Commons:Featured picture candidates/File:CzechRepublic-geographic map-en.svg

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

File:CzechRepublic-geographic map-en.svg, featured[edit]

Voting period is over. Please don't add any new votes.Voting period ends on 17 Dec 2017 at 12:59:25 (UTC)
Visit the nomination page to add or modify image notes.

Full SVG geographic map of Czech Republic
@Ikan Kekek: That's right, it is a very common map. The beauty is not only in the presentation but the details and the format.--Ikonact (talk) 07:42, 9 December 2017 (UTC)[reply]
There's nothing unusual about that, either. I've been a geography-lover since I got my first atlas at the age of 6. I'm very experienced in looking at maps, and I just don't think this is that special. If it were 5 or 10 times bigger and more detailed, like the paper maps I had from the National Geological Survey and whatever the Malaysian equivalent was in the 1970s, I would vote to support without a second thought. -- Ikan Kekek (talk) 08:53, 9 December 2017 (UTC)[reply]
@Ikan Kekek: , I'am not sure you understand the work involved in this map. You indeed just uploaded one file to commons, and doesn't seems to have practical map making experience. Ikonact gathered and processed various massives sources, from NASA to OSM, with multiple different data processing. His map is 100% svg, respects Wikipedia map conventions, for the hardest type of maps we have, the "(Topographic) Exchange maps". If your point is that Ikonact, alone, doesnt equal the dedicated team of full time trained cartographers and painters of the 70s, it's kind of telling SpaceX its space rockets sucks because NASA made larger space rockets back in the 70s, when NASA was 2% of the USA's budget. This map is technically one of the best map made and published on commons, and such works have a label, its Featured image / map / picture. --Yug (talk) 15:42, 18 December 2017 (UTC)[reply]
I don't know why you're pinging me after the fact, but on FPC, the whole point is to compare images to the best available in their category, and if that means that really diligent work by Commons photographers in reproducing paintings gets voted down because it's not as good as photographs by the museum itself or by Google Art Project, or in this case, that very good mapmaking by an individual gets voted down because it's not so special in the universe of existing maps, that's what should be done. In this instance, however, a consensus disagreed with me, so your post smacks to me of unnecessary triumphalism. -- Ikan Kekek (talk) 17:35, 18 December 2017 (UTC)[reply]
@Ikan Kekek: There is no really anything about triumphalism. I can just imagine the feeling of @Yug: , who is a known cartographer in Wikipedia, when he sees the statement that this is a very common map. My first reaction was somehow similar but I take your comments as very constructive. Personally, I agree that visually there is nothing exceptional in the map if compared to big editions of Atlas of the World but I do not think that there are too many maps of this quality in Wikipedia and in particular in SVG format. I thank you for your comments because they will help me to generate better maps in the future. I really hope that one day my maps will be able to meet your expectations. Regards --Ikonact (talk) 09:31, 19 December 2017 (UTC)[reply]
Me too. That would be great. Best, Ikan Kekek (talk) 09:41, 19 December 2017 (UTC)[reply]
I confirm it was not about post-vote triumphalism, I very simply didn't noticed the deadline for voting. Aside, as a Wikipedian cartographer knowing the workload involved, I wanted Ikonact's work to be judged fairly, including on underlying technicalities which he pushed pretty far. My vote below (not taken into account for the FPC decision due to its date) also explain both skepticism and support for the work. Last, I views the FPC as in the context of wikimedia Commons, while you seems to see it in a broader space, also explaining our diverging appreciation of the work done. Yug (talk) 12:03, 19 December 2017 (UTC)[reply]
@Benh: I use fonts that are supported in Commons. But I can put Helvetica as fall back font. --Ikonact (talk) 07:42, 9 December 2017 (UTC)[reply]
