Interviewer: First if you would just talk a little bit, each of you, about where you are from and a little bit of your career background and what lead you to the National Weather Service and AWIPS if you would? Ronla if you want to go first…
Ronla Henry: I can go first. I am originally from Roanoke, Virginia. I am your traditional Meteorologist. I have been interested in weather ever since I was a little tiny girl. I can remember when I was younger begging my parents to get home so we could go watch the weather report and know what that the forecast was, and going outside and observing the weather and just, kind of the traditional stuff you hear from a lot of Meteorologists. I announced in Middle school to my parents that I wanted to be a Meteorologist and my parents looked at me and said 'A what?' So I went to the University of North Carolina in Asheville for my undergraduate degree. I have a Bachelor’s in Atmospheric Science and then attended the University of Maryland where I earned a Masters in Meteorology. When I was up here, at University of Maryland, I co-oped in what is now OCWWS and got my foot in the door of the Weather Service that way. I worked for ten plus years down in MDL developing software which is on AWIPS and then went up to SEC for about a two year period to manage projects on AWIPS and now am in the Deputy Program Manager role. So I have been with the Weather Service and now my entire career which is approaching 18 years
Ronla Henry and John Vogel meet to discuss AWIPS II.
Off the record comments. John?
John Vogel: Actually I go back in weather a long ways although I don’t recall any particular interest in the weather as a kid I see it seems like that’s a really common thread. She must have read Don's interview. When I came into the Air Force many, many years ago, I was given a test and they gave me some different carrier fields that I might be interested in, and weather was one of those and off I went. As soon as I got involved, I did get deeply involved in it. But even when I was in the Air force doing DOD Weather, even in my first years, I was working with National Weather Service on severe weather stuff. We were collaborating with the radars of the times. They were a far cry from what we have now [Indiscernible] [00:02:47].
After a full career year in the Air Force, all in weather, I went to work still supporting DOD as a contractor with one of the local companies and actually it's headquartered in San Diego but big portion of it is here, and I did a fairly long stint at that. And the last job that I had was moving away from DOD support and supporting National Weather Service at the test facility out in Sterling and I ran the contract shop out there. Folks out there worked for me and I started working with some of the people here in the Headquarters and became more, and more interested in what the National Weather Service was doing. Particularly in AWIPS because it is clearly the premier program in National Weather Service at this time. When this job opportunity came up I applied for it and luckily I was selected and I have been a National Weather Service employee for three months. My career with National Weather Service is new, although I have been working in weather longer than Ronla has been alive, although I won't go back that far.
Interviewer: Well, having a fresh point of view is always a good thing too as far as Weather Service operations go. You people are the head of the program management office for AWIPS. Could you describe what your role is in development of AWIPS II?
John Vogel: Well we’re the Program Mangers for AWIPS. AWIPS II is a project within AWIPS, so it is an enhancement, a very significant enhancement. I think I’ll let Ronla talk about some of the big changes that we are seeing there. We went basically from the original AWIPS, a kludge of stove pipe programs. There was a lot of redundancy in the way it was put together. A number of things had to be done many different times, many different ways. And under this new services-oriented architecture, which is AWIPS II, they built the foundation first and then these applications will come in and access common databases within the system it will become much smaller, much more nimble, and much more expandable for future growth. I think Ronla can probably explain it technically much better than do.
Ronla Henry: Basically that is the concept. Back when we first re-did the contract for AWIPS it was recognized that the software, because it was stove piped, it was taking a lot of time, O&M time, and cost and money to be able to fix things and inject new science and technology and all that stuff. And it was based on, basically1990's technologies, so times have moved on, so it was an opportunity to kind of get modern technology, modern techniques into the system and to be able to make it, they call it the "ilities"--adaptability, the flexibility, all that type of stuff to incorporate that so that we can with long-term goals can make the transition of science into operations easier and work with our partners, either in Universities or other federal labs, to get better science and technology to the forecaster.
Interviewer: It sounds like this from what you're saying, the first time around AWIPS was sort of stuck together with things that we had and it was sort of built around adapting to existing programs and not thinking so much about what the forecaster needs. But it sounds like the new version is going to shape things more based on what the end result should be rather than having to accommodate all these existing things.
Ronla Henry: Yes. Probably a way to say it is that we’re going to be able, by using the common databases and the common, what they call services--the data services, dissemination and display services, ultimately we’ll be able to give a better look and feel that is more consistent across applications instead of a collection of individual applications. You won't be going, where is this button? It’ll be a more common look and feel and we will also be able to increase the raw abilities in response to the requirements that we get, for where our mission needs to go and what products and services we need to provide as the weather service. It hopefully sets us up to be able to be little more responsive in that area.
