Re: Ringmark is now open source

On Apr 11, 2012, at 19:12 , Charles McCathieNevile wrote:
> Certainly, there are some classes of application that won't currently run on Opera Mini. But there are some kinds of applications that do, and run very happily.

And having re-read this thread I'm pretty much certain that everyone's said that Opera Mini can run some kinds of applications very happily. In fact, I've heard nothing but praise for it. I really love Mini. That said, there's also a reason why I stopped using it when I got a slightly more powerful phone...

> I don't think it is a problem because it is "elitist", but because the group characterises its goals in a broader way - in other words, it is misrepresenting, or allowing too easily a basic misinterpretation of, what we are actually doing here.

That's fine, if we can simply rephrase a few sentences and move on to getting work done I'm all for it.

> This is the key question - what should be in Ring-0? And this is where I think we are starting in the wrong place. Beginning with the browsers made by the people who make the platform we're trying to bring to the Web is, in my opinion, getting a little too cosy for comfort. Most of the Web is not specific to those browsers, and our goal is presumably that none of it will be. Otherwise why would people whose goal is to build the Web be interested in this work?

Sure — I don't think that anyone wants to make the Web specific to these browsers. In fact, making sure that we can get over the WebKit monoculture would be quite valuable. That said, I really don't see how we could be helping by lowering the bar.

The Web ought to work at all strata. Targeting content at all strata can be difficult, and involves a number of trade-offs, but that's not the problem we're here to solve. The Web works actually pretty great in the feature phone stratum. It works pretty damn well on the desktop too. In both of those one can find decent levels of interoperability (for well-crafted content) and graceful degradation, as well as a reasonably healthy diversity of implementations.

There is, however, a stratum in the middle where things aren't as rosy. We have the weird conjunction of an implementation dominating the market *and* feature fragmentation at the same time (I don't think anyone saw that one coming a decade ago...). Producing better alignment there will make a lot easier to target content at that stratum, and in doing so will *also* make it a lot easier to target content across the board, with much better progressive enhancement capabilities since they will be much easier for develop to.

Boiling the whole One Web at once is too hard, but we can actually keep it much more "One" by boiling smaller pieces. This happens to be an important piece, too.

> If this group simply documents some lowest common denominator of Safari and the Android browser, that isn't a useless task. But it is something that a large proportion of developers are not very interested in (since they are required to serve a broader market that includes other phones). Browser makers other than those two will only very peripherally bother with it. The risk is that this leads to the group remaining a documentation group for those two browsers, and thus withers.

Hence the idea that we can dance around that shortcut baseline. But that dancing should be limited for R0 if we don't want to rathole as so many mobile groups have done before.

> I suggest we either lower the first bar to represent the interoperability of popular browsers today (and unless we're going to write ourselves out of global relevance, I think that means about 7 or 8 including Opera Mini, UCWeb, and Nokia), followed by rings that are indeed aspirational but that we can expect to be realised at a rate of one per year, or we drop the baseline browser idea and start with an aspirational ring based on something we expect to achieve in multiple browsers using real standards.

So to be clear and to make your proposal more concrete, you'd suggest delivering R1 say right after summer and dropping R0? Would that R1 exclude the likes of Mini? Because if it's for it to be the same discussion, I don't really see the point. What sort of implementation commitment would we be looking at?

-- 
Robin Berjon - http://berjon.com/ - @robinberjon

Received on Thursday, 12 April 2012 12:33:52 UTC