 |
 |
Thanks Alan for sharing the motivations of the segment representation! These motivations are also those of the point-based space syntax. The points in point-based space syntax can be either segments or junctions. The key to the point-based space syntax is that the point-point (or equivalently segment-segment) relationship are established through axial lines or natural streets. See details in the following two papers (as cited previously): https://www.researchgate.net/publication/220605896_Integration_of_Space_Syntax_into_GIS_New_Perspectives_for_Urban_Morphology https://www.researchgate.net/publication/1918773_Self-organized_Natural_Roads_for_Predicting_Traffic_Flow_A_Sensitivity_Study Cheers. Bin On 8/10/2016 5:31 AM, Penn, Alan wrote: Dear Bin, Not quite. There were several reasons we developed the segmental representation, including: - empirical phenomena such as land use, movement flows and development densities, often seem to vary along the length of an axial line, and so having a representation that varies at the segment scale is worth having to see if we can improve our understanding; - measures of betweenness inherently involve trips that traverse only some segments of any axial line, and so are really only meaningful in terms of a segmental representation; - the elegance of the angular weighting on a segmental map is that it retains a zero depth along the length of an axial line (hence a topological view) whilst allowing for differentiation between smiths winding streets and orthogonal turns. Carlo Ratti's critique was so elementary in its flaws that it was not a factor in this at all so far as I am aware. I'll take another look at your papers and think further. All the best, Alan Sent from my iPhone On 9 Aug 2016, at 22:11, Bin Jiang <[log in to unmask]<mailto:[log in to unmask]>> wrote: To supplement the previous message... To my understanding, segment analysis was developed partially to defend space syntax, because of the long-standing criticism on its lack of geometry. I think it is exactly the lack of geometry that makes space syntax unique and powerful. This is my way of defending space syntax. In my humble opinion, Ratti's criticisms on space syntax are mostly flawed. Ratti C. (2004), Space syntax: some inconsistencies. Environment and Planning B - Planning and Design 31 (4), 501–511. See my detailed arguments in this paper: https://www.researchgate.net/publication/265911553_Different_Ways_of_Thinking_about_Street_Networks_and_Spatial_Analysis The new topological representation we developed recently can be considered to be the same way of defending space syntax - the kind of topological analysis. https://www.researchgate.net/publication/305638074_A_Topological_Representation_for_Taking_Cities_as_a_Coherent_Whole I look forward to your comments and criticisms in particular. Thanks and cheers. Bin On 8/9/2016 9:40 PM, Bin Jiang wrote: Thank you Alan for the comments! My discussion is purely scientific, but my apologies if you or anyone else got offended. My previous message was put in general in terms of goodness of topological analysis: natural streets > axial lines > segments (where > indicates "better than"). Based on our studies cited in the previous message, and in my humble opinion, segment analysis as implemented in depthmap is a kind of damage for space syntax. As for roundabouts and traffic island, Axwoman provides some simplification, which was just recently added into. The previous studies I cited in the previous message did not have that simplification. These studies prove that a few roundabouts and traffic islands do not matter much in the topological analysis. Cheers. Bin On 8/10/2016 3:06 AM, Penn, Alan wrote: Bin, I was advising on a purely pragmatic basis. How to get a model together quickly for a thesis deadline on a city the size of NY. You are absolutely right that an axial map contains the detailed geometric information required as a basis for segmental analysis, something lost in RCL data. We might also observe that RCL tends to over complicate representations around features such as roundabouts and traffic island (although I think the Tiger dataset is pretty sparse). Anyhow for a basic analysis of NYC I still suspect it would do. By the way it's worth asking Mark David Major (aka outlawurbanist ) if he has an axial map of NYC. Alan Sent from my iPhone On 9 Aug 2016, at 20:13, Bin Jiang <[log in to unmask]<mailto:[log in to unmask]>> wrote: I have to disagree with Alan on the reasonable approximation! Among the three methods of segment analysis, only the angular method can achieve almost same results as axial lines. However, the three segment representations are essentially geometry based rather than topology based. By topology, I mean it enables us to see the underlying scaling property of far more less connected than well connected; see this paper (the second paragraph in particular): https://www.researchgate.net/publication/305638074_A_Topological_Representation_for_Taking_Cities_as_a_Coherent_Whole Someone may argue that segments are better than axial lines, because metrics are assigned into individual segments rather than axial lines, so fine resolution. This sounds a right argument, but I cannot stand by. To setup segment-segment relationship, it must be done through axial lines (this is the reason why angular method can obtain the reasonable results). This idea was implemented in our point-based space syntax: https://www.researchgate.net/publication/220605896_Integration_of_Space_Syntax_into_GIS_New_Perspectives_for_Urban_Morphology Later on we further extended the idea into natural streets; see this paper (Section 4.3 in particular): https://www.researchgate.net/publication/1918773_Self-organized_Natural_Roads_for_Predicting_Traffic_Flow_A_Sensitivity_Study In summary, natural streets representation is better than axial lines representation, although both are topology oriented. There is little doubt that natural streets predict traffic far better than axial lines; see evidence in the studies cited above. Recently, we used millions tweets locations in London as a surrogate of traffic, and get the same results as in the above studies; R square between metric and tweets locations up to 0.55 for natural streets, but only less than 0.2 for both axial lines and axial lines' segments. We are thinking of whether write a paper. It seems unnecessary, since the results are the same as before. Thanks and cheers. Bin On 8/10/2016 1:23 AM, Penn, Alan wrote: If you use the Tiger road centre line dataset and use angular segment analysis in DepthmapX you will have a reasonable approximation. The problem will be that really you need to unlink segment at underpass/ overpasses and that could be a bit tedious. Alan Sent from my iPhone On 9 Aug 2016, at 14:01, Anurodh Khanal <[log in to unmask]<mailto:[log in to unmask]>> wrote: Hi Rajesh, I would be willing to give it a shot, however i have to defend my proposal in 2 weeks, so it might take some time for me to do it. I am using both ArcMap 10.2 and Axwoman 6.3 for my study. I would also take this opportunity to enquire with the wider space syntax on a approach i took to generate axial lines using street centrelines as a trial. 1) Generate .dxf file of street centreline map in GIS 2) Use Adobe illustrator to draw axial lines - guided by the rules in developing axial model - i followed the street lines and drawing least number of lines to cover every street - every line as long as possible 3) exported the axial model from illustrator to depthmap 10 and run analysis My question is, can this type of approach be taken? If not using software tools, can axial model be different for the same street network? I would appreciate comments, suggestions, or solutions to generate axial model using street centreline data but not using depthmap or Axwoman but manually? Thank You, Best Regards, Anurodh Khanal On Tue, Aug 9, 2016 at 5:17 PM, Rajesh Chaturvedi <[log in to unmask]<mailto:[log in to unmask]>> wrote: Hi, Axwoman is an extension to arcmap 10.2.0 which can used to generate axial line map from road centreline map. Unfortunately, I do not have licence for ArcMap 10.2 .0. if any of you folks having ArcMap 10.2.0 licence can voluntarily