AO AEE ZK APPLICATION DEVEL IEH RISC-architecture "OU applications, ss well a8 the development K ge Oč1O rmipelivi Komne leto bas. CD Mgot new pnek 84 k j- ROM providing as much aš 650M bytes a konat CASE pooiš ME ih IE ea o rta pi available as d primary mediumi for the 4" toward development of nev applica- ži large information bašes. ' —Yšptions, they fundamentally change future - ' maintenance costs by bullding A late o IE ita (iva be: '|1980s include widespread distribution of SFused in a forward-engineering process to | integrated CASE tools, introduction of poi generate normalized data definitions "| ' and structured code for the same or a fi Currently, tools exist that can restruc- ture spaghetti code, but do not generate ' high-level specifications of the code. MU nologies such as third-" and fourth-g ation languages are being replaced by: much more powerful technigues. doji most promising new development tech: nologies are: highly parallel computers; ig parallel inferencing technigues; high'854" , speed optical fibers; optical disks; — vanced microchip technology; | neurocomputers; integrated CASE JE), information engineering; expert: " system tools; advanced development ; methodologies; object-oriented l: Si jand textbase management systems "' šošij This series of columns, which began des week with "Modeling Technology: guide to the Future". (Page 43, Oct. 17); |. examines the major changes that aret 100 bips 10 bips 1 bips 100 mips| V Future Implications pected developments in individual coj > ponents of computer technology and' ze 1995 2000 2005 2010 2015 ed into a formal model of future tech: ": nology, whose primary function is to ex; plore the implications of expected - ware and software technology. si 5 CASE workbenches will pomi all as: er available tools can convert, data defi- »pects of, the application development nition language (DDL) statements into puter systems will operate in the follow ,, ing range of speeds: low-priced li process, including Bysterh planning, ei) € high-level data structures, but do not re: - computers: 1 million instructions no per A lesa ome pevibii documenta- " structure process code. In the near fu- second (mips), medium-priced personal (ij Z tion generation, dka nae generation nandi; - ture, tools will be available that. convert - computers: 3 mips; ek priced personal "/š project management. ''/' %Y4WWfii'Y>'5 057 both low-level data definitions and pro- workstations: 10 mips; top-of-the-line SEZ) The front-end poime analysis and ij: cess code into high-level structures that mainframes: 80 mips; top-of-the-line ' breča design components of these products ''.2can then be used to regenerate the ap- ; supercomputers: 4 billion floating; polntst s will be closely integrated with the back: , plication or maintain the application at operations per second (bflops); and '"$'% end code, databaše and o nalation: va high level. Current CASE tools provide Pori a portion of this future capability. ' Expert-Systems Integration: In the luč next three years, expert. systems will eering tools prornlece to ae »—, have Soupeted te transition to ai in- ' large artificial- rintelligence inference en: - gines: 50 ni logical inferences per.i ije Reverse Engineering: The Bucesstul v mi | second 1 fjEY introduction of forward/rev " (Th opne oa pli ko ink ie generation facilitieg.)! vee5' rali V Pace 35 tegration with the conventional business environment. a ie Curreni pensive, run on personal computers and are easy to use. They interface closely with conventional operating systems and TP monitors, access corporate databases and integrate cledniy with the me si standard systems are becoming a com- ponent of data-processing applications. Standards: The movement toward standard computer-system environments is accelerating. SOL will become a wide- ly used standard for access to relational databases. The introduction of IBM's System Application Architecture (SAA) will provide standard ways for applica- tion software to access databases and networks and will support standard forms of user access to computers. Information-systems professionals imay use the Presentation Manager and Communication Manager components of 08/2 to impose controls and standards on distributed networks of PCs. For many organizations, the rapid proliferation of personal computers has resulted in chaotic management of PC resources and lost control of distributed PC networks by professional IS manage- ment. |. OS/2 offers both end users and IS W managers the opportunity to control the operation of distributed PC networks within a professional, well-supported en- ; vironment. Users will be asked to oper- jate within the constraints and standards J enforced by OS/2 and SAA. In return, . users will receive professional m ment of file servers and communication interfaces, as well as controlled access to corporate databases and distributed communication networks. The appropri- ate application of standards provides both end users and IS managers with a win-win situation. Strategic Business Systems: In the near future, organizations will increas- ingly regard computers and information systems as a strategic weapon. Applica- tions will be sought that enable a corpo- ration to use computer systems to gain a strategic advantage. The term "mission-critical" is used to describe on-line strategic systems which integrate critical aspects of the opera- tion of an enterprise. If the system fails, a vital portion of the enterprise stops functioning until the system is back in operation. Mission-critical systems re- auire highly reliable computers, includ- ing networks and databases which oper- . ate continuously.E The James Martin Productivity Series, an information service updated. guar- terly,i is available through High Produc- tivity Software Inc., of Marblehead, Mass. (617) 639- 1958. For information on seminars, please contact (in the Unit- ed States and Canada) Technology Transfer Institute, 741 10th St., Santa Monica, Calif. 90402 (213) 394.8305. In Europe, contact Savant, 2 New St., Carn- forth, Lancs., LA5 9BX United Kingdom (0524. 734 505.