[Qgis-user] mapping bus routes

Sat Oct 9 15:46:28 PDT 2021

I think you can get pretty far with aggregate expressions and geometry
generators, but I think at some point the complexity just gets unwieldy and
using a graphics program is more pragmatic. I even started building
a dendrogram legend within a map frame using QGIS at one point.

I've never done transit maps before, but one thing I've pondered before is
why they always tend to use offsets rather than staggered and overlaid
dashed lines. I imagine there has to be some design or ux element about
tracing the line with one's eyes that the broken line interferes with
tracking the route, but I'd appreciate a link/ article if anyone knows of
one, or just some resources that outline why the conventions are what they
are.

On Sun, 10 Oct 2021 at 00:27, Sam S <ssnellings at gmail.com> wrote:

> I built a number of bus maps in the past and kept everything inside QGIS
> using offsets in symbology. It worked because it was a simple dataset where
> the number of routes sharing paths was very minimal, so I just styled each
> 'unique' segment individually. I also did not have a problem with aligning
> with the street pattern below (David's noted issue) because I styled the
> routes larger than the street symbology so any 'incorrect' alignment
> between the routes and the streets was hidden by the larger route symbology
> on top.
>
> With a more complicated set of routes I would have probably graduated to a
> workflow that included a graphics application as the final step to make
> manual tweaks.
>
> At a certain point of complexity/geographic size most transit maps stop
> being truly scalable and become more abstract. A good example of the debate
> is NYC's vignelli vs hertz. However, most transit maps (in terms of
> 'quantity of maps') are for small operators where scale can be maintained -
> so any automated tool might not 'work' for very large not-to-scale systems
> but would be the most useful to those with the fewest resources (small
> systems).
>
> Anyways, tough problem.
>
> Sam
>
>
> On Fri, Oct 8, 2021 at 7:48 AM Nicolas Cadieux <
> njacadieux.gitlab at gmail.com> wrote:
>
>> Hi,
>>
>> Maybe Anita Graser has something on the topic? https://anitagraser.com/ She
>> does a lot of work in visualization mass transportation. Yes, looking at
>> line azimut by color was the way I did it in the past.  It’s still a good
>> way when the network has a lot of snapping errors.  My algorithm with
>> permit you find those problems. If you see a network problem my algorithm
>> does not find, please tell me and I could work them in.
>>
>> Nicolas Cadieux
>> https://gitlab.com/njacadieux
>>
>> Le 8 oct. 2021 à 02:54, Andreas Neumann <a.neumann at carto.net> a écrit :
>>
>> 
>>
>> Hi,
>>
>> Besides on what was discussed here on the list already, I can add the
>> following information: the Swiss QGIS user group has mandated Nyall Dawson
>> to carry out a couple of days of research exactly on this topic - kind of a
>> feasibility study if it would be possible to draw such network maps "on the
>> fly" / automatically - with transit maps and hiking maps (multiple parallel
>> hiking routes) in mind.
>>
>> I believe Nyall will work on this before the end of the year.
>>
>> Of course we cannot promise anything - but based on the outcome of this
>> study, there might be follow-up work pending. If there is interest from
>> other users we might want to organize an institutional crowd funding
>> initiative.
>>
>> This won't help you immediately - but it might be good for you to know
>> that the problem is on the radar. If there is a positive outcome in the
>> study, there might be implementation work pending next year.
>>
>> Greetings,
>>
>> Andreas
>>
>> On 2021-10-07 23:53, David Strip wrote:
>>
>> For a transit map, it seems the offset would not be constant. Consider
>> two bus routes that overlap along a single street in their routes. On that
>> street, the two lines typically would be offset to opposite sides of the
>> line representing the street. On the rest of the route, however, the route
>> would be shown as lying directly over the centerline, not offset to either
>> side. Life gets even more complicated when multiple routes overlap with
>> different numbers of other routes at various points on the map.  Or at
>> least that's what I've typically seen on transit maps.
>>
>> On 10/7/2021 3:30 PM, Nicolas Cadieux wrote:
>>
>> Hi,
>> You can also trim the lines in the symbology (at least in QGIS 3.20) as
>> well as giving it an offset.  That will insure that intersection are not
>> overlapping in the offset.
>>
>> Nicolas Cadieux
>> https://gitlab.com/njacadieux
>>
>> Le 7 oct. 2021 à 17:23, Nicolas Cadieux <njacadieux.gitlab at gmail.com>
>> <njacadieux.gitlab at gmail.com> a écrit :
>>
>> Oups!  Forgot the list...
>>
>> Nicolas Cadieux
>> https://gitlab.com/njacadieux
>>
>> Le 7 oct. 2021 à 17:23, Nicolas Cadieux <njacadieux.gitlab at gmail.com>
>> <njacadieux.gitlab at gmail.com> a écrit :
>>
>> Hi,
>> You can automatically draw an offset by adding a offset in the
>> symbology.  Things get more complicated when line vectorization changes
>> directions as the offset flips side depending on, for example, if the line
>> is drawn from A to B or B to A.
>>
>> In the QGIS models, https://plugins.qgis.org/, go to models then Fix
>> Directional Network models 1,2 and 3.  That will help you find problems
>> (like disconnected lines) but also identify all the lines that are flipped.
>> You can then flip them manually or figure out an algorithm to do it
>> automatically.
>>
>> You will also find a boring YouTube video on my YouTube channel.
>> https://youtu.be/v61PafSByvM
>>
>> Nicolas Cadieux
>> https://gitlab.com/njacadieux
>>
>> Le 7 oct. 2021 à 14:17, Hugh Kelley <hghklly at gmail.com>
>> <hghklly at gmail.com> a écrit :
>>
>> Hi all,
>>
>> I'm wondering if anyone has any good tricks for mapping bus routes. The
>> key problem is defining how to offset lines where the routes overlap.
>>
>> I've handled this by chopping up the lines and offsetting in postgis
>> according to some st_dwithins and st_intersections but wanted to see if
>> there is anything in the wide world of QGIS that I'm not aware of.
>>
>> My understanding is that it's a fairly significant challenge to do a
>> great job with it. A company called transitapp had an interesting blog post
>> about it a while ago.
>> https://archive.transitapp.com/transit-maps-apple-vs-google-vs-us-cb3d7cd2c362#.e9kcryzg1
>>
>>
>> Hugh
>>
>> _______________________________________________
>> Qgis-user mailing list
>> Qgis-user at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>>
>> _______________________________________________
>> Qgis-user mailing listQgis-user at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>>
>>
>>
>> _______________________________________________
>> Qgis-user mailing list
>> Qgis-user at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>>
>> _______________________________________________
>> Qgis-user mailing list
>> Qgis-user at lists.osgeo.org
>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
> _______________________________________________
> Qgis-user mailing list
> Qgis-user at lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20211010/efb30056/attachment.html>