So
File:Sesame Street logo.svg was already a SVG but it was PD-textlogo and about to be moved to Commons.
File:IStock LP 2014 Corporate Logo.png was also PD-textlogo and ready for Commons but it was a PNG. I took care of both of them, but for some reason I couldn't upload either as a SVG (the uploader changes .svg to .svg.jpg). I put them in their respective articles, but the SS logo was RV'd for being lower quality. I'd like to convert them both to SVG.
Hop on Bananas (
talk)
18:15, 27 September 2015 (UTC)reply
There is a problem with the first Image, made by
Spitzl, as used on Wikipedia. First of all, weirdly, it is an SVG file, with only Bitmap content, the lines are not vector. But this problem is minor. The bigger one is the data itself. I went to
http://stats.oecd.org/Index.aspx?QueryName=426&QueryType=View&Lang=en which is given as the source and downloaded the data myself, then made a new graph using the same data. But my graph looks different, which (given it is supposedly the same data from the same source) should not be the case.
In the second image I overlaid my own graph (thin lines) on the image currently used, trying to match the percentages and the years as exactly as possible. Sorry about Ireland, forgot about them. Everything is off, sometimes substantially so.
In the third image I did the same again, but this time disregarded the X- and Y-axis labeling (I know, you should not do this!) instead trying to match the lines as good as possible. Here most work out well, but France and Portugal are still very much off.
I would love to upload my ODS file, but commons does not accept ODS files. Instead I describe what I did. I downloaded the annual growth rate table from the URL given above, then set 2000 as 100, and used the following formula for the next years: =LASTYEAR/(100+GROWTHRATE)*100
Can anyone help making sense of this? Did I do something really wrong? Did Spitzl do something really wrong? Is there another explanation? Thanks a lot in advance.--
Lommes (
talk)
10:38, 1 September 2015 (UTC)reply
Dear Lommes, when I made this graph I used the figures from "Total Economy". That means, when you go to the OECD data source, you need not change not only the measure drop down menu to "Index OECD base year (2010=100)" but also the sector to "Total Economy". Hope that solves the problem. If you still get different results, feel free to replace the graph with your own. Best wishes, --
spitzl (
talk)
20:24, 1 September 2015 (UTC)reply
For graphists: per the
SUNY Poly Standards Manual the colors are PMS Blue 288 (hex #002C73) and PMS Gold 124 (hex #EDA900). The fonts are Friz Quadrata (custom kerning between "Y" and "T") and Azo Sans, with recommendations to use AGaramond and Helvetica as close substitutes. Cheers,
Mliu92 (
talk)
03:44, 13 September 2015 (UTC)reply
@
Kopiersperre: I'll take these. I have a couple ready to go, but the vertical continuous retorting furnace is taking some thought, and my current plan is to combine the given drawing with the original patent held by New Jersey Zinc Company (1948), which is slightly more detailed: U.S. patent 2,457,552
@
Kopiersperre: Sorry for the long delay. I have some drawings for #2 up now, based on the patent writeup and illustration. Please make sure my translation makes sense.
Vox has a fantastic visualization of what the U.S. would look like when each state is scaled proportional to population. I think it would make a great illustration for our article if someone was willing to make a similar, but free use vector image. –
czar19:28, 21 August 2015 (UTC)reply
@
Czar: This is an interesting request, and one that was solved mathematically using the data at the linked article. By definition, the population density is the population divided by area, or:
or, equivalently,
We want to resize each state according to its population, so each state's new area will be defined by a new (constant) population density, :
We can substitute for in the above expression to get:
This means the new area is simply the existing area multiplied by an areal scaling factor . Since area is the product of two dimensions, we can apply an equal scaling factor in each dimension (which is what the SVG expects) by taking the square root of the areal scaling factor:
I took the new population density to be equal to the 2013 population density of New Jersey, 1,210.1 persons per square mile, which in essence normalizes the scaling factor to New Jersey. For instance, California has a population density of 246.1 persons per square mile, so:
That means that I can apply a scaling transformation factor of 0.451 to the California SVG object in each (x,y) dimension. When all is said and done, though, I had to end up translating most of the state SVG objects as well, so it ended up being a matrix transform (scaling + translation). When the population and/or land areas update, the updated data could be used to adjust the scaling factors in the SVG directly without having to go through a graphical editor. Please feel free to adjust the SVG file to taste, as the scaling also affects the border size of the file.
