Daring Fireball: The Location Field Is the New Command Line:
The conventional wisdom was in fact correct — the web has turned into a popular application development environment. Where I’d gone wrong was in getting hung up on the idea of it needing to be high-quality before it could become popular.
I was thinking in terms of the apps that I used every day, circa 1996: BBEdit, QuarkXPress, Photoshop, Eudora. There was simply no way that a “web app” could ever provide the same quality experience as the “real” apps I was already using.
Amazing that someone would think that their application experience — running graphic design applications with no mention of anything related to business software — was typical and that, based on that, web apps were doomed to failure. Did John Gruber not realize why Netscape came in every flavor of OS under the sun? Had he never heard the quote about relegating Windows to a “poorly debugged set of device drivers?”
My own experience is somewhat different. When the web first came to my attention in 1994, with the release of Mosaic, I was working at a small manufacturing concern, in the throes of a major overhaul of their business systems. They had elected to move to a platform neutral system called ManMan/X (for Manufacturing Management) that ran in X-Windows: it could run on anything that supported X, which was, well, everything. Even Windows 3.x would be persuaded . . . .
Obviously, the right idea, just a little ahead of its time: in 1994, CGI was all there was for interaction (PHP and MySql weren’t even a gleam in anyone’s eye: I think Linux was still pre-1.0), so even if someone had twigged to the idea of the web, making it happen would have been hard.
Of course, later in 1994 and into 1995, this little company called Netscape (née Mosaic Communications) came along, as serial entrepreneur Jim Clark saw the potential of a deployment platform that rendered the OS irrelevant. The idea of everything running in a browser window — office applications and the like — still seems a ways off, and it’s obvious that word processing and spreadsheets have run their course, as far as innovation goes. But data collection, reporting, and presentation, database interaction, all the various facets of CRM, HR, and a host of other acronyms — including the kind of basic data management we were working with 10 years ago with an X-based solution — all work just fine in the browser.
I think John hits it dead on, as he points out how MSFT may have won the battle but lost the war, but I’m still surprised he missed this early. And it’s never a good bet to count MSFT out of the race.