IFAS COMPUTER COORDINATORS
NOTES FROM February 10th 2006 REGULAR MEETING
A meeting of the ICC was held on Friday, February 10th, 2006. The meeting was chaired and called to order by Steve Lasley, at 10:00 a.m. in the ICS conference room.
PRESENT: Nineteen members participated.
Remote participants: Bill Black, Dan Cromer, Marion Douglas, Mike Ryabin, and A.D. Walker.
Tom Barnash, David Bauldree, Ben Beach, Dennis Brown, Francis Ferguson, Chris Hughes, Wayne Hyde, Nancy Johnson, Winnie Lante, Steve Lasley, Chris Leopold, Donna McCraw, Joe Spooner and Jennifer Xu.
Guests: Marc Hoit.
STREAMING AUDIO: available here
Agendas were distributed and the meeting was called to order pretty much right on time.
There are several new members:
Steve introduced Tom Barnash, who will be taking Wayne Hyde's old position at Wildlife Ecology and Conservation. Tom moved here recently from St. Petersburg. He writes:
"I ran a computer magnet program called the Business, Economic and Technology Academy and high school. I was a one man team in my role at that school and another I ran in past years, so I have done it all, from rolling out Active Directory to a few thousand users to fixing paper jams. My interests really cover everything that has to do with computers, but I like desktop support because the user interaction is rewarding, and I also enjoy web design."
Steve also mentioned that Daniel Halsey recently joined the ICC list. Dan is housed in Building 162 and works .25 FTE with IFAS Software and .75 FTE with SARE. He has been involved recently in developing a web interface for the IFAS Remedy system that will permit OU Admins to control ticket assignment/notification. For those who do not know, our current listserv hardware and software are a product of our association with the USDA through SARE.
Dan Cromer then described changes in our District support staff. Josh Wilson has resigned for a much better paying position closer to his family in Vero Beach. Josh was truly excellent at his job and will be greatly missed. The Northwest District will be looking for a replacement soon.
Francis "Fergie" Ferguson has replaced David Ayers in the Central District. Fergie has had a long association with IFAS and it is great to have him back and working with us.
Bill Black was introduced and spoke a bit about his background (via Polycom from the OIT conference room). Bill is taking over the South Central District support and likely will be housed in Hillsborough County at Seffner. Bill said that he loves computers and especially enjoys working with end users and helping them be productive. In the past, Bill has worked for Disney, and for Fish and Wildlife. He last position was with the Department of Transportation. In his last two jobs he specialized mainly in videoconferencing and wireless networking.
Recap since last meeting:
Steve pointed folks to the notes of the last meeting for context but did not go into particular details.
A visit with Marc Hoit:
Steve introduced Marc Hoit, UF's Associate Provost for IT. Marc had visited with us last in May and we were very pleased that he had agreed to return. Steve hinted at some big changes which were occurring regarding centralizing IT at UF--a matter which Steve had first officially heard about at Wednesday's ITPAC meeting.
Marc began by saying that a key point is that there are NO changes which have happened yet. He then provided a handout of a Draft IT Re-Organization Plan which is a conceptual or visionary plan which covers changes which he believes we SHOULD make. Marc stated that this is a process for which considerable input will be sought before any actual changes will be made.
On the other hand, Marc said that they are going to attempt to "fast-track" it and make those changes move along. The Draft IT Re-Organization Plan has apparently been distributed and discussed with the Senior VPs, who all have approved the matter in concept. Marc took our group on a quick walk-through of that document.
Marc again stressed that this is a conceptual plan and not a detailed plan. No detailed plan currently exists. As we know, UF has a very distributed IT system that has grown up over many years. Those years have led to us having a huge number of small entities, and IFAS is likely a good example of what can be done with such an IT structure--centralizing some services and distributing others.
The UF IT re-organization is focusing on two primary principles: quality of service and security. There is a vast range in quality across our campus. There are groups that have amazing support and there are groups which are lucky if their e-mail works today--and Marc does not feel this is acceptable. Consequently, we want to gather services which are of a more commodity nature and bring those into a structure which allows uniform high-quality base services for all. Marc said that this matter has necessitated looking into a restructuring.
Marc has a $4.5 million short-fall between AT and CNS due to a loss of revenue from mainframe charges and testing services. Many of the clients for those systems have left and nothing was done to restructure our revenue sources for those facilities. Marc has been careful to note in his discussion with the VPs and Senior VPs, however, that there may actually be no cost savings in this restructuring. We may only get improved services--which would be worthwhile in and of itself. While Marc hopes there will be some cost savings, that cannot be determined until we get to the details.
The plan is divided into three phases and the first phase will be the quickest and easiest to accomplish. There are a number of groups under the Provost's office which are all being pulled together under a more uniform organization. This is mostly already accomplished, with the UFAD group (currently under Bridges) being the remaining group left to do. As part of this, it was decided that we would restructure around services. Groups will be reorganized so the particular services provided will be their focus. Marc is looking at Service Level Agreements (SLAs) for those groups. The plan draft includes 15 services which are provided by these groups. While Marc currently has just CNS and AT reporting to him, this restructuring doesn't necessarily mean he will now have 15 groups reporting, though it will be somewhere between 2 and 15--that is part of the details which will have to be worked out. In any case, we are going to re-organize around providing high-quality services across UF.
