FORBES.com
"Online broadcasting"
Om Malik, 05.13.98, 12:00 AM ET

Technology veteran and former venture capitalist James E. Long has seen many streaming media companies come and burn out like shooting stars. In this business, you have to sift through the hype to get to the real deal.

As the chairman of Starlight Networks, a Mountain View, Calif.-based software company, Long has never lost sight of his main objective--to develop streaming products for corporations like Boeing, General Motors and Bloomberg.

Streaming media technology takes a sound file, photograph or video and breaks it down into small bits and pieces, compressing and sending it over the networks as a steady stream instead of one gigantic file. This means the user on the other end of the pipe is able to view or listen to the file as it transmits, not after the entire file has been downloaded and then executed locally on their PC.

Streaming media companies such as Real Networks, Microsoft, VXtreme (owned by Microsoft) and Vivo (now part of Real Networks) are users of the Internet. Long's target market is internal corporate networks.

If a company's CEO could be simultaneously broadcast over the network to every employee's desktop, why bother spending money on renting an auditorium?

His logic in going after corporate customers is this: Since corporate networks have ample bandwidth--they can move traffic at speeds of around 100 megabytes per second--they could easily handle broadcast-quality videostreams. Moreover, Long thought, these companies could save a lot of money if they used video-streaming technology.

If a company's CEO could be simultaneously broadcast over the network to every employee's desktop, why bother spending money on renting an auditorium? All it would take would be some software and a $500 videocamera.

Now dubbed IP/Multicast, this was a radical idea in 1991, those pre-Internet times when streaming technology companies such as Real Networks rnwk (rnwk) , Xing Technologies, Vivo, Vxtreme and VDOLive had not even been conceived.

Long and Starlight cofounder Charlie Bass (who also started Ungermann-Bass, one of the first Ethernet companies) pitched the concept to venture capitalists. [along with cofounder Mark Gang.]

Basing their technology on MPEG1 and MPEG2 video, the Starlight team came up with a proprietary technology to achieve high-quality video streaming. (MPEG--Moving Pictures Experts Group--is a standard for compressing video, and MPEG-2 is used in DVD movies.)

Excited VCs pumped in $2 million in seed money, and over the next six years the company raised an additional $20 million in five rounds of funding from VCs like Sequoia Capital, InterWest, Accel Partners and Star Ventures.

"Everyone loves Jim because he is not caught up in his marketing hype and is a pretty serious fellow," says Joan-Carol Brigham, an analyst with International Data Corp. (IDC), a Framingham, Mass.-based research firm.

Fast-forward six summers and streaming audio and technology are the hot buzzwords, and the stock of Real Networks has gone through the roof. Thanks to some acquisitions, even Microsoft Corp. (msft) has jumped into the fray. (See Trouble ahead for Real Networks.)

Even though the promise of the consumer space seems very intoxicating, Starlight has not strayed from its course, and has left the likes of Real Networks and Microsoft to figure out the puzzle.

Does that mean the company shuns the Internet? Not at all. "Thank God for the Internet because it would be difficult to do this business without it. Now customers are more aware of streaming media technology," says Long.  [Later Starlight incorporated Real Networks public internet video into their StarLive presentation product.]

Long explains that it will be some time before the Internet has enough bandwidth to stream broadcast-quality videos over the net (30 frames per second (fps), versus the current 28.8 RealVideo fps rate of 3.)

"The consumer market is a real dogfight," Long says.

The company has developed a $50,000 software package dubbed StarWorks. The software has already won widespread support. There are 300 companies--100 of which are already clients and 200 that are testing Starlight's software to stream video over their networks.

Even though the promise of the consumer space seems very intoxicating, Starlight has not strayed from its course.

One such paying client is Bloomberg Financial Markets, which uses Starlight products to cheaply deliver broadcast quality video over its network to its proprietary financial terminals.

The success at Bloomberg has also helped the company rack up contracts with marquee names like SmithBarney, Volkswagen and the Pentagon.

Prior to using Starlight's software, Boeing used to create 7,000 videotapes of news programs for worldwide distribution to its employees. Which mean buying videotapes and mailing them out to various locations--an expensive exercise.

Using Starlight's products, the same tapes are now accessed on demand from central video servers by employees--a move that saved Boeing approximately $136,000 per month in 1997 in reduced video replication and distribution costs.

The financial services industry has also embraced Starlight with open arms, representing almost 50% of the company's sales. (Government and education sectors come in a close second and third.) Brokerage firm SmithBarney is going to replace its hoot-and-holler trading system with Starlight's software--a contract that could be worth a couple of million dollars for a company which did upwards of $5 million in sales in 1997.

Long says he is hopeful that the company's revenues could touch $10 million by end of 1998, and it might be able to eke out a small profit as well. That could come in handy when the company goes public later this year, an event the VCs are looking forward to.

Long believes that this is just the tip of the iceberg, as more and more companies are likely to use streaming video and audio technology to train their employees. Even though the total industry sales of streaming software were only $100 million in 1997, Long predicts the market could grow tenfold over the next five years.

If anything, the strategy has certainly impressed Brigham, who thinks that the company has been smart in "finding a niche where Microsoft has no interest." Microsoft is busy promoting its NetShow video-streaming product in the consumer space, and as a result has left the high-end corporate market alone.

If the corporate market is where the money is, then it won't be long before competitors start making their move into Starlight's space. Real Networks is already eyeing the market and has started to make a push into the intranet space.

Not taking any chances, Starlight has added support for Real Networks and Microsoft's NetShow formats in its products. Using Starlight's software, companies can stream the video as either a Real Video or a NetShow stream, which can be seen via a web browser.

But that's not all, as a new threat is emerging in the shape of Percept Software of Palo Alto, Calif.

The company, which was acquired by Cisco Systems in March for about $85 million, is involved in IP/TV technology--a client/server application that sends live or prerecorded digital video and audio to a large number of users over any IP-based local- or wide-area network. Given Cisco's size and lock on the networking market, this software could prove to be an even tougher opponent.

"They (Starlight) have a small window of opportunity which can slam shut any time if they do not push hard enough, or Microsoft gets serious," says Brigham of IDC.

Long says the only way they can stay ahead of the game is by keeping a technological edge. For how long? That is anybody's guess.

​