Description: "Do you have dozens of PDF files from your favorite scientific articles scattered on your harddrive? Do you also try to desperately organize them by renaming and archiving them in folders? But like the piles of printed articles on your desk, you can't keep up with all the new papers you download, and despite all your efforts it has become impossible to find that one article."
This is how the authors of the Mac software Papers introduce their software. I believe that it is important for the scientific Linux community to have a similar application. Given that I'm not able to do such a software, I propose the idea and hope that some experimented developpers would work on that!
I don't have much time (being a grad student and a new dad) and I'm merely a competent coder, but this is such a fantastic idea that I'm going to (slowly!) start working on it. I'll start it as a PyQt project and extend with C++ in the event that I need to do heavy lifting.
I'll report back in these comments when I have everything hosted (presumably Google Code, but probably a private svn repo for starters).
E-mail me if you want to contribute and/or have ideas/requests.
I would like to start working on this project. I was thinking of doing it, and then I saw this proposal. If you have already started working on, may be I could start contributing to your code. Otherwise, I would like to start separately. I would also like to use PyQT
Hi!
I have started a project like this 2 years ago and moved it quite a bit forward to the point when it was usable and contained quite a lot of features. It depends on Qt4 only. You can find it at:
http://sourceforge.net/projects/armaria/
Unfortunately, I could not continue with the development due to lack of time. Still, I would be very happy if someone could join or take over the project! Of course I would also be very happy to help.
Let me know if you're interested!
Cheers,
Andrzej
What about implementing it as a KIOSlave?
Just enter PDF:// and you have every PDF in the index stored in a three.
And when you click "open file" un KPDF, it defaults to "PDF://", hehe.
On the other hand, I think that Nepomuk + Strigi are about to fix this in KDE4, if I don't misunderstant.
Other related software to get ideas:
1. Jabref - Java application like kbib but also has reference grouping. It uses the bibtex file format and stores the additional info in a comment at the end.
2. Zotero - Firefox plugin. Haven't used it yet, but it can extract the reference information while browsing a page (journal databases, newspapers even amazon). Also supports tags, notes and other useful stuff. It can export to bibtex.
The proposal made here sounds like a great idea to me. Actually, I have been looking for a program like that for a long time, and found nothing I am happy with.
Two more comments:
1) From my experience in different theoretical physics departments, I can tell that most of them are running their computers with Linux/KDE, and are lacking kpapers or something similar. In this area, a good interplay with online preprint archives like http://www.arxiv.org, literature databases like http://www.slac.stanford.edu/spires/ and citation search engine like http://www.citebase.org/ would be great and also important to make maintaining ones local library a manageable task.
2) A somewhat related JAVA program has been created by Christian Sämann, and can be found at http://www.itp.uni-hannover.de/~saemann/celsius.html. For a screenshot, see http://www.itp.uni-hannover.de/~saemann/images/screenshot.gif. Maybe this can serve as inspiration for kpapers.
I really hope kpapers will be produced somehow!
Having something like this would be very cool.
I don't see the reason for this being an application though. Wouldn't it be better to have it integrated with okular or maybe even konqueror/dolphin?
Don't forget about books, it will be rather cool to see the Contents(stored in PDF), access(and index by themes/keywords) particular chapter and so on.
Cool app. =)
Take a look at the very lovely user-mode iso file handler acetoneISO here. When one starts this thing up from a terminal, there is a lot of activity going on. What's happenin' ?? One's home is being searched for iso images for this program to access. This without one noticing it without the terminal output.
Kpapers/Qpapers would search (and index!!) the pdfs in the same way, nicely threaded, and build and maintain the "library" without much ado.
JxJigsaws looks for pictures this way, as well.
Personally I use KBib to manage both the references and files. It's not perfect but it works. You can add references to actual files in KBib for each entry.
A very interesting and useful piece of software indeed.
Regarding the possible start comment, it seems to me that UI-wise the Papers people got it right the first time, e.g. tabbed-browsing and pdf previews among other things. Personally I wouldn't go with the thumbnail approach of Referencer. Thumbnails are ok for photos but convey no information when you're looking for a specific paper. A list view where you can filter according to author, title, year, journal, etc. would be better.
I hope somebody starts a project in QT/KDE soon. Unfortunately I'm not very familiar with QT or KDElibs (and don't have the time to learn) but I'd like to help out where I can.
No, I would not use thumbnails here. My librarian is a graphics librarian where thumbnails are the way to go.
