Page 18 of 85

Re: [Script] WME Color Highlights - 1.8 Feb 2014

Posted: Fri Mar 14, 2014 3:12 pm
by doctorkb
Kuhlkatz wrote: When unsplitting roads that does not meet the requirements and was split a while ago, having the speed info at hand makes decisions on which segments to keep much easier. You cannot always go by editor name and creation date of the segments in this case.
Except that unsplitting shouldn't result in the removal of any roads... when you follow the instructions here:
https://wiki.waze.com/wiki/Best_map_edi ... -Way_Roads

Re: [Script] WME Color Highlights - 1.8 Feb 2014

Posted: Tue Mar 18, 2014 5:22 pm
by doctorkb
Kuhlkatz wrote:
doctorkb wrote: Except that unsplitting shouldn't result in the removal of any roads... when you follow the instructions here:
https://wiki.waze.com/wiki/Best_map_edi ... -Way_Roads
That section is to preserve the House numbering.
I'm referring to mostly main arterials in suburbs that usually has very little or no house numbers assigned to it (in my case). In this case I'd prefer to keep the original speed info for both A->B and B->A on the original segments instead of merging the 2 and polluting the A->B and B->A for the resulting segments after the unsplit & merge.

If you merge them the 'Wiki way' to retain the House Numbering, only the speed info for the longest of the 2 segments seems to be preserved, which looks like it screws up historic data for either the A->B or the B->A if one of the newly added segments was the longest.
It is likely this sort of assumption that has had Waze remove the "average speed" information from the editor. We don't really know if it removes the historical data. It stands to reason that if it preserves house numbers, it would also preserve other shared data like speeds.

Of course, you could follow the instructions given, and alternate the "longest segment of two" between sides so as to resolve the issue in either case, if you're that concerned.

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Fri Mar 28, 2014 12:52 am
by doctorkb
SuperDave: I'm pretty sure that was already noted -- the feature was removed when the beta went live because it was a fundamental change.

Unfortunately, given that Timbones is based in the UK, I suspect we're still a while before the new system is implemented -- "Places" only went live on the NA server. The ROW and IL servers are still a few weeks away, apparently.

I hope Tim will choose to reimplement that feature -- I really liked it, especially for finding smudges being caused by landmarks. We will have to see, though!

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Fri Mar 28, 2014 8:39 pm
by doctorkb
what is an "incomplete turn" and how is it different from a "soft turn"?

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Fri Mar 28, 2014 11:01 pm
by doctorkb
sketch wrote:Seems to be the same thing. Correct me if I'm wrong, but isn't the soft turns/revcon/selfcon function in WMECH limited to rank 3 and above?
You are correct. Along with highlighting of recent edits and edits by specific user.

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Sun Mar 30, 2014 5:58 pm
by doctorkb
qwaletee wrote:That's an extra step for someone to remember during scanning. And one that most editors won't do -- Shift-Z mode is great when I remember to do it, but most of the stuff I find with WMECH and Validator are incidental to editing things nearby to the issues they highlight for me. Normal editing has Shift-Z off.
Shift+Z is part of the base editor.

It is not advisable (nor prudent) for an editor to be using extensions until they've mastered the base editor.

I don't think this should be highlighted. There's been argument about not highlighting uturns, even -- that didn't get removed from WMECH, but there wasn't an easy way to notice them.

I think there are more important things to be highlighted -- I've modified my WMECH to highlight streetnames that don't conform to abbreviation standards, highlight end-point junctions that have u-turns or soft turns set. That said, I'm not publishing these changes -- if Tim wants to include them, I'll share the code, but they're features that help my editing, but I don't think they should be made available.

This fits in that category, too. If you feel you need it: fine. I trust you're actually watching what you enable.

The biggest issue I have with this is that there are probably as many intersections that have restrictions as those that don't. I don't want to see something highlighted that is probably correct.

If you wanted to get fancy and could somehow highlight it when disabled straight-through and right-turn connections are highlighted (but not when a left-turn connection is the only one disabled), then that would be useful and probably worth integrating.

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Sun Jun 08, 2014 10:45 pm
by doctorkb
daknife wrote:I don't like private channels, I grant some use of them for discussing how to deal with problem editors. But any and all discussions as to how the map system works must be at least readable by all editors, limit comments by non champs, but stop hiding new details from us.
FWIW, this particular topic is in the Champs-only forum because it pertained to as-yet-unreleased material, and a lack of public disclosure was requested by Waze staff.

Given that some of the material has been released, perhaps we can find a public thread to include the information on... but please don't get your knickers in a knot over this -- it wasn't a Champs decision to have that in the private area.

Re: [Script] WME Color Highlights - 1.81 Mar 2014

Posted: Mon Jun 09, 2014 4:15 pm
by doctorkb
Split FC Highlights discussion off to here: viewtopic.php?f=819&t=95473

Re: [Script] WME Color Highlights - 1.91 July 2014

Posted: Sun Aug 24, 2014 9:27 pm
by doctorkb
As of last night, the node highlighting (of soft turns, u-turn, revconns) is no longer working.

Re: [Script] WME Color Highlights - 1.91 July 2014

Posted: Tue Aug 26, 2014 9:51 pm
by doctorkb
Timbones wrote:
doctorkb wrote:As of last night, the node highlighting (of soft turns, u-turn, revconns) is no longer working.
I feared that might happen. Given that Validator shows these too (?), I'm considering dropping this feature from my script...
:(

My experience is that Validator is very "heavy" -- whereas WMECH is rather "light"... that is, it seems to take considerable processing time to load Validator each time, but not WMECH.

This is significant when you're dealing with multiple update/unlock requests, etc...