John Vogel: A common day to day example would be Microsoft Office. It consists of a number of independent programs, Microsoft Word, PowerPoint, Excel, and others. Those all started out as standalone programs and when Office was first hammered together there was a lot of – it was very bloated program and there was a lot of redundancy. If you had a list of names or something you wanted, you might have to triplicate it because the different programs have their own way of getting access to them. So as time has gone on, this has become simplified. As Ronla says, the way you access these things, the same keystrokes get you to the same kind of information. You don’t have to have all different sets of keystrokes for each different program it has.
In some ways that is the way we are going. We have these different applications that do different things, maybe culminating in a severe weather warning or something. One of the things that it may need in the process is calculating relative humidities, for example. Well there may be another program that doesn't have anything to do with severe weather but it also needs relative humidities, so it goes and it calculates them also. So instead of having each application do that, we will have this database of relative humidities. That is probably not the greatest example, but it’s what comes to mind that is the difference in services-oriented architecture.
Interviewer: How would you characterize the progress of AWIPS II so far? Are we on schedule? Are things going okay generally?
John Vogel: That’s a tough question. I’ll let Ronla address that! First of all let me say that the simple answer is 'Yes'. We have been marching along with specific task orders that our contractor Raytheon has been responding to. It is difficult to measure exactly where we are because these functionalities that we had before--in the new system they don’t work yet because we have to build all these layers of this onion first and then we start putting the functions in. And those are the ones that are starting to come on now, so all this other stuff is being built and you have to look at this big picture and say, okay, all these functionalities and these layers are there in place, this is good solid architecture so I am in good shape.
What we have done more recently, there has been other Red team activity between Raytheon and National Weather Service. We’ve looked to see where we could make improvements in guaranteeing that the system is going to be ready and we are going to add more testing, and it could possibly extend the time. But we have not come to that part yet where we have looked at and actually how this going to impact the schedule. We are going to reduce risk by additional testing and I think I feel very comfortable that this is a sound program and it is going to be great enhancement to the National Weather Service.
Ronla: I don't know that have anything to add to that this once.
Interviewer: As deployment moves along, how will your roles change? When AWIPS begins to be deployed what will your roles be then?
Ronla: In a lot of ways it continues, because we have responsibility across the entire program and there are lots of little projects. Some are more little than others. But there are lots of little projects going on at one time that we had to kind of keep in constant balance for costs and schedule and basically the whole triple constraint that you care about when you doing Program Management. My particular role, because I do have a product improvement lead, I will start changing gears in the next little bit and start looking out at what are we going to be adding? And how are we going to be taking this new architecture and adapting it for future requirements, and what is the next big thing on the horizon that we need to incorporate and deal with in the program. So my focus will change a little bit away from just getting AWIPS II out into the field and deploying operation, to what are the next enhancements that we are going to be adding from a system point of view, from an architecture, infrastructure point of view so....
John Vogel: I think Ronla will also be able to, kind of ease back into a nine or ten hour day.
Ronla: (Laughing) That would be nice!
Interviewer: How do you plan to resolve potential problems that come up with AWIPS II?
John Vogel: Well of course to resolve a problem we have to identify it first. I think that is the kind of stuff that we want to do, and we do as much risk mitigation as we can. We get a lot of help on risk mitigation from our engineering partners downstairs. And it is additional testing, and planing, and experience with other programs that tells us the kinds of things we have to, which mole we want to knock down first before another one pops back up again.
Ronla Henry: Using the contract to our advantage, working with the contractor in a partnership mode to identify and address problems and mitigate the risks. They are a key component in this and we wouldn’t have AWIPS II without the contractor, so they have to be a key partner in going through addressing these things and dealing with any conflicts or any problems that arise as a result.
John Vogel: Furthermore we spend a lot of time with them. We have several meetings a week, every week. Some of them are small sized meetings at the program management level, Ronla and I the representatives from Raytheon, to larger representation that we have over here, where they bring their folks over. Standard formalized briefings that they present to us.
And one other thing that Ronla is doing with this particular task order is that we are doing these technical interchange meetings where we are having our subject matter experts sit down next to the Raytheon developers up in Omaha. Because these guys can see something on paper but they’re not meteorologists. They're Raytheon developers. So we need these meeting in order to make that code get into something that's workable. Ronla is probably thinking of buying a timeshare up in Omaha. [Laughter.] She's been out there three weeks out of the last five, or something like that.
Interviewer: Isn't there forecast office staff out in Omaha that are working with this as well?
