This page, part of the Graphics Lab Wikiproject, is an
archive of requests for July 2008.
Please do not edit the contents of this page. You can submit new requests here.
Graphist opinion:
There's an outdated template on it as far as determining usage rights? The image seems to be different from what it originally was; vector-images.com has a different version (
http://vector-images.com/image.php?epsid=76) and this is obviously not a .png version of that particular file. Where is the current version from?
The outdated template is the one that used to say "This is a raster from VI, and they allow free use of the rasters", however further enquiries into the situation produced such hopelessly contradictory replies that Commons just gave up and is attempting to get rid of them as fast as they can get new, free replacements for those images, hence the "this tag is obsolete" (Sorry if you already knew this).
That's fine and all, but the current image is not a raster of the VI version; it looks like it's been replaced last month, without updating the license. So where'd the current one come from?
MissMJ (
talk)
22:23, 3 June 2008 (UTC)reply
That would be me, the source has been fixed but I'm not convinced the licens is the right one since the latest version was created by the New Zealand government not the UK one. The best way to get around this is to create a free version based on the blasoning only. /
Lokal_Profil22:59, 4 June 2008 (UTC)reply
Arms: Quarterly, Azure and Gules on a Pale Argent three Lymphads (sailing vessels) Sable. In the first quarter four Mullets in cross of the last each surmounted by a Mullet of the second (representing the Constellation of the Southern Cross); in the second quarter a Fleece; in the third a Garb (wheat sheaf); and in the fourth two Mining Hammers in Saltire all Or.
Crest: On a Wreath of the Colours a demi-Lion rampant guardant Or supporting a flag-staff erect proper thereon flying to the sinister the Union Flag. Supporters: On the dexter side, a female figure proper vested Argent supporting in the dexter hand a Flag-staff proper, hoisted thereon the Ensign of the Dominion of New Zealand, and on the sinister side a Maori Rangatira vested proper holding in his dexter hand a Taiaha all proper.
Motto: Onward.
The present New Zealand arms are a 1956 revision of the 1911 design. The crest was changed to the Crown of St. Edward, the quarterings were redrawn, and the supporters, instead of facing the front, now faced each other. The motto was changed to “New Zealand”.
Request: The .png standard is derived from the original image. I think you cannot do anything about the princely crown, too small on the original but both "A"s are connected together in quite a different way. Can something be done about it ?
Hektor (
talk)
21:15, 9 June 2008 (UTC)reply
I hate to be the one to bring it up, but the image is fair use so it shouldn't really be here... On the other hand I don't see that SVGifying it would do any harm.
Time3000 (
talk)
14:29, 10 June 2008 (UTC)reply
What is the track record of fair use raster to svgs surviving? I would think it's clearly a violation of the terms of (fair) use, since you can scale it to any resolution you like. We had this discussion about the Interpol logo, and it didn't fly.
Dhatfield (
talk)
14:52, 10 June 2008 (UTC)reply
Question A : The second file is in Commons, so it clearly is not fair use ? Or it should not be in Commons ? Or can we just correct the second file to the right shape, and keep it in .png format (no .svg) so that it remains non scalable... ?
This fairuse image was
uploaded in a raster image format such as
PNG,
GIF, or
JPEG. However, it contains information that could be stored more efficiently and/or accurately in the SVG format, as a
vector graphic.
If possible, please upload an SVG version of this image. After doing so, please replace all instances of the previous version throughout Wikipedia (noted under the “
File links” header), tag the old version with {{
Vector version available|NewImage.svg}}, and remove this tag. For more information, see
Wikipedia:Preparing images for upload. For assistance with converting to SVG, please see the
Graphics Lab.
If you are right, that it is forbidden to do a .SVG from a fair use .gif, the template above {{SVG|fairuse}} should be deleted, since it serves no purpose.
Hektor (
talk)
18:06, 10 June 2008 (UTC)reply
All I'm saying is that an svg (or png) designed from a fair use image remains copyrighted by the copyright holder of the fair use image. Whether the new image qualifies for fair use or not I don't know. /
Lokal_Profil18:25, 10 June 2008 (UTC)reply
So it is allowed to do a .png out of it, upload it on English language wikipedia claiming fair use - since it is non scalable contrary to .svg - , not uploading it to Commons ? Can I ask for that here ? and what about a .svg ?
Hektor (
talk)
18:30, 10 June 2008 (UTC)reply
Just had a thought, could this be recreated in SVG freely in the similar style a flag or COA can be? What I'm thinking is of calling this a flag as surely, its used in that style, similar to the
Queen's standardMangwanani(talk)18:51, 10 June 2008 (UTC)reply
The differences between CoAs and this image is that CoAs are defined by a
blazon. That means that a free version can be created from the blazon (not from an existing interpretation of it which is why
Image:Coat of Arms of Iceland.svg is in trouble). Simply recreating an image in svg does not delink the svg image from the copyright of the original, this is why images such as Coat of Arms of Johannesburg above are in trouble. /
Lokal_Profil12:29, 11 June 2008 (UTC)reply
As I understand it the template {{SVG|fairuse}} is indeed a completely pointless contradiction. "...information that could be stored more efficiently and/or accurately in the SVG format". While arguably technically true, functionally nonsense. The fair use rationale stipulates that the image must be low-res. So even if you make a vector version you still have to render it in low res. The "efficiency" difference between a 30kB PNG and a 20kB SVG is negligible, and the rendering resources needed for SVG do away with that benefit. Accuracy is irrelevant, because you still have to deliver it to the viewer at low accuracy (ie. low res). Someone could download it themselves and view it in high res, but then they'd be in violation of fair use. No matter which way you slice or dice it that template is an invitation to waste your time.
Dhatfield (
talk)
18:17, 11 June 2008 (UTC)reply
That being said, I'd be most thankful if someone cares to produce a nice fair use .PNG of the standard of Prince Albert II.
Hektor (
talk)
18:27, 11 June 2008 (UTC)reply
Find a blazon and create the image springs to mind, bearing in mind the cliché "easier said than done". The Joburg raster was uploaded by Flagmanbruce who says he is the FOTW Southern Africa administrator (for want of a better word). Mangwanani(talk)19:41, 11 June 2008 (UTC)reply
But thats just the thing. Coat of arms (and coat of arms like flags) have a blasoning, logos and monograms normaly don't, thus making free versions of the impossible. The Joburg raster is the same as that
here and is thus not an "original interpretation of the blason" but rather a copyrighted interpretation from 1997. The fact that Flagmanbruce has scanned it gives him no copyrightrelated rights. Thus since the new vector version are a derivative of these they are also copyrighted by whoever created the raster. In order to create a truly free image the new one would have to based only on heraldic rules and the text:
ARMS: Vert, a fret couped Or, the mascle voided Gules, between in chief and base respectively four shield thongs Argent; behind the shield a spine erect Or, plumed Sable.
SUPPORTERS: Two young lions Or, armed and langued Gules, each gorged of a beaded collar dancetty throughout Gules and Azure, the triangles fimbriated Or.
SPECIAL COMPARTMENT: A ground, the flanks Azure and the centre Sable, separated from one another by means of narrow piles inverted, Argent.
MOTTO: UNITY IN DEVELOPMENT.
The new image would have to be created without any regard to the "official" interreation in
Image:Jhb Arms.jpg. As it is now the new vector versions (although nice and impressive) are fair use at best and should probably not be on Commons. /
Lokal_Profil22:07, 11 June 2008 (UTC)reply
Back to the Joburg Arms, it isn't entirely based on the raster. My trace obviously was but the rest of the work done by other users was an interpretation of the raster... Mangwanani(talk)15:51, 12 June 2008 (UTC)reply
I know it isn't entirely based on the raster (let's call it the "official interpretation") but both the trace and iterpretations of the "official interpretation" are processes which create works that are derivatives of the "official interpretation", and those covered by it's copyright. As an example if you watch all of the Star Wars movies and then draw a free hand sketch of the Millenium Falcon then that is stil a derivative of the copyrighted movies.
What's needed in order to make the image free is basically that you completely forget about the "official interpretation" and do a new image based on the blazon only. As for the blazonings being written in code I wholehartedly agree, it's hard enough to understand them when they are written entierly in one language (swedish for the ones I've done) but the english ones always seem to contain about 50% french in them. There is a heraldic glossary somewhere on Wikipedia (lost the link) but to start I'd recommend asking one of the heraldicaly experienced editors to translate the description to plain english (e.g.
SanchoPanzaXXI), after a while you'll start understanding the code.
Anyhow this discussion (although important especially for a graphics lab) seems to have trailed of from the original discussion about the Monaco monogram. /
Lokal_Profil12:15, 13 June 2008 (UTC)reply
Template:Blazon Is a good starting point for translating the words. Yes CoAs are a complicated business but it's an oportunity to create freely licensed alternatives to copyrighted images which is always nice. /
Lokal_Profil00:00, 15 June 2008 (UTC)reply
Request: The Roanoke Star was lit in white following the shootings at Virginia Tech. However, the only free images I know of showing the star in this configuration are the ones I took, and parts of the star were not working correctly, as this was the first night showing the star in this configuration. This was also the regular color configuration prior to 9/11, when the pattern was changed to red-white-blue. Would a graphist be able to "fill in" the remaining parts of the star to complete the image? It would make a great addition to
Mill Mountain Star, but I'm loath to add it in its current state.
SchuminWeb (
Talk)
02:07, 12 June 2008 (UTC)reply
Graphist opinion:
Doesn't look like an easy thing to do. Only because of the awkward perspective you took from the right. I'll try this. But I can't promise anything.
XcepticZP (
talk)
08:44, 12 June 2008 (UTC)reply
Seems doable, though as XcepticZP says, it'd take some work. However, the image would lose a lot of its documentary value: for the amount of retouching it'd take to add in the missing parts, we might as well just take an image of the star in ordinary red-white-blue colors and whiten it. In fact, that would probably be an easier and less invasive approach, if you really wanted an image of the star all lit in white. Personally, I don't see why we couldn't use your image as is: that's what the star really looked like on the first night, after all. —
Ilmari Karonen (
talk)
22:46, 13 June 2008 (UTC)reply
History is always better in its raw form. I prefer it like this, it's more real; I would suggest not editing it to fill in perceived gaps. --
Golbez (
talk)
23:24, 16 June 2008 (UTC)reply
For the SVG1, the center white square is white, while the four white lateral triangles are transparent and NOT white, is this ok ?
Hektor (
talk)
20:33, 12 June 2008 (UTC)reply
In all fairness, an interpretation on a blazoning isn't going to be too different to that image as its pretty plain and simple. Not like a COA with intricate detials... Mangwanani(talk)15:29, 13 June 2008 (UTC)reply
True, but this can only be done if the flag (and the Coat of arms) is in fact based on a blason rather then just being a logos in the shapes of Flags and CoAs. Considering the nation in question it would surprise me if they were based on blasons. /
Lokal_Profil16:07, 13 June 2008 (UTC)reply
Sorry for this late reply, but could you remove the clear edge around the flag, it makes boxing the flag look strange. --
SelfQ (
talk)
13:12, 19 June 2008 (UTC)reply
Request: The article
A Tour of Somerset Towers has been split from
Somerset Towers as the page was too long. It contains approx 50 images of different church towers illustrating the architecture of the towers. The current challenge is to select the "best" to go onto the
Somerset Towers article. I'm not photography expert (as will be seen from some of my pics on there) & help with choosing the best (and deleting the worst) would be helpful. In addition many of the photos are dark or have slight angles - any help with improving them would be great.—
Rodtalk13:15, 20 June 2008 (UTC)reply
Graphist opinion:
Have started improving several of the images by correcting perspective, channels, lighting and colour-correction etc will continue editing others.
Tango22 (
talk)
23:25, 20 June 2008 (UTC)reply
Request:This COA is horribly inaccurate, see here (
[1]) how it sould look and is used on everything from documents to vehicles.
If someone could make a accurate version that would be greatly appreaciated. Note: the tekst "Hilversum" and the Blue surounding it sould not be included, that is just a part of the website, see
[2]. Another note: I am going to iceland tomorrow and wont be back for a week, so I wont be replying this week. --
SelfQ (
talk)
11:06, 21 June 2008 (UTC)reply
As long as it follows the blasoning it's correct. The version on the webpage is only one of many interpretations of the blasoning. if someone reads duch ther might be more info on
[3]. /
Lokal_Profil17:24, 21 June 2008 (UTC)reply
Graphist opinion: There was already an SVG version in use, but it looked traced. Rather than overwrite it, I uploaded a new one converted from a unicode font with burmese characters in it — ₪₪ch1902₪₪10:56, 25 April 2008 (UTC)reply
I don't think we can overwrite since the images have different licenses (I don't think you can put a license on letters and geometric shapes...) can't we just use #2 with the existing name? — ₪₪ch1902₪₪17:07, 29 April 2008 (UTC)reply
The thing is, #1 is absolutely horrible wrong, it needs to be either deleterd or overwritten. If overwritten, the licenses can also be overwritten very easily. It needs to go, it is incorrect and has no place on the 'pedia.
Chris (クリス • フィッチ) (
talk)
04:53, 30 April 2008 (UTC)reply
Before anyone can work on that svg, they need to know what the text in that COA is. The second word looks like YERS or YERI. It is ambiguous and it would help if the exact word and spelling was known.
XcepticZP (
talk)
23:55, 20 May 2008 (UTC)reply
No response. Do you think we should just mark this one as "Resolved"? The request has been completed, so letting it go stale seems like a misfile...
68.39.174.238 (
talk)
15:28, 22 June 2008 (UTC)reply
Request: Improve this NASA patch by making the colors more solid, remove Collectspace logo and convert to the right image format.
Hektor (
talk)
23:24, 9 May 2008 (UTC)reply
Graphist opinion:
Comment Can't a better version be found at the NASA website? On another note,
Mysid (
talk·contribs) is our resident mission patch maven—she's done dozens of these in SVG, maybe she can work her magic on this one :)
Fvasconcellos (
t·
c)
13:04, 10 May 2008 (UTC)reply
Answer Unfortunately I haven't found it anywhere so far except in two places : hanging on a wall in a Lockheed building I was visiting in Denver last February, and on this Collectspace web site where it was taken from. However, maybe that in a few months, when the actual flight is going to draw closer, NASA is hoing to make this patch broadly available ?
Hektor (
talk)
13:50, 10 May 2008 (UTC)reply
Ok, then in this case I remove the "resolved" tag and if someone can put DEVELOPMENT FLIGHT TEST with the Skia font...
Hektor (
talk)
06:48, 31 May 2008 (UTC)reply
Not to be picky or anything, but it looks like it should be a smaller point size with more letterspacing. The 'T' on the original patch lines up with the rocket thing.
MissMJ (
talk)
16:06, 1 June 2008 (UTC)reply
It's possible... Just a pain in the butt? xD Plus it seems silly to bother when it's so much easier to just set the type in that font. Type on a path takes about 5 seconds; provided you have the font.
MissMJ (
talk)
06:25, 8 June 2008 (UTC)reply
Was thinking mostly of the bottom ones ; it seems that everybody agrees that the top ones are commercial Skia font, but I have seen no hypothesis regarding the bottom ones (ARES I-X).
Hektor (
talk)
21:28, 8 June 2008 (UTC)reply
Now let's hope someone who has Skia on her/his computer can adjust the top text so that the 'T' lines up with the rocket...
Hektor (
talk)
17:33, 10 June 2008 (UTC)reply
Since this doesn't seem to be going anywhere and it is completed to a sufficiently high level of accuracy (my opinion), I'm marking it as resolved.
Dhatfield (
talk)
12:15, 28 June 2008 (UTC)reply
Request for SVG and modified icon for Flagged Revisions
not sighted
checked for absence of obvious vandalism
Proofed version
Unreviewed
No Vandalism
Approved
Article(s):
These images are used with the new FlaggedRevisions (currently beta-tested in German WP). They are available in png and need svg versions — but this would have some time. However, I am bringing this forward here because of my hope that at least one of the icons could be improved to a better metaphor. Everyone responsible at German WP is very busy fixing other bugs, but at the same time a lot of bad blood about "white revolutions" / "admin versus contributors" is going on on German WP. I personally believe that the wording of text messages and the icons contribute to this. Non-reviewer contributors are told that the quality of their contributions is unknown and – even though the first step review only looks for obvious vandalism – the impression is that after review the quality is ok. This is not how Flagged Revisions are intended, just unfortunate wording. The FlaggedRevs-1.png enforces this: I believe it should not be a "minus" or "forbidden" sign! I cannot do nice icons. I got positive feedback from others for my ideas but with the catch: If you can find someone who can create new icons. Can you help here?
Request:
Instead of the gray minus, please create a gray question mark icon or (perhaps even better) any icon that suggests "draft" version (a pen?).
Consider whether the yellow eye is OK, or whether a more neutral metaphor can be found. The icon should make clear that the article was checked for obvious vandalism, but not for hidden (values wrongly changed) nor for any quality aspect of the article. It has not been scrutinized, just a plausibility check was made. The wide open eye may suggest more scrutiny than was done; but I cannot think of a better metaphor myself.
The yellow eye is ok and the green sign too, but the grey minus could be a grey question mark. Furthermore the current icons have a wrong shadow effect, they should better look like buttons instead of holes. --
Nichtich (
talk)
15:56, 27 May 2008 (UTC)reply
Graphist opinion:
How about these? Reasons for selecting these icons:
Gray question mark seems perfect. No value judgement is implied and a pen will look like a blob at 18x18 pixels.
The gray-green "V" is between gray (unknown) and bright green (approved). "V" stands for Vandalism, obviously - does the German term also start with a V? If there is confusion, people will quickly learn what it means. I don't like the eye because in my opinion it is too
Eye of Sauron /
"Big Brother is watching you" for a sensitive editing environment.
The green tick is bright and easily recognised.
A comment on icon size: 18x18 pixels is very hard to work with. As you scale up, the images immediately become clearer and easier to identify, especially for people with eyes like mine - we're not going to run out of pixels so consider going a little bigger. Images at different resolutions are shown below for reference.
Dhatfield (
talk)
14:00, 14 June 2008 (UTC)reply
Many, many thanks for the good work! I like all three images. The V does work in German ("Vandalismus"), but it may not in other languages. However, it may be acceptable in many language to learn that it means no vandalism. One point: A crossed-out "V" would intuitively make more sense / be easier to remember (i.e. no' vandalism). However, I imagine this is graphically undesirable (to many lines at small angles?). --
Vigilius (
talk)
10:11, 15 June 2008 (UTC)reply
The "Approved" icon is good and even to recognise when very small: . The hook could be slighlty broader.
No problem, will do tomorrow morning.
The "V" first depends on culture and second two kinds of green are too complex. Every assumption that contains with "people will quickly learn" is an assumption too much. An eye on orange is better, the idea "... is watching you" is wanted, that's the point.
surely you are trying to say 'checked', 'passed', 'not vandalism', not trying to intimidate contributors that are not vandals. It should be a neutral or even friendly icon. Can we come up with another alternative? Blue?
The question mark is good but could be more recognisable when very small. Compare and or . This can also be read when very small: .
Thanks for the revisions Dhatfield! With the bolder font the question mark indeed has become much more readable at small sizes. And as said in the introduction, I personally disliked the eye a bit myself; I would vote for the V for the time being, with the option that someone comes up with a n even brighter idea. Some comments on German WP suggest to me that the eye is being interpreted as an indication of much more scrutiny than is actually happening. -- I have tow minor technical points: perhaps the gradient of the outer ring in ? and V could be a bit less contrasty. In small size at 1 o'clock it looks like a spot on my monitor on the ? (much less so on the V). Secondly, the OK sign () might have accidentially become non circular, both in total and with respect of the width of the outer ring. --
Vigilius (
talk)
21:38, 18 June 2008 (UTC)reply
Graphist opinion:
I created the left side of the image, I'm not sure what to do for the right side. Is this what you're looking for?
Emok (
talk)
18:31, 25 June 2008 (UTC)reply
That looks great for the left - have to run it past a physicist to make sure, but looks excellent. Only edit I'd like is the text in Verdana - it's more 'neutral' in appearance. The right image is just a couple of plates and some ellipses with radial gradients. I'll take it from here if you can't do elliptical gradients in (package X).
Dhatfield (
talk)
20:48, 25 June 2008 (UTC)reply
Sorry I didn't mention it before, can you please increase the font size so that the labels will be visible in a 200 pixel thumbnail. I think "Vacuum fluctuations" could also start with a capital. The colour of the plates in both images should also be consistent to imply that they are the same thing. Thanks. After that, we replace them in the original article and ask for comments.
Dhatfield (
talk)
12:06, 27 June 2008 (UTC)reply
The font size for the left image is larger. I tried to edit the right image to correct the plate colors, but for some reason couldn't do it without Inkscape crashing. Either MissMJ will need to do this, or I'll have to revert to the previous version with the ugly bubbles. While we're at it, shouldn't the bubbles between the plates be noticibly smaller?
Emok (
talk)
13:42, 27 June 2008 (UTC)reply
I can fudge with the colors later today. I thought the bubbles looked small enough, but I can play with the size more as well.
MissMJ (
talk)
16:43, 27 June 2008 (UTC)reply
Request:
Considering the important articles where this image appears, it deserves to be some of our best work. Initial suggestions: use different inner and outer gradients for the cones to improve contrast. Bring the time axis in front of the back of the cone as it projects into the future. Create a sense of scale to space and time in the image - I suggest the space axis should be on interstellar scales and the time axis should be in years because this image misleadingly suggests that our usual perceptions of space and time 'fit' with the speed of light.
Dhatfield (
talk)
13:10, 24 June 2008 (UTC)reply
Graphist opinion:
Hey, something I could actually possibly do! I see somebody's been perusing one too many science articles. >_> Clarification on the scales of the graph please: what do you mean by "interstellar scales" for the space axes, and what sort of intervals do you have in mind for the time axis? One year? Ten years? A hundred? Would the observer be at 1 and go down into BCE and up into
CE, or be at say... 2000?
MissMJ (
talk)
21:35, 24 June 2008 (UTC)reply
I started and it's looking pretty good, except Wikipedia won't display it (
Image:World line2.svg). =( I'm not sure what's wrong, although I have a sneaking suspicion that using Gradient Mesh to shade the cones is what's causing it to crap out. But, it's the only way I know how to get the gradient to be a shape other than straight stripes or a circle so... If anyone has any ideas, share 'em.
MissMJ (
talk)
02:12, 25 June 2008 (UTC)reply
I know what is wrong, you used adobe illustrator, and the output is not valid SVG, but instead it is a collection of SVG and png files embedded in an SVG file. If you open the file in notepad you will see there is a very large payload of useless binary data.
Jackaranga (
talk)
13:40, 25 June 2008 (UTC)reply
1. There is a load of useless binary Adobe illustrator data included in the file
2. The cones are png images embedded in the SVG file (probably you created them using a tool in Illustrator that has no equivalent in SVG)
3. The text is saved as paths instead of plain text, making it much more difficult for a third party user to translate or modify the text, negating one of the main advantages of using SVG
I don't know if it is Adobe that sucks or the users who are not using it correctly, but every time it is the same thing, Illustrator creates huge files, when a small one could be made, and often looses many of the advantages of SVG as pointed out above. In short this file has only the disadvantages of SVG and none of the advantages. The original is full of useless Adobe crap also, the file is unnecessarily large. Damn I hate that software.
Jackaranga (
talk)
13:57, 25 June 2008 (UTC)reply
Well somebody's got a chip on his shoulder.
How Adobe Illustrator codes SVGs is not something I can fix; I don't write software, I draw images.
That's probably the Gradient Mesh tool. It's the only way Illustrator can make nonlinear/nonradial gradients and it probably got rasterized when saved as SVG.
Certain label text is converted to paths because of the way it is angled along the plane to make it look three dimensional. In order to do this in Illustrator, the text loses its ability to be edited.
You can hate it all you want, that's your prerogative, but considering it is the design industry standard for editing vector images, it's really not going to go anywhere anytime soon.
MissMJ (
talk)
22:42, 25 June 2008 (UTC)reply
Wow! That was an amazingly unhelpful rant from our friend
Jackaranga. The one I just uploaded renders in the direct link
Media:World_line2.svg, but not in thumbnail. 0_o Anyway, it's the non-linear gradients on the cones. Can you try changing them to linear and maybe saving as plain SVG or similar? With the scale idea, I was thinking of some stars and galaxies scattered around on the plane - we know how much you love stars ;), and put "(Years)" on the vertical axis so you get the sense that the plane dimensions are on the order of light-years. Doable?
Dhatfield (
talk)
21:34, 25 June 2008 (UTC)reply
The labeling is doable, the gradients won't look as fancy... but they'll work, which is the important thing, I guess. >_> Stars and galaxies... Hm, that would look infinitely better with raster images, but I'll see what I can do.
MissMJ (
talk)
22:42, 25 June 2008 (UTC)reply
Great! Your first version looked better, but what can you do. You're right, a raster shifted into perspective would be the best option, but we'll get some smallminded induhvidual esteemed coeditor who will no doubt complain. Would you give it a try - it really may give great results and I have had success with embedded images in
this. I re-organised the comments so we can focus on being artists, hope you don't mind.
Dhatfield (
talk)
08:20, 26 June 2008 (UTC)reply
Working with raster images in Illustrator isn't optimal honestly, gives some really screwed up output sometimes. I just drew some stars/dots to make it look like deep space. Looks a bit cutesy, but eh. >_>
MissMJ (
talk)
20:11, 26 June 2008 (UTC)reply
I think that's a huge improvement. It definitely implies the correct scale. Let's see what the physicists say. I'll replace some of the existing images with the new one and ask for comments.
Dhatfield (
talk)
11:49, 27 June 2008 (UTC)reply
No offense guys, I think the original one was better. The gradients look amateurish on the new one (they aren't correct for the forms), the stars don't mean anything, and the original one had a nice simplicity to it. The original one looks like it could be in a science textbook. The new one looks like it is something someone made for Wikipedia. Just my two cents. When I am looking for a light cone image to use, I'm going to use the first one, not the second one. Specifying "years" for time is totally inaccurate. In fact the whole thing has gotten more inaccurate. The light cone is not just meant for interstellar distances—it's a fundamental concept to all of special relativity, and applies even on small scales. Ugly and inaccurate -- bad combination! --
98.217.8.46 (
talk)
15:10, 12 September 2008 (UTC)reply
I second all of this. The two things that were added—the "(years)" label and the stars in the hyperplane—are wrong, however well intentioned, and need to be removed. The new image doesn't look perspective-correct to me. The gradient should be more subtle so that it gives a 3D effect without being distracting. The sides of the cone should ideally be inclined at 45° to the axes, not at some random angle. With all of those changes I could accept the new image, but the result of making all of those changes would pretty much be the original image. Not that the old image is perfect, but it's a better starting point for tweaking than the new one. For the time being I'm going to switch back to the old image. --
BenRG (
talk)
16:02, 2 December 2008 (UTC)reply
Request: Is there any program (paid or free) that will convert
PDF maps to
SVG vector maps? This would greatly help various map-related projects. I will link to those conversion programs from
commons:Map resources, the above-linked WikiProjects, and other places.
PDF maps are great vector maps that can be enlarged to any size, and to zoom in on any area of the map. This is especially useful for maps. Try it in the PDF file, and see what I mean if you aren't already familiar with this (increase the percentage to 200% or 400% for example). I picked the above map at random from
commons:Category:PDF maps as an example.
To see the map you may have to download it first. For some reason I can't click on it in either the
Firefox browser, nor in MS
Internet Explorer, and have it show up in
Adobe Acrobat Reader. Once downloaded though it shows up fine when I click on the downloaded PDF file. To download it in Firefox right-click the above image, and then click "save link as" in Firefox. In Internet Explorer click the image, and then on the image description page right-click the link below it,
APISmap1.pdf, and then click "save target as." Or just click on the link and your PDF reader should open it.
I categorize many maps on Wikipedia and the commons, and good SVG maps are a great help. They allow maps to be used in other-language wikipedias just by changing the map labels in the SVG code. --
Timeshifter (
talk)
08:52, 21 June 2008 (UTC)reply
Graphist opinion:
Inkscape, with the right set of extensions and associated programs, can convert pdf to svg. You'll need phyton (unless it's being bundled with inkscape now), pstoedit, ghostview (and thus also ghostscript). You may also have to update your environment variables so that pstoedit get added to the path. I can give this one a try when I get to my laptop. Not all pdf's will be convertable to svg since some contain embeded raster images rather then vector images. /
Lokal_Profil13:03, 21 June 2008 (UTC)reply
Thanks! It is good to know that it can be done at all. I have left a link to here from
Wikipedia talk:WikiProject Maps#PDF map conversion to SVG. I, or others, might copy this thread to a subpage of the Graphic Lab, or somewhere else, later on, and use it as the beginnings of a tutorial. We can link to it from the relevant WikiProjects, etc.. There are many great vector illustrations of all kinds in PDF files. --
Timeshifter (
talk)
13:35, 21 June 2008 (UTC)reply
You can just open a PDF in Illustrator and then save it as an SVG. No extensions or associated programs needed. ;)
MissMJ (
talk)
03:40, 22 June 2008 (UTC)reply
That is great. 5 seconds! Are there any free programs that will convert PDFs to SVG? Also, how can I view the actual downloaded SVG file? I downloaded the file, and clicked it. My image editor,
IrfanView, just shows a very small image that can't be enlarged. Firefox (v2.0.0.14) and MS Internet Explorer will not open the downloaded file. I have been doing some reading concerning the Adobe SVG Viewer
[6], and Firefox 3. Inkscape opened the SVG file, but I only recently installed Inkscape, and have no clue how to enlarge the small image that I see in it. Is there any SVG viewer that will zoom as easily as Adobe Acrobat Reader does with the PDF file? And why is the SVG file 10 times larger (in kilobytes) than the PDF file. Is that due to the font problem you mentioned? --
Timeshifter (
talk)
08:38, 22 June 2008 (UTC)reply
Re: file size: it probably is a font thing. =S The map used Helvetica Neue (which actually wouldn't display for me at first because, although I have the fonts, they were named differently; had to substitute them) and since I doubt many people outside graphic design own that typeface and I have no idea how to embed a typeface into an SVG (is there even a way?) I chose the "Convert font to outlines" option when saving. It's still editable as type if you own the fonts, but I guess it creates outlines if you don't, making the file really big. No idea how to go around this, honestly, except for completely changing the typeface to something generic, which would then mess up the placement and spacing of the labels.
MissMJ (
talk)
17:50, 22 June 2008 (UTC)reply
Is it possible to increase the size of the SVG map at full size? (note: Those reading this who are not familiar with the full-size SVG map please go to
Image:APISmap1.svg with the Firefox browser, and then click the image to see it at full size). The smaller black text is hard to read at the current full size of the map. I experimented with sizing, and found that at least 1100 to 1300 pixels wide was the minimum needed. Copy [[Image:APISmap1.svg|1100px]] into a sandbox and see what I mean. The same is true for the PDF map. That needs to be viewed at the 125% setting to be legible enough. Click:
Yep, you can just open it in an image SVG editor, select everything (Ctrl+A, usually) and proportionally resize it. In Illustrator, there's a menu option to uniformly scale objects where you just type in a percentage (125% for example), I assume Inkscape probably has something similar? You could also do it manually by dragging the corner of the selection box, just make sure that it's not distorting it (in Illustator, again, you could just hold down Shift while dragging, Inkscape probably has an option to do that as well). Then just Save, and voilà! Larger nominal size (since vectors can be scaled to any size, the "full size" you are referring to is commonly called nominal size).
MissMJ (
talk)
21:16, 24 June 2008 (UTC)reply
Sure, I put some instructions on that page. Also nominally resized the map so it's bigger. Purple lines are still there though. :(
MissMJ (
talk)
22:26, 25 June 2008 (UTC)reply
Thanks for the instructions! The map looks almost exactly the same though, size-wise, to me. But I am not sure this map is worth any more effort. Due to the text problem. From
Wikipedia:WikiProject Maps/PDF map conversion to SVG#Warning is this: Some PDFs create very "bad" SVG files when converted automatically. A "bad" SVG can be identified because it will convert slowly (more than 5 seconds), it will be very large for an SVG (over 1MB) and it will render very slowly on WP - there is a long wait while the text and other images on the page are loaded, but the SVG image does not show up. Bad SVGs put a lot of strain on the Wiki servers. If you suspect you have a "bad" file, go to a Wikigraphist for a manual conversion. These will almost always be much better than the automated versions in terms of being smaller, better drawn and easier to translate because the text is created properly. --
Timeshifter (
talk)
20:49, 26 June 2008 (UTC)reply
I think that paragraph refers to PDFs that have both raster and vector images in them; Wikipedia can't display those, so the file is "bad", and big. I think this particular map will be pretty large any way you do it, just because it is so detailed (and manually tracing it would be rather tedious and useless). At the end of the day, not every SVG is going to end up being less than 100kb. *shrug*
MissMJ (
talk)
21:21, 26 June 2008 (UTC)reply
May I note that whoever did this particular map is a complete idiot (then again it is the government -_-')? The labels for the rivers and channels are actually individually placed letters instead of angled type, or type on a path. So much for easily editing them.
MissMJ (
talk)
05:27, 22 June 2008 (UTC)reply
I read somewhere that Firefox 2 does not implement all of the latest SVG code. I also read that MediaWiki implements SVG differently too, and that one has to carefully choose which format to save the SVG file in before uploading to the commons. See:
User:Phidauex/SVG tips. Maybe those are the problems. I don't know. I haven't installed Firefox 3 yet. I have been doing some reading trying to figure out how best to view SVG files. See:
Well, the original question was "Is there any program (paid or free) that will convert PDF maps to SVG vector maps?". =P Of course there's always more than one way to do something, and the free version is probably preferable for Wiki purposes/more accessible to people in general, but it seems complicated and requiring some knowledge of software and what to install. I highly doubt that I'm the only person in Graphic Lab who uses Illustrator. Perhaps along with a table of Wikigraphist ability we could have a table of what kind of software everyone uses/has access to, so that people with Illustrator could be contacted to quickly do a PDF-SVG conversion?
I thought Timeshifter was looking for a DIY guide, so I chose software that is easily available to everyone. Somebody who has Illustrator probably doesn't need a tutorial on this stuff ;) With regards to converting we just need to make it known that we are happy to do them here at the lab. I think it'll be better if the mapmakers know how to do simple conversions and bring the challenges to us. Next thing you know we'll have gone from COA Central to MapsRUs.
Dhatfield (
talk)
19:37, 22 June 2008 (UTC)reply
Thanks for the tutorial! I started a subpage. Please see:
Request:
This has been sitting around with a request for svg conversion. Vector graphics aren't my thing or I'd do this myself: original artwork from multiple sources, obvious encyclopedic value, and engages the viewer's interest in the subject. A bit small for FP candidacy in its current incarnation, but that could be addressed during the conversion. Any nibbles?
DurovaCharge!21:45, 24 June 2008 (UTC)reply
I tried for a direct copy. Please shout if you want the teeth to be perfect (lots of work for marginal overall improvement) or you want any other changes eg. size, scale lines or anything else. Cheers.
Dhatfield (
talk)
13:02, 26 June 2008 (UTC)reply
Hmmm, not sure it'll make it, but you never know with that crowd 0_o Why not? Would you leave a note when it's up - I try to avoid the FPC page due to Wikistress? Thanks.
Dhatfield (
talk)
22:27, 28 June 2008 (UTC)reply
As a veteran of FPC I can say that it will have a much better chance if you take the time to smooth out the rough edges--literally. There doesn't seem to be a curve in the entire image! Make things less blocky and you'll increase you chances dramatically.--
HereToHelp(
talk to me)22:32, 28 June 2008 (UTC)reply
Can I see the scars? That's what I get for being lazy and tracing the bitmap :) Thanks for the heads-up. I'll hand-tune it and get back to you with something better once I finish mucking about with the Utah teapot.
