This morning, Tim O’Reilly and David Pogue addressed the more than 700 attendees of the O’Reilly Mac OS X Conference in Santa Clara, California. O’Reilly, founder and president of O’Reilly & Associates, discussed what Mac developers can learn from the UNIX/Linux world and what UNIX and Linux developers can learn from the Mac. Pogue’s keynote was a lighthearted but sometimes keenly insightful history of the Macintosh and where it may go in the future. The audience was composed of both the Mac developers and the UNIX and Linux coders that this conference is intended to unify.
| <?php virtual(“/includes/boxad.inc”); ?> |
Speaking first, O’Reilly began his discussion by tracing the few truly revolutionary ideas that have shaped the course of both scientific discovery and computer technology. Each of these ideas triggered a paradigm shift, and O’Reilly said that paradigm shifts in technology have lead to changes in the way people use computers.
Citing Clayton Christenson’s work, “The Inventor’s Dilemma,” O’Reilly outlined how there are sustaining technologies that make existing methods better and disrupting technologies that are completely new ways of doing things.
Disrupting technologies often start out not working as well as the technologies they replace. New users regard them as toys, as was the case with early personal computers or web browsers. However, disrupting technologies eventually get better than the original.
Using the disruptive idea, O’Reilly went on to say that numerous paradigm sifts have influenced computing, ultimately bringing us to the open source UNIX model that we see today on Mac OS X.
For example, the personal computer commoditized computer hardware. People used standard hardware instead of specialized hardware that required specialized code. Early PC companies looked for the “killer app,” or the application that would make the PC so important or useful that people would buy it.
“We have still not yet shed the mentality that the killer app does not have to live on the PC,” O’Reilly said.
O’Reilly went on to ask if a web browser was the killer app or was it the content that it browsed.
While refraining from an open source sermon, O’Reilly said that there are a few things that Mac OS X developers should adopt from their UNIX and Linux counterparts.
“What’s fundamental about Linux is that it is commoditizing the software,” O’Reilly said. “Licensing is a holdover from the last paradigm.”
Most importantly, Mac developers and Apple should work to keep the Mac OS X Kernel small, with a clean, clear separation from programs. “I still like the idea that you can see through the OS,” O’Reilly said.
Further lessons that Mac developers and Apple can learn from UNIX and Linux is to adhere to standards and keep things small and modular. Also, Mac developers should document as they go with MAN pages and RFCs.
O’Reilly likes Mac OS X, namely its UNIX underpinnings, Apple’s commitment to wireless with Rendezvous and the digital hub concept, the i-apps, and Sherlock.
With Mac OS X’s UNIX core, O’Reilly said that Apple’s best move was to embrace a protocol centric model instead of APIs as Microsoft has. “It’s not just UNIX underneath,” O’Reilly said, “it’s the philosophy at the heart of Unix that’s important.”
In conclusion, O’Reilly said that Mac OS X is a great platform. To keep it that way, he said developers need to think platform, think network, think open and play well with others.
Pogue followed up with an entertaining history of the Macintosh, most of which should be well known to the Mac Central audience.
He then outlined the development of the Mac OS from System 1.0 to Mac OS 10.2. This included conventions that were lost between OSs that Pogue would like to see back. He also listed a few things that Mac OS X could learn from Windows XP and hundreds of items that XP could take from the Mac OS.
Pogue also gave predictions for Mac OS XV, such as documents yellowing to show their age, icons that better reflect their contents and size and more.
For the “deep future,” Pogue plotted curves that said by 2020, RAM vendors would pay us to take RAM, we will have 55″ monitors, and processors so fast that, “they get as hot as the sun and melt towards the core of the earth.”
On a more serious note, Pogue predicted more use of RAM drives that simply connect to the back of a Mac on a “nubbin” similar to how USB key drives work today. He also said that speech recognition is not the future and that battery problems will not go away in the near future.
Finally, Pogue said that Bluetooth is a “sure thing” for future development, and he went on to describe how it works.
“I can’t say what, but in the next two weeks, one of the major makers of something important in your life will start making something that has Bluetooth standard,” said Pogue.