Page 2

By eweek  |  Posted 2004-09-27 Print this article Print

So now youve got Seagate and some of these other vendors that are at least eyeing this space. Whats stopping people from taking the bad habits of the desktop space and applying them to the mobile market—the cutthroat pricing model, for example? I think before the IT industry learned to be cutthroat the CE industry learned to be cutthroat, out of necessity. Years ago, I worked for Kodak. We were getting into the videotape business, buying on an OEM basis tape from a Japanese supplier. We got into the camcorder business for a short while. I picked up a lot of things about the consumer electronics industry at the time. They lost their pennies a long time before the IT industry lost their pennies on each little screw, each plastic component and stuff. So I think that the purchasing, the engineering [departments] have grown very attuned to cost, theres nothing new here. Some of the practices have grown up with slightly different supply chain management warranties, and all that, but those differences sort of pay off in the long run.
Lets talk about flash memory for a bit. I think the assumption in the industry is that the disk drive industry must provide a significant measure of additional capacity compared to flash, or that flash is going to win. What sort of safety margin does magnetic storage have to maintain to prevent flash from displacing it?
I dont think we have any market research on that, so I dont think we have an exact answer. Certainly we have some customer feedback. On the Microdrive product we actually have some idea. We innovated with the 1-inch drive in the year 2000. We started shipping that in the summer of 2000, and into the consumer business there. We introduced the drive at $399 [as a 512M-byte] product, and almost three years later it was down to a $99 product. But in this case we had a large lead, and there was no other direct competition for a few years. We basically could observe over time and see what consumer behavior was like, because the original Microdrive had a 4-1 lead, what the capacity was compared to the price, and by the time it had reached end of life it had a 15 to 20 percent advantage. Flash had caught up—we hadnt moved. So basically by the time we went end of life we were pretty much at the breaking point—we couldnt really afford to sell it for much less, not much of anything was left, and flash was kind of taking most of the business. So, I think clearly a ratio of 2-to-1, 3-to-1, 4-to-1 is increasingly a comfort. If you get lower than 2-to-1—and this is my opinion—you can become increasingly uncomfortable. Now I think that the drive companies have to think about form factors as well as cost, and 10-, 15-, 20-year investments for the new form factor. You may say that it takes nine months to tool and build a factory and start making something, nine months to design the drive, plus youve got to factor in fixed costs and capital equipment. Theres a direct cost of all your product development at a point in time. But actually, for a product to have an economic return, you have to be innovating each year, nine months, 12 months, whatever cycle youre on, bringing the cost down while seeing the volumes come up and getting a return on all the investment that youre making as well as your capital additions youre making along the way …. So you need that time in order to pay [the investment] back. So if you look at the flash relationship, in a certain time well see a 2-1 ratio or better, but you have to see that continuing with confidence, over time. And you see the curves converging. I can see them getting closer, but I dont know how fast theyre getting closer—so am I really going to get 15 years out of the drive? The one-inch drives Im pretty confident of, were shipping a 4GB today, Seagates shipping a 5GB in relatively low volume, so that I guess with flash were holding that roughly 4-1 ratio, like we did when we innovated five years ago. So that gives you a sense of "I can do it." But to balance smaller drives, while watching the prices of flash memory—well, that gets pretty tricky. In desktop drives there once seemed to be a point at which there would be enough storage, a "good enough" point. Then digital video arrived on the scene and blew that away. Is the same trend going to occur in microstorage, or will we hit a "good-enough" point? I have a very simple answer, a very quick answer, that might require a little scrutiny. I think the handheld devices will track with the home; basically, its the supplementary use of the same content. Maybe some of the digital content comes from handheld devices; a digital camera creates digital content, a digital movie camera the same, maybe even a mobile phone with camera features. I think youll see more and more content stored in the home, in stationary devices or roughly stationary devices like PVRs and PCs and so forth, that will fuel more demand for handheld devices. In addition theres what Microsoft is saying—[a need for] higher-quality content, HDTV video and music with less compression. Microsoft has made two other points: that users will eventually sync all of their fixed and mobile content with content providers, ensuring their licenses are up to date; and that portable storage might not be necessary in a world where content can be streamed through a ubiquitous wireless network. Do you agree with Microsoft on this? Those are both fairly meaty points. The thinking here, I think, is that Microsoft is pitching this behavioral model, which maybe to some degree is tied into their shopping model. I think that its tied into their Janus [DRM] service. I think what they see is what Napsters doing, putting out a subscription service, putting all the content on the PC, and then transferring it onto handheld devices. Just because its possible doesnt mean people want it. Im not disputing it, Im just thinking that some people like the browsing and shopping experience just for choices sake. I think that Microsoft has got something that theyre evangelizing, something that theyre selling. We really dont know yet what the behavioral model is going to be like. So syncing, yes its important whether youre synching to play huge libraries that youve downloaded, you know, or whether its your playlist. So syncing is important, as the staged and portable host get more capacity, you can pull more and more data in. I think that freedom of choice is something that people prefer, especially when theyre away from their home station. So I think that that ease of syncing is going to be an important capability. I also think that the wireless piece is tied into views of the behavioral model that will prevail. ... Weve had the debates internally since we came out with the Microdrive, and we continue to have this debate about bandwidth, about how much local storage people will need and will there be a future where they wont need any local storage. Whats actually happening is that more and more OEMs are talking to us about embedding local storage in devices, devices that we never actually thought about embedding with drives. … So I sort of see a near-term view that OEMs are not going in that view but in the opposite direction, putting more storage internal to the device. So its a little bit of a cop-out; I dont have a projection here. But I do think that the wireless capabilities as well as high-speed syncing are both phenomena that will drive more use of the devices, and I personally believe that most devices are headed toward higher local storage. And, in fact, we talked about this earlier. A PVR relies on local storage. The PVR model says you dont want to be tied toward enjoying the content on a localized basis. So even if its content you can call up on demand, do you want to watch it for that streamed period of time or do you want a delay in it? Do you want to program your life so carefully that I want to download this movie or this album or this TV show, and say, OK, "Im going to sit down and just watch this." Youve paid for it; it would be nice to have a model that allowed you a few options. The last question I wanted to ask was about the Endurastar, Hitachis disk drive for cars. Tell me how many youre selling, what OEMs are using them for. A: Well, Im not going to tell you how many were selling. What are OEMs using them for? The majority of OEMs are using them for optional aftermarket navigation or entertainment systems. I dont know whats predominating—navigation or entertainment. A lot of the activitys in Japan, and thats kind of the leading market. Theres some you can get in the U.S., but not in high volume. But if you look at a navigation system, I tend to think of them as being used in high-density urban areas, where you cant memorize all the streets. With an interstate highway, you can pretty much memorize it. Beyond that, you need more detail&. #133; We know that major auto manufacturers are watching this closely, and we believe in future model years these will be built in. Check out eWEEK.coms Storage Center for the latest news, reviews and analysis on enterprise and business storage hardware and software.

Be sure to add our developer and Web services news feed to your RSS newsreader or My Yahoo page


Submit a Comment

Loading Comments...
Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel