
	Weekly meeting #20: Friday 27th August 1999

Participating: Peter Kretschmar, Allan Hornstrup, Carol Anne Oxborrow,
	Sami Maisala, Juhani Huovelin, Niels Joergen Westergaard.
	
Subjects discussed: Updating ISDC Libraries
	Deadlines for executable updates and first V1 release
	Travel plans for Helsinki meeting Sept. 22-23 1999
	Network problems at Geneva Observatory

Action Items:

AI990827_1	ALL 	ASAP	Send travel plans and hotel room requirements
to Juhani for Helsinki meeting 22-23 Sept 1999
OPEN



*** IRC session log started at Fri Aug 27 09:39:15 1999.
H oxborrow     #jemxadr 0 Carol Anne Oxborrow          ~oxborrow@tenma.dsri.dk
H allan        #jemxadr 0 Allan Hornstrup                ~allan@apollo.dsri.dk
H Peter@       #jemxadr 0 KRETSCHMAR Peter         ~pkretsch@isdcul10.unige.ch
315 apollo.dsri.dk * End of /WHO list.      
#jemxadr> Hello chaps! Sorry I'm late. There's no good excuse I'm afraid. 
#jemxadr> Stefan sent me an email to say he's ill in bed and won't be joining us.
+ I hope someone from Helsinki will be with us soon. END
<Peter:#jemxadr> Being gentlemen (sometimes), we'll forgive you and try not to lo
+ok to smug ;-) END
*** smaisala <~smaisala@taurus.astro.helsinki.fi> joined channel #jemxadr
#jemxadr> Hi Sami! Did you have a good holiday? ENd
<smaisala:#jemxadr> Hi there !
#jemxadr> Let's start since we've got one person from each group. First:
#jemxadr> Any comments on the component status anyone? END
<smaisala:#jemxadr> Yes it was really nice holiday ! END
#jemxadr> My personal feeling, which I know is shared by NJW is that we've had a 
+lot of learning to do
<Peter:#jemxadr> Nothing new from my side. But the status of the others looks qui
+te good. END
#jemxadr> to make the V1 components (ObsSim in NJW's case) and that though there 
+is very little actually coded at the moment on the V2 
<Peter:#jemxadr> I can also repeat CAO's statement. END
#jemxadr> components, these should be relatively easy since we now have learned t
+he DAL&Co language. It seems that the vast majority of my coding/testing/debuggi
+ng time
#jemxadr> has been taken up with ISDC I/O and error checking and documentation.
#jemxadr> Processing has been only about a third or maybe less of my developement
+ time.
#jemxadr> There is possibly a down-side corollary of this intense learning that h
+as 
#jemxadr> occurred in the last few months:
#jemxadr> If Don, Bruce, Stephane and friends add all sorts of whistles and bells
+ to DAL/DAL3 theres a good chance that these `helpful' things won't get used now
+ that we've learned to do things the older supposedly more difficult way.
<Peter:#jemxadr> I don't think this is necessarily true:
#jemxadr> A `difficult' method one can already apply is far superior to an `easie
+r' way one still has to learn. END
<Peter:#jemxadr> (a) A lot of the above mentioned gentlemen's work is in direct r
+eaction to a real or perceived lack in functionality
<Peter:#jemxadr> (b) the major part of the ISSW is still to be written and my fee
+ling is that it is there that any new functionalities will come into use
#jemxadr> Added functionality I can live with: tinkering around with stuff we alr
+eady use is REALLY going to complicate matters. END
<Peter:#jemxadr> (c) Actually there is a serious effort at ISDC __not__ to add "b
+ells & whistles" to the libraries any more but just to react to real problems.
<Peter:#jemxadr> This is also to finally free some time again for other duties.
#jemxadr> That's very reassuring to hear. I imagine that with their workload the 
+ISDC developpers don't have time for frivolous stuff. END
<Peter:#jemxadr> For me personally, the only thing that has changed between DAL 1
+.2.7 and DAL 1.3.1 is that my OMC software now works because I can use one new f
+unction Don wrote (the sorting problem
<Peter:#jemxadr> we spoke about) END
<Peter:#jemxadr> Where I foresee new additions is mostly in DAL3JEMX:
#jemxadr> I do have one more quite important question: next week I'll be deliveri
+ng version 2.0 of JGainFitting, the luxury cadillac, all the functionality-you-c
+an-dream-of version, however,
*** huovelin <~huovelin@taurus.astro.helsinki.fi> joined channel #jemxadr
<Peter:#jemxadr> Once we have the daat formats for analysis products (images, spe
+ctra, lightcurves) settled, we will probably qant common functions to read/write
+ those. END
#jemxadr> Even the most cursory discussion with people from the instrument team m
+akes me think more changes will be required, possibly involving changing process
+ing that was previously changed to accomodate their wishes. Question to Peter:
#jemxadr> what is the deadline for submitting absolute final versions of V1 deliv
+eries that have already been delivered and integrated once? END
<Peter:#jemxadr> Probably I should discuss this with Tim & Roland, but my impress
+ion is that there is no firm deadline in this sense
<Peter:#jemxadr> We may always need to update and improve existing software.
#jemxadr> It might even be conceivable then, that after launch and with real-data
#jemxadr> experience we could be changing processing during the mission? END
