In-Depth

September Response Time

• 09/01/1999

Thank you for clearing up the EJB conundrum in Mark Cathcart’s "OO" article in the April edition of Enterprise Systems Journal ("OO Alternatives for O/S 390," page 34). Perhaps you can suggest what we can use now, while we wait for EJBs on OS/390?

Well, in the interim, there are Java product gateways for CICS, IMS and DB2 that allow you to build new programs that exploit those systems and new or existing applications and data. In addition, you can use VisualAge for Java and its Enterprise Access Builder to build CICS, IMS and DB2 gateways and Java programs that use the Common Connector Framework (CCF). CCF is a standard set of interfaces and services that allow a programmer to minimize the number of new Application Programming Interfaces (APIs) that they have to learn in order to communicate with existing data and applications on OS/390. It is our intention to provide a migration path from CCF to EJB.

*****

I think that is an interesting premise you present [in the June ESJ Editorial, "Y 2 Bother," page 6], people buying up front and not buying at the end. However, the reason companies are doing so is because they are gearing up new systems. Y2K is as good a reason as any to revamp legacy systems and bring in new technology that will allow them to have or maintain their competitive edge in the new millennium. The mention of e-commerce drives my point home.

The older systems, which some companies have in place, could not possibly handle the new data loads they are expecting after the turn of the year or even after 3/2000. Boards of directors have been tight on the money up to this point, and now IT directors and PMs are killing two birds with one stone.

The "storm before the calm" may not be too far-fetched either. A lot of spending is happening now, but when it is all said and done, I expect spending to drop off a little or a lot, depending on how much "wound licking has to be done." That is to say, companies will be assessing the damage and rerouting funds to fix things that may have been missed so they don’t fall too far behind the power curve and are able to move forward with that "e" initiative.

IBM is taking advantage of that circumstance and giving the best they can right away; that way, they have a cushion when things start to fall off. Really, it is hard to tell, at this time just how much spending is going to drop off. But, I think it will. Just watch.

*****

As a long-time reader of ESJ, I appreciate the changes you’ve brought to this publication. It seems more relevant now, fresher. Is that what you’ve been aiming at?

I’m teaching a course from time to time called "SAS For COBOL (And Other) Programmers." That old question has come up again in approximately this form: "Isn’t mainframe usage declining, so why should I bother learning another mainframe language?" I honestly don’t know the answer. What would you say?

Thanks for the kind observations, and yes it has been our goal to tighten the mainframe focus of ESJ, while moving the publication forward.

In an effort to determine the continuing role of mainframe computing in today’s enterprisewide IT environments, ESJ performed an extensive research study conducted by independent researcher, Martin Akel & Associates. Fifteen hundred IT managers, senior IT managers, corporate managers and IT staff were randomly chosen from ESJ’s subscription database and mailed a comprehensive questionnaire.

When asked how significant mainframe computing is to achieving the organization’s mission-critical objectives today compared to three or four years ago, 35% of the respondents said it was more significant, 49% said it was the same and only 16% said it was less significant. When we posed the question regarding their outlook for the next three to four years, 70% said it will be more significant or the same and 30% said that mainframes would be less significant. Clearly, a majority of respondents continue to rely on mainframes to achieve their mission-critical objectives.

Reasons why organizations invest in mainframe computing instead of, or in addition to other platforms include:

• Mainframes’ degree of reliability and performance: 88%

• Mainframes provide a highly secure environment for mission-critical applications: 80%

• Many users can be supported via a mainframe: 70%

• Centralizing data on a mainframe is the least expensive way to empower users: 59%

• The costs of acquiring/using mainframes have declined: 45%

• Mainframes spawn a wide variety of access modes: 41%

For the complete findings and a report written by ESJ’s Publisher, Calvin Carr, visit the Merant newsletter site at http://216.32.20.157//industry_1.asp?sessionid=B43DUC0P83G83Y3Q&dropid=F2.

ESJ invites readers to assist Mr. Zirbel in his questions by submitting their responses to ESJ Editor-in-Chief Charlie Simpson at simpsoncm@esj.com.