Extracting data from copyrighted images in a paper ?

[rogerharris]

I am using quite a few images from other neuroscience papers or medical books in my paper. I went through the recommended procedure of writing to the authors and copyright owners about six weeks ago.

All the authors wrote back right away and said ok to use. The copyright owners: journals and medical publishers have for the most part not replied. Except some automated replies. One journal publisher did reply and asked for a copyright fee. Another wanted a fee to fast track the copyright application as well as copyright fee.

In light of this blank wall of approval I am thinking i could use some of my graphic skills to "extract data" from images ? That is i do a line drawing of the relevant parts from the image using a graphics tablet. I'll then put a reference under the image..(Line drawing of ----- )

I was told here that i could "extract data" from images and not break copyright. SO i guess what I'm asking is making a hand written line reproduction extracting data, or am i still breaking copyright ?

 

[Mapes]

Talk to your university's intellectual property office, or take the easiest approach to get a complete draft (e.g., include the original images with credit and references, but without permission) and let the accepting journal handle the details. My recommendation is to save your time for your research, not for working through the minutiae of copyright law.

I realize I'm not exactly answering your question, but every journal, university, and country is different. Talk to the people whose job it is to help you personally.

 

[Manchot]

FYI, there's a free program called Engauge Digitzer that allows you to easily extract data from an image. It's basically point-and-click: you define the axes of whatever it is that's plotted, and then select whatever data points or curves you want to include. It will then export that data in a variety of formats. Pretty awesome, if you ask me.

 

[rogerharris]

Ok thanks for replies peeps. graphic analyser looks interesting.

Its a lot harder to locate the right person in a university than it is to go online to physics forums. Even if you locate them, then there is often a delay period. Perhaps the universities could learn a thing or two from what goes on here. Set up university forums etc. Anyway that's another subject.

guess i'll just have to take this up with the journal editor.

 

[Andy Resnick]

You are dealing with a very tricky issue, and I'm not surprised the Journals did not respond. I would proceed *very* carefully- you could easily end up in an bad position.

I would first speak with your faculty advisor/department chair. After that, you may want to approach someone in library services- they are surprisingly familiar with copyright and free use rules.

Copyright and free-use issues are highly volatile right now due to the development of open-access. Again, I would recommend against simply pirating data from a paper- which is how it will be presented at your disciplinary hearing.

 

[Vanadium 50]

Again, I would recommend against simply pirating data from a paper- which is how it will be presented at your disciplinary hearing.

I agree.

Normally, if one wants data that were published in the form of a graph, but in (e.g.) the form of a table, one writes the original authors asking them for it, cites them properly, and acknowledges them for it in the paper.

 

[rogerharris]

You are dealing with a very tricky issue, and I'm not surprised the Journals did not respond. I would proceed *very* carefully- you could easily end up in an bad position.
.

why are you surprised they did not respond ? All the journals are concerned about is time and money.

they don't want to spend time answering permission requests because there isn't much money. All the replies from journals are from managers regarding how much to charge. How big is the image, how many times will it be reprinted in another journal, do you want it fast tracked.. all leading to this will cost you this much.

by contrast the authors who spearheaded the research I'm quoting, instigated it, sweated over it...all the authors replied promptly and gave permission when they knew it was being cited in a respectable journal. Some had questions to ask, but mainly about how the data would be used.

so to sum it up

Authors agreeable to the scientific cause but concerned about representation and use.
Journal managers do not care about represenation / get in the way of research and concerned about how much they can charge.

The journal management are just blocking science. They arent concerned at all about how the work will be represented. There isn't any difference between citing data from a paper and using an image from it ,except that by copyright law you can get money for an image. All the image preparation is paid for and done by the authors. The research to make the image paid elsewhere. I was in the scenario of being asked by one elsevier journal to pay them a fee so an image can be re-used in another of their journals. Yet the journal didnt do anything except take free ownership of the image.

I would recommend against simply pirating data from a paper- which is how it will be presented at your disciplinary hearing

perhaps it might be an idea to get a hearing on this. What the journals are doing is blocking the spirit of science.

 

[rogerharris]

Copyright and free-use issues are highly volatile right now due to the development of open-access. .

ITs volitale because the same blockage happens in journal access. all the services geared up towards taking money but not to help researchers do research. Have you tried accessing journals using the academic logins ? We have one called athens in the UK. Often it doesn't work. I mean nobody bothers to sort the login software, to a particular publisher unless enough people complain. All the credit card payment options work, but if you want to read an article under this basis you have to submit to the university why you want to read it. So you "might" get to read it 4 -12 weeks later.

Scientists have to submit to this system for free, sometimes having to give the journal a fee, then face problems using it. A lot of researchers just decide to pay to read articles out their already low wages or student grants.

