Unfortunately most of us here are very far from experts on this subject, so I presume this has been left here because it is too difficult for us to interpret exactly what you want. Is there any existing tree elsewhere that we could use for inspiration? Could you perhaps sketch out the basics for us? Even a crude drawing would be most helpful and give this request a better chance of being fulfilled.
NikNakstalk -
gallery18:05, 28 March 2015 (UTC)reply
I'm afraid I am also far from expert on this subject. But there is a recent open-access
citrus taxonomy article on BMC Genetics (plus a CC_BY
citrus article and a CC-0
citrus article on
PLOS One). Hurray for
open access! So the easiest way would be to just copy the figures from these articles. I particularly like figure three of the BMC paper; I've added it above and in the
Citrus and
Citrus taxonomy articles. But it's not a family tree, so it doesn't meet the request.
There are several phylogenetic tree diagrams in these three articles. On the other hand, these figures are a bit indigestible. The scientific names are sometimes not annotated with the common names, there are no pictures, that sort of thing. Ideally, one could represent this information as a cladogram using the
cladogram template (
example), which would let names be links, with the current page's citrus fruit highlighted, etc..
WP:Biology and
WP:Genetics might be able to help. It would probably be good to run any modified figures past the corresponding authors of the academic article they are modified from; their e-mails are listed on the articles, any they could spot any errors.
Looked back at this and I'm talking through my hat. What I originally wanted was something that would show the network of hybridizations in Citrus. A straight phylogeny or cladogram is not going to work, because citrus varieties hybridize so much.
For example, suppose you have four original species, A, B, C, and D. You also have E, which is a cross or hybrid made by interbreeding A and B, and F, which is a cross of B and C, and G, which is CxD. Then you have H, which is GxA, and I, which is FxG, and J, which is CxH. And so on. The challenge is to plot all this intelligibly.
In real life, you have at least four putative pure species:
citron,
pummelo,
mandarine, and
papeda (~Citrus medica, Citrus maxima, Citrus reticulata, and Citrus micrantha). The hybrids are things like common oranges, grapefruit, etc., and they are listed in the
Citrus and
Citrus hybrids pages.
I'm afraid I'm even more confused than I was to begin with, if that were possible! As far as I can tell, the only way a cladogram would really work is to cut it down to the original species, their hybrids, and one further layer of hybrid-hybrid/hybrid-original combinations below that. Anything more would be totally unwieldy and frankly I think even that would be difficult. The alternative "family tree" option might be worth a try, but again, there are probably too many for it to look reasonable. I don't think a 3D solution would be any better and would be just as confusing to understand, I think.
Given you want a solution that shows most of the species, I think the figures you've found are the best possible solution. For individual fruits, perhaps we could have the originals and hybrids, and then only one or two fruit below them (the one in question and direct ancestors) with its ancestry and link to the original species. I don't know if I've explained what I mean particularly well, but that's all I can think of.
NikNakstalk -
gallery17:27, 6 April 2015 (UTC)reply
@
HLHJ and
NikNaks:I took a run at this, to try to get a feel for how manageable/readable this would even be. Unfortunately the diagram quickly became erratic and another problem I had was trying to gather all the information. It could be that research is still on-going or just the information is missing from wikipedia. I had more data to plot but I kept running into missing or conflicting information so I stopped adding to the chart. While it would be nice, I think there is too much missing information and not a great way to plot all of this data. Ideally this would be easier to plot and navigate this data in a interactive javascript based chart (i.e. like vis.js), but at this time that is not something that can be done through wikipedia. Here is the image that I attempted.
Offnfopt(talk)07:18, 22 April 2015 (UTC)reply
@
Offnfopt and
NikNaks: The graph looks good. The conflicting data sounds tough, I'm sorry. It may still be possible to make the graph usable and useful; it looks like the research is advancing fast enough that being a bit out-of-date is inevitable. I've added a good additional source at the top of the request, which may resolve some of the conflicts in the data; if you post anything that still has conflicting information, I will try and get a solid source to decide one way or another, or we can just take it off the chart. Do you think this would work? How many conflicts are there? Alternately, we could put anything that hasn't got solid genetic evidence for its ancestry with different lines (gray? solid for established parentage?). It is also possible to say which parent was the pollen parent in some cases. I will try to get some better data when I have time; there are good sources out there.
