[306] in Tooltime
Scopus Implementation Meeting Oct 22, 1996
daemon@ATHENA.MIT.EDU (Steven Wade Neiterman)
Wed Oct 30 09:28:12 1996
Date: Wed, 30 Oct 1996 21:30:08 +0100
To: tooltime@MIT.EDU
From: Steven Wade Neiterman <wade@MIT.EDU>
Scopus Implementation Meeting October 22, 1996
Attending: Goguen, Rosato, Smyser, Wade, Yanow
(A number of team members were at NT class)
Agenda:
1. Acceptance Testing
2. Bug list and Tracking
3. Technology Transfer Update
4. Old Meeting Notes (10/8)
5. Next Steps
6. Roundtable/Status/Action Items
-----
1. Acceptance Testing:
- Wade handed out copy of all feedback received. Need to write
executive summary for Bill Hogue as sponsor of the project.
- Keeping in mind that acceptance testing was designed to answer
how well it met the functional specification, it appears that there
are no showstoppers that would result in the primary functions not
being accomplished.
- Acceptance testing did not attempt to answer when the application
is ready for rollout since this has a separate criteria and a set
of principles to base a decision.
- Major highlights include 1. Appear to be done as compared to
functional specification, 2. There are rollout issues, including
stability testing on the Macintosh, 3. There are ergonomic and
other changes required within a short time of release 1.0 to
enhance the usability and functionality of the system.
2. Bug list and Tracking:
- Wade handed out a copy of the Scopus data model used for bug
tracking, along with annotation on possible field definition and
tracking process.
- Rob and Peter Lee will begin to review comments sent to tooltime-bugs
for the purpose of creation of a list of items to be tracked.
- Rob offered to define a Filemaker database to track the information.
Other alternatives considered are 1. Excel, 2. Scopus, 3. PowerBuilder.
- Goal is to generate first pass of the list for meeting in two
weeks - November 5th.
3. Technology Transfer Update:
- Tom is waiting for release 1.0 to be completed (which should now be
the case), so that Lynne and Tom can review the tools and system.
- We have stated that learning the tools and system is a pre-requisite
to attending the CA class.
- Tom needs a new machine to run Scopus, preferable one that runs NT.
This will allow Tom to run ScopusWorks instead of using Unix.
Wade will check with Bob Ferrara/Roger Roach on status.
- For now, it is agreed that Tom will evaluate Design Team on Unix
(Athena) in order to quicken the learning phase.
- Since Athena supports Tcl, this is an option to begin the learning
- Tom will report at the next tooltime meeting on progress with
learning the system from Lynne.
- Should help-it-dev be moved from Lynne to Tom?
4. Old Meeting Notes (10/8)
(Some items covered on 10/8, meeting notes not previously sent)
- Agreed that we are ready for training since:
1. 1.0b7 has enough functionality to train
2. Training document ready
3. Known issues can be addressed during training and are not
thought to be showstopers.
- Help Desk staff to be trained (introduced to Scopus) on 10/10,
students on 10/15.
- Help Desk staff to help with student training since there will
be about 50-60 people. Attempt to pair only two students per
machine.
- Also need to train DCS and Transmissions
- Issues: Should keep training close to the actual rollout for
effectiveness.
Need some sort of "makeup" for students who miss initial
training
- The following comments were made with regards to acceptance testing:
- Best results occur when release 1.0 is frozen or as close as
possible - (pakulat)
- Acceptance testing without good criteria becomes subjective since
people have different minimal requirements (allen). Tolerance
is subjective (pakulat)
- Acceptance testing answers "would I buy it?", does it meet
minimum requirements (pakulat)
- current state is OK to perform acceptance test (smyser)
- Searches is important for acceptance testing (Lee, Goguen)
- Based on the above comments, acceptance testing was delayed until
10/16 for both a select set of students and staff in 16-720.
5. Next Steps:
- Next meeting agenda will include at least the following:
- feedback on bug list gathering
- technology transfer update
- Macintosh stability testing
- schedule refinement
- criteria for rollout
- status on release 1.0 application changes
- roundtable on all action items
6. Roundtable/Status/Action Items
-----
Wade:
- Started to test SNS encryption for Windows, Mac and Unix as a first
step to install SNS on the server (SNS is Secure Network Services
and performs encryption). Also, worked on install programs.
- Will meet with database group on other issues (see action items)
- Visitor Center demo for E40 and E19 on 10/23
Not previously record - from meeting 10/15
- Performance with 4k logs on test showed slow results in some
search cases, but not all. New theory may be related to how
Scopus displays long result sets. A better test of database
performance is to look at a query from sqlplus contrasted
against the test and production database.
