third tn a sertes
of articles on
the recentliy an-
nounced, IBM
Repostory, part
of IBMS appli-
cattons-develop-
ment strategy
Jor the 19905.
IBMs an-
nouncement of
AD/Cycele is
prompting many
managers to re-
assess their ap-
plications-development, strategies. But,
what is the significance of AD/Cycle,
and how will it change the development
process? Will it provide an open applica-
tions-development, environment in which
tools from multiple vendors can share .
design information in support of all
phases of the development; life cycle?
What should a manager do to take max:
imum advantage of new development
environments, methodologies and tools?
The details of the AD/Cyele announce-
ment provide a clearer view of how ap-

plications will be developed in the 1990s .

in the IBM world. AD/Cycle provides

Ms AD/Cycle Will Bring On a New

ment, user-interactive processing is per-
formed on the PS/2, and application-spe-
cific data is accessed through data serv-
ers distributed throughout the network.
AD/Cyele is ideally suited to support:
an applications-development environ'
ment based on the client:server model, '
which consists of PS/2-oriented CASE
tools and distributed repositories of de-
sign information. Using CASE tools that
are compliant with AD/Cycle, each %
member of a design team interacts with
a personal repository used to store de- -
sign information in a standard format.
Specifications from analysts are consoli-
dated in a departmental repository lo-
cated within the LAN. Corporate-level

AS/400

APPLIED INTELLIGENCE

AD/Cyele Moves Application Development
To the Desktop |
PS/2s Generate Code;

( h
'4

Li

Be

BR nij

J J ;

H Snamčal

4d di 319
pot
k

O

based on identifying the best; architec-

tures, methodologies, tools and manage-
ment technigues available. Strategic is-
sues are important in defining how ap

ia, '
4

a k
F

plications-development; technologies like %

CASE will be implemented within an or-

 ganization. A strategy for CASE must ji
| directly address critical success factors,

of the organization, such as the use of '

the use of computerized systems to lock
in customers and lock out competitors. |
An effective plan for CASE should be
tied directly to the critical success fac-"
tors for the business. '/'/ k
One of the most important critical. '
success factors for many businesses to-

Hosts Aet as File Servers

vsebi. | me Corporate Repository (

the architectural framework reguired to — | zij

build applications efficiently within Sys-

tems Application Architecture (SAA), '' s

It's based on a common repository used .
to store design specifications in a stan-.
dard format. De
Applications-development tools that
comply with the standard Repository
lormat can operate in an open environ-
ment and can share design information.
This has important implications for
vendors of computer-aided software en-
gineering (CASE) tools. CASE tools will
no longer be based on proprietary, mu-
tually incompatible repositories of de-
sign information. Instead, all compliant
tools will follow a common definition of
the objects used to specify an applica-
Kon; the tools will compete in offering
users better ways to specify and manip-
ulate these objects in support of the ap-
plications-development, process.
The battle among vendors of applica-
Uons-development products will be
fought on the desktop, not in the main-

frame arena, As shown in the figure, ap-

"and more to the deskton opa s O?

away from

sof Uware-development cycle anje

desktop-oriented development cycle.
SAA, as an architecture, supports the

Computers. IBM is movin
the conventiona] mainfran

distributed client-serve
| | rver enviro
shown in the figure. In this envinoi?

zna
le umi

a

ga ii ;

z

CASE tools won't be ba

ta
|

— definition of the objects,

SER UROEU

SEV Niri ovi,

ta models and process models are typ-
« Yy accessed from a Corporate reposi-
loh a host computer, m;

- Development environments of the fu-

sei based on AD/Cycle and SAA, will

grated CASE poakon ME high-end, inte-
Therefore, a strategy
development in the 1990«

fesign information. Compliant tools will follow a. commo -

$rowing more efficient due to the wide-

dl ket. much

ž a
Kimi pe

- k i -. s
nie. |
ai h John Avaklan
i]
|

DA km V — m a m! : MU
ZA k ae a NE EO, a Či
zfa di ri La na —% gim si S H $ "

Prej pa ! I
" ločne po 2 ?
; ' ia <a f
[A Ne Fi

m '
i 4 za
i
ii J na | | ! i |

NA | d |

ue di k - je MI a d
ovaj ib (ii Ni

, i i di
K ua | b

ik hja a? OTA EJ ME i
day is support of busi sli change. Ev- |
erything in business is accelerating. The
business metabolism is increasing and <

spread use of computers and the in-' |

, | Sreasing speed of applications develop- ! '
x, ment. As the windows of opportunity ! ' 4
—, - shrink, organizations that take three / /

years to build misslon-critical applica- "|.
li do prvidina their competition" !
a ma
slikah Cases, the inefficient applica- Je
Ste da methods used in infor-

information for strategic advantage and .': mission-critical systems that run major

«strategy should be based on using the

signed to build detailed data and pro-

modeling. It moves through successive EU;
: levels of analysis and design, building up " UE:
' sufficient information to generate code 4
— for an application automatically 2.

, trained analysts. A great deal of man- |. j | li

structured development technigues. Its
|. terconnect. Most competitive firms are
- software connectivity throughout the

Value chain, which they cannot achjeve

- ward support of disciplines such as stra-

' projects and rapid applications develop-

to speci | Oni ne | - ment. If an organization is not usi
te HM! ; | oi a li« pdlali) | h šč | in

; embodied in information engineeri
ri Z a | | Č it
 lsata strategic disadvantage. Šteh el

in developing a strategy for CASE
s Next, Week, I will discuss a ara
, or introducing CASE tools that are
4 Compliant with SAA and AD/Cycle.a
To learn more about the subject of Sa
| please James '.
| Report, an information

For information on seminars, p

< 741 10th S

COCTOBER 16, 1989

trategy for CASE

technology and development environ-
— ments, neh as AD/Cyele, to support busi:

ness change and ensure that the applica-

tions of the future will be able to evolve

as guickly as the business reguires.

One of the most important compo-

nents of a CASE strategy is high-speed

development, including the rapid devel-
'opment and maintenance of complex

' portions of the enterprise. The CASE

most powerful integrated CASE tools
available and embedding those tools
within a development: methodology de-
signed to build strategically important
applications at high speed, in support of
business change. |
'. The most appropriate methodology
'for building strategic systems across a
corporation is information engineering.
The methodology was specifically de-

cess models that can be converted auto-
matically into code. Information engi-
neering typically starts with high-level

: strategic planning issues and enterprise

Recent extensions to the information-
engineering methodology have provided
support for high-speed applications de-
velopment, with small teams of specially

agement attention is devoted to provid-' (H
ing the teams with clear goals, a high. (|!

tion for excellence. | 1
Turning from Tradition IH.
Were rapidly moving to a develop- | m

ment environment in which businesses "| ' V
can no longer afford to use traditional //.|

Na PUB

no longer possible to use separate teams
to build separate projects that, don't in-

now saying they must have application

without information engineering.
Information engineering is oriented to-

tegic planning, enterprise modeling, com-

plex data-modeling technigues
ode: | es, business
area analysis, tight integration across

nigues are of ( undamental] importance

Tiarteriy, at (800) 242.1240, |
contact (in the United States ani eu,
ga ie ales and Can-
gol eTboR zi a unajer Institute,
AJŠA »dnta Monica, Cali bi
oObOR (ia) zah asno ica Caf.
tact Savant, 2 New St, Carnforth,
Lancs., LA5 9BX United Kingdom ]
ME NE AE

k pi] 1 A 5 14
a f ae
| d j
li 1 | ih
| HI
] | |