FAQ
From Open Clip Art Library Wiki
How do I download the clipart?
The library has grown to a size where packaged releases may not occur in the future. At this time, all art since January 2007 and parts of the older collections must be downloaded as individual files from the website.
Older packages are available here:
For Linux users, the collection is available in RPM form:
- http://rpm.pbone.net/index.php3/stat/4/idpl/4456828/com/openclipart-png-0.18-3.noarch.rpm.html
- http://rpms.mandrivaclub.com/rpms/mandriva/official/2006.0/i586/media/contrib/clipart-openclipart-0.17-2mdk.noarch.html
A collection of recent contributions (up to Jan 2007) is on http://commons.uncyclomedia.org/wiki/Category:Openclipart
See also: http://www.openclipart.org/downloads/index.php for the last official release.
Where do I save it to? How do I import it into OpenOffice.org?
See http://openclipart.org/wiki/?title=OpenOffice.org_Integration for info on conversion to OpenOffice-compatible formats.
What is the project's official name and branding?
The project's official name is Open Clip Art Library. One tangible is that the website is www.openclipart.org.
- "Drawing Together."
What is clip art?
Wikipedia says it best:
"Clip art, in the graphic arts, is the use of images either copied or physically cut (hence the term) from pre-existing printed works, either books that have entered the public domain, or books specifically published for such use (which, if they contain images that are not in the public domain, include a license fee in the cover price)."
The American Heritage Dictionary defines it as:
"Ready-made pieces of printed or computerized graphic art, such as illustrations, borders, and backgrounds, that can be electronically copied and used to decorate a document."
Why is this project Open?
The usage of the name is to convey connection to Open Source software and culture. Also, the word conveys the concept of the project lowering barriers for participation for submission of artwork, development on the project, use of the site and use of the clip art stored.
How can I help?
Find out more on how you can help.
The best way to help is to create and submit new clipart.
But if you're interested in helping the project more broadly, there's several areas you can get involved with:
- Join the Open Clip Art Library Release Team, and help in preparing the clipart package releases and announcing the releases throughout the open source community.
- Update pages in the Open Clip Art Library website. Post news items, review/revise existing pages, add new tools, etc.
- Help existing clipart package creators with organizing their files, conducting releases, and other general maintenance work.
- Seek out new clipart that we can use under Public Domain terms.
- Mention Open Clip Art Library to applications you think could benefit from having access to a rich source of clipart.
- Strive to increase publicity for Open Clip Art Library through registration in search engines, linking to us from your website, addition of the Open Clip Art Library URL to appropriate websites, and other creative marketing ideas.
I just have a few clipart items to contribute; how can I contribute them quickly and easily?
You can upload one-off clipart images fairly easily. First, you need to dedicate your work to the Public Domain; go to the Creative Commons website and fill out their form for doing this. Next find the incoming clipart page on the website and upload your work, including the Public Domain statement.
I have a bunch of clipart items and intend to add more later; how can I do so with minimal hassle?
If you intend to submit a lot of clipart, instead of filling out the Public Domain statement and uploading each piece individually, you can establish your own 'clipart package'. This way, you can fill out the Public Domain declaration once for the package. Adding new material is as simple as copying it into the package archive and re-uploading it. This feature is currently not working. In the meanwhile, please upload you clipart as described above.
What is the preferred form of contribution?
The format we support for clipart is SVG as it is an open scalable vector format and allows the most flexible use of the clipart. All submissions enter the Public Domain. When saving please try and optimize your SVG for best compatibility. Keep the SVG as standard as possible and try to avoid features that are specific to your prefered choice of the software. At this time only individual files can be uploaded at http://openclipart.org/. We hope to support bulk uploads in the future.
Why is it called a Library?
A library is a place where media is stored, also called a collection. While this could be an individual's collection, as several references mention, the interpretation of the term as an institution, "Library", is most accurate, as it refers to a publicly accessible collection of media. The Open Clip Art Library is a publicly accessible collection of clip art.
To take this metaphor further still, we can map the role of librarians onto software developers. Librarians help organize, repair, check-in, and check-out media. Similarly, the Open Clip Art Library's core developers, or libraries, have these same roles in relationship to clip art.
Why is it important to contribute my artwork to this project?
