Talk

Web browsers have become huge, swell by the day, and we keep on stuffing them with ever more “features” at a very fast pace. The ultimate goal seems to be for the web to behave “like native” (in terms of performance and capabilities) — or directly building “an OS on top of an OS”. This is detrimental to end users, makes security much more difficult, creates the need for an ever-growing stack of tools, libraries, tricks, polyfills, etc. and therefore shortens the lifespan of the average web developer 4.2 years. OK, this last part may not be true. Definitely for developers, there may be frustration in trying to keep up with so many recent developments (APIs, tools, paradigms, jargon), as opposed to more constrained development platforms.

We argue that instead of pursuing that path, browser vendors and web developers should focus on the strengths of the web: URLs, simplicity, immediacy, platform-independence — and be more critical of the current trend. Particularly, the web could benefit from the UNIX philosophy: to combine “small, sharp tools” to accomplish larger tasks.

In this talk we'll survey the status of the Open Web Platform (OWP), and also showcase, tongue-in-cheek, some applications that take the web way too far. We'll look at the web ecosystem, its weaknesses and strengths, the handful of languages it's built upon, and the myriad of APIs it now includes. We'll see how this could be simplified, and whether it makes sense to keep on replacing so much native software with “web applications”.