Cladograms: You're right, a cladogram for hybrids totally wouldn't work. I did do a rough phylogram for some
Australian limes (which together with
trifoliate oranges and
kumquats are also minor ancestral taxa; trifoliates are just used as rootstock, but
calamondins are a kumquat hybrid). I also have
a paper with a phylogenetic tree of some of the naturally-occuring taxa, but I'd want to check the grouping against another study or two, as some branches were pretty tight. I tried putting it at the top of Offnfopt's graphic, which seemed to work visually if the lines were distinctly trellis-like, but I think it might be better as a phylogram directly in the
Citrus taxonomy page. It would be possible to insert causes of speciation on such a graph.
Lime ancestry: The name "lime" seems to get attached to almost anything.
This paper says "Lime is often considered to be a chance seedling, with citron (C. medica) and papeda possibly being its parents"; it means "Lime (synonym: West Indian lime, Mexican lime, and key lime),
C. aurantifolia (Cristm.) Swingle". But they have no data on that, it's just background.
This paper has actual data supporting the statement. I think we can safely call
key lime a citron x papeda hybrid.
Changes: One possible improvement would be a visual separation of pure species from hybrids, or marking those that breed true (similar results, I think; most hybrids are asexually propagated). The only obvious changes are tidiying the key lime into citron x papeda hybrid parentage and promoting kumquats to the top level (and possibly the addition of the
Australian limes at the top level).
HLHJ (
talk)
11:34, 24 April 2015 (UTC)reply
@
Offnfopt: Thank you very much for your second draft. I've edited the
Key Lime article to make it clear that the first is a papeda x citron hybrid, and the
Lime article to show that "limes" are a totally un-phylogenetic, paraphyletic group; so you can amalgamate the boxes for "Lime" and "
Key Lime" and call the box "Key Lime". Australian limes do not have papeda parents; they are non-hybrid species, as far as I know. I'm sorry to change my mind, but I think it would be simpler to get rid of Australian Lime, unless you want to show their hybrids, which aren't very economically important yet. Mikan is a Japanese word for mandarin; best as I can tell there are Onshu, Unshu, Kishu, Ehime etc. mikans, all apparently named after growing regions. You could place "Mikan" under mandarins. Ponkan seems to be a mandarin-pomelo hybrid [1]. The limetta's maternal line is descended from a pomelo [2], and as it is sweet it must have mandarin ancestry (only mandarins are sweet, of the main ancestral species). Does that tie up the loose ends on the graph?
HLHJ (
talk)
23:07, 25 April 2015 (UTC)reply
@
HLHJ: I'm still going through a bunch of data, I just uploaded a new version before I saw your ping, problem is there is just so much data to go through to try to sort all this out. I end up spending more time going through data and only a few minutes on the actual diagram. I'm sure I'll be uploading a newer version soon enough so expect it to change.
Offnfopt(talk)23:22, 25 April 2015 (UTC)reply
@
HLHJ: I went ahead and made those changes, I also went ahead and changed Mikan to the alternative name "satsuma mandarin" to make it easier to search for. Unfortunately after I uploaded the recent revision I found
a news article from 2012 about some additional research done, seems I will need to redo parts of the diagram again based on that information. I also need to find where to link "Trifoliate Orange" to.
Offnfopt(talk)04:43, 26 April 2015 (UTC)reply
@
Offnfopt: Trifoliate Orange is easy; it's top-level, just like kumquats. It is also less realated to all the others than they are to each other[3], although the clustering within the others is not yet known (lots of papers disagree).
The LA times article seems a bit odd. There's a
Nature paper on the
full-genome sequencing of the sweet orange (CC, but sadly NC, so we can't use their images). It says something quite different (in section "Heterozygosity and hybrid origin of sweet orange"): "we hypothesized that sweet orange originated from an interspecific hybridization with pummelo as the female parent and mandarin as the male parent followed by a backcross with a male mandarin (sweet orange = (pummelo × mandarin) × mandarin)". This is shown in their
Figure 3. Their evidence seems fairly strong. I haven't been able to find Gmitter's paper, even on his personal site, so he probably hasn't published yet, so I can't tell if there might be anything to account for the disrepancy. I note that the full-genome paper was published two days after the LA article, so Dr. Gmitter probably hadn't read it when he made the conference talk. Also, according to an article published two years after he spoke, the Ponkan unexpectedly contains some pomelo genes ("two homozygous fragments for a C. maxima haplotype (ma1/ma1) and one fragment heterozygous for two C. maxima haplotypes (ma1/ma2) were found in ‘Ponkan’ mandarin."[4]), so the hypothesized ancestry would not work. I suggest we take the full-genome paper's interpretation and ignore the LA Times article.
