be a problem in a few months, but not right now. END
> So how's image reconstruction and binning and sources coming along - hey all
> you guys with `extra time'? END
<njw> Image binning: ...shadowgram has been delivered in a version for consolidation.
<larsson> I looking for my extra time must have droped it somewhere. END
<njw> It seems to work with event selection on PI and time (IJD).
<njw> The background correction is not implemented yet but interfaces (i.e.
<njw> parameters) are there.
<njw> Image reconstruction: ...basic_recon has been delivered and seems to work
<njw> except for proper WCS keywords.
<njw> ...src_find is in the workshop right now (I'm fighting with a 'segmentation fault'
<peter> Btw, the GTI problem has been resolved. I've tested a pre-delivery of the new DAL3GEN yesterday succesfully. END
<njw> ..._cor_intens is a minor executable and it will be ready next week.
<njw> END
<larsson> I have not yet redelivered j_src_lc.
<larsson> I prefer to make some tests and updates of j_src_spectra first
<larsson> (so I dont have to redo too much work).
<larsson> A problem is realistic and complete test data.
<larsson> I have got some new simulation from NJW which I am working
<larsson> on now.
<larsson> I will keep working on both up until end of June.
<njw> Peter, did you test it with j_ima_shadowgram ?? END
<peter> Yes, with your newest version. END
<njw> Good END
<larsson> I'll negotiate the delivery date of j_src_lc with Peter. END
<peter> j_image_bin works, but I had some problems with ISDCTask use, which are generic.
<peter> j_imaging and j_src_ectract_spectra are preliminarily reworked but not tested in any way. 
> Are we allowed extra time for developement if we run into `generic' problems
> as opposed to ones we've created ourselves? END
<peter> In both cases there is delay which shows we were too optimistic in our schedule. I'll have to think about what we can manage by mid June and then presnet at the Consortium Meeting and what we can not. 
<peter> I'll discuss these questions with Isabelle Lecoeur and Roland and tell you, when we have an agreement.
> While I'll agree we were too optimistic with our plans, I don't know  how
> we could have been more `realistic' in our time-requirements estimates: we
<peter> Inmy opinion we should make all effort to have an integrated chain with limited functionality, e.g. not treating REST events correctly or no bkg reduction, now and plan for another round of updates in parallel  to the first ...
<peter> integration of the 'rough' chain. END
> just continue as fast as we can, and this is the result. END
<peter> But we aren't finsihed with V2:
<peter> BIN_S and BIN_T:
<peter> only one program has been redelivered j_bin_bkk_spectra. While I'm sure that I almost completely reworked j_bin_evts_spectra already I seem to have wiped that out very succesfully (a backup also did not help much). 
<peter> So these will be delayed further, I _might_ have them redelivered in limited form by end of the month, depending on how
<peter> much trouble the integration is. END
<peter> the scripts for these have not been updated. I'm considering moving the whole block to 'after June' as it's not required for basic QLA and SA. END
> Dear Peter, it's very re-assuring that you're as far behind as the rest of
> us! We will all struggle on a little more cheerfully knowing that we're not
> alone - I know I will at least! So we're a little delayed, but bear in mind
> quality work requires time - so perhaps we're doing a more thorough job of
<peter> Well, I did rewrite several scripts in the meantime :-) END
> updating our components than we'd originally planned - I know I did!!!! END
> Let's soldier on manfully (and womanfully) and we may still be the first
> team to complete integration END
> Okay, next on the agenda: ADD updates. I'll be making a new version of
> the ADD for release in time for the JEM-X science meeting here next
> Monday and Tuesday (how many of you coming to that?) - and it will form the
> basis of my presentation to the assembled throng. I'm assuming you've all 
> been good and updated your component contributions with each delivery. If
> not be sure to have all updates in the CVS by 9.00 Monday morning when I'll
> check out the repository for texing for the last time before release. The
> onus is on each developper to keep their contributions updated as well as
> possible. END
> Comments? Anyone still not know how to use the CVS repository? END
> Last on the agenda: error code proposals. Shall we bite the bullet and go
> with a major overhaul of the actual numbering system? If yes, I suggest that
> I also add a piece in the DD part of the ADD to describe what the system is.
<peter> Since I proposed it, I'm naturally in favour :-) END
> Personally, I'd rather see this overhaul done now than later, IF this is
> the last major structural change to these codes. eND
<njw> Is it really a structural change? I had the impression that we were to
<njw> update the include files (as well as jemx.h) ?? END
<peter> If your personal include file is correctly written it should not need a change in most cases (except where CAO
> Only jemx.h would be changed - the names of the defined constants remains
<peter> chose shorter names). But the .txt file needs update. END
> the same. Documentation for each component wherein we disclose what each
> error number means will also have to be changed. It doesn't in anyway affect
> the functioning of the programs. END
> I vote for Peter's proposal, and an addition to the ADD by me. What about
> the rest of you? END
<larsson> It looks reasonable to me too. END
<smaisala> Fine to me. END 
<njw> OK from me too END
> Also, the new code numbers will only affect those programs that use the
> latest version of jemx.h
> Okay, I'll make the changes to jemx.h, and circulate the new file to you
> all a little later today. END
> Next: AOB? END
<njw> N
<peter> No
> Any requests for the colours of the flowers at the meeting? END
> There will probably also be a chat meeting next Thursday, where we can discuss
> the results of the science meeting where scientists interface with ISSW.END
<peter> Flowers: black, red and golden for Germany, blue and yellow for Sweden, blue and white for Finland and Danish colours too :-)
<peter> I.e. red and white :-)
> I'll try my best, boys. Got to run now. Have a good weekend. Bye END
<njw> Bye bye END
<larsson> Bye. END
<smaisala> Bye. END
<larsson> bye
*** njw has left channel #jemxadr : (njw)
*** Signoff: larsson ()
*** Signoff: smaisala (using sirc version 2.211+4KSIRC/1.1)