<Peter:#jemxadr> The only caveat is that anything delivered later than beginning 
+of September may not find its way into the first installation of the V1. END
<Peter:#jemxadr> To CAO: Yes, actually everybody expects that to happen.
#jemxadr> Thanks for giving me one firm date to work with :-) END
<Peter:#jemxadr> The important thing is to have some __working__ software but it 
+will be reworked and improved with 99.999% probability. END
#jemxadr> I just wanted to check that such changes wouldn't come as a horrendous 
+surprise to you chaps: I can feel them coming already. END
<Peter:#jemxadr> It's more a problem for you really - balancing the time you use 
+to update exsiting code versus the time to write new contributions. 
#jemxadr> Given that flexibility I can deliver next week the version 2.0 of JGain
+Fitting. END
<Peter:#jemxadr> Tim warns about the trap many people fall in of fiddling too lon
+g with stuff that already works. END
<Peter:#jemxadr> Sounds good. David Landriu will spend all of September at ISDC a
+nd the expectation is that at the end of the month, a patchy but recognizable V1
+ system exists. END
#jemxadr> I agree that's a serious problem - we all feel we can improve our code 
+- but me real concern comes from the IT's demands. END
#jemxadr> That's all I had to say about developement and delivery in general. I h
+ope I haven't bored those people who have kept quiet. END
#jemxadr> Any more comments? END
<allan:#jemxadr> ( - leaving , dentist appointment. Happy chatting) END
*** Signoff: allan ()
#jemxadr> Next on the agenda: travel plans to Helsinki. END
#jemxadr> Sami will be going to Geneva in early september to learn about making
#jemxadr> grey filter history tables. Allan and I have agreed to take some respon
+siblity
*** njw <~njw@heao1.dsri.dk> joined channel #jemxadr
#jemxadr> for this functionality, so he suggests that I stay an extra day at the
#jemxadr> Helsinki meeting to discuss this with Sami and Eero. How does that soun
+d Sami? END
<smaisala:#jemxadr> Yes that sound like a good plan END
#jemxadr> Can I make it an action on all of us to send our travel plans
<huovelin:#jemxadr> I have received wishes for hotel only from Peter and Stefan s
+o far, so others please respond as soon as possible END
#jemxadr> including number of nights to stay in Helsinki to Juhani so that he can
+ get on with finding hotel rooms? END
#jemxadr> Next on the agenda: contributions to the monthly report. PLEASE give me
#jemxadr> some juicy coding/developing exploits to send to ISDC end
<Peter:#jemxadr> Ahem, I fear there will be nothing new from my part - other duti
+es taking too much of my time. END
#jemxadr> I'll include a cut down version of the component status summary I sent 
+you all yesterday, but what about documents read/produced?
#jemxadr> Libraries installed? DAL bugs uncovered? Meetings attended?
#jemxadr> Test data simulated? Code debugged? Prototypes converted to C?
<njw:#jemxadr> You may mention that I've just submitted a preliminary version of 
+the
<njw:#jemxadr> JSIM raytracer executable to Davide to see if he can compile and l
+ink it.
<njw:#jemxadr> The rest of the suite is standing in line if it proves to be OK EN
+D
#jemxadr> That's great, but does it count as JEMX ISSW? END
<Peter:#jemxadr> I'd say yes. END
<njw:#jemxadr> No, sorry, no END
#jemxadr> Goody! That goes in. Anything else? END
<Peter:#jemxadr> Hmm, probably Davide should mention it in his report, but OTOH I
+ see nor reason why you shouldn't mentionit here as well. END
#jemxadr> If there's really nothing else for the monthly report, what about AOB? 
+END
<Peter:#jemxadr> Sami - are your travel plans for Geneva fixed now?
<smaisala:#jemxadr> To Peter: Do you still have computer problems there in ISDC b
+ecause I can't log in ?
<Peter:#jemxadr> Probably yes. I'll check with Alain. 
<Peter:#jemxadr> It's probably linked to the full-sclae overhaul of the internal 
+network at the observatory which upsets things around here too.
<Peter:#jemxadr> END
<smaisala:#jemxadr> Ok So we just have to wait ... END
#jemxadr> Does that mean it could be a problem connecting remotely next week? END
<Peter:#jemxadr> Hang on for a minute, I'll see if I get him on the phone ...
<huovelin:#jemxadr> I'll have to get on with other tasks now, so bye bye everybod
+y ! END
*** Signoff: huovelin ()
<Peter:#jemxadr> Can't get him right now, but I'll raise the issue.
<Peter:#jemxadr> Anything else?
#jemxadr> Well, I think that's everything: we'll chat again next week.
#jemxadr> I think it helps to keep in touch and find out what problems other peop
+le are having. END
#jemxadr> Have a good weekend everyone. END
<Peter:#jemxadr> Good. Enjoy your weekend. END
<smaisala:#jemxadr> OK. Let's 'see' you next week. Have nice weekend! END
<Peter:#jemxadr> Sami: remember to tell me your travel plans! END
<smaisala:#jemxadr> Peter: I send you a mail.
<Peter:#jemxadr> Fine - thanks! END
*** Signoff: Peter (Leaving)
<smaisala:#jemxadr> BYE! END
#jemxadr> Goodbye END
*** Error: Closing Link: oxborrow[~oxborrow@tenma.dsri.dk] () 
*** IRC session log ended at Fri Aug 27 10:27:08 1999.