Well that's a bit more complicated I think. The version we see on the thumbnails (and the PNG restarised by Commons' servers) will use those fonts supported. But putting another font of your liking (in my case Helvetica, but don't feel obliged) will only choose it in priority over the other ones when viewed on a device/engine which supports it. Note that Helvetica is available in Mac, but not necessarily in Windows or Linux (the latter has equivalent, and the former can always fallback to Arial which is similar to Helvetica). Also note that I don't know which font you choose (I haven't opened the sources) but the names of the major cities render with a serif font on my Mac. The font should be chosen wisely! - Benh (talk) 09:19, 9 December 2017 (UTC)[reply]
The last font in the list is 'sans-serif'. I wonder if someone has a serif font in their browser configuration for the default sans-serif.--RaboKarbakian (talk) 03:10, 10 December 2017 (UTC)[reply]
@Benh: , I use Deja Vu Sans Condensed. This is a font supported in Commons. I usually put sans-serif as a fallback font and thus the system can choose the one defined by default. I can put Helvetica and Arial as fallback too. I will check the issue with this file. May be there is something with the fonts --Ikonact (talk) 09:05, 12 December 2017 (UTC)[reply]
It seems silly and impossible to account for faulty browser configuration. The sans-serif in the list should allow people to see the font they prefer for all sans-serif purposes.--RaboKarbakian (talk) 02:43, 14 December 2017 (UTC)[reply]
@RaboKarbakian: There was a problem with the font for the the names of the cities but not for the towns, so @Benh: was partially correct :-) I repaired this and I left Deja Vu Sans Condensed with sans-serif. --Ikonact (talk) 11:51, 14 December 2017 (UTC)[reply]
@Ikonact: Ahha, a lesson for me against spot-checking. My apologies to @Benh: not just for being wrong, but for enjoying discussing it. Personally, it is good to know that Benh the wikimedian has not degraded over the years. :) --RaboKarbakian (talk) 15:00, 14 December 2017 (UTC)[reply]
@RaboKarbakian: The advantage of doing these maps as .SVGs is that the text can be independently edited, making translated versions much easier to make. Daniel Case (talk) 04:43, 9 December 2017 (UTC)[reply]
@Daniel Case: That is true, and this map might only need the legend details translated for many languages. If it does end up as the POTD, it should have more.--RaboKarbakian (talk) 03:10, 10 December 2017 (UTC)[reply]
I planned to translate it at least in Czech... It is not difficult as the format is text. That's right that there may be little to translate as the names of the cities (with few exceptions) do not change. However, this may be a challenge for non Latin script languages. --Ikonact (talk) 09:05, 12 December 2017 (UTC)[reply]
@RaboKarbakian: I made a CZ version. I will try to make others. --Ikonact (talk) 11:51, 14 December 2017 (UTC)[reply]
1. I think there's something wrong with the railroads. The symbol suggested for rail tracks by the legend is alternating white and black segments of equal length. That's good, as that's how this is typically done. However, in the map itself, the length of the black and white segments is totally random. It took me a while to figure out that these are supposed to be railways, I first thought it was another two types of road not mentioned in the legend.
2. You have 2 types of road in the legend, differentiated by color (yellow and red). But in the actual map you have 4 types, differentiated by color and width: yellow-thin, yellow-thick, red-thin and red-thick. Either explain the difference between thin and thick in the legend or make them all the same width. The legend not fully describing the map's elements is a serious no-no.
3. Similarly, neither the triangle used for mountain peaks nor the airport symbol show up in the legend.
4. The placement of text labels (especially for the towns) looks like it was done automatically, which leads to lots of problems. For example, there is an unreadable town name hidden behind "Prague". Prague's airport marker collides with the town Hostivice. Štětí north of Prague disappears in a mess of railroad. In some more densely-populated areas it's very difficult do figure out which label belongs to which town marker (e.g. around Ostrava and south of Zlín). There are many more things like this that need manual adjustmends, and I think clarity could benefit from leaving out some of the smaller towns.
