Terry Floyd has built a past and proposed a future around 3000 manufacturing. The founder of ERP integrator and consultancy the Support Group inc., Floyd is marking 25 years of experience with the system this year, and making plans for far more. He cut his professional teeth when IMAGE was new on the 2100A, the precursor to the HP 3000. Floyd began his career as an MIS manager in 1974, but within seven years he was working at ASK Computer Systems, the creators of the MANMAN MRP application suite. He came up through the MIS manager ranks using MANMAN at a Texas manufacturing company, using background in math, science and accounting as well as experience with FORTRAN. In the 1980s, Floyd was managing the southern region of the ASK technical force after having “learned it the hard way,” since ASK didn’t have any training classes. It meant Floyd learned the massive application from the internals out, and gave him a grounding in it that only its development staff could match. His years of work as regional technical manager for ASK gave him a developer’s knowledge of the software and basis to create his first company, Blanket Resources, and an EDI application after he left ASK. That first company was named after the small Texas town where Floyd grew up, riding and roping calves on a ranch.

Between being a crusading gadfly for MANMAN’s vitality, and pushing HP and Oracle back into an e3000 development plan, Floyd is working with plenty of opinion and passion for ERP customers on the platform. As he headed to the annual IFS conference with several prospects from the e3000 world in tow, we asked him where the platform’s manufacturing is headed, and how it might play a bigger role in the system’s renaissance.

What’s happening to manufacturing choices for the HP e3000 customer?

It’s slowly moving. There’s not a lot of enhancements in MANMAN since 1985 that were addressing the fundamental problems with it. ASK was interested in replacing MANMAN in the early 1990s with a product called Advance that killed their company, and CA bought what was left. I wrote an article for my own newsletter last December which attempted to be critical enough to get someone to invest more in the new releases of MANMAN. The last releases haven’t done much in the way of addressing the 1,500 open enhancement requests that ASKUS left on the table when they quit working on enhancements.

Every year there was a new set of people at CA who said “We’re going to do this,” and the next year they were gone and new set of people said “We’re gonna do that” — and this and that never happened. Perhaps something is going to happen in Release 12 now. It seems like there’s a lot of activity to provide real functional improvements in that release.

What’s your view of the other choices for e3000 ERP?

MM has become eXegeSys eRP, and both MANMAN and MM are IMAGE-based products. It’s kind of hard to convince new customers to move to a platform that is conceived of as being as proprietary as MPE with the IMAGE database on it.

Is the HP 3000 still a system that can carry an ERP solution into a site as a new 3000 sale?

I’m positive that it can. It’s happened — rarely, I suppose — but the eXegeSys people are bringing a few. MANMAN in the last five years has sold some new sites, but I’m aware of one I worked on last year that bought it. They were already HP 3000 fanatics. If you have people who are pro-MPE, they’re looking for some solution. It’s being perceived as being a sideways move to go from MANMAN to eRP from eXegeSys, or from GrowthPower to MANMAN. There are three or four packages that run on the HP 3000, and many people think of them as being old.

Can manufacturing can play the same kind of role Smith-Gardner or HBOC software plays: bringing new customers to the platform?

To be able to do that, you’d have to have an application that was stellar and that only ran on an HP 3000. I don’t know where that’s going to be coming from. Your answer for that is eXegeSys right now. I hope they’re attracting people with their product and its wonderful enhancements. I fear that CA is not doing much to attract people to the HP 3000, or that they very much care about the HP 3000.

You said people think of these applications as old. Do you think the 3000 is being perceived as old?

Yes. I’ve talked to so many people who think that it’s an old machine. And when you tell them it’s physically the same hardware an a 9000, and do you think a 9000 is old, they say “Oh no, we think that’s state of the art.” It’s the same box. People outside of our readership here think of it as old.
The way I’m heading is to bring a new, modern ERP package to the HP 3000. We are doing that with the IFS ERP package, which we are now marketing. We did the port, and there’s not actually a lot of port there. When I’m out there talking to non-MANMAN customers about IFS on a 3000, if I’m talking to a GrowthPower user who likes the 3000, they are extremely interested in IFS running on a 3000. If I’m talking to someone out there who has never heard of an HP 3000, and they are talking about running Oracle, they are much more interested in NT or HP-UX. They don’t understand the things we do — about how robust and easy to run MPE is, and how IFS will really be more robust and easier to support on HP 3000s. That’s kind of a hard sale we’re still working on.

What does IFS offer that’s 100 percent unique for the MPE/iX marketplace — not duplicated in another e3000 ERP package?

