CIT Meeting Minutes
In Attendance:, Allison Snow, Peter Ayoub, Stan Thomas, Tommy Murphy, Richard King, Bill Markham, Jay Dominick, Ralph Kennedy, Bryan Locco, Susan Smith. Guests: Ching-WanYip, Pat Morton
Topics for Discussion
1. Follow up question for Pat Morton from Previous meeting: “Is there hope for becoming less dependent upon wired connections for software updates and other critical applications such as password updates?”
IS has a solution. A batch file has been written that will work over wireless to run updates. One issue being addressed: way to provide alerts that your password will expire. If you log onto VPN, you get prompted. They are looking at ways to look at the active directory to send an email to remind people. At this time they are not sure if can do anything to notify over the wireless.
Discuss of topics related to Standard load:
Pat Morton presented his recommendations concerning IS recommendations in regard to:
1. Hard drive partitioning
2. Migration from Mozilla 1.7 to Thunderbird 1.0 for email
Refer to the attached “Considerations for the 2005 ThinkPad Base Configuration” for detailed explanation.
Discussion on browser issues:
The question was asked about the viability of running Mozilla/Firefox at the same time. Pat has found they crash when he tests both simultaneously.
Discussion on hard drive partition:
Stan suggested getting students involved in testing next summer with Linux partition test.
Allison reported not a being fan of hidden drives. She, and others, want to know what’s where and size, etc.
Stan asked for movement regarding IS recommendations on browser changes and hard drive partitioning for this year’s standard load. Ralph moved to accept the IS recommendations. Jay seconded the motion. It was approved by all present.
3. Software that might be removed from standard load but still available for individual install.
There was a discussion about the feasibility of reducing the number of discipline specific software titles preinstalled on the standard load. Pat reported that some applications, such as SPSS, require pre-configuration of WFU specific info. Yip reported that some applications, such as MS Word, are very hard to install and configure so are best done on the standard load. He also mentioned that until now, there was no way to install keyed programs over the network. Now there is a way to autoinstall keyed. However, attempting to install many of these applications over the wireless would be very difficult and too large a process to take place during class.
An alternate approach, the removal of unneeded software, was discussed. The point was made that students may be unsure what’s safe to uninstall. Pat says they could include a “ReadMe” to give instructions.
Next week’s planned meeting will be cancelled.
Considerations for the 2005 ThinkPad Base Configuration.
24 February 2005
The two largest concerns for modification to the 2005 ThinkPad LoadBuild are the possibility of partitioning the Hard Disk Drive and the move from Mozilla as the default Email to Thunderbird Email.
I am including my recommendations and the rationale of each recommendation. I have made these recommendations based on my own research but also based on the opinions and requests of various members of the 2005 LoadBuild team and closed conversations with IS and ITG personnel.
1) Hard Drive Partitioning.
The recommendation is YES but with implementation on the 2006 ThinkPad. This decision is based on overwhelming YES responses from the LoadBuild group, HELP Desk and Tech Shop requests and BCS requests over the past year. In addition the Computer Science Department has been requesting some type of partition for a while. I would like to put together a team to investigate and develop guidelines and best practices for implementation of a partitioned HDD in 2006.
Recommendations are based on several factors:
1) A partition would allow for a more rapid reimage of the OS without touching the individuals Userdata
2) Separating the Userdata out on another partition will help reduce collateral corruption should the OS corrupt or suffer from some other debilitating breakdown
3) General ease of backups and finding data via non traditional explorations (such as using a PXE, using Linux as the OS or using a Samba Share along with Boot Array.)
Partition format is ultimately to be determined by the LoadBuild team and approved by JLD and the CIT. The most likely scenario will be one to offer the most flexibility and least amount of confusion to the end user. Potential options could include visible partitions, a hidden partition containing the Userdata or use of a PXE backup file contained in a hidden service partition.
The reason for implementation in the 2006 load is that the largest issue at hand will be that of user education. Users that have been used to seeing the entire HDD as a single volume, will now have to be reeducated to seeing OS and userdata separated into separate logical drives (drive C and Drive D) or it the event of a hidden partition a substantial loss of HDD space (containing the Userdata) This issue is the one that needs the most consideration from JLD and the CIT. in terms of overall user experience and end user training.
Another issue will be application behavior in a dual partition environment. Many programs available as a third party install have application and save parameters hard coded. This may make the use of a second partition difficult if not impossible for some applications. Implementation on the 2006 load will allow a test group to start identifying problems that may arise due to a partitioned logical drive.
Another implication is actual partition size and format. This also includes development of contingency plans for resizing partitions and deletion of the second partition all together if requested by end users.
2) Migration from Mozilla 1.7 to Thunderbird 1.0 for Email.
The overall feeling between me and Ching-Wan Yip is to NOT migrate to Thunderbird at this time. However, a migration path from Mesilla to Thunderbird is not out of the question. I would like to put together a small test group to look into the plausibility of migration from Mozilla to Thunderbird.
Current rational to stay with Mozilla follow:
1) We are seeing reports of missing features, features that don’t work as expected and features that cause problems with programs
2) We have reports of leaks in the download manager that will cause it to bog down unless the history file is purged manually by the end user.
3) The bookmark manager drag and drop doesn't work in its window, only works if you use it in sidebar.
4) The latest clean release of Thunderbird is usually 2-3 months behind stable releases of Mozilla thus making the next revision available too late to be included in the standard load for 2005 (mozilla 1.8 releases in march/april, thunderbird 1.1 releases in may/june)
5) Mozilla remains an All-in-One product with a very robust feature set for both Email and Browser. Thunderbird is deliberately designed to be a very lightweight, slimmed down version of the mozilla email client and as such has removed a good deal of the feature set that our users may have become accustomed to in Mozilla.
6) As a standalone Email client, Thunderbird will require its own upgrade and patch path. This coupled with a separate browser will mean TWO separate programs to track for viruses, bugs and upgrades. This will put an increased burden on both the HELP Desk and the end user.
7) Several critical university wide applications have been fine tuned to work with Mozilla; these include Crystal Reports and certain Banner interfaces. Switching to Thunderbird would necessitate locating and potentially writing custom plug-ins to accomplish what we have already done for Mozilla.
8) Thunderbird is a standalone Email package. As such, its adoption necessitates us looking in to a browser to replace the browser built into Mozilla. Remember that Both Mozilla and thunderbord/Firefox use the same backend engine (gecko) and as such you cannot have both Mozilla AND Firefox/Thunderbird running at the same time.
9) User Re-Education will be a large part of the migration. Many end-users have just now become adept at the capabilities of Mozilla and to move them to a new system will cause downtime as they reorient to Firefox (these transition issues were seen in the move from Lotus Notes to Netscape and from Netscape to Mozilla)