We are going to have to offer a number of new campus-wide services. There are four which are obvious to Marc: e-mail, collaborative and file services, web hosting and network wall plate services. There may be more or less. The timeframe for implementation of these varies. Some may start in six week to a year and some may start in two years. That is part of the detailed issues which are yet to be worked out.
Universal e-mail: We currently have 150 registered e-mail servers on campus. Marc was told yesterday that there are actually 750 on campus. That doesn't seem like a very sensible plan for this university. Marc has been given estimates that with $600K as an annual expense (FTE and hardware) we can provide universal e-mail using MS Exchange. This could cut overall e-mail costs to the university by a factor of about 3, but we can't really recover that because it is distributed across individuals as fractions of an FTE. This is an example where a lot of people would benefit, but the central services would get no cost savings.
Collaboration and file services: We do not have a good collaboration group on campus. A researcher who wants to share files has to start their own web server or build an FTP site to accomplish that. A lot of smaller departments have trouble with just file services--sharing files within their department. The move in the industry is to go to collaboration services which supply that. The one which is on the top of Marc's list, but not necessarily the answer, is MS Sharepoint. It integrates very nicely with all the Microsoft products which we already have a license for. It allows for ad hoc groups, portals, file sharing and pretty much everything we need in this environment. The proposal is to build a large bank of servers which provide this service universally.
Web hosting: We want to be careful that it is just web hosting. Marc had estimated we had 250 web servers, but he was told yesterday that this number is more like 2 thousand. It seems wrong to Marc to have this many servers which must be maintained and kept secure. You can get 50MB of hosting space on the net currently for about $9/month. Although there are certainly some groups which do need their own servers for this, a great many could be eliminated by consolidation and centralization.
Wall plate: We have a $5 million central networking infrastructure that doesn't include all of IFAS's networks or those of the Health Science Center. The big problem with wall plate is that people don't want to pay for it; but for security, continuity and quality of service purposes we need something better than what is going on currently. Marc is proposing that we take over and provide network service to all of central campus; he would love to bring IFAS and HSC into that. Those discussions will be made. Payment will be off the top. People currently paying wall plate charges would be able to keep that money.
The key to the restructuring process is to locate early adopter units. The goal is to find units that are interested and which have sufficient IT resources to combine their resources with the central services to offer this new structure. Consequently, Marc sees IFAS as a perfect example. His proposal to IFAS administration is that if we have people who are doing e-mail, providing file services and web services, we would give Marc some of those people and in return have those services hosted centrally and provided for free. Marc mentioned that such sharing has been successful already in that IFAS has provided part of an FTE for videoconferencing services. Who, when and how many is a matter for discussion. As an example, if there were five people providing those services for IFAS, maybe Marc would get three of them and the remainder would stay in IFAS to provide specialty services--thus IFAS would be able to re-purpose those positions.
The difficult aspect of centralization, but one which might promise some of the greatest cost savings (estimated at between $5-20 million), is desktop support. How do you handle the diversity which is out there is one big question. The quality of support varies greatly.
Marc mentioned that his plan is being discussed to the VPs as the "CIO" model. The first 18 months will involving getting the mail servers and collaboration servers up. We will start bringing the early adopters in and start transferring over. It will not be where everything is turned on in one day. The move-over will be gradual. We will have a good transition plan and backup plans in case something doesn't work. Old mail servers will continue to operate until the transition is complete so there will be no loss of service.
The second phase, after about the first year and one half, would be to bring in the rest of the units and likely set a deadline; sorry, no more mail servers after some given time. The same thing would go for other services. Marc stated that the Health Science Center is very interested in doing this and that all the senior VPs and VPs have basically endorsed the plan. The issue is the details.
Phase three would be to look at desktop support. Marc does not believe that the UF Help Desk and Remedy systems are nearly at the level where they should be. While "standardization" may be too strong a term, Marc thinks we need to consolidate, provide a wider level of support and get closer to industry standards of having one support person for every 76-200 people. We need to start using automated techniques, make sure the quality is there and that we have the right people in the right places. Marc believes the university really needs to look at a desktop replacement model. This is a $2 million/year budget item, but it seems ridiculous that we have people in departments with seven-year-old computers running Windows 98. Dr. Hoit is not sure how to consolidate this and pull the resources to do this, but he feels that $2 million out of the $2.5 billion budget that this university has, is not a lot of money.
Budget: The university realizes that additional funding will have to be put into place in order to make this plan work. Then, over time, there may be savings. Next year we will have to come up with $6 million. The year after it may be less, or it may be the same. The year after that the savings might be starting to be realized, and over a 5-7 year period we might get down to where the $4.5 million shortfall gets erased, but the quality of service goes up. Maybe. Until we get the details, there is no real way of knowing.
