Prototype of version 0.4
uuellbee
http://prototype.bycycle.org/tripplanner/
- It uses Oct '06 RLIS data prepared by Jack [Newlevant]
- Data for regions other than Portland hasn't been imported
- It's a little slow, especially for route lookup
- There are several "known issues"
- I haven't tested it in Internet Explorer at all
Most of the issues should be resolved fairly soon. In the meantime,
please give it a try and report any issues, opinions, etc. We'll be
able to change some things in this version, but other things,
especially new features, will go into the next version.
__wyatt
Jack Newlevant
reactions.
Suggestions:
- Put the caret in the typing field below the Search-Map &
Find-Route tabs, both when the page 1st comes up and after I
change tabs.
- Multiple Geocode (go-away, sub-)panels on the left would
be more useful if I could drag them to the Find-Route fields.
- The Route panels heading bar should include 1) a sample of
the color used to portray the route on the map and 2) the
choice of Route-Type.
- Including "toward [No Name]" isn't at all useful (in fact,
I don't find "toward" anything to be helpful).
Bugs:
- "change | Help | Feedback" buttons send me back to same page.
- after I enter an address & it geocodes ok, I see a map-pin
symbol which looks clickable but does nothing. Only the red
bulls-eye (in which the map pin is "stuck") brings the
desired pop-up.
- Geocode go-away sub-panels on the left can be duplicates.
- from ne grand & irving to 1904 se hemlock 97214 (Normal)
looks wrong going down Sandy & 7th. Its clearly shorter,
faster, less traffic to go the way I ride: jog on Davis or
Couch, then straight down 14th.
Cheers,
Jack
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.409 / Virus Database: 268.13.29/520 - Release Date: 11/6/2006
Cone, Paul
Can you give us a quick run-down of how you built this new data file (i.e.
is there metadata)? I think you might have mentioned it at the presentation
at Metro but I didn't take notes. I remember something about it being a
"rebel" network or something like that, but nothing specific.
Paul
Wyatt Baldwin
Thanks for all the feedback. I'm going to reply to some things in line.
On 11/6/06, Jack Newlevant <Ja...@newlevant.com> wrote:
>
> Congrats, it looks great overall! Below are some initial
> reactions.
>
> Suggestions:
> - Put the caret in the typing field below the Search-Map &
> Find-Route tabs, both when the page 1st comes up and after I
> change tabs.
+1
> - Multiple Geocode (go-away, sub-)panels on the left would
> be more useful if I could drag them to the Find-Route fields.
Do you mean drag the text from a multiple match into the input fields?
That sounds a bit complicated, but it would be cool. I was planning to
add links for "set as start or end" for single matches; adding them to
multiple matches would be good too, and maybe do the fancy "web
twennie" drag-n-drop at some point in the future.
> - The Route panels heading bar should include 1) a sample of
> the color used to portray the route on the map and 2) the
> choice of Route-Type.
+2 (I really like these ideas). I would also like a "show on map" link
for routes.
> - Including "toward [No Name]" isn't at all useful (in fact,
> I don't find "toward" anything to be helpful).
I agree about the [No Name]s. I think the solution is to make sure
every segment has a name, especially since I strongly disagree about
the usefulness of "toward" :-).
> Bugs:
> - "change | Help | Feedback" buttons send me back to same page.
Whoops. Haven't given a thought to those links in a while. I'll make
them link to where they did before for now. Eventually, I want to
create a simple inline help/feedback system that doesn't take you away
from the trip planner.
> - after I enter an address & it geocodes ok, I see a map-pin
> symbol which looks clickable but does nothing. Only the red
> bulls-eye (in which the map pin is "stuck") brings the
> desired pop-up.
Lauren and I were *just* discussing that. The plan is to implement
that and all other functionality in the current live version in the
new version by the end of the week, tweak some things for performance,
then go ahead and go live with it. (And then start adding cool new
features.)
> - Geocode go-away sub-panels on the left can be duplicates.
I'm not sure that's a bug. It would be somewhat complicated and maybe
bad for performance to check for duplicates in the front end (but
maybe not), and I think it's not something users will do too often.
OTOH, I do like the idea of checking for a dup and then moving the
duped result up; I think that's going to be version 0.6 thing though
(unless someone else wants to implement it).
> - from ne grand & irving to 1904 se hemlock 97214 (Normal)
> looks wrong going down Sandy & 7th. Its clearly shorter,
> faster, less traffic to go the way I ride: jog on Davis or
> Couch, then straight down 14th.
No comment. Actually, it's probably true, but I'm not too focused on
route selection issues right now. I haven't changed the selection
algorithm in a long time (it's the same as is being used in the
current live version--I wonder what difference, if any, the new data
makes).
This feedback is great. If anyone else has any more ideas, etc, please
don't hesitate to throw in your two (or more) cents. Oh, and if
possible, along with sending those "ideas, etc" to this list, please
enter issues into the issue collector. It's really easy--just navigate
to http://trac.bycycle.org/tripplanner/newticket, login with
username/password ticket/$ticket$, and fill in the short form.
__wyatt
Wyatt Baldwin
>
> Jack,
>
> Can you give us a quick run-down of how you built this new data file (i.e.
> is there metadata)?
I'm curious about this too. I didn't look at the new data all. One
thing I can say is that it imported without any significant trouble
(the only thing that was a problem at all was a few fields that had
their names changed).
> I think you might have mentioned it at the presentation
> at Metro but I didn't take notes. I remember something about it being a
> "rebel" network or something like that, but nothing specific.
It's a pirate network. Arrr...
__wyatt
Cone, Paul
something else? Do you have some segment IDs as examples? If so then I can
take a look at them and see what can be done about giving them names.
