On Mon, Aug 19, 2013 at 2:57 PM, Frederik Ramm <span dir="ltr"><<a href="mailto:frederik@remote.org" target="_blank">frederik@remote.org</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
--> Going off on a tangent here and leaving the scope of immediate iD improvements - someone else has posted that a while ago in a different discussion. Maybe we are far too obsessed with trying to make sure nothing is ever broken in an edit session. Maybe we should focus more on post-processing of edits. Give users the option of saying "I'd like someone else to review my edit". If user does that, a special tag ("review=yes") is set on the changeset. A list/map of such "changesets for review" could then be generated and processed by users who are interested in helping. Before too long we'll have feature where changesets can be commented/discussed which would go nicely with this.</blockquote>
<div><br></div><div>+100 on this. OSM editing for a first timer is lonely.</div><div>There's no evidence anyone in the project cares until you make a mistake (then you get pillored).</div><div><br></div><div>Mistakes themselves are not a problem in a system with undo. </div>
<div>But in a database with millions of nodes, something has to focus checking, else bad edits just drop as needles into the haystack.</div></div>