I'd be very grateful if a skilled user could expand the abbreviations used in the above image 9used in
Lung), so that it can be made more readable. --
Tom (LT) (
talk)
23:51, 30 August 2015 (UTC)reply
Please update the colors in the file to #0021A5 and #FF4A00 . The colors have since updated since this file was originally uploaded in 2011. Source can be found
here. -- Corkythehornetfan18:10, 11 September 2015 (UTC)reply
The file of Seborga's flag doesn't look right, It looks like it's been traced poorly. Can someone please retrace the crown and the shield in the flag? After that, can you also vectorise the coat of arms using the elements from the flag? Thanks in advance. --
109.77.100.184 (
talk)
16:20, 12 September 2015 (UTC)reply
@
Fry1989: Let me know what you think about this draft. I reused the central portion of the State Department Seal with some edits (updated colors, gold border in the field of thirteen stars). The typeface is Century Schoolbook, but it appears the weight used in the actual seal is some extra-heavy weight that I don't have installed. I adjusted it to space roughly identical to the spacing around the rim, although there ended up being more space between letters than the original, as a result.
Would someone be able to create
this symbol? I can't seem to figure it out myself. It has a height of 140 units and width of 120 units, and a radius for each curve of 110, but I don't know how to do the curves so they end in points like that. Fry1989eh?17:45, 1 October 2015 (UTC)reply
@
Fry1989: I think this is the one you were asking for ... How does this look? I also uploaded a file that illustrates how I created this shape, four overlapping circles with radius = 110 units. You can use Inkscape's "intersection" (two shapes at a time) to create it. Let me know if you want me to add text as well (using Roadgeek typeface?). Cheers,
Mliu92 (
talk)
13:58, 7 October 2015 (UTC)reply
So
File:Sesame Street logo.svg was already a SVG but it was PD-textlogo and about to be moved to Commons.
File:IStock LP 2014 Corporate Logo.png was also PD-textlogo and ready for Commons but it was a PNG. I took care of both of them, but for some reason I couldn't upload either as a SVG (the uploader changes .svg to .svg.jpg). I put them in their respective articles, but the SS logo was RV'd for being lower quality. I'd like to convert them both to SVG.
Hop on Bananas (
talk)
18:15, 27 September 2015 (UTC)reply
There is a problem with the first Image, made by
Spitzl, as used on Wikipedia. First of all, weirdly, it is an SVG file, with only Bitmap content, the lines are not vector. But this problem is minor. The bigger one is the data itself. I went to
http://stats.oecd.org/Index.aspx?QueryName=426&QueryType=View&Lang=en which is given as the source and downloaded the data myself, then made a new graph using the same data. But my graph looks different, which (given it is supposedly the same data from the same source) should not be the case.
In the second image I overlaid my own graph (thin lines) on the image currently used, trying to match the percentages and the years as exactly as possible. Sorry about Ireland, forgot about them. Everything is off, sometimes substantially so.
In the third image I did the same again, but this time disregarded the X- and Y-axis labeling (I know, you should not do this!) instead trying to match the lines as good as possible. Here most work out well, but France and Portugal are still very much off.