It’s very simple: a relational database system, and all of the things that are around the edge of a relational database system. IFS is, as far as I know, the only relational database ERP system for the e3000 on the market. It’s the obvious reason why there are not more ERP packages running on the HP 3000 other than the four or five ones that have been there forever.

If we go back and quote the slides from the 1994 All-Texas RUG conference and a speech by [HP’s] Dave Wilde, he had a slide with “HP 3000 New Applications.” On this was a list of companies we all wish had new software on the HP 3000: Datalogix, Delta, SAP, PeopleSoft, MANMAN/X, Oracle Applications, Baan, Progress, QAD, Filenet, Ross and many more. None of them were available on the 3000 then, and they never were. What happened? I look at what I’ve done here, and I wonder how to get Oracle8 on the HP 3000 without paying for the port myself.

Do you need an Oracle8 port to the HP 3000 to continue offering IFS to ERP sites?

Yes. IFS runs on Oracle, therefore it runs on some 90 different platforms. When I go to talk to someone about keeping them on an HP 3000 and I talk about IFS, I have to say, “Oh by the way, IFS currently uses all the features in Oracle 7.” And Oracle8 has been out for a couple of years now. IFS in their newest release does not take any advantage of anything in Oracle8, so I have no problem with the latest version of IFS running on the current release Oracle 7. But when I get to the next release, I’m told that some of the features of Oracle8 may be used. So I will be forced to investigate how to turn an HP 3000 into a 9000, so I can get IFS running on Unix — since I may not have Oracle8.
I know I have to work on Oracle for this port as well. I hear the reasons that Oracle 7 has failed to gain the market share hoped for in 1994 was that Oracle really wasn’t too involved in the port. Oracle has never really had much interest in helping the cause of the HP 3000.

What roadblocks do you see in getting IFS accepted in the 3000 manufacturing community?

The questionable future of Oracle on the 3000 is the only one I know of. I can see us not needing Oracle8 until maybe the end of 2001 or the beginning of 2002. We will be working to see what the problems are related to making that version run on Oracle 7. But since I assumed I’d have IA-64 HP 3000s by 2002 or 2003, and IA-64 seems to guarantee I’ll have Oracle8 back again, I thought I wouldn’t have to worry much about it. Now IA-64 seems to have been pushed out further than that, and I do have a gap in here where I probably need Oracle8. I’ll see the plans for the next IFS release at the IFS conference.

So what do you think would motivate HP to do the next port of Oracle?

They’ve told me, “show me the market and we’ll do it.” I’ve pretty much done that, I think. I have quite a list of people now who are interested in moving from MANMAN to IFS and staying on an 3000.

Where are you on the acceptance curve for IFS in the 3000 market? Are prospects looking over the software today?

Absolutely. We’re about to close some deals. By the time this goes to print, we’ll probably have our first customers. There’s a ton of interest, at least 100 sites we’ve identified right now that want to leave MANMAN. We didn’t talk them into that; we don’t want to talk them into that, but that’s what they have already decided. They are leaving MANMAN and there’s nothing anybody can do about that.

Is there a compelling reason to move away from MANMAN today?

Not really, in my opinion. In the long run, the reason to leave MANMAN will be the ease of interface to new subsystems and modules that fit around the edge of MANMAN. Some of the things that are going on now in HP to make it easier to interface are good. But still, it’s not the HP interface that makes it difficult to interface to MANMAN — it’s the internals of MANMAN and how the database is designed. Not the technology provided by MPE and IMAGE; the bad things about MANMAN have nothing to do with IMAGE and MPE. They have to do with programming techniques that were used by programmers in the 1980s on MANMAN. Other more modern packages have that kind of mish-mash. When I look under IFS, I don’t see that mish-mash. I see a consistency that’s compelling.

Why have those hundred companies told you they want to leave MANMAN?

I think it’s a technology thing, mostly concerning DBMS and GUI. It’s ease of use and training: computer-based training, long-distance learning, online help that’s context sensitive, the ability to train new users on MANMAN is not easy or intuitive.

What kind of application lifespan do you expect MANMAN sites can hope for when planning for their ERP needs?

It depends on their own internal expertise. The more self-sufficiency we find in our user base — the ability to handle LANs, networks, Microsoft and Novell-centric things — helps them with the ability to interface things to MANMAN. The little shops have a lot more struggle than the big shops in keeping MANMAN. Tiny companies that are happy with MANMAN are using only the financials parts of the system — not the MRP or Master Production Scheduling or capacity planning.

