To: PDG who wrote (682 ) 7/27/1998 2:52:00 AM From: Kashish King Respond to of 5102
PDG's Problematic Examples HP is one a few companies in a position to unilaterally influence the direction of Java for several reasons. First and foremost is that the company has a highly proprietary leaning in everything from operating systems to languages. Whether it's HP UNIX or HP BASIC, they clearly have penchant for the proprietary. They were poised to take the same path with Java but apparently the market has convinced them otherwise. If you've ever used HP's proprietary systems you will join me in cheering that policy reversal. We really don't need C++ only better, Java only better, IIOP only better, TCP/IP only better or XML only better. U c, u mai b beter or mor fishent but at wut price? We can take just about anything and make it better until we're blue in the face; what we'll have is a proprietary hack which nobody will want to use. The standards body is there to ensure that what goes into the language/environment is in the best interests of the users and not one particular company. The way to make your system "better" is to add "better" components built using the latest Java standards. That is, you adhere to the standard and compete above it. Now, as for examples. There are no restrictions on what can be done with a Java technology versus C or C++ in terms of applications. That should be obvious. On the other hand, nobody is saying you can write excellent software using SmallTalk, C with ActiveX or Visual Basic with ActiveX. I certainly think that my ActiveX and C++ based products are of excellent quality. The question is how much time and effort is required to achieve the same results as those working with Java? How much time and effort is required to extent those results in future releases? Whether you're talking about stability, security, threading, networking, or reusable components, Java wins hands down. In terms of support for the object-oriented and distributed computing paradigms, Java wins hands down. In terms of mapping designs into code, Java wins hands down. What the customer does understand are costs and time to market with new or improved products which meet the required level of quality. So, I don't see what examples have to do with it. It's never been an end-user issue. I hope nobody ships software that they know has some serious bugs or performance problems that make it unusable regardless of what language/environment they're using. It's not about the final result, it's about the cost of delivering it. It's about what you can deliver under given set of financial and chronological constraints.