- Some of the migration concerns include:
- Force password changes on existing accounts for production
- Availability of reporting (printing or via Brio Query) as needed
by netinstall and others.
- Need better numbers of Filemaker data for migration, including
UNIX/VMS.
- Need to determine date for Filemaker change to 'read-only' mode
- Determine everyone who needs to be notified of the switch date
- Availability of some Help Desk staff to be available when data
is migrated from test to production to confirm the readiness of
the system.
- Training completed, Help Desk full time staff available for QA
- Limit "test" database account and usage to developers and testers
- Process issues: adding/deactivation of accounts, checking R1.0
usage of warehouse lookup, pick list (software) updates, etc.
ACTION ITEMS:
- Refine SLA with Facilities Management on production machine and
coordinate loading of Scopus software on the production machine.
- Confer with database services to install SNS on test machine after
production release. After initial testing on test, migrate SNS to
production machine.
- Automate warehouse feed
- Collate post release 1.0 "todo" list to ensure stability
- Work with database group on account transfer from test to prod.
- Waiting for arrival of ScopusWorks
- Need better counts of final Filemaker data for production import
- Need to check on Windows machine for Tom
- Follow up on Scopus bugs,
21009 - InformTEAM, Terry called this in to Melissa.
23591 - Macintosh SNS issues
- Need to complete write-up for acceptance test findings
- Wade to work on vision statement and direction for next 12 months,
including potential customers (e.g. Local Experts use as a backend
communication and searching). Need plan, pilot?, requirements.
- Need to resolve problem with adding accounts to production - getting
database constraint error
- Further investigate performance as stated above
- Investigate resources needed to produce release 2.0
- Check timeline for payment of next maintenance contract?
- Write architecture document
- Obtain latest documentation on InformTEAM and SupportTEAMplus
(a.k.a. Scopedia) 510 597 5612 Joel@scopus
- Work on rollout schedule
- Continue on test and machine cleanup and install documentation
- Index mit_lock_user on production (was done on test)
- Check on next release of Scopus, version 5.0
-----
Rosen (via Wade):
- Submitted response to Scopus on InformTEAM issues (8/29). Brett
sent a clarification of the Scopus response, but Scopus closed the
log.
- Possible future items may include looking at improved searching
and email/oracle publishing
ACTION ITEMS::
- Follow up on InformTEAM. Terry Arnett pointed out that we had
outdated documentation, Wade to follow up. Terry submitted a
call to Scopus, Wade and/or Brett to follow up on issue number.
-----
Sousa:
- Completed 1.0b7 (10/18) - see motd for changes. Many changes
were agreed to on 10/8, including: resolve duplicate button GPF,
remove "no changes made" dialog, pin result set from My Logs
function, remove 'No Changes Made" on New function, a number of
clipboard cosmetic changes.
- Working with Tom on technology transfer issues
- Working with Peter R. on Macintosh stability testing
- Recording known problems and issues that need to be addressed
in release 1.0b8 (if there is one). What are know issues
(e.g. is duplicate button working?).
ACTION ITEMS:
- Report to team on Macintosh stability issues.
- Lynne will complete the Programmer's Guide with the latest 1.0b7
changes and send back to Esther. It should then be complete and
ready for the Web.
-----
Ragone:
- Currently helping with Macintosh stability testing, will report
by November 5th.
ACTION ITEMS:
- Continue exploring Brio Query, report progress to team
- Help with QA on training document and 1.0b7 testing
-----
Smyser:
- Currently working on bug list gathering and Filemaker form for
storage and reporting.
- Completed software list update. Need to send wade updates to
Filemaker. Also, need to check other fields, e.g. hardware
ACTION ITEMS:
- Helping out with training document (if there is feedback????)
- Continue with QA on the 1.0b7 changes
-----
Yanow:
- Interested in updating training guide, need feedback
ACTION ITEMS:
- Agreed to add information on Housecalls, Netinstsall, and a table of
contents. Additionally, an appendix (e.g. FAQ) will be included.
(Do we still need information on a FAQ and TOC? - wade)
- Check with Lynne on completion of the programmers guide and
incorporation in the project notebook.
- Agreed to download Scopus from the web and test it in E19
-----
Rosato:
ACTION ITEMS:
- Become familiar with the Scopus environment. Begin technology
transfer by using the design tools, writing a "sample" program.
- Next status update due on November 5th
-----
Lee:
- Two unknown issues from 10/8 include: multiple history/summary search,
and Data Probe. Research into both completed. Need final
summary to team.
ACTION ITEMS:
- Help with bug list gathering
- Help with gathering better Filemaker load counts, supply Filemaker
data for production load, possible advance test.