Clip art is extremely important for users of vector art programs (and all graphical programs), because it can save a lot of time when making certain kinds of drawings and provide examples users can start with when making their own creations.
What inspired this project?
The Sodipodi flags project was the original inspiration for this; at Inkscape we found that users were submitting clipart and examples, so we thought it'd be worthwhile to have an independent clipart project that could provide useful content for all open source drawing tools. A side-effect of these goals is that the repository will be accessible and the graphics reusable by all (from the web).
Who is this clipart being provided for, and why?
The clipart that the Open Clip Art Library provides is intended first and foremost for users of Open Source applications such as Inkscape, Sodipodi, OpenOffice.org, AbiWord, but also for any and all other applications which allow SVG to be used. The reason for providing this clipart is to make these open source applications more useful to users; for some users, the availability of good clipart that can be quickly dropped into a document to dress it up can be as important as any other feature in the application.
I'm still using a non free word processor (e.g. Microsoft Word). Will I need to switch to a free word processor to be able use the cliparts?
Nope, but you will need to make sure your word processor supports SVG, in order to use them. We hope eventually all word processors will support SVG, but we'll have to see...
How did the Open Clip Art Library Start?
This project began as a spinoff from Inkscape by Bryce Harrington and Jon Phillips as a way to help consolidate SVG images contributed by Inkscape users with similar collections from other projects. It was directly inspired by the Sodipodi Flags Clipart project from 2003.
What are Open Clip Art Library's Development Goals?
- Serve as a collection point of Public Domain clipart for Open Source applications.
- Establish standards for how clipart is installed, and how applications can access it.
- Establish a universal metadata schema for use with clipart.
- Promote use of open standards for graphic formats such as SVG and PNG.
- Promote interoperability between different open source applications so that standards-compliant clipart can be easily shared from one application to the next.
- Help identify or create tools and libraries to convert clipart from one format to others.
Why does the Open Clip Art Library require artwork be covered as 'Public Domain'? What about GPL/LGPL/other?
Many of us are familiar with the Gnu General Public License (GPL, LGPL) and other open source licenses through our software development work, so it's logical to wonder if those licenses would be similarly helpful in promoting open clip art. Those licenses are valuable for software because usually when someone modifies your software, the result is more valuable than the original, so you want to ensure it is given back for others to modify and use.
However, with clip art the situation is different. Clip art is intended to be used in larger compositions, presentations, books, and so forth, and it really doesn't matter to us if those are open source or not; indeed, in many cases such things cannot be open source due to other issues (such as containing sensitive data). In other words, the result is not as valuable to us as the original.
There are some licenses that don't require this Copylefting behavior. Typically these require the user to do something, such as provide attribution. However, this reduces the usefulness of the clipart, because the user would have to carefully include the author names, etc. for each piece of clip art they may use. This quickly gets considered an "obnoxious advertising clause" and the artist will seek out clip art that is more Royalty Free elsewhere.
The main reason we favor Public Domain is that for the end user it is very simple. There are no conditions they must adhere to. They don't have to worry if they're allowed to use it for a commercial purpose, nor follow any particular rules if they modify it, nor have to worry about giving credit when they use it.
Of course, Public Domain has some downsides to it. Registration of Public Domain requires slighly more work than licensing, but Creative Commons provides a form that eases the pain a little bit. Also, copyright law varies from country to country, so making something Public Domain could be more complicated for some people. However, despite these potential issues, it is currently felt to be the best option for the project and its users.
Even though users are not required to provide attribution, Open Clip Art Library strives to keep track of the submitter information. There are several reasons for this:
- First, we want to ensure that there is a place where the author's name is displayed, so they get credit for their work and can be contacted in case of questions.
- Second, we feel it is more legally clear if there is traceability of the authorship back to the artist, so any questions of legality can go to them.
- And third, we want to ensure a way that fans of an artist's work can find other art they have contributed.
Why can't I contribute work under a Creative Commons license (such as BY-SA) ?
