As far as why you are getting a turn right at 44 degrees, I like theory A. But I don't have any further information. You should reach out to milkboy and ask.

I understand where you're coming from. But I'm still leery of overwriting information that waze staff added to the wiki, even if we know some is misstated and they haven't confirmed the current accuracy of the information. Basically, although the wiki is a community endeavor, I still think waze should ultimately be responsible for the the accuracy of the information provided. At the very least we should give the staff at last chance for input - "We will change the wiki to this proposed page unless you tell us not to."

qwaletee wrote:I don't care what the source is

We'll have to differ on this. I do care what the source is.

Yes.

kentsmith9 wrote:they had me create this message box and add it to the main page.

I find a bit of tension between "they had me create" and the added message. If waze doesn't want to publish a product manual presumably they have a reason why they believe their users don't need such guidance.

I have no problem with:

kentsmith9 wrote:Waze staff . . . have turned it over to the community for us to manage and maintain.

If this happens:

kentsmith9 wrote:There is no problem overwriting content if the old content is determined to be inaccurate based on confirmed data, especially if someone from Waze staff can confirm it.

We are still volunteers and unless there is some way for those that are actually responsible for waze to confirm information in the wiki or unless there is some other source of official product information then I don't want to change what they say in the wiki.

It is not fair for waze to place the entire burden for documenting their product on the volunteer community. Nor is it fair to users to have no recourse to challenge waze itself for inaccurate information provided about its product.

I still believe that when we have substantive changes to the descriptions of the detailed operation of waze, we should at least provide waze staff the opportunity to correct us.

Kent seems to have the best channels for wiki review at this point. But I'm happy to ask for review if necessary.

One thing I would like to note is that I do not find it easy to retrieve any part of the old version of the wiki at any time. For instance finding the original waze statement on what the qualities of a major highway are is not easy. The statement has over time been edited to what is on this page, but finding the original page is not easy. Similarly finding waze's original statement on what information can be referenced as set forth here is not easy. Waze's original content is likely buried somewhere in the wiki history, but it is not easy to find.

sketch wrote:The Wiki must be current and accurate. It doesn't matter if staff posted the original page, if the original page is wrong and we're right, change it.

If only right/wrong was actually binary.

sketch wrote:It's very easy to find old revisions of a page. Just go to a page, click "History", click "oldest", and click whichever revision you like. You can even compare them with the "cur"rent page.

But over time the information tends to migrate from page to page, so the history of a particular page ceases to be enough to find the original waze statement.

sketch wrote:It can be done. It doesn't need to be dead simple. If there's a compelling enough need to find old info, someone who knows the Wiki well will be able to track it down. (I had no trouble finding the old "How to Label and Name Roads" article.)

Ok, I surrender. It just needs to be dead simple for me to able to find it. (I wasn't finding it because I wasn't paging through all the pages of revisions).

Wow, impressive chart. It does make the instruction selection look complicated. But it does look comprehensive. My comments after looking at the chart:

• "Are s-in or s-out part of a roundabout?"
  This makes it sound like you get an instruction at the entrance to the roundabount, at the exit from the roundabout, and at every junction from roundabout segment to roundabout segment when rounding the roundabout.
• "Is the only other unrestricted sN <45° overlapping s-out?"

  PesachZ wrote:

  CBenson wrote: Do you know if overlapping segments suppress the instruction if there is no best continuation? In other words if both segments would give keep left/keep left instructions if not overlapped, are the instructions suppressed if overlapped?

  No it doesn't, they both get announced. But if the left branch is overlapped it gets a keep right instruction instead of keep left.

  I had thought this meant that even if an sN overlapped s-out that under the right circumstances you could still get an instruction.
• "Are the name AND types of any sN & s-out matching?"
  In other blocks you say "sN <45°." Shouldn't that be the case here as well?
• "Is there an sN to the left of s-out"
  Should this be the "Is s-out overlapped" box? There is no "Yes" from this box. Again should this be sN <45°?
• "Is s-out overlapped?"
  Should this be the "Is there an sN to the left of s-out" box?
• "Is s-out a primary road or ramp?"
  Both occurrences, there is no "Yes" path from this box.

Shouldn't the Yes paths from "Is there an sN <45 to the left of s-out?" and "Is s-out overlapped?" go to "Is this in a RHT country?" while the No path from "Is s-out overlapped?" go to "Is this in a LHT country?"?