”SoftOne”
Sören Janstål

.   Tidigare gallups
Fun

Data Research DPU ab
Torsvikssvängen 34
181 34 Lidingö


Tel +46 (0)70 727 67 95
Skype sjanstal

eMail till Sören Janstål


The Enterprise in Transition
the debate rolls on

The Enterprise By Other Means is the best selling industry report Bloor Research has ever produced, having sold thousands of copies. When first published it provoked an immense level of press comment and prompted a debate which is still running. Our analysis contained much material, but the most dramatic proclamation it contained was that we are witnessing the death of the PC - at least, in the corporate environment. However, the findings of the report cannot be expressed in a single phrase. In this short article, we would like to record some of the reports key arguments and address some of the questions that they have provoked.

The Primary Proposition: There is now an established move back from client/server to centralised computing. This trend is driven by three key technologies - maturing MPP hardware, which can replace multiple servers with a single machine; the Java language, which can run in any environment that supports its interpreter; and the emergence of thin client hardware. Given also the overwhelming driving force of the Internet, we are moving towards the global integration of computing. The corporate PC will largely be superseded by thin clients. The browser will become the GUI paradigm. Software development will be based on Java, or at the very least on Javas mode of operation. Finally, the corporate server environment will be consolidated into a smaller and smaller number of boxes.

It should be obvious that these changes have profound implications for IT vendors and corporate IT functions. They are also fundamental to the enterprise as a whole - hence the reports title. But are our basic premises correct? The debate has already moved on.

The Debate

Java

Objection: Java is a low level programming language. It is simply not suitable as a universal development environment. Far more importantly, it will be just the latest example of todays strategic technology becoming tomorrows crippling constraint.

Response: The real point about Java is not the language, but the Java Virtual Machine and its portability. The Java Virtual Machine frees us from operating system lock-in. As for programmers - Java  is a cut down version of C++ and people from that background are actually using it with great enthusiasm. Furthermore, any application development tool which generates C++ can easily be modified to generate Java. This has already happened in many cases.

Network Computers

Objection: Thin clients or network computers - NCs - have a high entry cost. This includes not just the NC hardware, but network infrastructure upgrades and major client/server application rewrites. Whats more, applications could be loaded on demand from the network just as effectively through PCs whose local caching would be invaluable given bandwidth limits.

Response: We take the NC entry cost question seriously. A smooth transition can be achieved by altering the approach to networking over time and gradually replacing PCs with NCs. Large sites should not forget the very high costs of continuing with PCs. Most PCs are over-configured. If the NC evolves to include local disk space, this will be only for caching purposes. The desktop device is a caching resource and must be managed as such.

Centralised Computing

Objection: If we return to a centralised architecture, we will be back to hardware vendor and operating system lock-in. Above all, only the distributed model gives businesses the flexibility that is now essential to them.

Response We believe that the NC/Java device will become the desktop standard. We then expect to see a high degree of convergence in the server space, with vendors cloning each others environments. Organisations will have to choose a prime vendor, but with secondary suppliers waiting to take over if the prime vendor behaves in too proprietary a manner. As for flexibility - a corporate strategy based  on the PSI factor (portability, scalability and interoperability) is the best way of achieving it. It is existing fragmentation that makes it difficult for many businesses to cope with acquisitions or major internal reorganisation.

Conclusion

The debate continues. We have barely touched here on the real significance to business of the new centralised computing paradigm. Follow the argument with us in our forthcoming report The Enterprise in Transition.



Närbesläktade rapporter:


ERP market, just this minute

Hmta faktainsamlingsformulr/ checklista fr utvrdering av affrs- och ekonomisystem utan kostnad.

Lmna din egen bedmning av ngon programvara eller system och erhll marknadsanalys och checklista utan kostnad!

Lmna din egen bedmning av en IT leverantr och erhll marknadsanalys och checklista utan kostnad!

Data Research DPU
fr utvrdering av IT - Informationsteknologi och Data Produkter


[ Bestllning | Mer info | Frslag ny utvrdering ]
[ Konsulttjnster | Utbildning | Priser | Distributionslista ]
[ Kontakt ]

Tillbaka till Data Research DPU startsida.


Data Research DPU ab - Torsvikssvngen 34, SE-181 34 Liding, Sweden - Telefon 070-727 67 95 Skype: sjanstal, SkypeIN: 08-559 25 900 - Kontakt (email)



Ataio

Plats nr 5 för sponsor/s


Plats nr 9 för sponsor - Ledig (s)


Metodika

Ändrad senast onsdag 8 maj 2002