Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Wind River going up, up, up! -- Ignore unavailable to you. Want to Upgrade?

To: Allen Benn who wrote (2901)	3/15/1998 11:57:00 PM
From: Allen Benn	Read Replies (1) \| Respond to of 10309

The Evolution of the Internet - Part 3: Implementation Early experiments with smart packets will use general computers as nodes and languages like Java with Just-In-Time compilation. But to be ready for prime time, smart packets probably need their own language and they will impose numerous requirements on the processing environment for intelligent nodes. Smart packets require a robust, relatively safe programming language that embodies functionality common for filtering data on the Internet, particularly when low-bandwidth is a hindrance. This suggests the language should be oriented toward handling multimedia constructs and screen layouts. In fact, PostScript might be a better starting point than Java. (An interesting digression is that Adobe recently chose to begin deploying PostScript for printers on top of VxWorks.) Robust automatic preparation of smart packets is particularly important, not to protect the Internet, but to make customized smart packets practical. Sending poorly performing or buggy packets off to do yeomen work will cause enormous aggravation. Perhaps the best way to perpetuate robust smart packet programs is to provide support for code-simplifying techniques like objects and aspects. Object Oriented Programming (OOP) is familiar turf, but Aspect Oriented Programming (AOP) is a new complementary approach that provides a means of succinctly specifying objectives that cut globally across most functional objects. Coincidentally, AOP is a research activity at Zerox PARC, funded by DARPA. (See darpa.mil ) For smart packets to reach their full potential, intelligent nodes must reside within a multi-tasking, real-time processing environment, available on most nodes at extremities of the Internet, including most sources and sinks (clients and servers). Further, the intelligent node environment must be practical to update in software, as new versions occasionally need to be perpetuated throughout the net. In short, what is needed is a ubiquitous, multi-tasking, real-time operating system that can implement intelligent node processing in software as tasks. By ubiquitous I mean it must be available for virtually any Internet device, ranging from PCs to NCs to handheld devices to Embedded Internet Devices to cable modems to routers to servers. Smart packet handling, the revolutionary next step for the Internet, clearly is a job for I2O and IxWorks. Low-cost, high-volume devices can integrate I2O onto multi-functional chips to reduce costs, but IxWorks would still be available. This means smart packet communication interfaces will be practical for all devices, not just relatively expensive ones that deploy separate I2O chips. Along this line, don't forget that the StrongARM I2O chip is specifically designed for low-power, inexpensive devices. While no one is predicting a huge market anytime soon for intelligent nodes (based on I2O or anything else), still it is interesting to contemplate future implications of such an evolution of the Internet. If this were to happen, I2O may well become the intelligent core connecting most Internet nodes off the backbone, including wireless devices, PC's, NC's, routers, intermediate servers, or anything else that accesses the Internet. In numbers, instances of I2O would skyrocket into the multi-billions. Although it is premature to build this expectation into the price of the stock, its nice to know there is a pleasant upside to an evolving Internet. Whether smart packets promulgate or not, they provide a more compelling example of why I2O may become ubiquitous than efficiencies derived from off-loading CPUs. Allen