An Interview with Jerrie Stewart, TPF Product Line Manager
by Jeff Robinson
Background: With IBM for over 20 years, Jerrie has a breadth of experience, including programming, management, and strategic planning. She has a technical background in distributed systems, systems and application development process, and emerging technologies. She was most recently the S/390 Application Technologies Manager responsible for enabling application development for S/390 platforms through tools and APIs; her primary focus was the delivery of object technology for MVS in support of the IBM Object Technology strategy.
Since joining the TPF community, Jerrie has focused on a TPF strategy and plan that will bring the best technology available to address the requirements of the TPF customers. She has traveled extensively and actively participated in the TPF and TPF/MVS users groups to better understand the TPF customers' businesses and their information technology needs. After her first year, she is confident the TPF strategy and plans are on course to meet the customers' needs. Recently, I sat down with Jerrie and interviewed her for ACP/TPF Today.
JR: I know you do a lot of traveling, can you tell us where you're going next?
JS: We compact a lot of the customer visits so we can get the economies of scale. So I may take a tour of the East Coast of the U.S. and hit as many of the customers in that area as I can, then connect a few on the West Coast; those are probably due again in the June/July timeframe like I did last year. In early fall I usually do a tour through Europe -- hitting a city a day almost. In between these tours are usually the user group meetings, so I get to meet some of the customers during that time period -- especially those I may not have been able to book when I was making a tour through.
JR: How much of your time do you spend in travel?
JS: Between a month to two months a year and that doesn't include the user group meetings. In February/early March I just did the Southeast Asia tour which was probably the most exciting for me given I hadn't been through Southeast Asia that much.
JR: What did you think of the recent TPF User's Group meeting in San Francisco?
JS: I'm getting a little more comfortable now; After I've met a lot of people in the TPF community, its a little easier for me to communicate with them so I'm finding the meetings a little more exciting. I probably challenged them a little bit because I was speaking up in some of the sessions; but I don't mind participating -- it gives me a chance to really understand what they meant by something or to get some more feedback. I guess one of the things I was really struck by this user group meeting was the difference in tone with what's important to some of them. I think even the users are starting to think about TPF in their business and what else it needs to do for them as opposed to just the traditional "lets tune and tweak the platform so that it performs better or we have more coverage tools". Don't get me wrong, those things are still important but I think they're broadening their scope. They know in order to develop new applications for TPF they need better tools, more standard APIs, better connectivity with other platforms in their environment. And so, since that's been the primary focus for me in the last year, its kind of good to get them realizing these things are important to TPF and their environment. It also helps validate the strategy we've putting together to meet those needs and its nice to get that rapport back from them.
JR: With all the hype over the Internet and the World Wide Web, does TPF have a place in that new technology?
JS: Definitely; when you look at the fastest growing industry in the World Wide Web today, it is the travel industry. And one of the key components in providing information technology to the travel industry is TPF. So we're finding a lot of our customers though TPF is not directly attached to the Internet -- are literally using their TPF systems in support of their interaction of providing services and goods on the Internet. Our customers are now starting to advertise and tout the fact that they are able to do work on the Internet. They may not always say that "TPF is that big, humming system in the background", but it's not too hard to understand that that's where they're coming from. In IBM's recent annual report, I thought it was kind of interesting that of all the ads that they show, there is a nice big two-page ad about JAL (Japan Airlines) using the Internet for their business. And they mention the reservation system in the background which they use in supplying Internet access.
JR: ...and JAL's is a TPF-based reservation system...
JS: Oh, definitely. So, its kind of interesting that they're finding ways both through our help and on their own to attach their systems to the Internet. We have a lot of work underway in the lab to try to help facilitate that even more and in some cases even provide server capability for TPF so that if you wanted to directly attach [to the Internet] we would be trying to make that happen.
JR: Industry insiders and outsiders have "prophesied" for years that TPF will "go away' someday. Where do you see TPF as being five years from now?
JS: I guess when I first came here I was really struck by the fact that there were those rumors still around. And in talking with some of the customers they've all told me that part of their strategy is to reassess the role of TPF; that doesn't necessarily tell me its going away it says they may change its role over time. I consider it my job to guarantee that TPF continues to have a role. To be quite honest, in the next five years there is not anything that can do the work that TPF does for them today. But during that next five years, it's my job to make sure we do all the things that make TPF a better system to play in their arena so that they don't have to worry about what happens to TPF ten years from now.
JR: So, in ten years you definitely think TPF will be around?
JS: Oh, definitely!
JR: A lot of TPF programmers are concerned about the future of their careers in this very competitive and ever-changing industry. Do you think they should be concerned?
