SDAST Chat Meeting #100: Wednesday 9th February 2005 Present: Carol Anne Oxborrow, Niels Joergen Westergaard, Jerome Chenevez, Peter Kretschmar, Silvia Martinez-Nunez, Stefan Larsson Summary of subjects discussed: Misalignment matrix: going well so far; CAO to use NJW's IDL tools for SCW selection Imaging development (NJW) still pushing ahead with no unforeseen problems though still has work to be done - SRCL-RES still to be output yet. Big back projection tables organised in own package for delivery. Mosaicking (JC) pushing ahead - will use both NJW's and NL's images as input and vignettting giving some I/O problems, but otherwise on track Source extraction (SL) having detailed collimator modelling included NJW working on a collimator fine model table with nominal tilt for SL and ISDC, and about to submit template to ISDC PK and SMN to make new scripts including new imaging and mosaicking tools Delivery checklist: updated and ready for comments before Friday when it will be sent to Isabelle and Roland; SVR should be added at lowest priority Smart Selection for radiusLimit now on back burner Flights to Madrid to be discussed during next chat (Weds. 23rd Feb.) Not much progress with AI list; CAO to send round personal lists to all Action Items: AI050209_1 ALL 2005/02/11 Email comments on updated delivery checklist OPEN to CAO before checklist is sent to Isabelle and Roland AI050209_2 SL 2005/03/01 Write SCREW for improved source extraction OPEN components to incorporate full collimator modelling **** Logging Started : Wed Feb 09 10:57:07 MET 2005 > Hi NJW! END Hello Carol Anne END *** jerome (~jerome@oso.dsri.dk) has joined channel #jemxadr > Hello guys - this is our 100th chat meeting!!!!! END > Let's hope we're not the only ones who turn up to celebrate this fact! END Hello! How do we drink champagne with Zircon? :) END *** larsson (~larsson@130.226.216.181) has joined channel #jemxadr Hi there. Afterwards we can have some sherry - I found a bottle during the cleaning up END Hi Stefan! At least someone outside from Dk. Hello Stefan, how are you? Good of you to take the ISDC turn at the end of the * njw month. END *** peter (~pkretsch@crab.unige.ch) has joined channel #jemxadr I notice a serious discussion (sherry). END I seem to be missing the fun stuff! END We will celebrate the chat number 100! END Hello Peter. We are dscussing how to celebrate this 100th chat meeting! END *** silvia (~silvia@uhuru.dsri.dk) has joined channel #jemxadr Buenas Dias Silvia! Bon jour everyone ! Good idea to celebrate the 100th chat - but what do we remote guys do? END I guess we will have to bring some for the March meeting END > I'm not even sure what us non-remote guys should do! Perhaps we should > sing the SDAST song (once it's composed!). Anyway, it's great to see > so many of us here for this, our 100th Chat meeting. Any comments or can > we dive right in? END Let's go! Let's dive in! END > First on the agenda: misalignment matrix. This doesn't seem to be too hard > and Pascal's written a good manual, the only real problem thus far is finding > all the necessary scws to use, and I'm spending a couple of days just on > that. Once I've done the sicentific analysis, no doubt there'll be interesting > things to report in the way Pascal's matrix-making software works (or rather Carol Anne: Please note that I have some IDL tools that might help you in that regard. END > doesn't work), but that will be in a few days time. END > Thanks, NJW, I'll look in on you this afternoon to pilfer some good tools. END > I'm really trying to get this done for Stefan's sake so that he can see if > a better misalignment matrix (to go with the half-pixel shift) changes his > results any. Any comments? END N N N No no - good work! On the imaging status then: j_ima_iros now exists in a version that runs with FITS I/O. The missing part is the writing of JMXi-SRCL-RES, but I hope to finish that this week. Then I think that we are ready for a first delivery to ISDC, however, there is a number of templates to deliver first and some IC files to discuss with Nicolas (the large backprojection lists, that are now organized in a manner similar to the IMOD group). Yesterday I've started a test series on rev_2 data from the Stockholm Archive - of about 20 SWIDs there was one crash which has to be investigated * njw more. END > Well that sounds all very active and moving ahead, even the crash sounds > like you're learning something new! Better to have crashes now than in > April. Any other comments? END Yes, I'm quite impressed too and I'm glad the SRCL-RES will be written :-) END > So, can we move on to mosaicking status? END Y Y > JC can you give use your status? END Ok. I am still deep in the puzzle of combinig images from either NJW (OSA4 style) or NL (IROS style) with several E- bands and mixing ScWs with or without sources... But it is going slowly ahead and I should have something running in the OSA environment soon. Then, there still will be to fix some issue regarding the vignetting but it should just be a question of I/O. END > Good stuff. That vignetting really is a problem - it sounds so sweet and > victoriana-ish, and it's just a nice name for a real