search

adrianba / webstandards

Proposals for developing web standards.
1 stars 1 forks source link

Channeling volunteers #2

Open rubys opened 9 years ago

rubys commented 9 years ago

Re:

I heard other people expressing the view that they would want to take advantage of the hard work Sam has done

I don't believe that we should focus on sunk costs.

The question I would like to tee up is how we channel people who are willing to do hard work so that the results are more productive (i.e., will actually affect the future evolution of the web).

I'll state my premise that I do believe that there is a role for people who are informed, engaged and independent (i.e., not beholden to any one browser vendor).

adrianba commented 9 years ago

First, my intent with this section (which I think I might need to re-word to clarify) was to point out that I had seen two approaches discussed in consideration of interop challenges:

  1. One way is "interrupt driven" where interop issues affecting web developers are discovered and need to be resolved. (either through existing sites that behave differently in different browsers or through examples of web developers running into problems during cross-browser testing of in-development sites).
  2. An alternative is to spot that there several are differences in how browsers process a particular feature and do an exhaustive search for all the problems (perhaps by fleshing out the test suite). This is what you did with URL.

I then went on to say that I imagine (1) or (2) will differ on a case by case basis and I also think that different browser engine developers will favour one or the other at different times too. Our experience is that these differences between browsers do eventually bite and the question is really whether that will be sooner or later. On the specific example of URL, Microsoft doesn't see this as a high priority and would put it in bucket (1) today.

For (2), what I heard was some people saying that they do want to take advantage of the investment and enthusiasm people put into what are real interop gaps. I think this is less about sunk costs and more about "we're going to have to do this eventually and if someone has done the work we should pay attention".

You raise an interesting question though: how do we maintain a backlog of this type of activity that needs informed, engaged, and potentially independent people to work on and to give those people something to invest in with maximum ROI.