Whit Andrews

It will come as no surprise to our clients or fans (we have fans!) that our internal discussion process is fecund but sometimes a tad groping. I’m looking at a 90+ message thread view of our (main) email conversation on Google Chrome OS that contains terse pith from some colleagues and scrollworthy analysis from others. While we’re working on how best to distill this into our lamentably brief First Take format and notes beyond that, I’m going to take a moment to mash this up into a collection of issues we are internally mulling.

1. Chrome OS, as proposed in Google’s spare blog entry, is not like any OS that has preceded it from a major vendor. One analyst puts this as “this is not your father’s operating system,” and the use of the trope made famous by the Oldsmobile ad featuring Lisa Marie Presley (oh sigh!) is a shortcut that is useful mainly as shorthand, but which stops short of insight. To take it farther, the point — as Ray Valdes expressed it to me this afternoon when I interviewed him for the First Take we are developing with Mike Silver and dozens of analysts’ input and review — is that Chrome OS is proposed as a vessel that is deeply associated with its contents.

I have tried to find the proper metaphor for this. (I am a literature major and I survive at Gartner and in IT on such somewhat reductive but fanciful thinking.) In essence, the idea is that Chrome OS presents a means of establishing stability and predictability that guarantees Chrome will always have a home within which it can serve users. The biology buff might see this as a sort of constructionist commensalism. In commensalist relationships, one species benefits from the relationship, and the other doesn’t much care. The Chrome browser would always benefit from Chrome OS, but the Chrome OS doesn’t get much benefit from the browser.

Why would this matter? Well, for one thing, it means that the Chrome Browser needs to be able to live as many places as possible. It is already an OS-like browser, as Ray explained in his own excellent (and less fanciful) blog entry. Such a development will come as no surprise to those who remember the fact that Netscape, as combined with Sun’s Java, was perceived as the very hopeful creation of a similar kind of commensalist relationship — client-side Java could live in Netscape or in other Java virtual machines. Java would always have the safety of the Netscape browser to retreat to, not unlike the clown fish can always retreat to anemones. (Whew. Got outta that metaphor with all my fins intact.)

UPDATE: Another way to characterize it, David Mitchell Smith notes in his blog entry, is that Chrome OS is the first OS in a long time not also to carry the burden of being a “platform.”

2. Chrome OS is currently what I think one might charitably call “postware,” as in “software which primarily is represented to the public as a concept described in a blog post.” I wanted to call it “blogware,” but Wikipedia says that’s one way of describing stuff you’d use to make blogs with.

There is nothing wrong with describing a piece of software in a blog post. I noted in my own post on this particular issue that there are at least two good reasons to inform everybody with access to the Internet why you are starting a new operating system. My colleague Mark Stahlman — who has astonishing depth in the impacts of technology companies’ actions on public markets — pointed out today in an internal meeting  that by telegraphing their plan to create a free operating system (NEW OS TO COME SOONEST STOP TELL WORLD STOP STILL WRITING IT FULL STOP), Google immediately puts pressure on Microsoft’s ability to command the price it will want to get from its partners for Windows 7 installations. And lo! Google itself has told us in its Google Chrome OS FAQ that it intends to work with, at least, “Acer, Adobe, ASUS, Freescale, Hewlett-Packard, Lenovo, Qualcomm, Texas Instruments, and Toshiba.” There’s nine vendors who will feel a little more comfortable waiting two or three rings before they pick up when the Caller ID shows a Redmond number than felt that way before that FAQ went up.

But postware brings with it some serious issues. Our advice is not crystallizing around any sort of frantic reaction to Google Chrome OS, especially for company IT people. I looked at our research last night (I love working here), and we rarely counsel people to skip OS generations because of the increased cost of leapfrogging. Meanwhile, Ray’s saying it’s going to be three years, at least, before this has any impact in the average enterprise. And yet our own phones blew off the cradles when Google posted the blog entry. The media, at least, is excited about the possibility of resurrecting the Operating System Wars (note to music: cowbell here).

