Get a sneak peek at whats next for Permanent Hazards on our April 7th Office Hours!
The place to get information and ask questions about everything to do with properly and successfully editing the Waze Map.

Use this forum for all general editing questions, and the sub-forums for specific types of Waze Map Editor features.
Post by krankyd
From what I can see, the map has updated since (see attached).
Can you please try and 'refresh map' on your Android? Or try and install the latest beta (if supported on your device) to see if this issue is still happening?
photo.PNG
(200.57 KiB) Downloaded 991 times
I don't think there's a general problem with all updates of a specific date.
krankyd
Posts: 939
Been thanked: 27 times
Send a message
wazing like a hurricane

Post by Omega-Supreme
Hi,
Shirli have posted about the issue at status.waze.com .
Hi guys,

I'll starts with the good news:

The map tiles are updated now as to map edits from November 9 ! Hurray :)

However, you might run into a bug: some roads might appear as missing from the client (even though they appear correctly in the livemap and in the map editor).

This issue might affect the routing as well.

We are verifying currently how many tiles were affected and working on fixing it ASAP.

We will keep updating as soons as we have more details.



Thanks,



Shirli
Wanted to make sure you've seen this.

Thanks,
Eyal.
Omega-Supreme
Waze Global Champs
Waze Global Champs
Posts: 16
Send a message

Post by petervdveen
This morning the routing and the map where perfect.
petervdveen
Coordinators
Coordinators
Posts: 10370
Has thanked: 212 times
Been thanked: 847 times
Send a message
Coordinator Waze Netherlands

Click here to sign-up for slack, our communication platform

Contact by email: NL - BE - Lux - Benelux

Post by redmek
I don't get a map at all since upgrading to version 3.0 on ice cream sandwich (Galaxy Nexus)
http://img804.imageshack.us/img804/6500 ... 315532.png
redmek
Posts: 1
Send a message

Post by rlpl
This issue has been also discussed here. I have posted several screenshots there. They are too many to check them all but none of them have been fixed with the newest map update. As always, I have forced the client tiles refresh.
Valter_Rehn wrote:A suggestion: Would it be possible to update the "last updated" date of all segments last updated around Oct 25th? I believe the next tile build would then fix the issues, wouldn't it?
I hope it would fix the problem. Maybe it would be less destructive to change the rule selecting the tiles to rebuild from "after Dec 29" to "after Dec 29 plus between Oct 11 and Oct 29" (or whatever dates are correct?) Or maybe it is possible to compare the road geometries as seen by the client with those stored in the main database and detect those which differ with no reason?
krankyd wrote:Or try and install the latest beta (if supported on your device) to see if this issue is still happening?
My device is not supported by the latest beta. But could this issue be version dependent? Does 2.9/3.0 series use different map tiles? Will the tiles for 2.0.x be left broken?
krankyd wrote:I don't think there's a general problem with all updates of a specific date.
Of course, that would be too easy to detect and maybe fix. :-) There were many (most? all?) tiles broken as reported in this status update. And then, as announced, several days later most of them but not all have been fixed. Then Unholy asked whether we still could see the issues and we answered that yes, we could see and posted many examples. I don't know if anything has been changed later at your side but except two locations I can't see any fix. As it has been mentioned before, routing works fine despite the broken tiles and the new tiles are correct if the roads have been edited after the issue. Of course I don't know all details so may be wrong but it really looks like forcing rebuild of the broken tiles should fix the problem.

EDIT: It turns out that 2 tiles which I mentioned as fixed are fixed because they have been edited on Nov 28.
rlpl
Posts: 495
Been thanked: 1 time
Send a message
My previous login name was rl_pl.