Ronla Henry: There is a Valley Nebraska forecast office that is basically just a few minutes down the street from Raytheon, so they've got a partnership worked out between them. The forecasters will come over and test the software and the developers will go over and observe the forecasters using the existing AWIPS. Because the initial goal of AWIPS II was to basically replicate, meet, or exceed what we have in AWIPS currently. So to be able to observe the forecaster and how they are using the system and in operations, gives the developers valuable insight.
And that expands into the next thing. Because in AWIPS, ultimately our customer is the forecaster on the shift, 24x7, all the time, they are key to identifying problems and issues with the software that will impact their ability to get it into operations. So, we've reached out to the regions and the various forecasters for testing support, evaluation support, the subject matter experts support, so that we can engage them as a stakeholder into this whole project.
Interviewer: Right. Earlier.
Interviewee: So they are constantly involved in they’re in a lot of our meetings as well.
Interviewer: Could you talk about some of the costs benefits of AWIPS II that you see, that we may see coming along? I understand that probably open standards is part of that right?
Ronla Henry: There is several, most of them are from a technical point of view. Ease of maintenance is one. Actually one of the main driving things was to drive down some of maintenance costs. In fact, a good part of this project is being paid for by savings in maintenance costs. So it's by consolidating the baseline code so you don’t have redundancy and making it a simpler code base, the costs of maintaining it are cheaper going forward, so then that allows us to recover some funds to apply to other projects and do other things with it.
So that’s one thing. One of the big benefits which we hope to realize is the transition from science to operations. That has always been a big issue, taking upwards of 18 plus months to get a new requirement into this system, just from the build cycle and validating the requirement and going through the development and the build cycle and deployment and all that. We’re hoping to compress that time down significantly, so we can be more responsive to the requirements that are coming out for new products and services. Then there is a lot of efficiency to be gained in data distribution and the way we just communicate amongst ourselves with the data. So those are the big areas for some of the cost benefits.
John Vogel: There is a great potential for some hardware savings as well. Because as we reduce the code it is going to take fewer boxes to run this and we have already found that, through smart ways of doing business and consolidating, Raytheon was able to save us, most recently, I believe, on a switch upgrade, by combining certain things we were able to realize a $2 million savings. Which rolls into additional hardware refresh, stuff like that. It wasn’t that Ronla and I got a big bonus. But it does mean that we are able to refresh things and get newer, faster machines out there a little sooner. I give Raytheon a lot of credit for that kind of innovative thinking that gets us there.
Ronla Henry: Right. And to do the – to write the software the architecture and stuff like that, just like John says, it's going to allow us to be more flexible in what our hardware footprint looks like and how we can adjust and adapt to that. So it does potentially have some serious cost savings.
John Vogel: Smaller footprint.
Ronla Henry: Smaller footprint, easier to maintain that type of thing.
Interviewer: Well that’s all I really had to ask you here. Is there anything else you guys want to talk about particular points you want to make sure that come out in the article.
John Vogel: I want to make sure that everyone understands that we look up to Jack Hayes as our mentor and he is our great leader.
Actually you better not put that in because he might retaliate. When I was at DOD, Jack and I worked together, and he punishes me at every opportunity as well.
Ronla Henry: Or even worse, Don may say, hmm, wait a minute. I’m going to punish you guys too.
John Vogel: I was on the teleconference with the training guys and he was out there visiting and he jumped in and he was telling these guys about some old nickname that I had. So nothing is sacred to him.
Interviewer: One of new big themes Weather Service has for the future is decision support services. And it sounds like AWIPS II is going to be a major player in that as we move along.
John Vogel: I think surprisingly so. We really didn’t get into that because that started to get into another area of where we’re going. But whenever we are talking about AWIPS II and this services-oriented architecture, we are taking about the 4-D cube, and we are talking being able to put up a standard product. Any real customer or stakeholder out there can apply his resources too.
NOAA has been interested actually in using this particular architecture again in other areas as well. So Mary Glackin is going to be looking at demo tomorrow of AWIPS II over at the Raytheon complex with Jack so….
As you know, of course, in her earlier career Mary was AWIPS Program Manager, so it is something she holds near and dear to her heart.
Interviewer: So to have her there in NOAA Headquarters these days has to be great for the National Weather Service.
John Vogel: Well, it gives us perhaps more visibility than we can stand (laughter), people up the food chain interested in your program.
Interviewer: Sure. But it doesn't take long to bring her up to speed.
John Vogel: That’s right. She knows how to spell AWIPS, that’s for sure.
Ronla Henry: I am a big operations focus person, so an overall arching theme is that we do projects, pick things, and plan them so that we can give better quality and service to the forecaster, to be able to support them doing their jobs. I think that we truly see AWIPS II taking a huge step in that direction. We will set ourselves up that we are able to provide tools to the forecasters ultimately that will help them and improve their operations.