Timeline: There is a proposed timeline which is very aggressive. Marc already knows that the timeline cannot be met because we are already into February and he cannot produce finalized budget plans by March. But those might be doable by the end of March or April--depending on how we do it. There will be implementation teams developed. There will be advisory structures developed. These are not set yet. There should be a technical implementation team for each of the services that will look across the services, to where they are being provided and decide the best mechanisms in light of industry standards. These groups would develop a full business plan and implementation plan for doing that on a technical level. There will be outreach groups that will work with faculty and staff to include them in these things, so that their needs are met. There will be higher-level groups which will oversee the entire process to see that we get input from all the affected groups. This whole structure is fluid right now. As soon as Marc finds out which groups are interested in starting early, they will be involved in the planning process and we will start moving forward.
Marc hopes, within a couple of weeks, after he starts talking to these groups, to find out from the deans and directors those who are interesting in being in the early adopter group. Marc said that IFAS is hopefully one of those who will be considered an early adopter and will be in on helping set the plans and directions and making this work.
Steve mentioned that when Marc met with us last, we were still waiting on the Provost position to be filled. At that time Marc had said that the attitude of the new provost to the structure of Marc's position--whether it was considered a director-level or top-level position--would greatly affect how things went ahead. Steve asked if anything had been decided on that topic. Marc responded that nothing has happened yet, but that this plan is being called the "CIO Plan"--meaning the university will have a CIO. Marc believes it safe to say that the university has bought into the fact that it needs one CIO who manages IT for the university. That does not mean that IFAS or HSC loses control--which is a key point. It does mean that there is a central point for keeping IT organized and for managing it, however.
Marc said that, as IT professionals, we all know that service, input and responsiveness to the needs of the users is the highest level of organization--and that is the quality of service which he mentioned. We have to develop structure so that the special needs of IFAS and the HSC can be met, while fitting within the structure. HR is worried about the same thing. They have to get payroll out. How do you make sure you get payroll out if they go to an IT committee saying we need more resources and the committee says we think IFAS's extension is more important, so now we aren't going to get payroll out. That is not going to work. Every group has this same need, that their services have to be met. In the past, the university hasn't had the funding to do that and Marc doesn't believe that people will magically come up with all the money we will need. There is going to have to be prioritization and decision making. The way we do that is going to have to be inclusive.
So, yes--there is a CIO model. Currently, as best Marc can tell, it would report to the Provost. UF has three Senior VP who have budget authority for the organizations beneath them. However, there is still a position called the Provost which is the main academic leader for the entire university. So it is the President, the Provost and the three Senior VPs. It happens that we have a Provost who is also a Senior VP. So currently, it is Marc's understanding that the position would report to the Provost and that all Senior VPs have bought into this model. The transition plan for how and when this happens is still up-in-the-air, so Marc could not tell us any of that. He does believe that he will be given the authority to begin to implement this plan. How that happens and what goes on after that we will have to wait to see.
Steve said that there are obviously a great number of details which need to be worked out, but he wondered who had been involved in developing the plan to its current state. Marc said this was a smaller group--mostly the IT directors--and his one-on-one discussions with other people. It does, however, use industry standards and practices. Marc said that a key thing to remember is that there is nothing here that is not obvious. We need some centralization in order to do a better job. It needs to be managed, and we need to organize it. E-mail is a commodity. We don't need 150 servers. It is the details that everyone is worried about and those are the things that are not formed. Management-wise, this is a fairly obvious structure. As a vision, Marc doesn't feel there is much controversy.
Steve mentioned that, in Wednesday's ITPAC meeting, Jan van der Aa expressed concern about the details of the governance model and how input would be obtained. Marc said that this is indeed a concern and that Marc currently doesn't have the answer. If everyone reports to a CIO, how do you make sure that the needs of HR, IFAS, HSC and everybody else are met. If they report through their own units, however, you don't have a centralized structure. Marc will be working with the VP of HR, Kyle Cavenaugh, and the other different groups and they are going to develop a structure which handles that. There has been talk of everything from dual reporting lines, which Marc does not particularly think will work, to having designated people who are responsible for handling the needs of each group, and there are discussions on a high-level board. The ITAC committee needs to be re-organized, but if you bring in one from every affected group, then you have a board of 50 that can't accomplish anything. So it is fluid currently and any suggestions on structure are being entertained.
Steve believes that the key to making any project work well is to get input from end users. This is very difficult to do, but one of the ways we try to do that within IFAS, with the ICC and ITPAC, is to make the process as transparent as possible. We try to make all the information, that we can, available to everyone who is interested. That is something that, traditionally, we have had problems with at the UF level. A good example might be the Active Directory ITAC sub-committee. We had actually offered to do streaming of their meetings for them and were declined. Getting information on what is going on so that people who are interested can see that and possibly get some input into it--through the proper channels, of course--Steve feels that this is key to getting something like this to work well.