Paul
-----Original Message-----
From: Wyatt Baldwin [mailto:wy...@bycycle.org]
Sent: Monday, November 06, 2006 5:32 PM
To: byCyc...@googlegroups.com
Wyatt Baldwin
<snipped Jack's feedback and my responses>
> Oh, and if possible, along with sending those "ideas, etc" to this list, please
> please enter issues into the issue collector. It's really easy--just navigate
> to http://trac.bycycle.org/tripplanner/newticket, login with
> username/password ticket/$ticket$, and fill in the short form.
I extracted the issues from this exchange and added them to issue
collector. (I may have missed a couple though.)
__wyatt
Wyatt Baldwin
> something else? Do you have some segment IDs as examples? If so then I can
> take a look at them and see what can be done about giving them names.
There are 7,178 segments in the PostgreSQL database that don't have a
name. Does someone know how to query a shapefile for that info? It
would be good to do a quick comparison there to make sure I didn't
screw something up in the shapefile=>Postgres import.
I've attached a file containing the LOCALIDs and street type code for
the segments that don't have a name.
__wyatt
Mark Bosworth
> at Metro but I didn't take notes. I remember something about it
being a
> "rebel" network or something like that, but nothing specific.
a Pirate without a cause ...
One question Jack: Did we get the Three Bridges "turned on" for
routing?
Mark
Mark Bosworth
Geographic Information Systems
Metro
600 NE Grand Ave
Portland, OR 97232
Tel: 503.797.1583
Fax: 503.797.1909
email: bosw...@metro.dst.or.us
web: www.metro-region.org
>>> wy...@bycycle.org 11/6/06 5:37:49 PM >>>
On 11/6/06, Cone, Paul <paul...@pdxtrans.org> wrote:
>
Mark Bosworth
I just did a quick query of the STR04 shapefile and got 6715 records
with FNAME = ' '. Attached is a low res jpg showing the records
highlighted in Cyan. Note that most are in rural areas and many are
unpaved forest service roads. Otherwise, they mostly appear to be
multiuse or minor multi-use paths that aren't streets.
Here is a frequency by Bikemode:
ca 4
ht 14
lt 8
mm 226
mt 100
mu 152
pm 21
up 1697
xx 4493
We should probably decide on some method for identifying those that
actually do have a name but we haven't captured it properly. And then
some wholesale approach to dealing with the balance. We need some kind
of naming convention for the paths that we are routing people onto.
Mark
Mark Bosworth
Geographic Information Systems
Metro
600 NE Grand Ave
Portland, OR 97232
Tel: 503.797.1583
Fax: 503.797.1909
email: bosw...@metro.dst.or.us
web: www.metro-region.org
>>> wy...@bycycle.org 11/6/06 7:50:09 PM >>>
On 11/6/06, Cone, Paul <paul...@pdxtrans.org> wrote:
>
Cone, Paul
project yet, and I would expect to. I will track them down and get
something in the centerline for them.
-----Original Message-----
From: Mark Bosworth [mailto:bosw...@metro.dst.or.us]
Sent: Tuesday, November 07, 2006 1:44 PM
To: wy...@bycycle.org; byCyc...@googlegroups.com
Subject: [byCycle-dev:21] Prototype of version 0.4
> I think you might have mentioned it at the presentation
> at Metro but I didn't take notes. I remember something about it
being a
> "rebel" network or something like that, but nothing specific.
a Pirate without a cause ...
One question Jack: Did we get the Three Bridges "turned on" for
routing?
Mark
Mark Bosworth
Geographic Information Systems
Metro
600 NE Grand Ave
Portland, OR 97232
Tel: 503.797.1583
Fax: 503.797.1909
email: bosw...@metro.dst.or.us
web: www.metro-region.org
>>> wy...@bycycle.org 11/6/06 5:37:49 PM >>>
On 11/6/06, Cone, Paul <paul...@pdxtrans.org> wrote:
>
Mark Bosworth
Paul.
The 3 bridges are represented in the "pirate" network, but they have
the attribute of "planned" so I assume that Bycycle drops them from the
route solution.
Mark
Mark Bosworth
Geographic Information Systems
Metro
600 NE Grand Ave
Portland, OR 97232
Tel: 503.797.1583
Fax: 503.797.1909
email: bosw...@metro.dst.or.us
web: www.metro-region.org
>>> paul...@pdxtrans.org 11/7/06 2:05:05 PM >>>
Cone, Paul
everything, especially if it has an address range (in which case it must be
named).
-----Original Message-----
From: Mark Bosworth [mailto:bosw...@metro.dst.or.us]
Sent: Tuesday, November 07, 2006 2:05 PM
To: wy...@bycycle.org; byCyc...@googlegroups.com
Subject: [byCycle-dev:22] Prototype of version 0.4
I just did a quick query of the STR04 shapefile and got 6715 records
with FNAME = ' '. Attached is a low res jpg showing the records
highlighted in Cyan. Note that most are in rural areas and many are
unpaved forest service roads. Otherwise, they mostly appear to be
multiuse or minor multi-use paths that aren't streets.
Here is a frequency by Bikemode:
ca 4
ht 14
lt 8
mm 226
mt 100
mu 152
pm 21
up 1697
xx 4493
We should probably decide on some method for identifying those that
actually do have a name but we haven't captured it properly. And then
some wholesale approach to dealing with the balance. We need some kind
of naming convention for the paths that we are routing people onto.
Mark
Mark Bosworth
Geographic Information Systems
Metro
600 NE Grand Ave
Portland, OR 97232
Tel: 503.797.1583
Fax: 503.797.1909
email: bosw...@metro.dst.or.us
web: www.metro-region.org
>>> wy...@bycycle.org 11/6/06 7:50:09 PM >>>
On 11/6/06, Cone, Paul <paul...@pdxtrans.org> wrote:
>