User talk:Strebe

From Wikimedia Commons, the free media repository
Jump to: navigation, search
Welcome to Wikimedia Commons, Strebe!
Afrikaans | Alemannisch | العربية | Asturianu | Azərbaycanca | Беларуская (тарашкевіца)‎ | Български | বাংলা | Català | Čeština | Dansk | Deutsch | Deutsch (Sie-Form)‎ | Zazaki | Ελληνικά | English | Esperanto | Español | Euskara | Estremeñu | فارسی | Suomi | Français | Frysk | Galego | עברית | हिन्दी | Hrvatski | Magyar | Հայերեն | Interlingua | Bahasa Indonesia | Íslenska | Italiano | 日本語 | ქართული | 한국어 | Latina | Lietuvių | Македонски | മലയാളം | मराठी | Bahasa Melayu | Plattdüütsch | नेपाली | Nederlands | Norsk bokmål | Occitan | Polski | Português | Português do Brasil | Română | Русский | Scots | Slovenčina | Slovenščina | Shqip | Српски / srpski | Svenska | Kiswahili | தமிழ் | ไทย | Türkçe | Українська | Vèneto | Tiếng Việt | 中文 | 中文(简体)‎ | 中文(繁體)‎ | 中文(台灣)‎ | +/−

-- 04:14, 15 August 2011 (UTC)

Babel user information
Users by language


Hello, I noticed your nice maps (projections). I'm myself writing down some GIS cartographic tutorials, and I'm interested to know :

  • where to download this 'skin', is it a GIS file ?
  • which software do you use ?
  • have you some specific understanding of geographic projections ? (I'm myself looking for a good projection for Mainland China)

Cheer, Yug (talk) 20:28, 10 September 2011 (UTC)

Apologies for missing your posting. I use Geocart for my maps. The base image comes with the software. I wrote Geocart, so I know quite a bit about map projections. What kind of map do you need to make of China? Strebe (talk) 05:53, 9 December 2011 (UTC)

File:HEALPix projection SW.jpg[edit]

I can't agree with your comment "Not a line drawing; subject is photographic data; does not contain significant compression artifacts." about File:HEALPix projection SW.jpg. Technically, it does contain line drawing, and is not really photographic data. However, the main problem is not the classification, :-) but the fact that while the original indeed does not contain significant compression artifacts, the thumbnailed images do show quite noticeable ringing artifacts (they are well seen in the illustration to en:HEALPix). Could you please recreate and upload a PNG version, if SVG is impossible? (And, after all, this "Blue Marble summer month composite" has nothing to do with the projection, so why not to take something general from Category:Blank SVG maps of the world and make a pure SVG? I think, it could be much cleaner and more usable.) — Mikhail Ryazanov (talk) 03:16, 9 December 2011 (UTC)