HLHJ (
talk)
09:38, 26 April 2015 (UTC)reply
For a very large key as to what is hybrid with what, this
study of a large citrus variety collection is a good source. It represents it graphically (Fig. 1), with the key (relating the numbers on the coloured bars to the variety name) in the supplementary material. Can you read it, or is it paywalled? It shows 370 varieties, obviously too many to fit in the chart, but for instance limetta (#58) is represented clearly as a pummelo(~1/5) x citron(~4/5), so my "mandarin" guess was wrong.
It would be important to note in the metadata for the image that the lines to ancestors just represent ANY genetic contribution, and do not represent the proportion (e.g. 1:3 for sweet orange, ~1:4 for limetta).
HLHJ (
talk)
10:39, 26 April 2015 (UTC)reply
You can also find some images at
[1], but also here they are not really large. But parts of it, especially the coat of arms, can be taken from other already vectorized images.--
Antemister (
talk)
08:54, 29 March 2015 (UTC)reply
Found this seal on wikimedia, but it's pretty fuzzy and many of the details are lost. Also, the vectorised elements of the Royal Coat of Arms can be found at the linked Commons Category, although none seem to match the lion and unicorn exactly.
Please can you move the dark grey squares to the centre of the image, as these are the 'Crossbenchers'. Also please move the yellow squares (Lib Dem peers) next to the red squares, in place of the grey squares, as this party is now in opposition. The blue squares can stay at the bottom. Thanks --
HazhkTalk17:59, 8 May 2015 (UTC)reply
Graphist opinion(s)
Based on your request I've uploaded a new file entitled 'House of Lords current (alt).svg' which moves the crossbenchers to the middle. Let me know if this is what you had in mind. I also note the current
House of Lords article has a count of 11 "other parties" so I've updated the colors accordingly. Cheers,
Mliu92 (
talk)
23:56, 17 May 2015 (UTC)reply
Could someone please create an SVG file of the Royal Standard used by
William III of England from 1689-1702 by placing the arms of
House of Nassau (above right) in the center of the 1603 Royal Standard (above left).
We need some examples to illustrate the principles of Mayan hieroglyphic writing. The best would be svg drawings of the different ways to write the word "b'alam" (there are five different ways to do so). They can be found in this
pdf on page 24 and 25. A separate svg file would be required for each of the five variants. --
·maunus ·
snunɐɯ·19:08, 22 May 2015 (UTC)reply
@
Maunus: Now there is
a draft to look at. I guessed that you wanted lines like I have done and not exactly as the diffused images. I have reused the different parts and scaled and edit them to some extent. If you want them in any other way just tell me, and I show them all together (you will get five separate files) so that you can compare sizes. Give me feedback on this, thanks. --
Goran tek-en (
talk)
14:25, 31 May 2015 (UTC)reply
The clarity is great, and the size is fine too. The one thing I would change if I were able would be to use lines of variable thickness like for example using a calligraphy pen or its equivalent digital tool. If this is too much work, the images can be used as they are.
·maunus ·
snunɐɯ·15:34, 31 May 2015 (UTC)reply
Yes I understand, to get it more handmade, lively. If you want me to use a calligraphy pen it would be good if you told me up front, now I will have to redraw it. The calligraphic pen doesn't always do what you want it to do, can bee tricky. I have made another test on the left figure
here for you to look at. It's more of just making the line more lively. I can redraw it with calligraphy pen but I will have to test first. Let me know what you think and want, thanks. --
Goran tek-en (
talk)
18:33, 31 May 2015 (UTC)reply
Yeah, sorry I didn't think of that untill I saw them with the straight line. The calligraphic pen would make it more liveli and also closer to the way the hieroglyphs look on Maya pottery where they were painted with brushes. I think these ones are better, except the first one which has a very shaky line for some reason. I think it would likely be more beautiful with a calligraphic pen, but the way they are now they serve the purpose of illustrating the principles of Maya writing, so I leave it up to you if you want to take the extra time to try out the calligraphic style.
·maunus ·
snunɐɯ·18:44, 31 May 2015 (UTC)reply
Yes, I can see the line gets a little shaky. I do like the more variable line though. But probably unless you can get it a little smoother we should stick with the first type - though maybe with a thicker line.