5. Many line features (roads, rivers, railroads) are much more detailed than they need to be for this kind of map. Sometimes, railroads are hidden behind parallel roads or rivers by railroads – straighten them out a bit and pull them apart to show them running in parallel. Again: What's going on in Štětí? Finding the right amount of simplification is crucial for map making, and I think this map would greatly benefit from more simplified/streamlined lines.
--El Grafo (talk) 12:04, 15 December 2017 (UTC)[reply]
@El Grafo: Thanks for the comments. I can clean up a bit the map by removing the mess of railroad and some of the town names. I will also improve the legend. However, I will not be able to repair the railroad issue. The railways are from OSM and are in small chunks and it is very difficult to stick all together. The only thing I can do is to propose a discontinued line as symbol. This will not solve the issue but will hide it. However, the biggest difficulty I have is your comment number 5. Personally, I believe that by simplifying, the map will loose its value. So if you believe that discontinued line for railroads and no simplification will be acceptable for featured picture I will be happy. --Ikonact (talk) 14:33, 15 December 2017 (UTC)[reply]
@Ikonact: I see the problem with the railroads, I've run into that issue with OSM data before. It's all chopped up into single line segments which makes it really hard to work with for this kind of thing. I wish I had a good solution for that, but I can't think of one right now (ArcGIS probably has some kind of tool for that, though). I'm not sure I really understand what you mean by "discontinued line", but if you can find some way to cover it up I'm fine with that.
I understand your concerns about simplification, but maybe look at it this way: Every map is always a simplification of reality, and lots of the art of making maps lies in choosing the right amount of en:Cartographic generalization (that's the term I was lacking earlier …) for the intended use (think about subway & metro maps). You can see that on online maps like OSM or Google maps, where the type of things shown and the precision e.g. roads are shown at depend of the zoom level. At lower zoom levels, linear features like roads are drawn much broader than they are in reality (otherwise you wouldn't be able to see any of them) and small bends and curves are straightened. They've put a lot of thought into how to do this automatically in software, but they will probably never reach the quality of a map hand-drawn by a professional. If in reality a road runs parallel to a river, the map should show it like that. If the chosen scale requires both to be displayed broader than they are in reality, so be it. If that means that they are too broad to be displayed next to each other, they have to be moved apart a bit ("displacement") because otherwise the map would be misleading (e.g. showing the river flowing beneath the road). You could say you have to lie a bit to get the message across, but keep in mind that you are not publishing raw geographic data here that is intended to be viewed at the scale of a few metres.
So long story short: The right amount of generalization is crucial for a good map (dozens of books and research papers have been written about that), and I think yours could use a bit more of it over-all. It won't lose it's value – quite the opposite, imho. But hey, I know I can be a pedant, so if the majority of the voters here think it should be featured I'm totally fine with that. Cheers, --El Grafo (talk) 16:11, 15 December 2017 (UTC)[reply]
Thank you @El Grafo: I am working on a new cleaned up version. I will try to address your comments but may be not everything on generalisation :-) Cheers --Ikonact (talk) 08:27, 18 December 2017 (UTC)[reply]
  •  Support. Take 5 minutes to examine the work of it, 100% svg, and what it means has earlier skills and processing, and you realize it's quite an impressive work relative to Commons.
    On the semiology, I must agree with El Grafo's points 1, 2, 3, 4, 5, but it can be fixed and should be. Railways (1) will be technically tuff to solve : OSM provide multi-lines paths, upon which your CSS/styling fails miserably. Yet, the work is already impressive and with the expected fixes at the top of Wikimedia commons maps. Yug (talk) 15:42, 18 December 2017 (UTC)[reply]
Confirmed results:
Result: 10 support, 2 oppose, 0 neutral → featured. /PumpkinSky talk 13:34, 17 December 2017 (UTC)[reply]
This image will be added to the FP gallery: Non-photographic media/Maps
Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Featured_picture_candidates/File:CzechRepublic-geographic_map-en.svg&oldid=273014502"