JS: I don't know if concerned is the right word; I think they should feel challenged a little bit. Because it's a small community, they need to pick and choose very seriously what it is that they end up doing. I've met quite few that aren't here in the lab that are part of the TPF community and know that they have skills and talents that aren't duplicated anywhere. It is a finite group. I worry that its finite because we have so much work going on and so much to do that I'm worried we don't have enough TPF skills to do it. Part of what we're doing in the lab is trying to expand that community by allowing for applications to be developed in "C" to provide better and easier tools for development so that the skill is not just the TPF Assembler skill anymore. So that if someone has "C" language experience they could get proficient with TPF especially with the new APIs we're putting on things like MQ, TCP/IP and those standard things they should be used to and familiar with. So I see the skill base growing if we do all the right things. And given that the system [TPF] really isn't going to go away I'd like to see the community continue to grow.
JR: Are you also saying that maybe they should branch out and learn other things like Java...?
JS:...And "C" and be proficient in technologies like TCP/IP,MQ Series, and maybe even start studying a little Object Technology and get as familiar with the Internet world as they can because I don't see that one going away in five years either.
JR: I know there have been a few new TPF/ALCS licenses recently; Are you excited about that and do you think there'll be more?
JS: Yes, I think its kind of interesting we've been able to position that part of our product family as well. Its primarily been in the financial community because applications exists there and those customers wanted to capitalize on those applications. But I think it just goes to show that from a TPF product line point-of-view, it still brings considerable value to the table either in applications that already exists or some skills that already exists. So, that if you have a problem that can easily be solved through an ALCS or TPF set of APIs that you can still do that and have a good business solution as a result. I don't see the base growing dramatically but I think its been one that's steadily building. Everyone once in a while we may migrate someone from one platform to the other (.i.e. from TPF to ALCS, etc.) or we'll see them do consolidation of platforms. But the number of customers seems to be increasing at a steady pace -- not leaps and bounds -- but at a steady pace with having TPF's family of products solving some business problem. Now, hopefully, one of the new things we're doing to the platform that ALCS can take advantage of -- especially in the new OS/390 arena-- is with parallel sysplex. [With that] I could see the base growing even more once we get customers to those latest technologies.
JR: What do you have to say to those TPF shops that are looking for alternatives -- even while admitting that there's nothing currently out there that comes close to its performance?
JS: I think its good business sense to look at alternatives all the time; if nothing else it validates that maybe what you're doing is right. But any IT (Information Technology) executive should be always questioning because of the rapid change of technology in our industry. So I hope that's exactly what they're doing. I hope that the technology we add to TPF keeps TPF as a valid player in those IT shops where I want them to be always challenging and looking. It also helps me prioritize which things are most important to them -- knowing what they're looking for out there in the future and how I can then supply that to meet their business needs. So hopefully they don't have to worry about where they go; but what they're asking for is what solves their business problems and not whether TPF solves their business problems.
JR: Will TPF ever become as popular as a product like Java?
JS: There's a slight difference in TPF in Java; Java is a new language, a new environment to create applications in. And yes, TPF provides a platform for applications to run on. I guess what I'd really like to see is a marriage of the two technologies. What would one say if there was literally a Java Virtual Machine running on TPF? So that you could write a new TPF transaction in hours or days as opposed to weeks, months, years.
JR: So, are there any exciting products being planned for TPF or that are coming out which you can tell us about?
JS: I think that probably the biggest, new exciting thing for us and what we talked about at the San Francisco user group meeting are those things which we will address at the Fall TPF User Group meeting in the area of application enablement. Lots of people know that we've been working with the University of Alabama [in Birmingham] on the TPF OR (Object Repository) that is now starting to come to fruition. We've got things running, we've got people using the system and so it's a little more than just hyper-vaporware now. We've got actual systems up and running. So we're pretty excited about that and I'm looking forward to sharing with the TPF family those new enhancements to the platform. And hopefully that will make their application development a bit easier for TPF. I guess there are a few things beyond just the OR activities that we'll be discussing there that might be exciting in the application enablement area as well. We're really interested in extending the language support from "C" to "C++". We know we need some file system support on the platform just for our sake in order to bring technology to the TPF community as quickly as possible. There are some other initiatives underway: the user group requirement for an MQ Server as well as client support is something we're working on. I think primarily in the application enablement arena there's going to be lots of new things that should make the platform really exciting for people in the future. We're also very interested in continuing our connectivity initiative. So, being able to attach a TPF system directly to the World Wide Web is something that we'll like to look forward to and what the role of TPF as a web server might mean -- or even a "Super Web Server"-- given how many transactions we can handle.
JR: Great, its sounds like a lot of exciting things are coming. Jerrie, thank you very much for taking time from your busy schedule and I wish you great success in your TPF strategy.
JS: Thank you. This is exactly what I'm here for -- I like talking about TPF.