 

[cristo]

We have one called athens in the UK. Often it doesn't work.

I've never had a problem with using Athens for off-campus access.

A lot of researchers just decide to pay to read articles out their already low wages or student grants.

Really? I don't think I've come across anyone who pays out of their own pocket for access to journal articles. If there's an article I want to read in a journal we don't have subscription for, I get a copy through the library, usually within a week, which is paid for by my department.

 

[Andy Resnick]

why are you surprised they did not respond ? All the journals are concerned about is time and money.

<snip>

Please read my post- I said I am *not* surprised the journals did not respond.

ITs volitale because the same blockage happens in journal access. all the services geared up towards taking money but not to help researchers do research.
<snip>

I hear what you are saying- and so authors are usually allowed to share their published papers (for example, PDF files) with whomever they like, or even post them on a webpage. That's different from saying *you* can take someone else's data and use it in *your* paper without giving the person/group who generated the data proper attribution.

You can do what you want- you asked if it was permissible to pirate ('extract') someone else's data, and you have been told it is not. You are, of course, free to do so anyway.

 

[rogerharris]

I've never had a problem with using Athens for off-campus access.
.

its getting sorted now. You can get into at least three of the main publishers. Previously it used to work with selected elsevier journals and that was about it. At the moment it seems to work in about half of the bioscience journals in about 2/3 of the main publishers. Its operational at about 24-30% at most. Maybe you have a different level of login. I though

Really? I don't think I've come across anyone who pays out of their own pocket for access to journal articles. If there's an article I want to read in a journal we don't have subscription for, I get a copy through the library, usually within a week, which is paid for by my department

Thats not always conveniant way to do things when you are snowed under with stuff to do and out of location. The idea of online access is not to go through all that.

 

[rogerharris]

Please read my post- I said I am *not* surprised the journals did not respond.

.

Ok maybe i took this wrong, i took it to mean the journals are right for some reason.

You can do what you want- you asked if it was permissible to pirate ('extract') someone else's data, and you have been told it is not. You are, of course, free to do so anyway

I doubt the authors would have any problems if their data were extracted as the idea is to reference them. I think it is going to need editor input to sort this.

 

[physics girl phd]

Since you say (in post #1) that all the original authors "wrote back right away and said ok to use" so why didn't you ask them flat out for the original data points as Vanadium suggested? This is the way I've gone about it in the past (when I wanted to add my data to a plot to show I had improved a materials process for better results of the desired materials characteristic). It's then also considered polite if you not only acknowledge them but also send them a copy of the way you've used their data... probably even for their review before submission to any journal.

Being an inconvenience when you are "snowed under with stuff to do" (as you say in post #11) isn't a valid excuse. Everyone's busy and yes, it does take time to get things out in the correct manner.

I'd also personally avoid bothering the editors of journals too much... it sounds to me like you've already bothered them substantially. As you've noted... most of them aren't responding anyway. If you really want to know about this, consult real legal counsel (for some consultation fee).

 

[rogerharris]

Since you say (in post #1) that all the original authors "wrote back right away and said ok to use" so why didn't you ask them flat out for the original data points as Vanadium suggested? ).

at that stage i didnt think the journal publishers would be non-co-operative with permissions.

I'd also personally avoid bothering the editors of journals too much... it sounds to me like you've already bothered them substantially. As you've noted... most of them aren't responding anyway.

i hadnt written to any editors regarding this. The journals are all centralized at their publishers. For example all permissions requests go to an email address such as permissions@elsevier.com or whatever the publishers permissions are. most don't bother to reply.

Before you submit an article the instructions state you to get publishers permission for images required such as fMRI. So this is what led to the problems. But i notice some of the instructions in elsevier are wrong. One section said to make tables without using grids. there isn't any way to do this, so i emailed them. A week of back and forth and it turns out there are errors in instructions. Well these things arent perfect. I guess it needs to be borne in mind Elsevier is now a pretty large organization.

 

[Moonbear]

If you use the exact figure from an already published source, you do need to pay the fees to use it. Some journals have well-established policies on this and it's fairly easy to obtain the rights to copy the work as long as you credit the original source and pay the fees; others might not agree.

If you can get the data from the original authors, you ARE allowed to create your own figures that are not identical copies of the originals, and cite it as "Modified after:..." or "Adapted from:..." and cite the original published source. Though, it's usually best if you don't need to copy someone else's figures, but just refer to the original paper in your text. The one case where it's sometimes useful to use the figures is in writing a textbook where you need the illustrations to help the students. But even then, you can often get away with a stylized representation of the data rather than a figure using the exact data.

To me, it seems perfectly fair to require a fee to use someone else's figures. But, it seemed rather obscene when colleagues of mine had to pay to use their OWN figures because they had already published them in a journal owned by a different company than the one they were doing a review article in.