·maunus ·
snunɐɯ·19:43, 4 June 2015 (UTC)reply
Can you please vectorize from this signature of Richard Armstrong from this
external image. It takes to January 9, 1860. Keep the downward swoop of the "g" but don't include "Rev." Thanks.--
KAVEBEAR (
talk)
05:38, 8 June 2015 (UTC)reply
This request is somewhat of a challenge; one that I can't take on myself because I'm more familiar with mapmaking than flag making. Here's the request:
1) I have found a fair use copy of the seal of the city of
Utica, New York, which I am requesting to be vectorized. I made it transparant and cropped it a little from where I originally found it (which was a fluke find in itself). The color of the seal should be black.
2) After this, I am requesting that a flag be made of the city by using the fair use image linked above, which I do understand is somewhat distorted and cutoff, but this is the best image I could find and has the basic proportions and colors in place (needed for the seal). The then-vectorized needs to be incorporated into the flag, because other than the dual yellow stripes there isn't much to it. I think the color picker can handle the color matching.
This is probably a cryptic request so if anyone has questions and they are willing to help, just ping me and I will respond immediately. --
Buffaboytalk23:55, 14 April 2015 (UTC)reply
Graphist opinion(s)
Buffaboy, here is the seal, if I get a chance, I'll try to do the flag. The
Template:Non-free flag uses the wording "low resolution" image, so a SVG would not be considered that. If I'm misunderstanding the licensing issue feel free to correct me, I'm not all that familar with these licensing topics. I can try to make a non-SVG (i.e. PNG) version if you want? (unless there is a reason that allows for a SVG)
Offnfopt (
talk)
03:21, 16 April 2015 (UTC)reply
@
Offnfopt: Thank you for creating the seal! O.K., I realized that there is a license issue. Technically, the image of the news capture is not the flag itself, therefore in retrospect the non free flag identifier should not apply. In addition, I believe that if the license were strictly interpreted, the news station would have been in big trouble for filming the flag as it is a representation being captured on film. If I am 100% wrong, somebody please correct me!
Buffaboytalk05:00, 16 April 2015 (UTC)reply
@
Buffaboy: I have a PNG version of the flag done, here is the link for it.
But I'm still not sure as far as the licensing goes. I don't have much experience with dealing with the non-free images. I don't know if since the flag/seal is my work that I can pick the license, or since the image portrays that of a seal/flag etc if that takes precedence so has to be set as non-free. I went ahead and set it to non-free to be safe, but if for w/e reason turns out I can put those under different licensing terms then I grant to release them to public domain, if that is even possible legally speaking.
Offnfopt (
talk)
05:49, 16 April 2015 (UTC)reply
@
Offnfopt: Wait, I have a better idea. I can request deletion of the news screenshot, but for sourcing purposes, I can link to the webpage where the image was found to base the flag off of. This presumably should clear up all of the copyright issues associated with the files. I (again) could be wrong.
Buffaboytalk16:11, 16 April 2015 (UTC)reply
@
Offnfopt: I was told at
WP:VPP that you can now upload the SVG file of the flag, and then attach a public domain license to it since the flag originated before 1923 (which it did). This way, there won't be any more licensing issues, we can delete the PNG flag and its screenshot counterpart, and link to the original webpage where the news screenshot was found.
Buffaboytalk20:27, 18 April 2015 (UTC)reply
It was suggested
at the photography workshop that I bring this here... I'm looking for a way to scale this image up cleanly to around 100-200px. Vectorising it seems like the most sensible solution, but I'm not actually sure it can be done while maintaining the distinctive aspects of the seal. Opinions and suggestions would be welcome; I'm clueless at this stuff.
Yunshui雲水21:26, 29 May 2015 (UTC)reply
Graphist opinion(s)
It's really hard from this very small image in px to do anything and still keep the aspects and be true to the original. Do you want a drawing with straight lines like I guess the seal had from the beginning or with uneven edges as every print from the seal will be. It's two completely different things, but you will still have to find an image with higher resolution. If one uses this as an "original" it would be a lot of guessing. If you want me anything you will have to ping me, thanks. --
Goran tek-en (
talk)
18:14, 30 May 2015 (UTC)reply
I figured the size would be an issue from the outset - I haven't been able to find anything larger. At the moment one of the photography workshop graphists has suggested creating a raster image, so I'll see how that works out before asking you to do any additional work. Thanks nevertheless.
Yunshui雲水11:20, 2 June 2015 (UTC)reply
The .jpg version is exactly what I was after (actually, it's even better). Thank you very, very much, but please don't feel the need to do anything further on my account (unless you're bored and actually want to).
