Although I can never quite remember from whom I pilfered the question – my best guess would be Jonathan – it’s one of my favorites. If you’ve seen me present more than once, you’ve likely heard it more than once – my condolences. Fortunately enough for you, the question is also an easy one: What is the most popular application in the world? The answer, by my – and, presumably, the originator of said question – most unscientific reckoning? Google. Or Google search, if you’re one of those types.
Whether you agree or disagree with that particular assertion is in some respects besides the point, because the implications are not, interestingly, all that dependent on the answer actually being, you know, correct. Even were it #2 or #3 or #5, it would remain simultaneously a powerful indictment of more formalized taxonomic discovery technologies (the basis for the application’s popularity) and a fundamental validation of the value of multi-platform, multi-architecture technologies (the basis for the application’s reach).
The latter points holds the greater interest for me. Not really because the popularity of and desperate need for search – at least in retrospect – is nothing if not predictable. But because the very underpinnings of Google’s success have had little visible impact on some project and vendor ambitions, and a major impact on others’.
Originally I’d planned to tackle what are admittedly orthogonal subjects in separate entries, but time is always against me. And they truly are related, at least peripherally. To see why, let’s turn to my old friend, the Q&A.
Q: I’m not sure I followed the thesis above: can you explain it in more detail?
A: It makes more sense when the individual datapoints are revealed, but the essence argument is this: Google’s search doesn’t care – at all – what your desktop is. The main reason I’ve declined to dismiss Google as a serious threat to Microsoft – among others – has been their tactical approach. Rather than trying to out-Microsoft Microsoft on the desktop, as a small horde of firms have attempted to do over the years, they leverage a new platform to make Microsoft less important.
Q: Less important how?
A: Google, along with more or less anyone designing applications that will run in multi-platform browsers, is diminishing the desktop’s role in the consumer and enterprise spaces. I don’t even consider this point debatable; look no further than the reactions Microsoft folks have had to Apple’s recent PC vs Mac campaign (this is far and away my favorite), or the fact that Dell will be shipping Linux desktops. You’re probably thinking, “but this means nothing from a marketshare perspective,” and while we could argue the semantics of your usage of ‘nothing,’ I’d generally agree. But it means everything from a control perspective. Dell shipping Linux for desktops would have been unthinkable a few years ago; these days, it’s merely unusual. I attribute much of that shift to Software-as-a-Service: much of what users used to transact via rich clients such as Office or Outlook Express is now done on the web.
Q: Ok, I understand the threat that SaaS generally and Google specifically play to traditional desktop vendors, Microsoft most notably, but how does that tie back to current vendors and projects? How have they failed to learn from Google’s success?
A: Failed is probably not the appropriate term here, but I am surprised at the continuing interest – fierce interest – in the desktop space. Where I’d believed that the trend towards the browser as platform would continue apace, there’s instead been a resurgence of investments – dollars and people – in non-browser application platforms. Platforms, essentially, that are one abstraction layer up from the operating system.
There are a handful of technologies that would see themselves crowned as the desktop of the future, or as Ted eloquently put it, the “Microsoft of the Web.” Which is a bit ironic, because while Microsoft is furiously striving to maintain its dominance in a world where what operating system you’re running on is often less important than which browser – see the generally unenthusiastic response to the near-decade-in-the-making Vista – a diverse set of players is seemingly attempting to achieve similar dominance, if not reimplement the Microsoft model one layer up.
Q: Can you provide examples of these “desktop” efforts?
A: Sure. Think Adobe’s Apollo, Eclipse’s RCP, Microsoft’s WPF/E, Sun’s J2SE and possibly even Mozilla’s XUL-Runner.
Q: Ah, so you’re talking about the RIA – the Rich Internet Application space?
A: Probably so, although I can’t say that I understand that term in any sort of concrete way. If the intent is to describe non-browser applications that are reliant on network functionality, then yes.
Q: So if I understand your argument correctly, you’re saying that these frameworks are pointless given the increasing popularity of SaaS type applications? That the writing’s on the wall, and it’s the browser?
A: No. There is now, and there will be for the foreseeable future, a need for rich client applications. Browser based applications have done a marvelous job over the past few years addressing some of their less forgivable shortcomings: interactivity (Ajax), featureset (plugins), and most recently offline access (Dojo Storage/Flash, Slingshot, Zimbra, Firefox 3.0, etc). But I’m not using one to write this post (that’s in Scribes), or handle my running IM sessions (those are in GAIM). Even should every conceivable technical limitation be remedied, however, there’s still user comfort levels and preferences. Rich clients, in other words, aren’t going anywhere.
Q: So what’s your contention then?
A: Merely that the level of interest and ambition is puzzling, as far as I’m concerned. Consider the following comments:
Adam Bosworth:
“My mother never complains that she needs a better client for Amazon. Instead, her interest is in better community tools, better book lists, easier ways to see the book lists, more trust in the reviewers, librarian discussions since she is a librarian, and so on.”
Tim Bray:
“On more than a few occasions—most recently in the context of Avalon—I’ve observed here that both IT admins and end-users prefer browser-based apps to traditional compiled clients, for everything except content creation. Every time, I get emails and incoming pointers from people saying ‘You just don’t get it, the Web interfaces are so tired, we really need a richer UI paradigm.’ The interesting thing is that these reactions are always—every time, without exception—from developers. Not once has an end-user type person written in saying they wished they could have a richer interface like the kind they used to have in compiled desktop apps.”
Then ask yourself the hard question: are these technologies in search of markets, or markets in search of technologies? Obviously I lean towards the former, but your mileage may vary.
Q: Are you contending then that the status quo is acceptable?
A: No, absolutely not. My objection is perhaps more appropriately defined as relating to the ambitions of some of the would-be players here. Put another way, I can’t say that I’m eagerly awaiting a future in which I need an Apollo runtime for one set of applications, an Eclipse runtime for another, a Java implementation for yet another, and a WPF/E runtime for yet another…oh wait, that doesn’t support my platform. And oh yeah, a browser too. Maybe two for the irritating one or two applications I use that are single browser targeted.
Q: What about the general ambition for local clients to become more network aware?
A: Could not be more strongly in favor of it. And I have been for a long time. Take the latest news out of GNOME-land: Havoc Pennington charts the course for a GNOME desktop version oriented towards online applications, manifested in a project called Big Board. While there is some dissatisfaction with the project’s lack of collaboration with existing solutions and focus on non-open source SaaS solutions, it’s clear to me that there’s an opportunity here, and a significant one. As I was discussing with Jeff the other day, why can’t EDS speak to Google’s Addressbook (particularly given that Joe’s got you halfway).
This was my precise train of thought, in fact, when I proposed that GNOME introduce Flickr as a potential backend for its wallpaper function (which Calvin Yu was kind enough to build).
As an aside, it’s worth mentioning that I think GNOME and other open source desktops have an inherent advantage here versus commercial OS’s like OS X or Vista, in that they’re not required to advantage certain SaaS options – quite the opposite. But I digress.
Q: So you’re in favor of network aware applications, just skeptical of the runtime ambitions?
A: Just so. Particularly lately, because seemingly every technology project and product on the planet is headed at best speed towards a similar opportunity. I’m never one to say “there can be only one,” but I think the four or more we’re likely to see is a few too many.
