Given the earlier statements regarding alignment, etc, of the road vectors, and the seemingly large amount of work to revert these changes, perhaps an incremental replacement of road geometry with TIGER 2011 data along with manual conflation of existing attributes is a more effective application of effort. This assumes, of course, the TIGER data (which I have not examined for LA County) is markedly superior to the existing data.<div>
<br></div><div><div>Best,<br><br><div class="gmail_quote">On Tue, Jun 5, 2012 at 6:25 PM, Charlotte Wolter <span dir="ltr"><<a href="mailto:techlady@techlady.com" target="_blank">techlady@techlady.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div>
Steve,<br><br>
<u></u> <u></u>Thanks for
your thoughts, and I hope others in the community will share their views.
My view would be that it should be OK to do such a reverse if the person
has not signed the new license, particularly when there has been such an
enormous number of questionable edits.<br>
<u></u> <u></u>If it is
decided to do reversals, I suggest we start with his most recent
changesets and work backward, little by little. Though most of his
"edits" follow a definite pattern, a few appear to be unique,
genuine edits much like the work we all do. I'll bet the most recent work
is what was done robotically.<br>
<u></u> <u></u>Also, to
give a little more information, many of the "edits" often show
up as intersections that must be deleted and redrawn. This is a
particularly time-consuming process, as, I believe, "blars"
knew. <br><br>
Charlotte<div><div class="h5"><br><br>
<br>
At 11:56 AM 6/5/2012, you wrote:<br>
<blockquote type="cite">
<blockquote type="cite"><u></u>
<u></u>I have been
working on the LA remap, which mostly involves correcting thousands and
thousands of edits by a user named "blars."<br>
<u></u> <u></u>The odd
thing about the "edits" that "blars" did is that they
seem to have been done almost robotically. Although many intersections
have been changed enough to be marked "modified" by the OSM
Inspector, it's difficul to see what, if anything, was done. The poor
street alignments from TIGER have not been corrected. Obvious issues,
such as a street that no longer exists, have been ignored. Almost every
single dead-end street or those with turning circles are
"modified," as are many intersections, but, again, bad
alignment has not been touched.<br>
<u></u> <u></u>All this
plus the sheer volume of changes makes me think that "blars"
used some kind of bot to make these changes. These "edits"
corrected or added little, except to make the map "modified"
for the new license.<br>
<u></u> <u></u>Perhaps
someone who knows JOSM or other tools better than I do could come up with
a way to make these repairs on a mass basis. Or, perhaps, there is a wasy
to reverse just what blars did within a certain time frame.<br>
<u></u> <u></u>I
appreciate any help you can give. I've put in hundreds of hours on this
and barely scratched the surface.</blockquote><br>
Hello Charlotte:<br><br>
I have been watching the fine efforts in greater LA, with impressive
progress and results (in a "slow and steady" way), so I first
wanted to congratulate you on your efforts, and especially reaching out
to this talk group. You have done much more than scratch the
surface, you (and others) are making real progress.<br><br>
One method may be to revert changesets by blars. Reverting
changesets can be controversial (it is polite to ask / reach consensus
with the original author, et cetera). However, in the case of an
editor who has not agreed to the new CT, it may be correct for OSM users
who HAVE agreed to the new CT to establish if/when/how this is an
OK/proper thing to do. And then feel OK doing it, and then do
it.<br><br>
Technically, reverting a changeset is rather easily done with a JOSM
plug-in, which you can read about on OSM's wiki. There is the
separate issue of IDENTIFYING which changeset to revert, which can be
"more difficult" and then there is the rather straightforward
procedure for doing it, which isn't difficult.<br><br>
But "socially," or more properly stated, in the context of
"reaching OSM consensus," what does our community think of
(rather wholesale) reverts of a contributor who has not agreed to the
CT? Are we OK with that? Apologies if this is already clearly
stated somewhere. But if so, I haven't seen it and it is high time we
freshen up how/where we are about this.<br><br>
SteveA<br>
California<br><br>
_______________________________________________<br>
Talk-us mailing list<br>
<a href="mailto:Talk-us@openstreetmap.org" target="_blank">Talk-us@openstreetmap.org</a><br>
<a href="http://lists.openstreetmap.org/listinfo/talk-us" target="_blank">
http://lists.openstreetmap.org/listinfo/talk-us</a></blockquote>
<u></u></div></div><p><span class="HOEnZb"><font color="#888888"><u></u>
Charlotte Wolter</font></span></p><div class="im"><br>
927 18th Street Suite A<br>
Santa Monica, California<br>
90403<br>
<a href="tel:%2B1-310-597-4040" value="+13105974040" target="_blank">+1-310-597-4040</a><br>
<a href="mailto:techlady@techlady.com" target="_blank">techlady@techlady.com</a><br>
Skype: thetechlady<br><br>
<b>The Four Internet Freedoms</b> <br>
Freedom to visit any site on the Internet<br>
Freedom to access any content or service that is not illegal<br>
Freedom to attach any device that does not interfere with the
network<br>
Freedom to know all the terms of a service, particularly any that would
affect the first three freedoms.<br>
</div><p></p></div>
<br>_______________________________________________<br>
Talk-us mailing list<br>
<a href="mailto:Talk-us@openstreetmap.org">Talk-us@openstreetmap.org</a><br>
<a href="http://lists.openstreetmap.org/listinfo/talk-us" target="_blank">http://lists.openstreetmap.org/listinfo/talk-us</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br>John Novak<br>Novacell Technologies and the Old Topo Depot<br><a href="http://www.novacell.com" target="_blank">http://www.novacell.com</a><br>585-OLD-TOPOS (585-653-8676)<br>
<br>
</div></div>