Yunshui雲水16:25, 2 June 2015 (UTC)reply
Unfortunately most of us here are very far from experts on this subject, so I presume this has been left here because it is too difficult for us to interpret exactly what you want. Is there any existing tree elsewhere that we could use for inspiration? Could you perhaps sketch out the basics for us? Even a crude drawing would be most helpful and give this request a better chance of being fulfilled.
NikNakstalk -
gallery18:05, 28 March 2015 (UTC)reply
I'm afraid I am also far from expert on this subject. But there is a recent open-access
citrus taxonomy article on BMC Genetics (plus a CC_BY
citrus article and a CC-0
citrus article on
PLOS One). Hurray for
open access! So the easiest way would be to just copy the figures from these articles. I particularly like figure three of the BMC paper; I've added it above and in the
Citrus and
Citrus taxonomy articles. But it's not a family tree, so it doesn't meet the request.
There are several phylogenetic tree diagrams in these three articles. On the other hand, these figures are a bit indigestible. The scientific names are sometimes not annotated with the common names, there are no pictures, that sort of thing. Ideally, one could represent this information as a cladogram using the
cladogram template (
example), which would let names be links, with the current page's citrus fruit highlighted, etc..
WP:Biology and
WP:Genetics might be able to help. It would probably be good to run any modified figures past the corresponding authors of the academic article they are modified from; their e-mails are listed on the articles, any they could spot any errors.
Looked back at this and I'm talking through my hat. What I originally wanted was something that would show the network of hybridizations in Citrus. A straight phylogeny or cladogram is not going to work, because citrus varieties hybridize so much.
For example, suppose you have four original species, A, B, C, and D. You also have E, which is a cross or hybrid made by interbreeding A and B, and F, which is a cross of B and C, and G, which is CxD. Then you have H, which is GxA, and I, which is FxG, and J, which is CxH. And so on. The challenge is to plot all this intelligibly.
In real life, you have at least four putative pure species:
citron,
pummelo,
mandarine, and
papeda (~Citrus medica, Citrus maxima, Citrus reticulata, and Citrus micrantha). The hybrids are things like common oranges, grapefruit, etc., and they are listed in the
Citrus and
Citrus hybrids pages.
I'm afraid I'm even more confused than I was to begin with, if that were possible! As far as I can tell, the only way a cladogram would really work is to cut it down to the original species, their hybrids, and one further layer of hybrid-hybrid/hybrid-original combinations below that. Anything more would be totally unwieldy and frankly I think even that would be difficult. The alternative "family tree" option might be worth a try, but again, there are probably too many for it to look reasonable. I don't think a 3D solution would be any better and would be just as confusing to understand, I think.
Given you want a solution that shows most of the species, I think the figures you've found are the best possible solution. For individual fruits, perhaps we could have the originals and hybrids, and then only one or two fruit below them (the one in question and direct ancestors) with its ancestry and link to the original species. I don't know if I've explained what I mean particularly well, but that's all I can think of.
NikNakstalk -
gallery17:27, 6 April 2015 (UTC)reply
@
HLHJ and
NikNaks:I took a run at this, to try to get a feel for how manageable/readable this would even be. Unfortunately the diagram quickly became erratic and another problem I had was trying to gather all the information. It could be that research is still on-going or just the information is missing from wikipedia. I had more data to plot but I kept running into missing or conflicting information so I stopped adding to the chart. While it would be nice, I think there is too much missing information and not a great way to plot all of this data. Ideally this would be easier to plot and navigate this data in a interactive javascript based chart (i.e. like vis.js), but at this time that is not something that can be done through wikipedia. Here is the image that I attempted.
Offnfopt(talk)07:18, 22 April 2015 (UTC)reply
@
Offnfopt and
NikNaks: The graph looks good. The conflicting data sounds tough, I'm sorry. It may still be possible to make the graph usable and useful; it looks like the research is advancing fast enough that being a bit out-of-date is inevitable. I've added a good additional source at the top of the request, which may resolve some of the conflicts in the data; if you post anything that still has conflicting information, I will try and get a solid source to decide one way or another, or we can just take it off the chart. Do you think this would work? How many conflicts are there? Alternately, we could put anything that hasn't got solid genetic evidence for its ancestry with different lines (gray? solid for established parentage?). It is also possible to say which parent was the pollen parent in some cases. I will try to get some better data when I have time; there are good sources out there.