Marc hoped that what we were seeing now, in him being here, will put a different face on how we are doing things. Marc has met with the Infrastructure Policy Council for the Faculty Senate. He has met with the Senior VPs and VPs. He is meeting with the deans on Tuesday and has met with ITAC-NI. A website is being set up where, as things get formed, more information will be posted. All the committee structures, the meeting dates, the minutes--all those things--will be placed there. There will be a central place where people will be able to track what is going on. Marc feels that the ITAC structure is a good basic structure; meaning that you have advisory groups and focus groups that are special to that service. In Marc's view, every service will have an advisory group. So if we have a networking and telecom group, there needs to be a diverse group which is there to provide advice, input and information to that group. Marc agrees that those structures are not in place yet, but they will be expanded and developed. Again, Marc is looking for input into what is the best way to do that. Marc feels that we have the equipment and there is no reason that we shouldn't stream anything and everything that is important. Marc related that, as we develop these things, it's all the past history that gets brought up way too often. Marc would like to put the history behind us. It is a new era with an open and inclusive project--and that is the only way that this is going to work.
When Steve suggested that this might be the triumph of hope over experience, Marc countered by saying that it has worked that way in everything which he has done. If you go back and look at the College of Engineering when he was chair, or the department when he was Associate Dean. That is the way he ran things. If you look at his short-term six month stint with PeopleSoft, that is the way that the part he was handling for student services was run. Marc feels that his history is a little different than the history of IT at UF.
Dennis Brown related that his faculty are scared of any kind of centralization. Dennis feels this is mainly due to their belief that someone is going to steal their research. Their reaction is to say that they are going to have to go back to saving everything on their own computer. Marc responded that there is indeed a huge trust gap at this university. Marc has had his own networks and has run his own file, web and mail servers. In his experience, central services have never been able to provide the quality of service necessary to do that. However, Marc also had lots of money from his research and could provide his own necessary level of services. That is not true generally across campus though. Marc would guess that most faculty saying they will do this for themselves are less secure than they would be if using a central service. He pointed out what happened at Tulane with the hurricane this last year. Nobody was prepared for that level of disaster and we are not prepared. Our disaster recovery plans are pitiful. This is a huge cost, but these steps can take us in a direction where we can have all these services properly backed up. If a web server goes down, we have alternate web servers and all the data is replicated somewhere else so that we have full backups. E-mail is a critical service that failed at Tulane. We have a backup system that runs in Atlanta across a 10GB connection. We are going to have hot backup available up there and eventually a mail server up there, so that if our mail server goes down--we may not be able to service the entire campus but--we will have a means of delivering important administrative messages. Marc feels we will be more secure, but that Dennis is correct in thinking we have something to prove. Marc mentioned that Service Level Agreements (SLAs) can set and enforce the expectations of the clients. This is not going to be haphazard, but will be run in a professional manner. We are going to prove that we can do this, and that is why we are going to start slowly.
Dennis was also concerned about the ability to locally control access in a centrally-run file sharing configuration. Marc said that they are working on a couple of things which tie this all together. Number one, they are working on an account management system whereby, when someone enters our organization, they get entered into PeopleSoft and that information is automatically transferred into Active Directory and their accounts get set up. We will still need people in the departments to assign access, however. In the support structure which gets developed there will still have to be people responsible for local groups to handle some of those issues. However, the collaboration servers automate and eliminate some of that setup. They will allow faculty members, for example, to set up and control their own collaborative environments. They can add documents and set up document aging so that those are available for review for a specified period of time. They can control what documents are readable and editable--all sorts of things that people would love to have, but don't even know exist. Some of the activities will be taken over centrally to assist in this--and some won't. The best way to manage all this is not yet known, and that is where the teams will come in to develop those processes. Marc encouraged Dennis to participate in those processes to be an advocate for his clients and to see that their needs are met. This is what Marc told the faculty senate. We are not going to have a faceless centralized system where we put everybody in one room and we all work remotely from terminals where you call us on the phone and will never see us. That is not the way to provide support. Marc stated that Dennis's concerns are real and we need to adapt and show that we can meet the needs of those people.
Dennis asked if Marc expects the people at the departmental level to be working for him or for the department. Marc answered that he did not know yet. The vision is that there are certain things that are commodity items that should be centrally handled and are not a problem. E-mail is probably the best example. On the other hand, if someone has specialty software which they use in teaching a course--that should not be done centrally. Thus, we need to differentiate between specialized needs and services, which are local, and the universal commoditized balanced systems. That dividing line is up for discussion. We might, for example, find out that there are 20 departments that want SQL server. Though not campus-wide, we might benefit from forming a unit that provides such services and have the departments which use it contribute extra money for that. We will always have that line to draw and it will be continually changing. Five years from now we will have high performance cluster (HPCs). Eventually, such things will be standardized and central.
Ben Beach asked about how Marc expects to handle extension, and Marc responded that he didn't know yet. That is where the people who are doing that are going to have to help, so we can find out what works. Marc doubts that we have more IT people than we need. We may have them in not-necessarily the right places or at the right levels in some instances. Perhaps we have a lot of temporary OPS people that we should consolidate into a full-time person. There is no way we are going to lay people off because Marc feels there is a glut of IT support personnel. Marc simply feels that we don't have it organized correctly.