Post by rlpl
foxitrot wrote:I suppose three different concurrent sets of "tiles" data are created from editors' output: 1) maps for routing servers, 2) maps for LiveMap (might use the 1) too?) and 3) maps for clients.
I agree, it looks so. Also there may be more tile sets like gray-on-white and white-on-blue city polygons. Every layer of Cartouche can be considered as the tiles generated from the maps: highlighted connectivity, turn restrictions, even the roads.
Just 3) seems to be broken.
Well, many times we have already experienced the cases when routing does not work although the map is correct. Not mentioning the turn restrictions layer which has been broken since forever. At few places I have also seen and heard the incorrect navigation commands which cannot be explained from the map, like if the routing server had a different map than visible anywhere. Actually, I am happy that these issues occurred because they look exactly how I imagined that the routing sever can see some roads. Now as the problems are clearly visible I only hope that the waze team will fix them all once forever.
rlpl wrote:
krankyd wrote:Or try and install the latest beta (if supported on your device) to see if this issue is still happening?
... could this issue be version dependent? Does 2.9/3.0 series use different map tiles? Will the tiles for 2.0.x be left broken?
I can not imagine Waze would maintain even more different sets of tiles 3a) 3b) 3c)... for various clients...
Why not? If they already maintain 3 or maybe 7 tile sets why not to have the 8th set? They said that 3.0 client is rewritten from scratch, why shouldn't it use a separate tile set? I'm not telling they really did it, I'm just telling that they had an opportunity. And if 3.0 and 2.0 clients use the same tiles then krankyd's suggestion to install the newest beta version would not help.

What is crazy here is that all tile sets seem to be generated by different and independent algorithms. They must be different if they produce so different results. Isn't it a pure waste of the CPU resources? I think that a better solution would be to generate one common tile set (or few tile sets but containing different data) and then generate the required final tile sets for the live map, mobile devices, editors etc. applying simple 2D post-processing methods, like blending or applying different color palettes.
Besides this, we have been testing the reported broken tiles on various clients, which were displaying them similarly.
The 2.x and even 1.x clients were generated from the same source code for all platforms, so unless you have the 3.0 client the differences between our clients were minor.
rlpl
Posts: 495
Been thanked: 1 time
Send a message
My previous login name was rl_pl.

Post by rlpl
Sorry for responding late.
foxitrot wrote:And I see it as totally inserious, that Waze withdraws such subtle details from the huge editors' community, letting them guess about the ongoing bugs for ages, yet being totally dependent on their precise work!! :evil:
It should not matter for us how many tile sets they have and how they maintain them as long as the sets are consistent and as long as the algorithm does not cause the process to run twice as long as it should. Unfortunately, that's not true.
Valter_Rehn wrote:Or, to restrict the proof of concept, just update the segments last updated by me and I'll let know if it worked on the next tile update:

update segments set last_updated=sysdate where last_updated between '10-11-2011' and '10-31-2011' and updated_by= Valters_ID;
I don't agree with the idea of restricting to your segment because not only your segments are broken. I'd rather call all segments updated between the specified dates suspected.

By the way, I confirm again that the tiles I have mentioned previously are still broken.
rlpl
Posts: 495
Been thanked: 1 time
Send a message
My previous login name was rl_pl.

Post by robin1979
Yeah, every other livemap update we are having this issue. This and routing server time-outs really make me want to throw the iPhone out of the window sometimes.

Lots of this this morning:
http://www.waze.com/forum/viewtopic.php?f=6&t=11009
robin1979
Posts: 3032
Has thanked: 1 time
Been thanked: 22 times
Send a message
MapSquadEurope
Area Manager - Europe

Post by robin1979
Seems like another map update is going on, navigation is impossible at this very moment again, lots of 'Proceed to highlighted route', 'Recalculating route due to map updates', 'Routing server time-out' and 'Routing server error'.

Please note: this is a basemapped country were routing is perfectly fine (except for some time-outs) when the map update process isn't running.
robin1979
Posts: 3032
Has thanked: 1 time
Been thanked: 22 times
Send a message
MapSquadEurope
Area Manager - Europe

Post by robin1979
petervdveen wrote:This morning the routing and the map where perfect.
It was, up until about 9:15 local time (UTC+1).
robin1979
Posts: 3032
Has thanked: 1 time
Been thanked: 22 times
Send a message
MapSquadEurope
Area Manager - Europe