Cladograms: You're right, a cladogram for hybrids totally wouldn't work. I did do a rough phylogram for some
Australian limes (which together with
trifoliate oranges and
kumquats are also minor ancestral taxa; trifoliates are just used as rootstock, but
calamondins are a kumquat hybrid). I also have
a paper with a phylogenetic tree of some of the naturally-occuring taxa, but I'd want to check the grouping against another study or two, as some branches were pretty tight. I tried putting it at the top of Offnfopt's graphic, which seemed to work visually if the lines were distinctly trellis-like, but I think it might be better as a phylogram directly in the
Citrus taxonomy page. It would be possible to insert causes of speciation on such a graph.
Lime ancestry: The name "lime" seems to get attached to almost anything.
This paper says "Lime is often considered to be a chance seedling, with citron (C. medica) and papeda possibly being its parents"; it means "Lime (synonym: West Indian lime, Mexican lime, and key lime),
C. aurantifolia (Cristm.) Swingle". But they have no data on that, it's just background.
This paper has actual data supporting the statement. I think we can safely call
key lime a citron x papeda hybrid.
Changes: One possible improvement would be a visual separation of pure species from hybrids, or marking those that breed true (similar results, I think; most hybrids are asexually propagated). The only obvious changes are tidiying the key lime into citron x papeda hybrid parentage and promoting kumquats to the top level (and possibly the addition of the
Australian limes at the top level).
HLHJ (
talk)
11:34, 24 April 2015 (UTC)reply
@
Offnfopt: Thank you very much for your second draft. I've edited the
Key Lime article to make it clear that the first is a papeda x citron hybrid, and the
Lime article to show that "limes" are a totally un-phylogenetic, paraphyletic group; so you can amalgamate the boxes for "Lime" and "
Key Lime" and call the box "Key Lime". Australian limes do not have papeda parents; they are non-hybrid species, as far as I know. I'm sorry to change my mind, but I think it would be simpler to get rid of Australian Lime, unless you want to show their hybrids, which aren't very economically important yet. Mikan is a Japanese word for mandarin; best as I can tell there are Onshu, Unshu, Kishu, Ehime etc. mikans, all apparently named after growing regions. You could place "Mikan" under mandarins. Ponkan seems to be a mandarin-pomelo hybrid [1]. The limetta's maternal line is descended from a pomelo [2], and as it is sweet it must have mandarin ancestry (only mandarins are sweet, of the main ancestral species). Does that tie up the loose ends on the graph?
HLHJ (
talk)
23:07, 25 April 2015 (UTC)reply
@
HLHJ: I'm still going through a bunch of data, I just uploaded a new version before I saw your ping, problem is there is just so much data to go through to try to sort all this out. I end up spending more time going through data and only a few minutes on the actual diagram. I'm sure I'll be uploading a newer version soon enough so expect it to change.
Offnfopt(talk)23:22, 25 April 2015 (UTC)reply
@
HLHJ: I went ahead and made those changes, I also went ahead and changed Mikan to the alternative name "satsuma mandarin" to make it easier to search for. Unfortunately after I uploaded the recent revision I found
a news article from 2012 about some additional research done, seems I will need to redo parts of the diagram again based on that information. I also need to find where to link "Trifoliate Orange" to.
Offnfopt(talk)04:43, 26 April 2015 (UTC)reply
@
Offnfopt: Trifoliate Orange is easy; it's top-level, just like kumquats. It is also less realated to all the others than they are to each other[3], although the clustering within the others is not yet known (lots of papers disagree).
The LA times article seems a bit odd. There's a
Nature paper on the
full-genome sequencing of the sweet orange (CC, but sadly NC, so we can't use their images). It says something quite different (in section "Heterozygosity and hybrid origin of sweet orange"): "we hypothesized that sweet orange originated from an interspecific hybridization with pummelo as the female parent and mandarin as the male parent followed by a backcross with a male mandarin (sweet orange = (pummelo × mandarin) × mandarin)". This is shown in their
Figure 3. Their evidence seems fairly strong. I haven't been able to find Gmitter's paper, even on his personal site, so he probably hasn't published yet, so I can't tell if there might be anything to account for the disrepancy. I note that the full-genome paper was published two days after the LA article, so Dr. Gmitter probably hadn't read it when he made the conference talk. Also, according to an article published two years after he spoke, the Ponkan unexpectedly contains some pomelo genes ("two homozygous fragments for a C. maxima haplotype (ma1/ma1) and one fragment heterozygous for two C. maxima haplotypes (ma1/ma2) were found in ‘Ponkan’ mandarin."[4]), so the hypothesized ancestry would not work. I suggest we take the full-genome paper's interpretation and ignore the LA Times article.
HLHJ (
talk)
09:38, 26 April 2015 (UTC)reply
For a very large key as to what is hybrid with what, this
study of a large citrus variety collection is a good source. It represents it graphically (Fig. 1), with the key (relating the numbers on the coloured bars to the variety name) in the supplementary material. Can you read it, or is it paywalled? It shows 370 varieties, obviously too many to fit in the chart, but for instance limetta (#58) is represented clearly as a pummelo(~1/5) x citron(~4/5), so my "mandarin" guess was wrong.
It would be important to note in the metadata for the image that the lines to ancestors just represent ANY genetic contribution, and do not represent the proportion (e.g. 1:3 for sweet orange, ~1:4 for limetta).
HLHJ (
talk)
10:39, 26 April 2015 (UTC)reply
You can also find some images at
[1], but also here they are not really large. But parts of it, especially the coat of arms, can be taken from other already vectorized images.--
Antemister (
talk)
08:54, 29 March 2015 (UTC)reply
Found this seal on wikimedia, but it's pretty fuzzy and many of the details are lost. Also, the vectorised elements of the Royal Coat of Arms can be found at the linked Commons Category, although none seem to match the lion and unicorn exactly.
Please can you move the dark grey squares to the centre of the image, as these are the 'Crossbenchers'. Also please move the yellow squares (Lib Dem peers) next to the red squares, in place of the grey squares, as this party is now in opposition. The blue squares can stay at the bottom. Thanks --
HazhkTalk17:59, 8 May 2015 (UTC)reply
Graphist opinion(s)
Based on your request I've uploaded a new file entitled 'House of Lords current (alt).svg' which moves the crossbenchers to the middle. Let me know if this is what you had in mind. I also note the current
House of Lords article has a count of 11 "other parties" so I've updated the colors accordingly. Cheers,
Mliu92 (
talk)
23:56, 17 May 2015 (UTC)reply
Could someone please create an SVG file of the Royal Standard used by
William III of England from 1689-1702 by placing the arms of
House of Nassau (above right) in the center of the 1603 Royal Standard (above left).
We need some examples to illustrate the principles of Mayan hieroglyphic writing. The best would be svg drawings of the different ways to write the word "b'alam" (there are five different ways to do so). They can be found in this
pdf on page 24 and 25. A separate svg file would be required for each of the five variants. --
·maunus ·
snunɐɯ·19:08, 22 May 2015 (UTC)reply
@
Maunus: Now there is
a draft to look at. I guessed that you wanted lines like I have done and not exactly as the diffused images. I have reused the different parts and scaled and edit them to some extent. If you want them in any other way just tell me, and I show them all together (you will get five separate files) so that you can compare sizes. Give me feedback on this, thanks. --
Goran tek-en (
talk)
14:25, 31 May 2015 (UTC)reply
The clarity is great, and the size is fine too. The one thing I would change if I were able would be to use lines of variable thickness like for example using a calligraphy pen or its equivalent digital tool. If this is too much work, the images can be used as they are.
·maunus ·
snunɐɯ·15:34, 31 May 2015 (UTC)reply
Yes I understand, to get it more handmade, lively. If you want me to use a calligraphy pen it would be good if you told me up front, now I will have to redraw it. The calligraphic pen doesn't always do what you want it to do, can bee tricky. I have made another test on the left figure
here for you to look at. It's more of just making the line more lively. I can redraw it with calligraphy pen but I will have to test first. Let me know what you think and want, thanks. --
Goran tek-en (
talk)
18:33, 31 May 2015 (UTC)reply
Yeah, sorry I didn't think of that untill I saw them with the straight line. The calligraphic pen would make it more liveli and also closer to the way the hieroglyphs look on Maya pottery where they were painted with brushes. I think these ones are better, except the first one which has a very shaky line for some reason. I think it would likely be more beautiful with a calligraphic pen, but the way they are now they serve the purpose of illustrating the principles of Maya writing, so I leave it up to you if you want to take the extra time to try out the calligraphic style.
·maunus ·
snunɐɯ·18:44, 31 May 2015 (UTC)reply
Yes, I can see the line gets a little shaky. I do like the more variable line though. But probably unless you can get it a little smoother we should stick with the first type - though maybe with a thicker line.