headache! But if you've > got it under control that's great. END Yes, in the present situation, it is really just a question of getting the right table from the right place in the right shape. END > Any more comments? Can we move on to source extraction status? END N n/y N/Y OK OK OK. I made some updates to the collimator handling to include the tilt information. I will complete it when the new IMOD is available. (No blaim NJ, I know you are working hard). I guess thats it. Next version early March. END Actually I'm working on this right now. There will be four columns in the new JMXi-COLL-MOD: XSECTBOT, YSECTBOT, XSECTTOP, and YSECTTOP with intersection points at the bottom (i.e. at the detector window) and at the top (the other end) of the collimator. For the moment I've just put in a tilt of 10 arcmin Is the 10 arc sec a "best fit" from Niels software? END Sorry, 10 arcMIN in the Y direction. Later on I'll update that value with Niels' best results but currently there is a discussion between Carl and Niels about the size OK. The actual tilt is not that important to me now. I just need the and direction. But you should be able to check that it has an effect, Stefan. END IMOD format to get the software in place. The actual model can be updated later. END > That's good news about the collimator modelling - they were quite concerned > about this at ISDC, whether it could be done in time. Any other comments? > END If you agree on the terminology I'll submit the new JMXi-COLL-MOD templates Only that I'm cautiously optimistic seeing all these efforts - it really looks good from here. to Marc Türler END END > Okay, any more comments before we move on to the delivery checklist? END N N N N Maybe a comment on the column names in IMOD for the collimator. Since the name for the bottom differ from the old names the update of my programs must go hand in hand with the new model. Is that a concern? END I could keep the old name (X_CROSS) for the bottom ones in order to avoid such transition problems. The names will become a little strange but is the * njw benefit greater than disadvantage ? END Since we are anyway working for a new major release, I would vote to just follow the logical order: new templates, new IC files, new software. Somewhere inbetween the software does not work, but that is by far not the first time we've had that problem. END OK, I'll stick with the new names END Fine. END > Well from the discussion above, I think I should add somewhat more specific > items about collimator model and back-projection packages to the delivery > checklist. Are there other things to add to the list? Who's going to deliver > new scripts to include the new imaging and mosaicking modules? END Good question! I will, maybe with help form Silvia. I will help Peter on this ! END I've agreed to as much in a phone chat with Roland. Niels was also asked but read his mail too late after all the changes in Copenhagen (are yous till in the same building :-) ) I'm very glad to hear this. Our only other way out would be to ask for help > I'll put your names on the checklist. END * njw from ISDC (Stephane Paltani perhaps) but this solution is better END Btw, I think you should include Stephane Paltani for mails on JEM-X Monday Meetings etc. > Yes I agree that Peter can probably do this most efficiently. END > Is there really nothing else for the checklist??????? I'd like to have it While he will _not_ do the actual integration, as Science Analysis Team Leader he should keep in touch with JEM-X developments and comment on any issues with ISDC's general framework and plnas. END > at detailed, file-by-file, package-by-package (smallest deliverable units) > so that we really can see what's been done and what's not, instead of lumping W.r.t. the checklis: we need an updated Scientific Validation Report where someone has to look at all these nasty Crab data Roland unearthed. END > everything together under one umbrella name. END Btw, "someone" does not include Germans working at Spanish ESA sites :-) END > Good Peter. That'll come after march 31st I think, where we've frozen and > delivered and then give the s/w a good old workout. Is that priority okay > with you? END Fine with me, I just thought we should show that we have thought about it - and make sure it is done! END This should start before. I will try to do some checks when my update is completed (early March in my famous plan). END If I stay in Valencia ( I think I will), I will help with the software validation END If you wish to make the checklist more detailed you could add template Famous plans are like famous last words? Often just ade up of thin air? ;-) ;-) deliveries (JMXi-BPL.