Thanks for your note. The ringing is from the downsampling applied by Wikimedia to create the thumbnail. The same thing would happen with a PNG of the same size. The data is indeed photographic—or at least, contains the same order of complexity of photographic data across the land areas: It’s NASA Blue World satellite imagery. A PNG would be much larger in file size to no benefit, so I do not understand why you prefer PNG. The reason for using the image is because it is used uniformly across the Wikipedia pages that display map projections. It is an informative image, and using the same image makes comparison much easier. I have no software for creating SVGs, so that is not a project I can work on. I can produce PDFs, including line art for those, but PDF is not a format supported by Wikimedia Commons. Strebe (talk) 06:09, 9 December 2011 (UTC)
I have converted the image to PNG to show you that PNGs do not suffer from ringing artifacts (check it in the article). The file size after direct conversion was ~3 times larger than of JPEG, partially due to JPEG compression artifacts; if created directly it would be probably only about twice larger (it is not "much", in my opinion :-) ). If you can produce PDF with vector grid, it can be easily converted to SVG by, for example, Inkscape. (In principle, Commons allow PDF upload, but for some reason it renders in JPEG only.) If you don't want to bother with conversions, could you please upload one PDF as an example, so that I can try to convert it to SVG? — Mikhail Ryazanov (talk) 07:26, 12 December 2011 (UTC)
You are correct; the PNG and JPG downsample differently. It’s odd that Wikimedia Commons uses different window functions for PNG vs. JPG downsampling. Clearly the PNG file is “softer”, and while I do prefer the look of the PNG downsampling, the JPG downsampling preserves more detail. Anyway, I can create a PDF image easily enough when I get time. Strebe (talk) 04:56, 13 December 2011 (UTC)
SVG version here. Strebe (talk) 03:27, 19 December 2011 (UTC)
Great! :-) I've experimented with it a little bit and found that black grid is much more contrast over snow and sand (and has approximately the same contrast everywhere else as the original light-gray), plus it's more consistent with the boundary color (which is a part of the grid). What do you think? (The recolored version is uploaded over your file, you can revert it after comparison...) — Mikhail Ryazanov (talk) 08:51, 21 December 2011 (UTC)
Thanks for the comments. I prefer the original color scheme. The thumbnail image in the article looks better with the black graticule as per your modification, but in the full-scale image, the graticule overwhelms everything. I do not think the outer boundary’s line color and weight need to match the graticule; that is not cartographic convention and in my opinion does not add any clarity. Also, all the other map projection articles use the original color scheme. I think consistency is important. The original color scheme was discussed publicly before creating the images so that the many needs could be balanced. Strebe (talk) 11:47, 21 December 2011 (UTC)
Yes, consistency is important. :-) I asked because if these maps are to be replaced with SVGs, it would be better to use good colors for all of them. Could you point me at the previous discussion?
Regarding that "the graticule overwhelms everything" — I thought, it was the main subject of these images. After all, the projection of a regular grid is more revealing than that of some complicated shapes. Figures in the main article en:Map projection suggests that this is indeed the case (and, for example, looking at "A two-point equidistant projection of Asia" illustration there I can barely see what is going on, because the grid is almost invisible).
Mikhail Ryazanov (talk) 15:36, 24 December 2011 (UTC)
We anticipated the need for actual maps, as opposed to just a caricature of the graticule. If the only need were for a thumbnail, then the images might legitimately be only a graticule, or possibly a graticule with a vector coastline. There would be no need for 2,000-pixel-wide satellite images. In most articles on map projections, the thumbnails are at 450 pixels in width, clearly showing the graticule as well as useful land-cover detail. It is true that the thumbnails are smaller in the article on en:Map projection, and it is true that the graticule is difficult to see on the two-point equidistant map of Asia. Yet it is not clear to me how helpful the graticule itself is in that image (actually, it is not even clear to me how relevant the image is to the article). In any case, the rationale is that you can always click on the thumbnail image to see whichever details are most relevant to you, the reader.
Also, most many projection articles include a Tissot indicatrix diagram in addition to the standardized imagery. (See, for example, en:Mollweide projection.) In my opinion that is a much better way to handle demonstrating the graticule and geometric relationships of the projection.
The discussion happened here and was announced on each map projection page whose image would be replaced. I see that I did not include an announcement on the HEALPix Talk page since it that article had no image originally. Strebe (talk) 18:02, 24 December 2011 (UTC)
Well, I see that in the "discussion" you basically showed only one variant, and there was just one reply "Good step", before the discussion slipped to the cutoff of Mercator projection. :-) No real discussion about the color scheme, not even anyone mentioned JPEG problems...
Anyway, if the real purpose was to show the landscape map in different projection, then your colors are quite good. And I totally agree that Tissot diagrams are indeed much more useful in articles about projection. Unfortunately, I haven't seem them there — probably because there are actually very few of them. Could you make such diagrams for the other projections? (And I'm going to revert the HEALPix to your original colors.) — Mikhail Ryazanov (talk) 03:08, 26 December 2011 (UTC)
All right, well, I am tired of this discussion. I am not responsible for how other people did or did not respond. I am not responsible for Wikimedia’s inconsistent thumbnail generation of JPEG vs PNG. I improved the images in 38 Wikipedia articles. I added images to pages that had none at all, including HEALPix. I announced the plans in advance on all the pages that had images that I was replacing. No one objected. It was a lot of work. If someone else wants to step up to help improve the situation, they are more than welcome to. Until someone else does, I infer that not many people care anyway, so it makes no sense for me to put even more effort into it, particularly if the only reward is to receive complaints and demands for yet more work. Strebe (talk) 05:39, 29 December 2011 (UTC)

Author of Geocart[edit]

Are you Daniel “daan” Strebe, one of Geocart author? If so, why do you chose to not license it freely? 11:56, 3 January 2012 (UTC)

Because it was a gigantic amount of work? Strebe (talk) 23:30, 3 January 2012 (UTC)

Yes, you are. Where did you find the formulas? Perhaps one of the sources is An Album of Map Projections? Refereced (talk) 06:05, 19 January 2012 (UTC)

Snyder’s “Album of Map Projections” was an early source. So was his “Flattening the Earth” and “Map Projections—A Working Manual”. Maling’s “Coordinate Systems and Map Projections.” Plus dozens and dozens of professional papers. And that’s just for the projection formulæ. Many of them can’t be used as-is for serious work because they contain singularities or numerically problematic regions. Strebe (talk) 08:24, 19 January 2012 (UTC)

Can you give me the formulas for 4 random projections? 08:29, 19 January 2012 (UTC)

“Can”? Yes. “Will”? What is this about? Strebe (talk) 09:34, 19 January 2012 (UTC)

Ellipsoidal projection in Geocart[edit]

Does Geocart support ellipsoidal projection or does it support only spherical projection? 06:28, 24 January 2012 (UTC)

See here please. Thank you kindly for your interest. Strebe (talk) 07:04, 24 January 2012 (UTC)

Strebe 1995 projection[edit]

Please upload this (the Strebe 1995 projection formulas here. 13:05, 25 January 2012 (UTC)

