Date: 8-Feb-2000
Attendees: C. Burkhardt, R. Douglas, K. Peterson, D. Shaw, H. Shukla
Most AIs from the last meeting were completed, except for the update to the IDE evaluation document to reflect the assessment of other, non-viable IDEs.
| Chris | Update the IDE evaluation document to include a summary of the non-viable IDEs. |
Rob and Chris evaluated a few UML tools for potential use on the APT effort, and have been tasked with writing up a detailed report. The upshot is that Rational Rose (98i) appears to be the most suitable for our work. While there was concern about being able to merge independent work on different parts of the data model into a unified whole, RR appears to be capable of handling that. Also, contrary to earlier belief, the output file formats for most of these UML tools appears to be incompatible. Since the Institute already has one site license for RR, it may that we need to purchase an upgrade, purchase additional licenses for multiple, simultaneous users, and move the whole thing outside the firewall.
Hemant reports that the UML tools run very slow on his workstation (a Sparc 5). The suggestions were to either a) upgrade, or b) telnet to aura.stsci.edu and display the session back.
The question of how to configure our tools and output products was raised, but not solved. We need to raise the issue with Tony.
| Chris | Investigate getting Rational Rose installed for general use on the science cluster, or at least on ra. |
| Dick | Investigate disk space and configuration resources on ra. |
| All | Learn to use Rational Rose effectively. |
We discussed Rob's initial work on the high-level data model. With a few revisions it looks like a promising start, although we each need to examine it in more detail to root out the deficiencies. Dick & Hemant discussed their work on the AstrObject, Telescope, Instrument, and Detector modelling, which is at an "intermediate" level of detail. It merged pretty well with the high-level draft model. Chris pointed out that we should all be thinking in terms of "interfaces" for all of our classes to make implementation much cleaner downstream.
| Rob | Post the working draft of the high-level model on the Web site. |
| Dick & Hemant |
Refine the AstrObject, Telescope, Instrument, and Detector packages, and create UML in Rational Rose. In particular, Hemant should revise the |
The next meeting (Feb. 15) will focus on comparing our refinements to
the rough draft of the data model description. We will also begin to
draft questions of the Requirements Team that must be addressed ASAP for
work on the DM to be concluded.
Copyright© 2000 The Association of Universities for Research in Astronomy, Inc. All Rights Reserved.