-GRP and its components) and IC file delivery Btw, all I know about Silvia's options here is that someone has taken a decision but they will definitely not tell us here yet ... END > I'll include you on the list Silvia. END It is so cold up here that air is almost freezing.:-) END > Perhaps, silvia you can contribute some results from your thesis? END (jmxi_bpl_grp_0000.fits). BPL means 'back projection lists' END > I'll circulate the checklist again. Please take a good look at it and let > me know about templates, files, IMOD data, software modules (think of all I think Carol Anne in the SVR we should include results with the new software (OSA 5.0) or not ? END > those outstanding SPRs!) that are not mentioned. END > Yes OSA 5.0 would be best, if you've got time to re-do some of your data. END I will try ! END Sure - the main idea is to have OSA5 results! END > One thing I've noticed from the checklist that no one's mentioned so far: As Stefan said, people should start checking their software when it is converging. But at the end a list of systematic checks must have been done - especially on the "bad cases" - and results published in the SVR. END > how are we all coming along with implementation of the Smart Selection > radiusLimit criterion/criteria? For my part I don't need it at all, what > about the rest of you? END It has too low priority at the moment END I guess you are thinking about DETECTOR radius limit (not images)? END We _should_ be doing it. It may remain a known issue for OSA5 in my case. END NJW: weren't you the one bothered by not correctly estiamting the effective area? END' Yes, but j_ima_iros takes care of this internally (that's what I understand OK * njw from Niels). When j_ima_iros is on the tracks I'll start on the other things END > In that case I'll remove specific mention of this from the checklist. > Can you all please look at the checklist and send me additions/subtractions > by friday, so that I can send it to Roland and Isabelle, to quell their > concerns about our status and deliverability? END Stefan, do you have a SCREW or SPR for the delivery of the new spectral * njw extraction which uses a better model of the collimator ? END I dont think so. > So that's an AI on you SL to write the SCREW, and an AI on all of you to I should check and submitt one if needed. END > return comments on the checklist by Friday END > All of which brings us to the AI list. I'll start the ball rolling: > I've got one open AI to add things to the instrument configuration table. > This table is SO much in need of an update that I'll probably do it one day > soon when I'm really mad with Pascal's program and need a break from > misalignment matrix making. What about the rest of you? Peter?END > NJW, how are your AIs? END OK, I'll start: Very bad, I can't add to what I've already said. END AI30_7 NL/PK Propose a new data structure template for Back Projection functions I _think_ this one is closed and now part of normal work led by NJW. Do you agree, Niels Joergen? END I do, yes END AI30_14 NJW/PK/NL Remove overlap in IMOD/detector data functions ... Currently considered low priority - maybe OSA 6??? Needs a new date quite a bit in the future, I think. END AI30_18 NJW/SL/PK write a proposal on how `Smart Selection' of radiusLimit should be implemented AI30_19 PK ... write corresponding SCREW. Also pushed back to more clement times, ins't it? END Y y > NJW, is this one closed or obsolete, or OBE? > AI30_17 NJW Xmas 2004 Send SM list of empty fields since last rejection OPEN criterion change in instruments, with which to make new background models obsolete, I guess END Agreed. > So your turn SL: AI status if you please. END Not much I am sorry to say. Maybe change 4.0 to 5.0 in AI27.22. END > Okay, guys, and guyesses, I've made these changes, but I will distribute > a personalised list to everyone who's got open AIs in an attempt to > stimulate some action, especially from the guys who don't chat (NL, SB and > CBJ). Now our favourite part: AOB? Anyone? I'd like to remind you all from > now on SDAST chats are every second Wednesday at 11.00, and I will send > an agenda and reminders on Monday and Tuesday. This seems to be the only > real way to plan around these chats, so I hope you will all do precisely Question regarding the SDAST Meeting: have you checked flights yet? > that. END > Yes, and they don't look promising! Personally, I have to say, don't count > on me turning up early - if at all. My father's been taken back into > intensive care, and my major worry is that just as I have to be with > Charlotte for her audition I'll have to rush off to Preston hospital instead. > sorry to let you down guys. END Sorry to hear that! Just hoping things will turn out OK. END So what are the flight options? Sorry to hear that Carol Anne ! I hope everything will be fine ! END > Of the flights I've seen, the ones that leave at a good time in the mornings > have terrible return times in the afternoons. Could we leave the details > for the agenda of the next chat? END Niels and I will go by car from Valencia either on Sunday afternoon or Monday morning depending on the agenda. END OK OK > Any other business? END n N n N > Have a good week then, and keep up the good work everyone! We next chat on > Wednesday 23rd February, which will be chat #101. END Sorry, I won't be around for the next chat (holiday in that week) - have fun - Perfet ! Take care Carol Anne ! Good week to every one ! Hej hej ... END Bye bye END *** njw has left channel #jemxadr : (njw) Bye! *** silvia has left channel #jemxadr : (silvia) *** Signoff: larsson () Good bye every one. *** jerome has left channel #jemxadr : (Quit) bye *** Signoff: peter (ircII/tkirc)