Friday, September 28, 2007

Sun's McNealy Returns

Well Scot McNealy of Sun Microsystems was back in town...and closeted with a lot of the cities tech big wigs (LUS, LCG, the University, and local business—tech enthusiasts) for a couple of hours before a press conference at LITE. Sorry I didn't get to this earlier, but I was mired in a recalcitrant web site that was too close to launch to neglect. But luckily the regional media covered it in force. What happened in that meeting—why McNealy made a return trip—was not immediately made public though hints could be gleaned from the reporter's coverage.

The Advertiser lead with and focused on the announcement of Lafayette's ranking on a jobs growth ranking and didn't mention the McNealy press conference, at which the ranking was mentioned, until paragraph five. KATC and The Advocate lead with the McNealy visit itself and didn't mention the job growth ranking which was apparently a reference point in the presentation. The two stories do dovetail, of course, but the focus of interest on this site is the technology issues implicated in the visit.

Seasoned readers will recall that McNealy made a supportive stop here right before the fiber referendum. He appeared on one of Joey's morning radio shows and was generally encouraging about our building a fiber system. Back then I laid out an enthusiastic, but I think still pretty accurate assessment of the potential of a Sun-Lafayette partnership. The gist is that LUS' big bandwidth, Sun's open source source software, and the immense potential of on-system storage and distributed computing in Lafayette's intranet has got to have smart companies like Sun thinking hard about using Lafayette as a test bed for new technologies. There really will be little to match the size and diversity of our user population, or the intranet-speed in-system bandwidth supplied between customers. That is a match made in heaven for those that have hankered after the bandwidth to make real changes in the (computer, video, cloud computing, name-your-techish-dream) area.

Sun's bread and butter has been building top-notch servers, and more recently, integrated server farms. That's a business built on the need for fast networks. Sun has in recent years pursued some pretty interesting ideas pretty relentlessly. Sun signed onto the open-source movement early. Free and more importantly open, readily fixable and extendable software is the result. Sun has also swum against the tide in insisting on a pushing a "network-centric" computing model. This involves big central computing facilities and distributed dumb terminals — though some Sun models can run as traditional independent stand-alone computers. Sun also has relentlesly pursued its vision for JAVA. The hope was for a platform for writing software that was independent of the underlying hardware and could run and interconnect processes on everything from toasters to big iron server hardware. JAVA has yet to becom the platform for realizing the more blue-sky versions of those dreams but much of the intuition is being realized in web-centric AJAX apps.

The potential of having a whole community with fast, cheap, universally available broadband capable of ripping the roof off the network limitations that have kept many of Sun's ideas barely viable has got to be tempting to the company. And the digital divide and development potential for Lafayette are obvious. There is surely partnership potential here.

But what is on the table now? I'd guess both LUS' fiber program and the city's computing needs.

Keith Thibodaux regularly complains about the need to update a creaky computer system. The dark lining on the silver cloud of having had an early strong computer department at ULL is that Lafayette's networks were developed back in the days of COBOL and significant portions of the city's core network runs in that crusty framework. Slipping in a modern Sun-based but still centrally organized, terminal-heavy system would allow that sort of mainframe-oriented system to move into the modern day relatively painlessly. As the tenders of that system reach retirement age (yes we are that far into the computer age) such a move might become critical.

The Advocate did a stellar job of focusing on the potential interaction of Sun and LUS' fiber to the home project. I recommend you go take a look. It is exciting stuff and doesn't bear much cutting here is a stream the good bits:

Durel said Wednesday that the project’s highest-profile cheerleader reinforced and supplemented the LUS team’s “vision” to not just provide “me too” products with the state-of-the-art network.

“It’s not just about saving customers 20 percent,” Durel said. “It’s much, much bigger than that.”

Durel said McNealy is a big fan of “open source” products, software allows tech-savvy users to upgrade and add their own innovations.

In an open environment, coupled with the vast bandwidth promised by LUS — which has said that traffic inside its network will be unlimited — there’s a great potential for people working out of their garages to develop innovative products in Lafayette, McNealy said.

LUS Director Terry Huval said McNealy talked about the potential for Lafayette schools to utilize curriki.org, which provides free, open source educational materials.

McNealy said Sun Microsystems offers a product called Sun Ray that could also be of great use with LUS’ system to help get more people using technology in their everyday lives.

Sun Ray is a simple, low-cost computer that serves as a conduit between the user and a massive server, where all information, software and processing power is stored.

The interactive display of Sun Ray is merely a way for the user to tap into the network, meaning that any user — with a pass code or swipe card — could use any Sun Ray to access their information, be it at home, work, the library or wirelessly, Huval said.

It's a grand dream and could get most of the city on the network in an extremely exciting and potentially sophisticated way. Serving (free) programs off a server to inexpensive computers is clearly the next step a city could take after offering cheap, universal, big bandwidth. Open source is the way to go and Sun is a leader. Partnering with someone who not ony cares about these ideas is a natural--especially when that partner has already bet the company on the ideas.

As always there are caveats, especially in the context of the digital divide: Sun's terminals are inexpensive--but no longer notably inexpensive in comparison to arguably more capable standalone computers. (And their standalones are more expensive.) The most price-attractive hardware is proprietary and not all open source material is ported to run there. It is a pretty closed ecology without the diversity found in the larger computer market. And it isn't clear what direction will be open to Sun as the mobile market continues to expand.

Without a doubt, it's all exciting and the relationship with Sun will bear watching.

No comments: