Netbooks: A Subsidized Outcome For Linux And ARM?

Every time I mention netbooks, ARM and its army of licensees predictably chime in, attempting to persuade me that the product category isn’t x86-only (or for that matter, Windows-only). I agree with them, but only up to a point. Statements such ‘x86-only’ and ‘Windows-only’ admittedly overstate the netbook current situation and future forecast…but not by much.

To understand the reasons for my guarded perspective, you need only look at the ARM-vs-x86 comments I made (for example) last June, or at my critique of Celio’s Redfly last October. And for added ammunition, consider the recent experiences of an online student at Madison (Wisconsin) Area Technical College, who was stymied when she discovered her Ubuntu-fueled Dell system was incapable of running either Microsoft Word or a Verizon DSL software install utility. Techies like me and (many of) you might be inclined to giggle at her predicament (OpenOffice, anyone?). However, as I reminded you just a few days ago, we are the exception…she is the rule. And such is the crux of my ARM-and-Linux pessimism. Here’s what I wrote in February of 2008, in appraising Palm’s dead-on-arrival Foleo:

If it looks like a laptop, is my contention, people expect it to be a laptop. If it’s not, but it’s not appropriately priced significantly below the laptop counterpart, it won’t sell.

You can now buy a Windows XP-based netbook with a 10" LCD and 120 GByte HDD for a bit more than $300 (a notable downtick compared to fuller-featured notebook alternatives). Linux-equipped systems, mostly still based on x86 CPUs today, sometimes dip below the $300 mark…but look at the fine print, because they’re also likely to contain sub-10 GByte SSDs, and many of them are also open-box or factory-refurbished (a reflection, I’d wager, of their previously customer-bought-and-returned status).

Admittedly, recent developments such as Qualcomm’s acquisition of AMD’s embedded graphics assets, with Snapdragon one likely destination, are encouraging news for the ARM platform’s performance-vs-power consumption potential. But would a switch from x86 to ARM make a substantial-enough improvement in Linux-based systems’ cost structure to cultivate a significant-enough price gap between them and their x86/Windows peers? I daresay probably not…

…unless, of course, other economics factors also come into play. Specifically, consider subsidies, a phenomenon well known both in the game console world (where hardware is sold below cost in the hopes that future, highly profitable content sales will more than make up the deficit) and cellular phones (where handsets are also sold below cost, but bundled with lengthy, highly profitable service contracts).

What’s this all got to do with netbooks? Consider that right now you can buy a Dell Mini system with a 9" widescreen LCD for $99 (after rebate, normally $449…supposedly)…bundled with a two-year-minimum LaptopConnect service contract from AT&T that costs $60 per month. The data plan seems pretty pricey to me, considering that the unlimited-data deal on my T-Mobile phone is only $20/month and that I can Bluetooth-tether the phone to any laptop (well, almost any laptop). But in saying that, I realize that both the idea and implementation of phone-as-wireless modem are beyond the grasp of most consumers.

Specifics aside, I’d argue that the subsidy concept which AT&T’s partnership with Dell exemplifies is fundamentally sound. $99 is less than 1/3 the cost of the earlier-mentioned x86/Windows-based MSI Wind. Most netbooks include a PCI Express Mini or other internal expansion slot, making it simple for an OEM to optionally populate systems with EV-DO, UMTS or WiMAX add-in modules.

Cellular service providers are likely to go with ARM/Linux-combo netbook designs in striving for lowest-cost hardware (thereby minimizing their upfront losses), as well as to reduce the likelihood of viruses, spyware and other customer-frustrating (and, don’t forget, network bandwidth-consuming) annoyances. And the resultant reduced functionality won’t be the problem it otherwise would be for consumers, both because the cellular service providers will position the subsidized offerings primarily as always-online communications platforms (web surfing, email, chat, social networking, etc), and because the systems’ ultra-low price tags will curb purchasers’ expectations of their capabilities.

Thoughts, readers?