Methods & Tools - News, Facts & Comments Edition - November 2001

*** Companies ***********************************************************

* SAP swings to Java

At the beginning of November, SAP, the giant German ERP software producer, announced that its new MySAP technology platform will integrate Sun's Java 2 Enterprise Edition (J2EE) software programming technology.

This is a new episode showing the coopetition (cooperation and competition) mode that influences moves of major software producers. SAP and Microsoft are cooperating to offer SAP on NT with SQL Server as a solution to decrease the fact that SAP software was often associated with an Oracle database, Oracle being a competitor of SAP in the ERP market. Java is a competitor of Microsoft's .NET environment, but now Microsoft is also an actor in the ERP market with the purchase of Great Plains, another company active in this area, and therefore a competitor of SAP. Like Microsoft removed Java Virtual Machine from its new XP OS to bother Sun, SAP has chosen Java to avoid giving a larger market to Microsoft. SAP has not completely left out .NET, because connections should be provided from MySAP to other programming technologies.

If this former paragraph seems complicated to you, I assume my responsibilities as editor... but the situation in itself is also complicated... :-]

* Database ups and downs

The recent quarterly reports from database vendors allow us to take a new look at this market. On the "looser" side we see Sybase (- 20% in licence fees from 2000) and Oracle (-8% in database sales). The "winners" are IBM which report growth of 19% and 36% in the last two quarters and Microsoft (+45%). Software AG also had good first nine months of 2001 with a solid growth its Adabas/Natural mainframe product and an increase of 41% for its Tamino XML server, even if this last product generate only 14.2 million Euro of revenues.

According to the Gartner Group, Oracle is still the indiscussed leader with a 34% market share for the global market and a 66% market share for the Unix segment. This comfortable number one position has not helped Oracle to attenuate its traditional arrogance... With time, the domination of Oracle has however decreased. With the acquisition of Informix, IBM has showed its intention to fight in the Unix area. Microsoft has gained the same market share level than Oracle in the Windows/NT operating systems. Even if Oracle has narrowed the pricing gap with IBM and Microsoft, the premium requested by Oracle will be an important influence to close sales in this weakening economy. The trends mentioned above will surely continue and lead to a more balanced market share distribution between Oracle, Microsoft and IBM. We think this is a good thing for buyers, as this should increase the importance of customer relationship in the vendors' attitude.

*** Numbers *************************************************************

* Do you really want quality?

Techrepublic.com conducted recently for Mks a survey with 550 respondants, mainly US developers, on software development needs. Here are some of the data you will find in this survey.

What percentage of the applications you build is Web-based?

• 100%       8%
• 76-99%    11%
• 51-75%    13%
• 26-50%    19%
• 1-25%     38%
• 0%        11%

Pick the description that best illustrates what is important to your organization?

• Software quality 52%
• Minimizing costs 23%
• Speed of delivery 21%
• Other 4%

What level of process management exists in you development organization?

• Highly regulated 6%
• Mature yet flexible 38%
• Loose and informal 46%
• Primitive 15%
• Nonexistant 3%

The answers to the first question mainly recognise the shift toward web-based applications. The software development world is trying to regain some of the benefits of the mainframe era, letting only minimal amounts of code out of the secure and managed area of servers.

The two other results are very suprising for me. It look like these were answers made mostly by managers... :-] If quality has always been highly placed in the Powerpoint presentations of software development projects, the reality speaks more about short deadline and understaffed projects. The time to "think right" the project and to assess its quality through an extensive control has commonly been minimal. Do you know many project with "serious" test plans? (I am not speaking of documents with just list of screen to test... and sometimes THIS is already a big achievement!). It is stranger to see that 52% of the participants say that quality is the most important thing for their organisation and that 58% have a software development process that is at best loose and informal.

Source: "Changing Application Development Needs", www.mks.com

http://www.mks.com/campaigns/wptechrep.jsp

*** In Other's Words ****************************************************

* The Future of OOP

" Stroustrup: In C++, without appropriate libraries, anything significant seems complicated. With appropriate libraries, just about anything becomes manageable. Building and using libraries will become increasingly important. This implies an increase in generic programming, because only through that can libraries become general enough and efficient enough. I also expect to see growth in distributed computing and in the use of 'components'.

Stroustrup: I suspect the reason for a lack of success in this area is that people - primarily nonprogrammers - expect too much from vague notion of 'components'. These people dream that components will somehow make programmers unnecessary. Instead of lots of unpredictable geeks writing code, neatly suited 'designers' would compose systems out of prefabricated components using drag-and-drop. For tool vendors, the added attraction to this scenario is that only they would retain the skills need to write those components.

The fundamental fallacy of this vision is that it is extraordinarily hard to design and implement components with a wide appeal. A single component or framework that does most of what is needed for an application or an industry would be attractive to its owner, and isn't technically too hard to build. But various players in that industry would soon realize that if everybody used those components, they'd have no good way to differentiate their products from those of their competitors. They would become purveyors of a commodity, and the main profits would go the component/framework supplier.

Tiny 'components' can be useful, but don't offer much leverage. Medium-sized, more general components can be very useful, but such components require great flexibility in their composition.

Lindholm: Java technology's success is in part of a consequence of what it doesn't do. The question that needs to be asked is, 'Is the feature essential, and what is the cost of adding it to the language?' Operator overloading in C++ is in some ways a strong feature, but it also adds a great deal of complexity to the language and is difficult for many people to deal with. By making an inspired set of choices among the various tradeoffs that could be made, Java technology hits a sweet spot of capability of need.

Lindholm: Java technology isn't revolutionary in itself, but rather evolutionary: almost all of its features have existed in at least one other environment prior to the emergence of the Java platform. Java technology's valuable contribution is the choice of the set of features and the tradeoffs that have been made in making those choices to produce something both practical and aesthetically satisfying. Java technology cradles - but does not coddle - the programmer.

Stroustrup: The key advantage of a formal standard, such as the ISO standards for C and C++, is that it makes it hard for a vendor to manipulate the language for its own commercial gain. It provides for multiple vendors, which lower prices for users and provides long-term stability.

The advantage of proprietary languages is marketing-fueled popularity, cheap facilities (until the users have become locked in), and rapid responses to perceived commercial needs and fads.

Formal/open standards primarily serve the users/consumers of programming tools. Proprietary 'standards' primarily serve the vendors."

Source: "The Future of OOP", Bjarne Stroustrup & Tim Lindholm, The Future of Software, Winter 2000/2001.

Bjarne Stroustrup is the original designer of C++ and Tim Lindholm was an original member of the Java platform development team at Sun. Their comments are interesting because it is true that the Internet age has brought to life a new wave of languages (like Java, C# (the .NET language), Perl, Curl, PHP, etc.) in a situation where the positions seemed relatively established with Cobol, C (++ or normal) or VisualBasic as major actors.

The true actual battle for the development of Web-based applications seems to be mainly reduced to a Java versus .NET fight. The commercial power of Microsoft should not be underestimated. It will be even freer to use it after the abandon of the anti-trust charges from the US Department of Justice.

*** Conferences *********************************************************

Given the recent events in New York, USA, Sinan Si Alhir (a regular contributor to the Methods & Tools newsletter) has freely volunteered to deliver two seminars in association with the "Thirteenth Week - On-line Seminars for the September 11th Fund" to raise funds to benefit the victims [http://www.thirteenthweek.com/september11.html]. The seminars are [http://www.thirteenthweek.com/seminars.html]:

* The Unified Modeling Language (UML) (Sinan Si Alhir)

- December 15, 11:00 AM - 12:30 PM EST

* The Unified Process (UP) (Sinan Si Alhir)

- December 15 - 1:00 PM -2:30 PM EST

Seminars on other interesting software engineering topics (eXtreme Programming, Project Management, XML, JAVA, etc.) will also be presented. You will find a complete listing at http://www.thirteenthweek.com/seminars.html