Steve pointed out that one of the problems in IFAS in trying to move toward centralizing some of these things is that we have a diversity in terms of financial support among our units. We have units that can afford to do a better job than we can currently do centrally. Consequently, they are not about to reduce their services by joining up with something like that. Of course, those are the units that tend to have the most pull as well. Steve assumed that Marc would run into this campus-wide. Marc agreed and Steve asked how he planned to address that. Marc said that one way was that this is to be a gradual transition. It is not an 18 month plan, but rather a 5-10 year plan. The actual implementation time will depend on feedback from the stakeholders. In terms of the diversity of our units, those that are better off may end up having extra support people for their localized support services than is the standard around campus. The commodity items, however, should have a universal standard that is acceptable to all. Those with extra research funds will always have special needs and will use those to support those needs; this plan, however, will raise the baseline level of service up to a level of quality and consistency.
Chris Hughes asked if central IT was then going to provide the budget and support to assure that the SLAs are met. Marc responded that the administration of this relies on the ability to gather some of the resources. The funding issue is understood, but when you compare our situation to that of others, we already have comparable numbers of people and comparable budget. The question is more of how are those people used and where is that money spent. The administrative challenge is to gather the various funds spent here and there on OPS, etc. into a centralized group. Paying for things off the top will mean that individual budgets will necessarily decrease--but it doesn't mean that more money will necessarily be needed overall.
Steve asked about the third FTE for UFAD. Marc responded that a person was earmarked for that position, but they were on active duty which keeps getting extended. Marc has decided that we can't wait on that any more. Now the difficulty is, that was an already paid position; so now Marc has to find somebody else and because of all this restructuring, he doesn't want to incur new costs. What he is hoping is that, in this restructuring process, he can get that part done more quickly. He doesn't know where he is going to get that resource from, but it is his hope to fill the position.
Steve then asked if Marc was aware or involved at all in the BizTalk project which George Bryan is working on. Steve thought that this project might have a tremendous potential for UF. Marc talked briefly about Service Oriented Architecture (SOA), based on web services. This is generally done through a messaging hub,--BizTalk being one of those. A BizTalk hub has been purchased and will be used as part of the account management system. The hope is that directory services will be incorporated into SOA. From there we are looking to do that with anything and everything that makes sense. This is an example of what a centralized IT structure can do. We can plan things universally for the campus and we can provide things that previously we could only wish were available. The BizTalk server has been sized appropriately for the needs of all campus. It is close to being on-line. Once that happens, anyone who wants to start using the hub can begin to work on that.
Steve asked Marc if he knew what specific project was foreseen for BizTalk that convinced administration to invest such a considerable amount of money. Marc said that this all started before the plan he was presenting. He believed it was based on forward thinking rather than having a single project in mind, although the account management portion was a big issue which they hoped to address with it. Steve mentioned its potential for handling purchases and other financial dealings with outside vendors electronically. Marc agreed that the potential there was phenomenal.
Nancy Johnson asked about videoconferencing. Marc said that this was the exact model which shows that centralization can work. Marc's understanding is that it is now working as well or better than it ever has. We now have a dilemma, in that we have triple growth in use of this service--80% of which is IFAS. We are now at the point where we need more support and again our budgets don't allow us to hire new people. This is an issue that the consolidation might be able to address, by having other users of the system dedicate resources to central support. Marc agreed that videoconferencing is critical. It is on the list as part of the collaboration group.
Nancy asked about the quality of the service--especially out to the counties. Steve said that build-out is part of a second phase and that resources for that are mentioned in the 2005 taskforce report which Joe Spooner will speak of shortly. Steve feels that simple usability issues are his biggest concern--things as simple as knowing who is participating in a given meeting, or the interface used for scheduling events across multiple sites. There is also a capacity issue when one begins to think about in-service training events which could potentially go out to all Florida counties at one time. Marc said that there will always be issues and that if those are things which need to be addressed, we will have a user group and an administrative structure to say that this is important and that we need additional resources. Marc hopes that we will have the maintenance and upgrade costs budgeted into these services--but there will always be additional costs and special needs that must be addressed.
Chris Hughes asked if Marc thought units are really all that willing to give up people. Marc responded that, among those he has talked to, they feel that if they get the quality of service they require and it is revenue neutral (meaning no extra money has to be put in), then he has seen a good willingness to discuss it. Nobody has signed, however. Chris then asked if he didn't anticipate having to build it first before he could get adopters. Marc sure hoped not, because the resources do not exist to do that. However, the Senior VPs and VPs have listened and endorsed it. We now have to take it out to the deans and departments. Marc thinks he has seen support necessary to get at least one piece of this started and going; maybe we will have to get the mail system going first to prove that this model can work. We just have to see who is willing to buy in. Marc added that the plan is to get implementation teams going and start the work. Even if everything fell apart, we would only be back in the same spot where we started. Nothing is being dismantled, so if the project takes an extra month longer than expected, things won't break.