You can stay on MANMAN, or you can go. If you stay on MANMAN, you have to invest in adding things around the edges of MANMAN yourself. If you go, it’s because you want to get access to the other things that MANMAN does not have yet, like a Product Data Management system. It’s an engineering kind of tool that allows you to pass drawings around. People want a user-friendly front end on MANMAN that allows them to show a picture of a part when they list a part in the system. CA is working on Web enabling and adding front ends to MANMAN, but at the rate they’re going, it will be ready by 2006. Other third parties are working on adding things around the edges of MANMAN, but if you buy a new system like IFS, things like PDM are included. That’s not something you can easily add to MANMAN.

Can you make the case to HP that IFS can leverage a new e3000 sale?

I think we can, after we demonstrate. In a few weeks we’ll be in the process of installing IFS on HP 3000s at existing MANMAN sites and begin doing conversions. I predict that by the end of 2000 we’ll certainly have a great story to tell about why you should buy IFS and implement it on an HP 3000 instead of NT or HP-UX. It will take a large company to see the advantage of the HP 3000. It’s these references we’ll be able to show people who have never heard of an HP e3000 that will show why they want an HP 3000 running IFS.

Why did the Special Interest Group of Interex change from SIG MANMAN/Choices to SIG-ERP?

The attendance at the meetings fell off. The reason to have a SIG meeting at Interex was because you had an advocacy issue. Last year there were five or six MANMAN users in a room of 50 people, surrounded by vendors, HP personnel, Interex personnel, other people with other packages wondering what was going on with MANMAN. For the last three years half the presentations we offered were alternatives to MANMAN. SIG-ERP includes a vocal group of MANMAN users on its board, and they are determined that SIG-ERP include a forum for MANMAN users to have discussions at HP World conferences. Maybe there will be more interest now that we’ve threatened to take away their meeting.

How effective is CAMUS, and what do you think needs to change about it?

If the users of MANMAN want to talk about MANMAN, they should be involved with the CAMUS users group. ASKUS changed their name to CAMUS, the Computer Applications Manufacturing User Society.

They’ve become more independent [of CA], and I’ve thrown my weight behind CAMUS. I’m trying to convince CAMUS they have interest in attracting users who are not on CA support, and who are on older releases of MANMAN.

CAMUS has an excellent meeting. This year’s will be in Boston in August. That’s the place I look to lobby for change at CA for future releases. It’s the best place to talk to someone about support for older releases of MANMAN, which CA now supports. It’s the best place to talk to other users who are using MANMAN. CAMUS fills the role we started with SIG ASK.

What kind of guidelines should an IT manager or VP look at to decide whether to enhance an existing ERP solution or move a new one in?

A lot of it has to do with how much of their budget they want to spend. It’s amazing that we’ve got MANMAN customers that have been on it for 15 years, it’s paid for, and they’re investing very little in it — a class of companies spending 0.2 percent of annual revenue on IT. Those people who continue to wring every ounce of revenue out of the 3000 will stay on MANMAN.

Companies spending 2-5 percent of revenue are the ones who recognize that information technology can change how their company works, and help them defeat their competitors. Somewhere between 0.2 percent and 5 percent is a happy medium, where people can enhance MANMAN and its use. Or they could spend that much money over a five-year period to replace MANMAN, and go to a new system.

With the future ASP movement of being able to outsource everything, people think they’ll be able to cut their budgets in IT. I’m questioning that in the short run here. Perhaps a few years from now, all of that will be reasonable. In the short run, the expertise you gain by having people on your site who understand your system and are able to modify it, enhance it, and produce reports as needed, interface it to other foreign systems is a great advantage that you don’t want to give up. Self-sufficiency in that area would be my recommendation.

Is the fact that you’ve built your own building an indicator of how healthy the 3000 manufacturing business is?

We’ve moved through two different sets of offices since we started five or six years ago. We’re proud of having built our own office building at Lake Travis now. It really has changed the attitude of the people who work here. We’re doing really well as far as being the last one standing in the MANMAN space. We’re still extremely MANMAN oriented, and will be for the next 10 years.

How do you think your MANMAN experience helps in promoting a future manufacturing solution like IFS?

It changes the P in ERP to an M, so it’s Enterprise Resource Management. The new systems emphasize execution of the plan. I’m trying to do anything I can do to convince people to stay on HP 3000s. People are going to be leaving MANMAN. We don’t try to convince people to do that, but we want to participate if they do. IFS has been able to maintain the feel of a small company, like ASK Computer Systems was in the beginning. We just want to continue a tradition of friendly service.