·maunus ·
snunɐɯ·19:43, 4 June 2015 (UTC)reply
Can you please vectorize from this signature of Richard Armstrong from this
external image. It takes to January 9, 1860. Keep the downward swoop of the "g" but don't include "Rev." Thanks.--
KAVEBEAR (
talk)
05:38, 8 June 2015 (UTC)reply
This request is somewhat of a challenge; one that I can't take on myself because I'm more familiar with mapmaking than flag making. Here's the request:
1) I have found a fair use copy of the seal of the city of
Utica, New York, which I am requesting to be vectorized. I made it transparant and cropped it a little from where I originally found it (which was a fluke find in itself). The color of the seal should be black.
2) After this, I am requesting that a flag be made of the city by using the fair use image linked above, which I do understand is somewhat distorted and cutoff, but this is the best image I could find and has the basic proportions and colors in place (needed for the seal). The then-vectorized needs to be incorporated into the flag, because other than the dual yellow stripes there isn't much to it. I think the color picker can handle the color matching.
This is probably a cryptic request so if anyone has questions and they are willing to help, just ping me and I will respond immediately. --
Buffaboytalk23:55, 14 April 2015 (UTC)reply
Graphist opinion(s)
Buffaboy, here is the seal, if I get a chance, I'll try to do the flag. The
Template:Non-free flag uses the wording "low resolution" image, so a SVG would not be considered that. If I'm misunderstanding the licensing issue feel free to correct me, I'm not all that familar with these licensing topics. I can try to make a non-SVG (i.e. PNG) version if you want? (unless there is a reason that allows for a SVG)
Offnfopt (
talk)
03:21, 16 April 2015 (UTC)reply
@
Offnfopt: Thank you for creating the seal! O.K., I realized that there is a license issue. Technically, the image of the news capture is not the flag itself, therefore in retrospect the non free flag identifier should not apply. In addition, I believe that if the license were strictly interpreted, the news station would have been in big trouble for filming the flag as it is a representation being captured on film. If I am 100% wrong, somebody please correct me!
Buffaboytalk05:00, 16 April 2015 (UTC)reply
@
Buffaboy: I have a PNG version of the flag done, here is the link for it.
But I'm still not sure as far as the licensing goes. I don't have much experience with dealing with the non-free images. I don't know if since the flag/seal is my work that I can pick the license, or since the image portrays that of a seal/flag etc if that takes precedence so has to be set as non-free. I went ahead and set it to non-free to be safe, but if for w/e reason turns out I can put those under different licensing terms then I grant to release them to public domain, if that is even possible legally speaking.
Offnfopt (
talk)
05:49, 16 April 2015 (UTC)reply
@
Offnfopt: Wait, I have a better idea. I can request deletion of the news screenshot, but for sourcing purposes, I can link to the webpage where the image was found to base the flag off of. This presumably should clear up all of the copyright issues associated with the files. I (again) could be wrong.
Buffaboytalk16:11, 16 April 2015 (UTC)reply
@
Offnfopt: I was told at
WP:VPP that you can now upload the SVG file of the flag, and then attach a public domain license to it since the flag originated before 1923 (which it did). This way, there won't be any more licensing issues, we can delete the PNG flag and its screenshot counterpart, and link to the original webpage where the news screenshot was found.
Buffaboytalk20:27, 18 April 2015 (UTC)reply
It was suggested
at the photography workshop that I bring this here... I'm looking for a way to scale this image up cleanly to around 100-200px. Vectorising it seems like the most sensible solution, but I'm not actually sure it can be done while maintaining the distinctive aspects of the seal. Opinions and suggestions would be welcome; I'm clueless at this stuff.
Yunshui雲水21:26, 29 May 2015 (UTC)reply
Graphist opinion(s)
It's really hard from this very small image in px to do anything and still keep the aspects and be true to the original. Do you want a drawing with straight lines like I guess the seal had from the beginning or with uneven edges as every print from the seal will be. It's two completely different things, but you will still have to find an image with higher resolution. If one uses this as an "original" it would be a lot of guessing. If you want me anything you will have to ping me, thanks. --
Goran tek-en (
talk)
18:14, 30 May 2015 (UTC)reply
I figured the size would be an issue from the outset - I haven't been able to find anything larger. At the moment one of the photography workshop graphists has suggested creating a raster image, so I'll see how that works out before asking you to do any additional work. Thanks nevertheless.
Yunshui雲水11:20, 2 June 2015 (UTC)reply
The .jpg version is exactly what I was after (actually, it's even better). Thank you very, very much, but please don't feel the need to do anything further on my account (unless you're bored and actually want to).
Yunshui雲水16:25, 2 June 2015 (UTC)reply