Marc made one last request that we all participate. If the only way you feel comfortable in doing that is to send a comment directly to him, then please do so. As these groups get set up, be sure to watch it and be sure to get your comments in. It is going to take all of our efforts to help do this. Your input will be listened to and evaluated. The decisions made may not be the ones that each of you think best, but your input will be listened to and you should be part of it.
Dan Cromer expressed his appreciation for Marc coming and sharing with us. Dan appreciates the open approach and certainly encourages that. Dan took a "soapbox" moment to say that he feels we in IFAS have gone through this in 1997 and in 2005 with the taskforce report and that he believes we have a good methodology and process for going through this with the ICC and ITPAC. Dan would be glad to share our experience. Dan also related that he had sat in on meetings with two Senior VPs, the HSC and Dr. Cheek, and they/we are supporting the plan. The Senior VPs support the concept of a CIO for UF. The details are still to be worked out, but we do endorse the vision.
Marc responded that Dan's support and input from all the folks here would be both necessary and welcomed. Dan encouraged the ICC to have some blunt discussions about this matter. Dan believes it is a good goal, but we need to develop the details of how we can get there while seeing that we don't lose any service or support--and possibly improve things. Dan is excited about the possibilities and the challenges and sees the people in the ICC as playing a key role in the process.
Report from the February ITPAC meeting
Steve related how ITPAC had reviewed the 2005 taskforce report and recommended that Pete Vergot request an audience with the deans (who meet every 2-3 weeks) to present the report to them. Joe Joyce had wanted to begin discussion of the report in the ICC because we want to figure out what can be implemented and how. The ITPAC is going to be the shepherd of this implementation process, so that group will have a lot of say, and the ICC will have input through that process as well in regards to how that is done. We need to get the support of the deans in concept and, of course, there will be many budget issues to discuss there.
Steve then turned the discussion at this point over to Joe Spooner, who was a key member of the 2005 IT Taskforce. Joe told us that the report is all but done. The report include a good 18 page overview of what the committee's findings and recommendations are.
Recommendations on Administrative Structure: The first recommendation is to review the IT structure of IFAS. We are looking to improve the efficiency, security and integrity of our systems. Of the action items specified on this issue, Joe believes #2, "Implement an industry standard IT governance framework to improve overall IT operations, policies, and procedures" could help us greatly with making changes. Joe feels that expediting changes is very important and pointed out that many of the 1997 report recommendations were never fulfilled. Joe also noted action item #7 which specified an increase in IT personnel for supporting the extension districts--there are budget issues involved there.
Joe noted that all the action items fall into three categories: those which can be done with little or no cost simply based on people agreeing, those which have some costs that may be addressed by shuffling priorities, and finally those items which are quite expensive and will be a huge matter for discussion. Obviously, some of these can be implemented more easily and quickly than others.
There is a recommendation (#9) for a Microsoft TAM. We need to look on whether it might be better to contribute to that centrally, but this level of support is definitely needed.
#12 fits in closely with our discussions today with Marc Hoit and is a particularly timely recommendation, perhaps. By "Divest from services that can be adequately provided by the University of Florida" we can focus on the core competencies of IFAS.
Recommendations on an IT Software Development Team: The recommendation here is to revitalize the software development team. Joe has seen no formal backing for this and doesn't feel that this particular approach has been well developed yet. He does feel, however, that this is a good start in recognizing that we do need to address this area. A lot of discussions will be needed here before anything should be done. Steve pointed out that currently, if someone needs an application developed, there is simply no where to go. Joe said that the problem goes beyond that there is no place to go, but that there is no process or methodology available in which such a service could be handled in a timely fashion. Joe knows there is a great need on the faculty side for applications--especially for simulation applications. There is also an even greater need on the server side for automation of servers and services and the system with which they coordinate. There are huge potentials for gains in efficiencies there if you were to automate some of the things that go on with the web and file and print services--even collaboration. Joe been impressed with what he sees coming in Sharepoint down-the-road and believes it has great potential for us.
Chris Hughes asked about which of these recommendations came from the users and which were simply a product of the committee itself. Joe said that these were not labeled as such, but thought it was a great question. Chris pointed out that most of the report was done before the survey even went out. In fact, the survey results are not included in the appendices nor are they available on the taskforce site.
Joe stated that the committee had felt rushed about completing the report and that led to some of these happenings. Joe mentioned that one clear goal of the group was to associate recommendations with action items and that those actions items have to related to a plan that can be executed and accounted for based on prioritization. That prioritization is coming from the ITPAC. Hopefully, the ITPAC or some sub-committee of the ITPAC will look through all these recommendations, prioritize them according to what can be achieved and what is important.
Steve related that he is not clear on how the ITPAC is going to work on that, but noted that ITPAC does have a new chairman now. Pete Vergot, who lead the taskforce, has agreed to chair ITPAC, replacing Jim Syvertsen.
Recommendations on a Recurring Funding Model: Joe said there is a strong belief that we should continue leasing hardware. It is a more cost effective model because it forces administration to deal with the fact of recurring costs in IT. The biggest challenge is to determine where the money will come to fund all these things. Right now, we don't know.