generate axial lines from road centrelines for me, please let me know. I will share the data sets immediately. I need axial map for New York city for my study and I can't figure out a way to generate it. Thanks in anticipation. -- Regards, Rajesh Chaturvedi -- Anurodh Khanal 33 Sexton Street, Highgate Hill QLD 4101, Australia 0404161352 [log in to unmask]<mailto:[log in to unmask]> -- -------------------------------------------------------- Bin Jiang Division of GIScience Faculty of Engineering and Sustainable Development University of Gävle, SE-801 76 Gävle, Sweden Phone: +46-26-64 8901 Fax: +46-26-64 8758 Email: [log in to unmask]<mailto:[log in to unmask]> Web: http://fromto.hig.se/~bjg/<http://fromto.hig.se/%7Ebjg/> -------------------------------------------------------- Academic Editor: PLOS ONE Associate Editor: Cartographica BinsArXiv: http://arxiv.org/a/jiang_b_1 Axwoman: http://fromto.hig.se/~bjg/axwoman/<http://fromto.hig.se/%7Ebjg/axwoman/> ICA: https://sites.google.com/site/commissionofica/ Geomatics: http://fromto.hig.se/~bjg/geomaticsprogram/<http://fromto.hig.se/%7Ebjg/geomaticsprogram/> RG: https://www.researchgate.net/profile/Bin_Jiang3 [Högskolan i Gävle] Högskolan i Gävle, 801 76 Gävle • 026 64 85 00 • www.hig.se<http://www.hig.se> För en hållbar livsmiljö för människan University of Gävle, SE-801 76 Gävle, Sweden • +46 (0) 26 64 85 00 • www.hig.se<http://www.hig.se> -- -------------------------------------------------------- Bin Jiang Division of GIScience Faculty of Engineering and Sustainable Development University of Gävle, SE-801 76 Gävle, Sweden Phone: +46-26-64 8901 Fax: +46-26-64 8758 Email: [log in to unmask]<mailto:[log in to unmask]> Web: http://fromto.hig.se/~bjg/<http://fromto.hig.se/%7Ebjg/> -------------------------------------------------------- Academic Editor: PLOS ONE Associate Editor: Cartographica BinsArXiv: http://arxiv.org/a/jiang_b_1 Axwoman: http://fromto.hig.se/~bjg/axwoman/<http://fromto.hig.se/%7Ebjg/axwoman/> ICA: https://sites.google.com/site/commissionofica/ Geomatics: http://fromto.hig.se/~bjg/geomaticsprogram/<http://fromto.hig.se/%7Ebjg/geomaticsprogram/> RG: https://www.researchgate.net/profile/Bin_Jiang3 -- -------------------------------------------------------- Bin Jiang Division of GIScience Faculty of Engineering and Sustainable Development University of Gävle, SE-801 76 Gävle, Sweden Phone: +46-26-64 8901 Fax: +46-26-64 8758 Email: [log in to unmask]<mailto:[log in to unmask]> Web: http://fromto.hig.se/~bjg/<http://fromto.hig.se/%7Ebjg/> -------------------------------------------------------- Academic Editor: PLOS ONE Associate Editor: Cartographica BinsArXiv: http://arxiv.org/a/jiang_b_1 Axwoman: http://fromto.hig.se/~bjg/axwoman/<http://fromto.hig.se/%7Ebjg/axwoman/> ICA: https://sites.google.com/site/commissionofica/ Geomatics: http://fromto.hig.se/~bjg/geomaticsprogram/<http://fromto.hig.se/%7Ebjg/geomaticsprogram/> RG: https://www.researchgate.net/profile/Bin_Jiang3 [Högskolan i Gävle] Högskolan i Gävle, 801 76 Gävle • 026 64 85 00 • www.hig.se<http://www.hig.se> För en hållbar livsmiljö för människan University of Gävle, SE-801 76 Gävle, Sweden • +46 (0) 26 64 85 00 • www.hig.se<http://www.hig.se> -- -------------------------------------------------------- Bin Jiang Division of GIScience Faculty of Engineering and Sustainable Development University of Gävle, SE-801 76 Gävle, Sweden Phone: +46-26-64 8901 Fax: +46-26-64 8758 Email: [log in to unmask]<mailto:[log in to unmask]> Web: http://fromto.hig.se/~bjg/ -------------------------------------------------------- Academic Editor: PLOS ONE Associate Editor: Cartographica BinsArXiv: http://arxiv.org/a/jiang_b_1 Axwoman: http://fromto.hig.se/~bjg/axwoman/ ICA: https://sites.google.com/site/commissionofica/ Geomatics: http://fromto.hig.se/~bjg/geomaticsprogram/ RG: https://www.researchgate.net/profile/Bin_Jiang3 [Högskolan i Gävle] Högskolan i Gävle, 801 76 Gävle • 026 64 85 00 • www.hig.se<http://www.hig.se> För en hållbar livsmiljö för människan University of Gävle, SE-801 76 Gävle, Sweden • +46 (0) 26 64 85 00 • www.hig.se<http://www.hig.se>