It’s just a blog post, so far. Reading my colleagues’ emails, I can tell you we’re fascinated by the very real possibility that something profound has changed. We see the potential for a lightweight OS that is mostly — well, probably TOTALLY — intended to support Web use as empowering a category of fantastical devices that Yefim Natis termed “browser appliances.” Such appliances, in just my opinion (not the company’s or any other analyst’s), would serve Google as a means of getting advertising umpty-bazillion places it isn’t right now. I see them on digital signage; I see them in toasters; I see them in shower TV’s.

But it’s all just a blog post. For now, it’s an energizing and exciting moment when we have nothing but the idea of a new operating system and a new browser from a popular company with bulging freighters loaded with cash circling the globe. But it’s just a blog post, and for many analysts, that alone is enough to demand not just caution, but active and intense skepticism.

3. There are a lot of reasons to start over with operating systems. Existing operating systems demand a bunch of electricity (at least collectively) and cooling systems. Ray did a back-of-the-email envelope calculation on what kind of code silhouette Google Chrome OS could present to manufacturers, and there’s the potential of a significant reduction in size, on the order of one-tenth the size of Windows 7, or about the size of Windows 3.1 NT. As Yefim Natis said, Google doesn’t want to battle Windows in its fortress redoubts; it simply wants to render them less relevant. And as Brian Gammage wrote tonight, Microsoft is far from intending to stand still.

Some analysts like to characterize this as a necessary element in what they see as the opportunity to create a “cloud operating system,” which is referenced in our written research mostly (as of this instant) in the context of Windows Azure. I couldn’t be happier than to tell you that I am weak personally on cloud computing, (do you see my literature degree? here, let me hold it up higher, above my political science degree) and when you combine that with operating system discussions, I skid fast. Let’s put it like this: If you, the user, can get to the cloud and orchestrate computing within it such that it can combine your identity, the application elements you need and the data to which you have the privilege of access, then it’s all good.

But technology is just the start of the reasons for having an alternative to Windows (and for that matter, Linux). Money’s also a factor. As Leslie Fiering pointed out, manufacturers are tired of paying a tithe to Redmond every time they want to sell users another device that runs useful software. At the best, an effective OS alternative would free them from the tax altogether. That’s not likely for a very, very long time, because as another colleague pointed out (not in contradicting anyone — just raising the issue) the substantial majority of client apps were written to run on Windows and its variants — not in the cloud. At the worst, though, Chrome OS would pressure Microsoft to charge less for Windows. Remember what Mark pointed out way the heck back up this sprawling entry — Chrome OS can attenuate Microsoft’s pricing power even while it is still “postware.” (Gotta trademark that.)

Let’s assemble all of this, now. (Remember: Analysts are skeptics for a living.)

1. Chrome OS is, as described, an operating system of an entirely new kind.

2. Chrome OS currently only can be detected in the actual world via a few posts on official blogs by one of the most powerful and popular companies in the world.

3. The idea of a new kind of operating system is extremely exciting for lots of reasons that are probably invisible to non-IT folks but aren’t to us (nor to you, probably, dear Reader), which is why we’re here to help companies understand how to translate those technological shifts into business opportunities.

So, this is how the story now reads: In a blog post, Google is describing a new kind of operating system with very spare functionality that as a result could have a very, very small footprint in a device and could profoundly change the market.

This is simple: Google’s posted something on a blog about a technology that very few people understand and even fewer people need to care about, as Jeff Mann points out acerbically and a bit dolefully. (Brian Gammage agrees.) It’s describing a fundamentally new strategy for such technology, which Ray Valdes is examining deeply. The strategy follows one of Google’s core tenets, which is that people want things to be as simple as possible. Possible impacts of the operating system — and even just the blog entry! — include deep shivers in one of Microsoft’s areas of greatest dominance, an increased number of devices that can effectively use the Web and Web applications, and a final draft of the Universal Theory. (Gotcha! You read this far!)

We’ll write research behind and forward of the paywall this summer. Stay tuned. (Or, perhaps, “Don’t turn off that feed.”)