Steve relayed Joe Joyce's conviction that support must come from the grass-roots. The role that each of us can play is huge in this, because if we get these IT issues being spoken about by our chairs and faculty, and they talk to the deans and vice versa--we can get these subjects being talked about all across the board. That is how we can get these things to happen. Otherwise it just goes under their radar and they don't even think about these things. Consequently, Dr. Joyce saw that our role could be very important in all this.
Marc mentioned that our strategic plan could fit well with his plans if we could work together on that. Joe Spooner feels that the biggest problem right now is not the plans or the people, but rather the funding model. Joe knows we have all experienced "spending emergencies" where funds suddenly become available at the unit level and must be spent immediately. What happens under that model is that resources are quickly spent at the unit level without considering the possible efficiencies involved in a centralization of those resources.
Recommendations on Bringing People Together with Technology: This recommendation focuses on the collaboration pieces that surround videoconferencing. Two items of particular note are the IFAS web-based calendar system and a system for on-line registration and credit card payment. We are looking for a website where videoconferences can be scheduled, and e-mails get sent to all the involved parties. Marc pointed out that, since videoconferencing is more centralized, that this is something we need to coordinate at a higher level than just IFAS. Regarding the credit card system, Pete Vergot is particularly interested in getting the capability of easily setting up conferences via the web with an on-line credit card payment system. Marc noted that we have a Department of Continuing Education that handled such thing currently. Why are we collaborating with them to do that? Certainly they would take some overhead, but what is the overhead to handle this yourselves? Marc's biggest hope is that we can get the lines of communication open so we can stop re-inventing the wheel, but rather leverage off existing services. Joe agrees that having processes in place for addressing recommendations could assure that we don't rush out and implement such things without investigating the collaborative possibilities.
Recommendations on Web Management: Joe said that much of this is already in place led by Ligia Ortega. Right now they are focusing on releasing Solutions For Your Life. That will be their primary focus until April or May. During that timeframe they are also going to release a few web site templates for IFAS to begin using. Hopefully those will be nicely standardized and compliant as we would like them to be. Marc again said that he would appreciate our cooperation on that, because this is a problem that exists on campus. If we have found a solution, he would encourage us to bring that back to him so he can see how we might improve campus.
An item here that Joe has been trying to gain attention for is the idea of content management for our web sites. Joe would really like to identify a system we could use for that, one which integrated with the other things we use such as MS Office, and something that is fairly open and not restricted.
Finally, Joe mentioned the need to deliver on-line non-formal education. There are already a couple of systems on campus that might address this.
Joe pointed out the listing of budget items which are associated with the various recommendations and action items. Attaching budget figures to these items was another major goal of the committee and the end result was that the recommendations would require almost a doubling of the current recurring budget. Marc suggested that we might be able to leverage most of that in the plans which he is proposing. He believes most of the recommendations fit very tightly with what we hope to do at the UF level.
Joe encouraged ICCers to take a close look at the report and provide your feedback directly to Pete Vergot.
Chris Leopold briefed us all on the plans for switch deployment out in the counties. He will be leaving this meeting to go to the Livestock Pavilion and meet with the district support people to deliver that equipment for dispersal. Chris had configured 54 (2 are spare) HP switches with non-site specific information and included a chart identifying proposed installers for each location. The switch installation check sheet includes:
- Note the switch serial #, building # and room #
- Replace old equipment with new equipment
- If it is an unmanaged hub, discard
- If it is a HP switch, return to IT/SA
- If it is a non-HP switch, keep for possible redeployment on site
- Install new UPS and plug in the switch and the site's MPS
- Don't forget to connect the USB shutdown cable to the MPS
- Make sure all equipment is in the battery backup side of the UPS
- If required, install GBIC in port 26 or 50
- Replace the fiber jumper going to the LIU
- Insert patch cables using new standard
- Uplink ports use 49/50 or 25/26
- Router, DC, MPS and Polycom are ports 1, 2, 3, and 4 respectively
- Client ports are 48 or 24 working numerically downward
- Test network connectivity
- If you really get stuck call Chris Leopold
- E-mail switch serial #, building # and room # to Chris Leopold
- Take a photo of ecah location for documentation purposes
- Place photos in "if-srv-file01\it\Remote Site Pics"
Unfortunately, the hard drive deployment to the multi-purpose servers could not be scheduled to coincide with the switch deployment. This means we will be needing a separate deployment schedule for that purpose.
Tom Barnash asked about port labeling. Chris explained that sites are currently labeled and the small size of most of these facilities did not necessitate that. Marc Hoit asked about a naming structure for the switch location information (building and room). Chris agreed that he would add that in. Chris intends to have all the work completed by the end of the month. There is some layer one work that needs to be done. He and Ben will address that, specifically in the St. Lucie and Indian River county extension offices.
Exit processes, NMB and permission removal
Prior exit procedure discussion. This topic was raised at ITPAC. Dan Cromer is working on this with Mary Anne Gularte, Director of the Office of IFAS Human Resources. The problem is that we have accounts now that are university wide, and when somebody moves outside of IFAS, they still have access to their account and to whatever resources they had been given. IT staff is not tied-in to the exit procedures sufficiently so they we are informed as people leave and can handle removing their access to various computer resources. Although Chris Hughes has created some excellent tools to help us locate and remove those permissions, we still need to be connected to the exit procedures. This is being worked on and hopefully we will have some improvement in that shortly.