Dhatfield (
talk)
12:07, 29 June 2008 (UTC)reply
Hmmm, just noticed that the renderer is up to its usual tricks again. Tried to fix the offsets (white 'drop shadows'), but no luck, they have a consistent offset with the rest of the image. I'm stuck. MissMJ?
Dhatfield (
talk)
23:17, 29 June 2008 (UTC)reply
Sorry but I don't see how a blurry yellow circle looks like a star. Maybe it can be used for a representation of a sun, not a star. Even though they are the same thing. I uploaded a little sample I came up with. I think one of the talented svg people here can make it much better?
XcepticZP (
talk)
19:35, 11 June 2008 (UTC)reply
Well, what do you want done to it? Looks good to me. Unless... Since the current image is of a star constellation, you could use an icon of the Big Dipper or something? I could make that.
MissMJ (
talk)
19:57, 11 June 2008 (UTC)reply
Hmm, the constellation thing is a really good idea. Better than a single star. Let's see how it looks? But I'm definitely leaning towards your constellation idea.
XcepticZP (
talk)
21:19, 11 June 2008 (UTC)reply
There. Though remember, when used as a stub icon it's going to be pretty tiny, so any minor changes made to it are unlikely to be visible.
MissMJ (
talk)
22:33, 11 June 2008 (UTC)reply
Honestly, what I think made that sample png image of a constellation look convincing is the fact that the stars were named. It's just how I'd expect to see a constellation. Also, the idea of putting a black background is also good.
XcepticZP (
talk)
20:11, 14 June 2008 (UTC)reply
How about an eliptical background rather then a square one (sure I've seen such stub images elsewhere), maybe even one which fades towards the edges? Dark background works well. /
Lokal_Profil00:03, 15 June 2008 (UTC)reply
Hm. It looks ok, but the problem with a circular background rather than a rectangle is that the crop area needs to be larger to fit the background shape, making the image in the center (which is really the important part) even smaller when scaled down. =S Maybe a rounded rectangle?...
MissMJ (
talk)
02:15, 15 June 2008 (UTC)reply
P.S.: A fade would take up even more room, plus, the only way I could do it is with a radial blue to white gradient in Illustrator which A. removes the transparency (Illustrator doesn't support gradients to transparency, AFAIK), B. looks funky because the gradient is circular and the background is elliptical.
MissMJ (
talk)
02:18, 15 June 2008 (UTC)reply
Request: Hi, please trim off the color all around edges and trim off the slogan in the lower right corner. Thank you!!!
Banjeboi00:37, 2 July 2008 (UTC)reply
Yes. Thank you - it looks perfect. I'll try to remember that the weird error-like graphic meant that a link was the actual work-around solution.
Banjeboi23:48, 2 July 2008 (UTC)reply
The Sri Lankan coat of arms svg is being screwy and I can't open it in Illustrator. If anyone can do it, you can copy the lion from that and stick it in the center. This is mostly done, but I don't speak whatever language is used in the banner so I don't feel confident tracing the letterforms; I can't figure out what a lot of the proper strokes would be and I don't want to accidentally spell something wrong. xD
MissMJ (
talk)
02:17, 10 June 2008 (UTC)reply
Added the lion. Managed to add the text as well but I'm not 100% "ලංකා" (in the center) is right. Also please feel free to improve the text alignment. The visible text is converted to paths but a unconverted backup exists to the left of the image (outside the displayed area)./
Lokal_Profil00:40, 3 July 2008 (UTC)reply
Request: Can we please have a labeled vector diagram of Schrodinger's Cat. The cat is both alive and dead, the bottle is both whole and broken, the poison is both in the bottle and spilled and the lever and hammer are both up and down.
Dhatfield (
talk)
22:41, 25 June 2008 (UTC)reply
Request: Would it be possible to vectorize these images (the ones on Commons, not the ones saved locally to en.wikipedia)? Also, please remove the funky border effect. — scetoaux(
T|
C)19:53, 25 May 2008 (UTC)reply
Thanks but the cadet insignia needs to be cropped so that there are no margins. Right now the way these are, they don't show up too well in image caption boxes. If you could crop them that would be great. I have no idea how to work with Inkscape so I'm afraid I can't do it myself (otherwise I would). — scetoaux(
T|
C)00:38, 31 May 2008 (UTC)reply
Fixed the source info. Don't forget to link to the components that you use to make up the image. not doing so is in violation of the hat images license . /
Lokal_Profil16:27, 7 July 2008 (UTC)reply
I removed the red underneath the black text since I assume it wasn't supposed to be there. Please revert if I misunderstood. Also why did you upload it localy rather then to Commons? /
Lokal_Profil02:05, 8 July 2008 (UTC)reply
Request: This graphic looks fine when viewed full size, but it's compressed on the image page. The
Nigericin infobox has room for a properly expanded image but for some reason this is squished. (My apologies if I'm being stupid, pls explain and be gentle.) --
JaGa (
talk)
02:18, 8 July 2008 (UTC)reply
Looks as though the only problem was that Width and Hight were given with the first letter capital rather then all lower case =) /
Lokal_Profil02:40, 8 July 2008 (UTC)reply
Sorry, was away for a bit. I did it in Inkscape, but not automagically: I just used the bezier tool, putting control points wherever I thought they'd be needed, and then went around with the curve editing tool fitting the curve as precisely as possible. To find the inaccurate bits, I put the jpeg in front of the curve, and deleted it, then went "undo - redo" to see which parts moved. --
Slashme (
talk)
10:41, 7 July 2008 (UTC)reply
I think this map will benefit from being turned 90 degrees counterclockwise and made bigger so that the labels aren't squished in. Lowercase labels will help too, IMO.