The functionality of a flat(ter) library structure than having to manually go through oodles of subfolders is what I illustrate here.
An expandable listreeview (ubiquitous) would be in order with some sort of preview pane.
Check out the Java frontend for xjigs jigsaw puzzles under other games here. (This is Java, no QT or KDE so probably did not belong here but that's where I put it a while back.)
Included is an image librarian. This will scan folders and produce a flattened menu of images (can even preview PDFs but not its strong point). Could easily be changed to PDF, RTF, etc., with more workable viewers than ImageMagick.
Looks like a tremendous package. Document management is one of the biggest areas today.
To start, specify and outline functionality desired. Maybe C++ and its ilk are not appropriate here.
Maybe there needs be a section on kde-apps for proposals, ideas and such without cluttering program listings. The foul language serves no one.
Crap like this has to labeled like "Idea" "request" or "Proposal" in the title.
Posts like these turn this site in a pile of random indistinguishable trash.
I'm gonna try jagref right now. Paper looks like very sophisticated and intuitive.
I believe that Linux should focus on the scientific community. There's plenty of app that could help genetics and molecular biology scientist and that would attract them (me included) to use linux platforms. Paper is the kind of software that definitely would do so.
As far as I know, Tellico can handle either PDFs as a File-Listing, or Bibtex-Files, not both of them.
As File-Listing, you get a list with meta-data and preview of the papers, as Bibtex the handling is comparable to JabRef and KBib.
Tellico.
It does almost all of the things described on the Papers website although this might not be visible on the first glance: It is highly configurable, you can add your own collection fields.
For example, you could add a URL field where you can drop your PDF files (or select them through a dialog) and get a clickable link to the file.
Tellico exports Bibtex and provides powerful sorting and filtering possibilities.
See
http://kde-apps.org/content/show.php/Tellico?content=10030
It doesn't have an integrated PDF viewer, though, but clicking on a URL leading to a PDF file (local or remote) opens it up in Okular with all its niceness :-).
Maybe one could contribute some Okularpart embedding to Tellico, though.
This is EXACTLY what I am looking for month now.
JabRef, KBib and Co. is good for Bibtex-Coordination, but not for Papers.
Gnome has a similar Program called Referencer. Unfortunately it does not work well under KDE.
Ratings & Comments
25 Comments
I don't have much time (being a grad student and a new dad) and I'm merely a competent coder, but this is such a fantastic idea that I'm going to (slowly!) start working on it. I'll start it as a PyQt project and extend with C++ in the event that I need to do heavy lifting. I'll report back in these comments when I have everything hosted (presumably Google Code, but probably a private svn repo for starters). E-mail me if you want to contribute and/or have ideas/requests.
I would like to start working on this project. I was thinking of doing it, and then I saw this proposal. If you have already started working on, may be I could start contributing to your code. Otherwise, I would like to start separately. I would also like to use PyQT
Hi! I have started a project like this 2 years ago and moved it quite a bit forward to the point when it was usable and contained quite a lot of features. It depends on Qt4 only. You can find it at: http://sourceforge.net/projects/armaria/ Unfortunately, I could not continue with the development due to lack of time. Still, I would be very happy if someone could join or take over the project! Of course I would also be very happy to help. Let me know if you're interested! Cheers, Andrzej
What about implementing it as a KIOSlave? Just enter PDF:// and you have every PDF in the index stored in a three. And when you click "open file" un KPDF, it defaults to "PDF://", hehe. On the other hand, I think that Nepomuk + Strigi are about to fix this in KDE4, if I don't misunderstant.
Other related software to get ideas: 1. Jabref - Java application like kbib but also has reference grouping. It uses the bibtex file format and stores the additional info in a comment at the end. 2. Zotero - Firefox plugin. Haven't used it yet, but it can extract the reference information while browsing a page (journal databases, newspapers even amazon). Also supports tags, notes and other useful stuff. It can export to bibtex.
I would really love to have some application like this! If there is a project to do it, I might try to collaborate!
The proposal made here sounds like a great idea to me. Actually, I have been looking for a program like that for a long time, and found nothing I am happy with. Two more comments: 1) From my experience in different theoretical physics departments, I can tell that most of them are running their computers with Linux/KDE, and are lacking kpapers or something similar. In this area, a good interplay with online preprint archives like http://www.arxiv.org, literature databases like http://www.slac.stanford.edu/spires/ and citation search engine like http://www.citebase.org/ would be great and also important to make maintaining ones local library a manageable task. 2) A somewhat related JAVA program has been created by Christian Sämann, and can be found at http://www.itp.uni-hannover.de/~saemann/celsius.html. For a screenshot, see http://www.itp.uni-hannover.de/~saemann/images/screenshot.gif. Maybe this can serve as inspiration for kpapers. I really hope kpapers will be produced somehow!