These formulæ describe the Strebe 1995 projection, which is equal-area. φ means latitude; λ means longitude. The plane coordinates are x and y.

x &= \frac {2D}{k_g} \cos \varphi_p \sin \lambda_p \\
y &= k_g D \sin \varphi_p \\
k_g &= 1.35 \\
D &= \sqrt {\frac {2}{1 + \cos \varphi_p \cos \lambda_p}} \\
\sin \varphi_p &= \frac {2 \sin^{-1}\frac{\sqrt{2} y_e}{2} + r y_e}{\pi} \\
\lambda_p &= \frac {\pi x_e}{4r} \\
r &= \sqrt {2 - y_e^2} \\
x_e &= 2 k_g \frac {\lambda (1+\cos \theta)}{\sqrt {4\pi + \pi^2}} \\
y_e &= 2 \frac {\sqrt \pi \sin \theta}{k_g \sqrt {4+\pi}} \\
\text{where } \theta \text{ is solved iteratively:} \\
\theta + \sin \theta \cos \theta + 2 \sin \theta &= \frac{1}{2} \left(4+\pi\right) \sin \varphi \\

Strebe (talk) 03:26, 15 February 2012 (UTC)

Please Upload Eckert II projection[edit]

It is equal area maping projection. the length of polar is half than equator. See This,[1]

File:Ecker II projection SW.jpg

--무소속 (talk) 07:54, 16 March 2012 (UTC)

Are you asking for an image? Formulæ? What do you want this for? Strebe (talk) 19:56, 16 March 2012 (UTC)
I want image --23:11, 16 March 2012 (UTC)
You gave a link to an image. Is there something wrong with that image? Strebe (talk) 23:15, 16 March 2012 (UTC)
I want Image like File:Ecker IV projection SW.jpg. -- 23:29, 16 March 2012 (UTC)
I want to. --무소속 (talk) 03:20, 26 March 2012 (UTC)
Done Strebe (talk) 04:04, 23 July 2012 (UTC)

File:Mollweide projection SW.jpg[edit]

Commons-emblem-issue.svg File:Mollweide projection SW.jpg has been listed at Commons:Deletion requests so that the community can discuss whether it should be kept or not. We would appreciate it if you could go to voice your opinion about this at its entry.

If you created this file, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it, such as a copyright issue.
Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!

Afrikaans | العربية | বাংলা | Беларуская (тарашкевіца)‎ | Català | Čeština | Dansk | Deutsch | Deutsch (Sie-Form)‎ | Zazaki | Ελληνικά | English | Esperanto | Español | Eesti | فارسی | Suomi | Français | Galego | עברית | Magyar | Bahasa Indonesia | Íslenska | Italiano | 日本語 | 한국어 | Македонски | മലയാളം | Plattdüütsch | Nederlands | Norsk nynorsk | Norsk bokmål | Occitan | Polski | Português | Português do Brasil | Română | Русский | Slovenčina | Slovenščina | Српски / srpski | Svenska | Türkçe | українська | Tiếng Việt | 中文 | 中文(简体)‎ | 中文(繁體)‎ | +/−

Kabyst (talk) 08:19, 14 September 2012 (UTC)

Request for Ordnance Survey grid projection[edit]

Note incomplete map outline at top left

Hi Strebe,

I came across your map projections and am fascinated by your work. I wondered if you could help me with an illustration I made to help readers understand the UK Ordnance Survey grid reference system, please:

I superimposed in green an outline of Great Britain which I found (I'm not sure if it's correct), but think it would be much better if the rest of Western Europe in the top-left grid could be shown.

Might you be willing to contribute a bit of your time to render in SVG format at any scale an outline of the land masses in the projection Ordnance Survey maps use? Alternatively, I will most appreciate a list of vertices which can be joined by straight lines, or quadratic Bézier or cubic curves.

Thank you in advance,
cmglee (talk) 21:30, 22 February 2015 (UTC)

Hello cmglee. I posted this: [2]. However, for reasons I don’t know, the outline varies slightly from the one you posted in these ways:
  • I had to scale the map left-right by 0.998 in order to match your image. I believe this is a flaw in your image.
  • The coastline does not match precisely in northeast Scotland and Cornwall, and to a lesser extent, eastern Ireland. I do not know the source of the discrepancy. I used vector data from WGS 84 coordinates and performed a Molodensky 3-parameter conversion, which might account for it, but the discrepancy seems larger than I would expect from that cause. Strebe (talk) 00:02, 23 February 2015 (UTC)
Thank you very much for your speedy response, Strebe. I've removed redundant Inkscape tags and attributes, and rescaled it to fit in my graphic. As the polylines are in parts, I couldn't shade the landmasses a solid colour, so just used an outline. You're right, there are flaws in my image as I traced the gridlines by hand and superimposed another map I found on Commons. Nevertheless, I think it's sufficient to illustrate the Ordnance Survey grid reference system. Thanks again, cmglee (talk) 23:21, 25 February 2015 (UTC)