I would love to upload my ODS file, but commons does not accept ODS files. Instead I describe what I did. I downloaded the annual growth rate table from the URL given above, then set 2000 as 100, and used the following formula for the next years: =LASTYEAR/(100+GROWTHRATE)*100
Can anyone help making sense of this? Did I do something really wrong? Did Spitzl do something really wrong? Is there another explanation? Thanks a lot in advance.--
Lommes (
talk)
10:38, 1 September 2015 (UTC)reply
Dear Lommes, when I made this graph I used the figures from "Total Economy". That means, when you go to the OECD data source, you need not change not only the measure drop down menu to "Index OECD base year (2010=100)" but also the sector to "Total Economy". Hope that solves the problem. If you still get different results, feel free to replace the graph with your own. Best wishes, --
spitzl (
talk)
20:24, 1 September 2015 (UTC)reply
For graphists: per the
SUNY Poly Standards Manual the colors are PMS Blue 288 (hex #002C73) and PMS Gold 124 (hex #EDA900). The fonts are Friz Quadrata (custom kerning between "Y" and "T") and Azo Sans, with recommendations to use AGaramond and Helvetica as close substitutes. Cheers,
Mliu92 (
talk)
03:44, 13 September 2015 (UTC)reply
@
Kopiersperre: I'll take these. I have a couple ready to go, but the vertical continuous retorting furnace is taking some thought, and my current plan is to combine the given drawing with the original patent held by New Jersey Zinc Company (1948), which is slightly more detailed: U.S. patent 2,457,552
@
Kopiersperre: Sorry for the long delay. I have some drawings for #2 up now, based on the patent writeup and illustration. Please make sure my translation makes sense.
Vox has a fantastic visualization of what the U.S. would look like when each state is scaled proportional to population. I think it would make a great illustration for our article if someone was willing to make a similar, but free use vector image. –
czar19:28, 21 August 2015 (UTC)reply
@
Czar: This is an interesting request, and one that was solved mathematically using the data at the linked article. By definition, the population density is the population divided by area, or:
or, equivalently,
We want to resize each state according to its population, so each state's new area will be defined by a new (constant) population density, :
We can substitute for in the above expression to get:
This means the new area is simply the existing area multiplied by an areal scaling factor . Since area is the product of two dimensions, we can apply an equal scaling factor in each dimension (which is what the SVG expects) by taking the square root of the areal scaling factor:
I took the new population density to be equal to the 2013 population density of New Jersey, 1,210.1 persons per square mile, which in essence normalizes the scaling factor to New Jersey. For instance, California has a population density of 246.1 persons per square mile, so:
That means that I can apply a scaling transformation factor of 0.451 to the California SVG object in each (x,y) dimension. When all is said and done, though, I had to end up translating most of the state SVG objects as well, so it ended up being a matrix transform (scaling + translation). When the population and/or land areas update, the updated data could be used to adjust the scaling factors in the SVG directly without having to go through a graphical editor. Please feel free to adjust the SVG file to taste, as the scaling also affects the border size of the file.
I'd be very grateful if a skilled user could expand the abbreviations used in the above image 9used in
Lung), so that it can be made more readable. --
Tom (LT) (
talk)
23:51, 30 August 2015 (UTC)reply
Please update the colors in the file to #0021A5 and #FF4A00 . The colors have since updated since this file was originally uploaded in 2011. Source can be found
here. -- Corkythehornetfan18:10, 11 September 2015 (UTC)reply
The file of Seborga's flag doesn't look right, It looks like it's been traced poorly. Can someone please retrace the crown and the shield in the flag? After that, can you also vectorise the coat of arms using the elements from the flag? Thanks in advance. --
109.77.100.184 (
talk)
16:20, 12 September 2015 (UTC)reply
@
Fry1989: Let me know what you think about this draft. I reused the central portion of the State Department Seal with some edits (updated colors, gold border in the field of thirteen stars). The typeface is Century Schoolbook, but it appears the weight used in the actual seal is some extra-heavy weight that I don't have installed. I adjusted it to space roughly identical to the spacing around the rim, although there ended up being more space between letters than the original, as a result.
Would someone be able to create
this symbol? I can't seem to figure it out myself. It has a height of 140 units and width of 120 units, and a radius for each curve of 110, but I don't know how to do the curves so they end in points like that. Fry1989eh?17:45, 1 October 2015 (UTC)reply
@
Fry1989: I think this is the one you were asking for ... How does this look? I also uploaded a file that illustrates how I created this shape, four overlapping circles with radius = 110 units. You can use Inkscape's "intersection" (two shapes at a time) to create it. Let me know if you want me to add text as well (using Roadgeek typeface?). Cheers,
Mliu92 (
talk)
13:58, 7 October 2015 (UTC)reply