Having something like this would be very cool. I don't see the reason for this being an application though. Wouldn't it be better to have it integrated with okular or maybe even konqueror/dolphin?
Don't forget about books, it will be rather cool to see the Contents(stored in PDF), access(and index by themes/keywords) particular chapter and so on. Cool app. =)
Take a look at the very lovely user-mode iso file handler acetoneISO here. When one starts this thing up from a terminal, there is a lot of activity going on. What's happenin' ?? One's home is being searched for iso images for this program to access. This without one noticing it without the terminal output. Kpapers/Qpapers would search (and index!!) the pdfs in the same way, nicely threaded, and build and maintain the "library" without much ado. JxJigsaws looks for pictures this way, as well.
Take a look in KBib, then suggest the features that you want to KBib developers. The chances that you achive your goal will be enhanced.
Personally I use KBib to manage both the references and files. It's not perfect but it works. You can add references to actual files in KBib for each entry.
A very interesting and useful piece of software indeed. Regarding the possible start comment, it seems to me that UI-wise the Papers people got it right the first time, e.g. tabbed-browsing and pdf previews among other things. Personally I wouldn't go with the thumbnail approach of Referencer. Thumbnails are ok for photos but convey no information when you're looking for a specific paper. A list view where you can filter according to author, title, year, journal, etc. would be better. I hope somebody starts a project in QT/KDE soon. Unfortunately I'm not very familiar with QT or KDElibs (and don't have the time to learn) but I'd like to help out where I can.
No, I would not use thumbnails here. My librarian is a graphics librarian where thumbnails are the way to go. The functionality of a flat(ter) library structure than having to manually go through oodles of subfolders is what I illustrate here. An expandable listreeview (ubiquitous) would be in order with some sort of preview pane.
Check out the Java frontend for xjigs jigsaw puzzles under other games here. (This is Java, no QT or KDE so probably did not belong here but that's where I put it a while back.) Included is an image librarian. This will scan folders and produce a flattened menu of images (can even preview PDFs but not its strong point). Could easily be changed to PDF, RTF, etc., with more workable viewers than ImageMagick.
Looks like a tremendous package. Document management is one of the biggest areas today. To start, specify and outline functionality desired. Maybe C++ and its ilk are not appropriate here. Maybe there needs be a section on kde-apps for proposals, ideas and such without cluttering program listings. The foul language serves no one.
Crap like this has to labeled like "Idea" "request" or "Proposal" in the title. Posts like these turn this site in a pile of random indistinguishable trash.
I think that the most important thing in Kpapers should be an "Import from PDF" function. You need it if yoou already has a collection.
I'm gonna try jagref right now. Paper looks like very sophisticated and intuitive. I believe that Linux should focus on the scientific community. There's plenty of app that could help genetics and molecular biology scientist and that would attract them (me included) to use linux platforms. Paper is the kind of software that definitely would do so.
Yes! That would be great, but not only PDF. I think we need support for PS, DjVu and have possibility to add support of whatever other formats
Having not <i>really</i> looked into this, i'm not certain of course, but could not Tellico be used for this sort of thing?
As far as I know, Tellico can handle either PDFs as a File-Listing, or Bibtex-Files, not both of them. As File-Listing, you get a list with meta-data and preview of the papers, as Bibtex the handling is comparable to JabRef and KBib.
Tellico. It does almost all of the things described on the Papers website although this might not be visible on the first glance: It is highly configurable, you can add your own collection fields. For example, you could add a URL field where you can drop your PDF files (or select them through a dialog) and get a clickable link to the file. Tellico exports Bibtex and provides powerful sorting and filtering possibilities. See http://kde-apps.org/content/show.php/Tellico?content=10030 It doesn't have an integrated PDF viewer, though, but clicking on a URL leading to a PDF file (local or remote) opens it up in Okular with all its niceness :-). Maybe one could contribute some Okularpart embedding to Tellico, though.
This is EXACTLY what I am looking for month now. JabRef, KBib and Co. is good for Bibtex-Coordination, but not for Papers. Gnome has a similar Program called Referencer. Unfortunately it does not work well under KDE.
Jabref may be what you are searching for. jabref.sourceforge.net