On 2 November 2014 16:01, SK53 <[email protected]> wrote: > So in summary: let all work on the documentation, discussions, but please > work for consensus and recognise that allowing time to build that will > result in an overall better dataset.
I agree that seeking consensus is important, which is why I stopped the vote (even though the proposal had 2/3 support at the point I stopped it). Given the comments, the proposal was clearly not good enough yet. I certainly don't think voting should be used as a substitute for seeking consensus. It is also clear that mechanical edits should not be carried out without community support. I don't think a single person objecting should be able to block a mechanical edit, but 'no significant objection' for a mechanical edit is not implied by a majority vote either. That's why I used 2/3 support as a threshold in my proposal, but maybe it should have been even 90% or 95%. In any case, I think we should have a way to measure the amount of support a proposal has in the community. I agree that the process of voting is far from optimal, but it's the best way to gauge the community's opinion we have. I think voting is better than looking at mailing list replies alone to decide whether a proposal is supported by the community, as 1) the number of voters is typically larger than the number of people taking part in the discussion, and 2) people objecting are more likely to comment on a proposal, making it harder to understand from mailing list replies alone whether an objection is 'significant' or not. I therefore think inviting list members to vote in order to make the position of the community explicit - in addition to taking comments on the mailing list into account, not as a replacement of it - is the safest way to proceed. -- Matthijs _______________________________________________ Talk-GB mailing list [email protected] https://lists.openstreetmap.org/listinfo/talk-gb