Trailer for IFAS-ALL-L explaining the nature of the list
Although the agenda does not reflect it, this item has already been addressed. There is now a footer attached to IFAS-ALL messages as follows:
***** IMPORTANT NOTE about this e-mail distribution list: *****
This e-mail distribution list is intended for official business purposes for
IFAS faculty and staff. Do not use this list for personal messages. If you
are a current IFAS employee, you will NOT be removed from the list. If you
are no longer employed by IFAS and wish to be removed contact the Helpdesk
at (352) 392-4636 or mailto: email@example.com.
Right now, Dwight Jesseman and Dean Delker are moderating the list. This is not appropriate, but it is a short term measure until the recent fuss subsides. If the list needs to be monitored, it should be monitored by someone in administration, because this is not a technical issue. Joe mentioned that another method of dealing with parts of this issue is to create a separate opt-in list for announcements such as plant and peanut sales; this is how such things are handled at the HSC.
Chris Hughes attended the air lift but did not feel that the training there was very impressive. It did highlight that SMS is going to take a much bigger role in deployment in the future. SMS 2003 will be the version that is used when Vista comes out. Consequently we are looking at deploying SMS. We have purchased the license and plan to deploy it in the next couple of months--starting with the servers. Chris is looking for units interested in early adoption of that, as we may have some test units which receive some benefits from that process. There will be SMS 2003 SP2 along with WDS on our DHCP server that will be able to deploy Vista. Chris intends to have the next beta up on the WDS server. RIS will be scrapped because it is being replaced by WDS.
Steve pointed that that documentation on his experiences with the Vista TAP are available (ufad\IF-ADMN credentials required for access). Steve encourages people to try and find the time to get involved in this and the documentation is there to hopefully make that easier. There are scheduled web-based sessions every two weeks on Tuesdays that provide information on Vista that are worth looking into. So far they have been very general, but those should become more detailed as we progress.
Removal of WINS
We had planned to remove WINS the first of February, but Chris Hughes discovered that a number of login scripts still used old netbios names. Chris assumes that other links like printers and drive mappings to those also exist elsewhere. Chris is going to work on a script that will dump all current mappings and he will send that to the OU Admins for correction before we move on this.
Listserv confirm settings
This project is underway. We have to come up with some documentation and then get out notification before doing that, but the project has been approved for implementation.
New File and SQL Servers
They have been requested to work on the SQL server first, but Chris imagines these will both come along quickly. We will be moving a couple of applications to the new 64-bit SQL server 2005 Std. Version shortly, including LiveStats.
Remedy Project Update
Steve reported that we had a meeting on the Remedy system with Dwight, Dan Halsey, Chris Hughes, Fran McDonell, Adam Bellaire and himself. As we had discussed at the last ICC meeting, what we believe is the last step before announcing the availability of this system to all of IFAS, involves creation of a means for OU Admins to control ticket assignment and e-mail notification for tickets which are generated by individuals within their respective OUs. There will be a web site created that OU Admins may go to for controlling this. If there is more than one admin within your OU, you will be able to select a single person to receive ticket assignments--or you can choose to have those go to the Help Desk. Those who elect to receive assignment will also receive e-mail notifications. If you are not having tickets automatically assigned to you, you can choose whether or not to receive just e-mail notifications. This will allow OU Admins flexibility in how they participate. That is being worked on and we expect that to be ready in maybe a month or so.
Proposal to temporarily delete all users’ Outlook.NK2 files via login script
This is a cache file (ufad\IF-ADMN credentials required) that manages the "Intellisense" feature of Outlook. Dwight had proposed deleting this file for everyone via login script but the feedback he received was not too favorable. Joe Spooner related that there are other issues with this feature. First of all, there is a 1000 name limit on that cache and that cannot be modified programmatically. Secondly, suppose you have a group in your contact list which you send to enough to make it into that n2k list. When you then change that group in your contact list, it doesn't delete out of your nk2 cache--though you may do so manually. Unfortunately, the file format is proprietary, so there is no scripting solution to alleviate this problem. Steve mentioned that he has stopped running Outlook in cached mode due to issues with the off-line address book. He stated that he is a late Outlook adopter who is still moving people over to that platform. In doing so, he has to be rather careful to avoid mentioning a number of these little problems if he is going to get people to buy-in to that new platform.
Update: User notification on forced reboots
We had left this issue with the idea that such events which occurred outside of the normal 2nd Tuesday window would be accompanied by notification to the ICC. Steve noted that there were a number of folks who had felt that notification should be to all. If anyone wishes to debate this point via the ICC-L, such a discussion would be welcomed.
Steve let everyone know that next month we will be having Mike Conlon here at our meeting; so please be thinking of questions and concerns which we may present at that time.
The meeting was adjourned early at approximately 11:50.