Think of the Year 2000 Problem – no, don’t groan or shut your eyes, it’s not that easy to make it go away – and chances are you think of large, MVS mainframe based Cobol applications. But if you think all estimated six billion lines of production Cobol just run on Big Iron, you probably missed the minicomputer revolution, or at least haven’t heard about it yet. Turns out there’s a whole boatload of non-mainframe Cobol systems still burning up electricity and doing good things for their users. So what are those users going to do between now and the Biggest Party Of The Century on December 31st 1999?
By Gary Flood
This neglected issue recently prompted Y2K guru Peter de Jaeger to write an open letter to Computerworld wondering why vendors were not thinking about the little guys in all this. Step forward NeoMedia Technologies Inc, which is positioning itself as offering the Y2K fix for the rest of us. Given that professional millennial scareologist Gartner Group is now talking about a possible one trillion dollars’ worth of potential litigation to do with the whole mess, this may come as good news to many a midrange systems manager still nursemaiding some mature but yet desirable system. Don’t be surprised at never hearing of NeoMedia before – as Cameron Jenkins, director of marketing and strategic alliances for the outfit, admits, the company has been doing no marketing or trade show participation or analyst/press work at all, giving rise to prospects asking it, Where have you been? Now that’s changing, with the firm trying aggressively to put itself on a neglected, but important, part of the landscape. In practical terms, what is on offer? Jenkins quotes analyst definitions of the five phases of the Year 2000 problem as being awareness (10% of all firms not even aware they have a problem yet, he opines); impact analysis; remediation; testing; and integration, both highly labor intensive process which will probably end up accounting for at least 40% of the entire Year 2000 fix. NeoMedia claims its product automates the impact analysis and remediation phases. The Adapt/2000 product set is based rules based/data dictionary using code scanning and parsing techniques, using seven rules to handle detailed impact analysis, automated data file conversion, automated source and copybook conversion, expansion/windowing (fixed/sliding) or contraction of suspect code, as well as a complete date routine library with source facilities, a full audit trail, and in-line documentation utilities for file browsing, fixing and testing.
Market gap
The next release, Adapt/2000 Release 3.0, due in the second quarter, is mainly targeting what Jenkins believes is a market gap in the way IBM Corp is handling some of its users’s Y2K concerns with its recently announced compiler-level Year 2000 fix, MLE (millennium language extensions), a windowing, non-code expansion solution. So what’s wrong with MLE? Jenkins, formerly VP marketing with Cobol specialist Acucobol Inc, claims Adapt/2000 provides automated source code conversion, whereas the MLE syntax can only be implemented manually; MLE automatically windows identified data fields for year 2000 conversion but it requires identification of all date fields and code changes in the Cobol language data field declarations for it to work, plus it only supports a limited set of standard date formats, overlooking the need to expand dates in file keys and sorts, according to NeoMedia. Adapt/2000 is further claimed to allow implementation of a combination of MLE, code expansion, and windowing of non-MLE date formats, while also providing inline bridging between date formats and the generation of data conversion programs where expansion is used. Neomedia claims Adapt/200 is aimed at solving the Year 2000 problems in the so- say orphaned markets of non-mainframe multiple Cobols, such as IBM DOS/VSE (though the product also covers the Y2K market focus of MVS), Wang, HP3000, Data General, Unisys, NCR, DEC VAX Cobol, and Tandem, as well as the PC-based Cobols of
Micro Focus Group Plc, Acucobol, Inc and RM/Cobol. There is no comparable competition here, says Jenkins. There is a prevailing belief that Y2K is just a Cobol on the mainframe problem. All the Year 2000 ‘Factories’ are targeting the big projects for big clients – an EDS Corp or a Keane Inc won’t touch a project that involves less than 10 million lines of code, because there’s enough business to pick and choose at the high end. He thinks there may be as many as one million non-mainframe Cobol systems up and still running out there, with the clock ticking. But surely this smaller, midrange market will be the first to dump these systems for packages? If it really was that easy, they’d have done it long ago, he counters, with fairly unarguable logic. Jenkins and the NeoMedia Year 2000 product range is a new venture for the company, which is better known as a player in the document management space, and comes almost entirely from its 1997 purchase of a small, closely held Y2K house called Allegiant Legacy Solutions, Inc. That outfit, whose offices are still in Cincinnati, Ohio, was only founded in 1996, and by the eight months ending in July of 1997 had revenues of just under $1m ($921,000) and net income of approximately $266,000, immediately prior to NeoMedia’s stock swoop on it, which has been accounted for as a pooling of interests, with the sale completing in the third quarter. In November NeoMedia announced figures showing revenues of $16.5m for the nine months ending September 30, up 29% from previous period’s $12.8m, on net loss of $4.5m, $2.3 in the third quarter, compared to a loss of $1.9m the previous year, with the company explaining the loss as based on the need to increase investment in infrastructure expenditure and expanding sales and marketing. The company says it has already seen significant increases in license and services fees as a result of the move into Y2K. NeoMedia, founded in 1989 and based in Fort Myers, Florida, went public in November 1996 (listed on Nasdaq’s small cap market), now has a total of 54 staff. Its primary business is in software for high speed and distributed printing systems featuring high density bar codes, secure document technology, and web thread technology. Jenkins estimates total fiscal 1997 may come in around the $27m mark.
Long term solution
Content from our partners
It needs to be noted that like all self-respecting Year 2000 vendors, NeoMedia is careful to position itself as a long term solution, lest anyone fancy it closes its doors on January 2nd, 1000. It describes this part of its business as covering legacy Cobol extension products and services, migration, and reengineering tools, offering not just Y2K migration and consulting services but Cobol to internet integration, i.e. other long term mass changes, such as large scale code maintenance, rehosting to open platforms from legacy systems, and a RAD environment for extensions and new development on these older Cobol systems, like the European Monetary Union challenge, telephone number digit and zip code extensions, as well as the potentially highly expensive change from fractions to decimals based share calculation that Wall St is starting to entertain. Adapt/2000 release 3.0 will be so priced as to be affordable to midrange users, with $20,000 as start price for NT and $30,000 for Unix, though the company in practice mainly sells through third parties and systems vendors. Current customers include Countrywide Inc, EDS (using it on a variety of customer apps including Unisys A-Series and Wang), Keane (HP3000), and Sterling Software (on VAX, Wang, and AIX systems). In December it announced secured Comsys Information Technology Services, a subsidiary of Corestaff Inc and Y2Kplus Inc have both agreed to sell the NeoMedia tool as their preferred product for midrange Year 2000 conversions. Adapt/2000 runs on Windows 3.1, Windows 95, Windows/NT, DOS and DOS networks, and Unixes including HP-UX, IBM AIX, DG-UX, Sun Solaris, DEC Unix, and SCO Unix. May be worth a look if you think you have a problem but don’t want to pay
EDS style prices.
