Web Maps???

Talk about whatever

34 posts • Page 2 of 3

Postby fraidykat » Wed Jun 09, 2004 1:16 am

hehe. we're filled with sarcasm.
it's mostly funny 'cause there's basically just the three of us and so all the sackings and promotions would be happening to the same person(s).

we encourage, in a mostly-unsupported fashion, playing with mappacks... search forums or see dutch's guide on netstumbler.com for details, and remember we switched X,Y from what one might expect. (long story).
I kinda figured that...

I checked the coordinates, They look fine to me, I was just comparing it to the cordinates in MS Mappoint and they were within .05 degrees of where I guessed at. close enough for me.

It's beyond me then. Is the parser throwing them off?

Good luck.

Postby fraidykat » Wed Jun 09, 2004 4:42 am

OK, I tried another mappack from a different area than mine, I know the rand Rapids Michigan area so I got Kent County. Its doing the same thing. has ANYONE else gotten this problem? or is no one noticing? maybe some only care about their quantity or APs and not the quality of the data. LOL.

I don't know. Did I get a bad download? could someone else try Grand Traverse county Michigan and compare it to the web maps? and/or also Kent County. You can clearly see a pattern in Kent County where people mostly follow the expressway through downtown. yet in my Dingle and Jingle maps they are way off the the left. but have the same pattern.

Then I think I will lay this thread to rest soon. unless someone knows how to fix my problem, I'll just wait until the web maps get updated.

Kat!

Postby uhtu » Wed Jun 09, 2004 10:38 am

everything you see in wigle are the collected observations of several hundred random people, just like yourself.

grand traverse has very few samples (vs. say, chicago) so there is not enough data to rule out bad gpses, highly directional antennae, amped rigs, bad stumble packages, etc.

there is a Quality of Service metric associated with points in wigle that describes a confidence interval for the data. it is very very low for that entire area. i don't actually think there is anything awry at all. there just hasn't been enough coverage to provide meaningful accuracy.
the single best way to clear up your issues with the data is for another person (or persons) to duplicate your results.

as near as i've been able to tell, all of the clients (web or otherwise) are providing the exact same view of the data (within the bounds of their various technologies.)

Postby fraidykat » Thu Jun 10, 2004 4:22 am

Ill try to show another example.

Kent county has quite a few more stumbled points. So I chose that, and I know the area well for landmark reference. Make note, note of these APs are from me.

This is Kent County on the Web Maps. Note the distinct pattern that follows US 131 down through the center.

Image

Here is the Dingle map of roughly same area. Note the same pattern. and in the picture after that I marked US 131 in purple.

Image

Here you can clearly see the APs pattern is off quite a bit. and it did not matter how far in or out I zoomed.

Image

As a check also, I did check Chicago area both dingle and Web. They looked correct, so It's not the application itself.

was this any help?

Postby uhtu » Thu Jun 10, 2004 11:24 am

spectacular, even. we have a suspect, and are investigating.

Postby fraidykat » Thu Jun 10, 2004 1:17 pm

Cool. I'm having a blast. I love finding new nerdy hobbies. I work in an IT department and everyone here thinks I'm weird doing this "Wardriving" thing.

They just have no passion.

Postby bobzilla » Thu Jun 10, 2004 3:08 pm

This is an issue with the .mappack not having aspect ratio correction applied to it. I'll fix in the next mappack version.

Postby bobzilla » Fri Jun 11, 2004 3:30 am

The new mappack version 2.4 fixes this aspect ratio error. I apologize for the problems.

Postby fraidykat » Fri Jun 11, 2004 1:03 pm

No need to apologize, I'm glad you knew what it was. Actually I am suprised you did it so fast!

Kat!

Postby fraidykat » Fri Jun 11, 2004 5:48 pm

Worked like a champ. Good work.

Postby Guest » Wed Jun 23, 2004 10:20 pm

The new mappack version 2.4 fixes this aspect ratio error. I apologize for the problems.
Is this for his specific mappack or for all? I was just noticing a similar problem with the data for Pima County, Arizona. While not nearly as severe, the APs are clearly offset from the path of the roads. The road at 32.244232,-110.85547 is a good example... Especially when you consider that, for example, the "City Limits" AP is on the other side of the road.

Postby Guest » Wed Jun 23, 2004 10:22 pm

The new mappack version 2.4 fixes this aspect ratio error. I apologize for the problems.
Is this for his specific mappack or for all? I was just noticing a similar problem with the data for Pima County, Arizona. While not nearly as severe, the APs are clearly offset from the path of the roads. The road at 32.244232,-110.85547 is a good example... Especially when you consider that, for example, the "City Limits" AP is on the other side of the road.
Sorry, using JiGLE 0.7.2.20040516203420 on Windows 2000 with JRE 1.4.1.

Postby uhtu » Wed Jun 23, 2004 10:24 pm

it would be for all post-2.0, pre-2.4 mappacks.

if you're still having the problem with a new version 2.4 mappack, please let us know.

Postby Guest » Wed Jun 23, 2004 10:46 pm

it would be for all post-2.0, pre-2.4 mappacks.

if you're still having the problem with a new version 2.4 mappack, please let us know.
The mappack is 2.4, I just downloaded it for the first time today.

Postby uhtu » Wed Jun 23, 2004 11:01 pm

Version 2.0 Raster MapPack for: Pima County, AZ (1476695 bytes)

i started a 2.4 which is being generated right now.

34 posts • Page 2 of 3

Return to “General Grabbag”

Who is online

Users browsing this forum: No registered users and 9 guests