MissMJ (
talk)
19:37, 7 July 2008 (UTC)reply
Yes, only I wonder if you could increase the size of the base map to as to avoid the appearance of all the text crowding it (Also it would allow writing "F.H." out) and secondly, the edges of the things like the North arrow would appear sharper and less "traced". Finally, I suspect the text should be "Jones' fleet" for the possessive. The idea of the turning it is probably a good one; if anyone later determines it would be better the original way, it should be easier to change an SVG than a GIF. Thanx,
68.39.174.238 (
talk)
21:10, 7 July 2008 (UTC)reply
Here you go. Oh and
Auawise, when making new versions of a file you may simply upload them over the previous image. This prevents Wikipedia from having a bazillion versions of one file under different names. If you're starting completely over or doing something radically different, a new file with a version number may be warranted, but for minor fixes (color changes, etc.) it's not necessary.
MissMJ (
talk)
21:56, 8 July 2008 (UTC)reply
Sorry, was away for a bit. I did it in Inkscape, but not automagically: I just used the bezier tool, putting control points wherever I thought they'd be needed, and then went around with the curve editing tool fitting the curve as precisely as possible. To find the inaccurate bits, I put the jpeg in front of the curve, and deleted it, then went "undo - redo" to see which parts moved. --
Slashme (
talk)
10:41, 7 July 2008 (UTC)reply
I think this map will benefit from being turned 90 degrees counterclockwise and made bigger so that the labels aren't squished in. Lowercase labels will help too, IMO.
MissMJ (
talk)
19:37, 7 July 2008 (UTC)reply
Yes, only I wonder if you could increase the size of the base map to as to avoid the appearance of all the text crowding it (Also it would allow writing "F.H." out) and secondly, the edges of the things like the North arrow would appear sharper and less "traced". Finally, I suspect the text should be "Jones' fleet" for the possessive. The idea of the turning it is probably a good one; if anyone later determines it would be better the original way, it should be easier to change an SVG than a GIF. Thanx,
68.39.174.238 (
talk)
21:10, 7 July 2008 (UTC)reply
Here you go. Oh and
Auawise, when making new versions of a file you may simply upload them over the previous image. This prevents Wikipedia from having a bazillion versions of one file under different names. If you're starting completely over or doing something radically different, a new file with a version number may be warranted, but for minor fixes (color changes, etc.) it's not necessary.
MissMJ (
talk)
21:56, 8 July 2008 (UTC)reply
Request: I recently added a png version of this graph, made by matlab, source code is in the png talk page. At the image page, an infobox appeared saying that the image would be better in svg format, so I tried to change it. Everything worked, except the aspect ratio: the svg version is wrong. But I also noticed that the svg file is larger than the png file. So I have two question: 1. Is svg really better in this case? I would think not, because the svg file is larger, but maybe that's not relevant. 2. Can somebody help me change the aspect ratio of the svg file?--
EdgeNavidad (
talk)
07:28, 8 July 2008 (UTC)reply
Graphist opinion:
Think something had got messed up in the aspect ratio where the image got forced from 399x306px to 200x200px. The quick fix is to wash it through Inkscape. I.e. open the file in inkscape and copy everything into a new file with the same size (which I did). The nicer solution is to remove the offending bit of code, this however normally requires a few tries since mediaWiki seems to render everything in it's own unique way.
Even if the svg has a larger filesize I'd still say it's preferable since it's easily translated and updated, but then that's just me./
Lokal_Profil15:13, 8 July 2008 (UTC)reply
Sorry, that became slightly more then a day. Anyhow a second couple of eyes catching any spelling errors is (as always) much appreciated. Apart from that I thin this is done. /
Lokal_Profil00:25, 10 July 2008 (UTC)reply
I worked 30 hours this 2 last weeks to create -in English- a really complete 15 pages long Tutorial:Cartography (basic).svg. This tutorial aim to take a perfect newbee in both Inkscape and map making, and teach him/her in 15 pages and one hour of explanations and exercices all what he need to know to make respectable, nice, and reliable wikimaps. This was made according to all I learnt on the French Map lab.
Despite I'm French, I wrote it in English since 8 wikipedians out of 10 can understand basic English, while only 1 out of 10 for basic French. Also, this Tutorial need first a big copyedit. Afterwhat: your advices are really encouraged ! (dowload and read the SVG version to test it)
So (please !), one volunter to copyedit this long svg (probably 2 hours of work, . (one only, or the second after the first).
Note : some indications are in the SVG ; the file have 11 sections (you can copyedit just 2: each help is welcome) ; 3 are missing and will be add later.
Okay, that's it down to the sources section. I am doing some rewriting as well as fixing language issues to make sure the intent of the text is as clear as possible. Also fixing layout issues. Onwards!
vlad§ingertlk02:53, 10 July 2008 (UTC)reply
...And, that's it for now. It's 1AM here and I'm off to bed. I went through every section, moved stuff around (though I may have neglected to add notes saying so). Somebody else is free to scan it again and fix anything I missed. Of course, I wasn't able to edit the parts that aren't written. I'm impressed with this - must have taken a while considering it took me ~2.5 hrs just to tweak.
vlad§ingertlk05:07, 10 July 2008 (UTC)reply
A big thanks for your tweak on Tutorial-cartography_(basic).svg , I think the final version will help map creation and wikicartographers' beginning greatly. After my big effort to create this tuto, a big thanks to you for this correction, I quickly overviewed it : that's really better ! I also appreciated when I seen you display good initiative by changing many sentences and expanding some explanations with your knowledge. This was completly my aim : make a wiki-SVG-tutorial, first started by one big effort, then improve by every cartographers when better explanations are possible, exactly like a wiki page ^-^y . For all this : a big thanks to you.
As I said, I will let other cartographer read it and expand it if they want. I will probably create the missing sections separately on my PC, then, in one or 2 weeks, I will include them and make some structural update (change some sections' order). I hope the version 1.0 (complete and corrected) will be available for August 2008 :]
Note on next steps
Copyedit = Done ; I mark the request as "Resolved" to avoid waste of time of other graphists.
3 missing sections = to do in the 2 next weeks, with little reorganisation (sections' order), and rewriting of the bolean section (cf Inkscape 0.46 new tools).
last copyedit: will then need a small copy edit.
Final release and review: August 2008, graphists will be welcome to keep this tutorial as best and as up to date as possible.