Both distributed computing and webapps rely on the network - that's why I think it's relevant. For example, "the network is reliable" is applicable, given outages. I'd be interested in you articulating why each of the 8 specific points are irrelevant to webapps.

> it would be equally applicable to "native" apps as webapps.

I think you're saying that webapps use the network, but desktop apps can too (e.g. ping, telnet, ssh, ftp, netcat, email clients, napster, etc), and so "what's the difference?". I find it an interesting perspective, because it's noting that the web stack is just a stack for network communication. There's nothing magic about it, except that it's broadly implemented (though there's perhaps some special merit in its REST architecture). Bits of the stack are even used locally e.g. HTML help docs; javascript scripting in Acrobat. It's no different from all the other (thousands of?) protocols and standards. I think this perspective, because it's looking impartially from the outside, and not assuming anything special.

Webapps of course don't have to use the network (apart from for delivery), such as a Javascript calculator (or game for that matter). And, with RIA, you can have webapps that have local storage, and theoretically are indistinguishable from desktop apps.

Is this what you mean? Have I articulated your position?

While I agree with you in theory, in practice most webapps do use the network; and most desktop apps don't. You might think this doesn't matter, but I put it to you that apps are not real things - they are merely shadows of the need they meet. Over time, they move closer to meeting that need. Thus, because webapps use the network, the standards, protocols, third-party tools and products, and developer wisdom and training all get better and better at that. While desktop apps take advantage of being local, and get better at that (though, I'd say this is mostly in the past, and I see little improvement these days). All the wrinkles and little problems and recurring deadends get ironed out (and often absorbed into the stack, so later developers are not even aware there ever was a problem). Webapps get network problems ironed out; desktop apps don't. As for RIA, it doesn't seem to have gone anywhere - silverlight is almost non-existent, I've actually never seen a javaFX app in the wild, nor an adobe AIR.

Therefore (assuming you accept my above argument that FoDC is relevant because it deals with the network), FoDC is less applicable to "native" apps than webapps, because webapps usually use the network, and "native" apps usually don't.

I've been speaking of "desktop" apps - I guess it might be different for iOS iPhone/iPad apps. Do they tend to use the network more? I would guess they might, for games' highscores and so on, but even if they use it more than desktop apps, the issue for my argument is that they use it less than webapps. And from playing with an iPad, that seems to be the case. After all, native apps have the advantage of being local, and so it makes sense to play to that strength.

I note that this would have been a much more interesting exchange (and you would have likely avoided downvotes) if you had articulated your position more clearly, instead of me having to do it for you. I hope you don't mind this admonishment, but i you have found this comment of mine at all helpful, I request that you at least attempt to put in as much effort in articulating your response as I have. It would be enjoyable to have an interesting discussion on HN, instead of the pedantic focus on peripheral points that one often sees - and as I said, you have an interesting perspective.