If you look into commercial clip-art packages, you will find that however the collections are licensed, they have one thing in common: the individual images are in the public domain. This is what it means for images to be clip-art. People have to be able to use the images in their presentations, signs, fliers, greeting cards, bulletins, and other documents without giving attribution. They can do this with images from traditional hard-copy clip-art books, and they can do it with commercial CD-ROM clip-art collections. They have to be able to do it with our clip-art as well. If you are not willing to allow people to use your images with no strings attached, then you must not contribute them to a clip-art collection, because that is what clip-art is all about.
What limitations are there in how I can use the clipart?
There are no limitations, you can do whatever you wish with it.
We select the Public Domain for the clipart in order to provide maximum flexibility and ease of use to the user. You can put the clipart into your own drawings without any affect on the copyright or license of your work. There are no requirements to include attribution of the clipartist. You can edit and modify the clipart as you wish and redistribute it under your own terms.
Can I modify some clipart and resubmit it to the Open Clip Art Library?
Yes, since the clipart is Public Domain you're welcome and encouraged to modify and resubmit it, however we do ask that you observe a few guidelines. First, if the changes are minor such as fixes or slight tweaks, please submit these back to the original author if possible, so we can avoid having several near-duplicates of a given image. Second, please try to notify the original artist of your changes to provide them the courtesy of knowing how their work will be reused within the project. Finally, in your README, please include a mention of the original artist's name and the identity of the original image.
What are some suggested viewers for displaying SVG?
Several browsers support SVG natively, including Mozilla 1.8+ based browsers, Firefox 1.5+, Camino 1.0+ and SeaMonkey 1.0+, which support most of the SVG 1.1 specification. Opera 8.x and 9.x support SVG Tiny 1.1 and SVG Basic 1.1 respectively. Safari seems to support SVG, either through the WebKit plugin, or apparently in Safari 3, as seen here, here, or here (Safari 3 is available in OS-X Leopard for OS-X 10.4.x through updates).
Other viewers include; Batik Squiggle is a popular open source SVG Viewer written in Java, meaning it works on all major platforms. It supports most SVG 1.1 features. Renesis Player supports most SVG 1.1 and some Draft SVG 1.2 features. It is closed source, but available on many platforms. Adobe SVG Viewer plugin is widely used on both Windows and Linux, though closed source. Adobe will drop support from January 1, 2009, although is currently still available for download. Corel have similarly removed their SVG browser from their site, though it may still be found elsewhere.
svg2png is a unix shell script for converting SVG files to png for viewing. librsvg is leveraged by a wide variety of other applications such as Nautilus for rendering previews of SVG files.
Note that there are some occasional discrepancies between how one app displays a certain SVG feature and another. Invariably this is usually due to one or the other apps not properly implementing the SVG spec; you can assist the SVG cause greatly by reporting these incompatibilies to the given apps whenever you find them.
See svg.org for more information.
What are some suggested editors for modifying SVG files?
A number of apps (both SVG and general drawing) are listed on the Inkscape wiki. In addition to Inkscape, other SVG-capable apps include Corel Draw, Sodipodi, Adobe Illustrator, Jasc Web Draw, Sketsa, Express Animator and Sketch (aka Skencil). Other SVG apps can be found by searching freshmeat.net for "SVG".
How can I configure my .mailcap on Linux or similar for easily viewing SVG files?
As superuser, edit mailcap with your favorite text editor and add these lines if you want to use Inkscape to open SVG files. If you would like to use sodipodi or another editor on Linux, then please replace inkscape with your preference:
image/x-svg; inkscape %s image/svg+xml; inkscape %s
To do this in redhat 9 - fedora core 2:
vim /etc/mailcap [insert] image/x-svg; inkscape %s image/svg+xml; inkscape %s [write changes / save changes]
To locate mailcap:
locate mailcap [or] whereis mailcap Setup ~/.mailcap for just your own user account.
What are the most important long term goals of the project?
We want to build a rich collection of reusable clipart that open source programs can leverage for their users. Currently, if one wishes to create a quick document using clip art, the easiest solution is to use a proprietary program since they tend to have clip art included along with them, or massive creative archives because of the longer history of support for the closed source standard design applications. We want to turn that around so that it ceases to be an argument for not using Open Source. Ultimately, we would like to achieve a situation where the Open Source world has orders of magnitude better clipart than the proprietary world (although our repository will be usable by both worlds).
How can I use these images in OpenOffice.org?
OpenOffice.org (up to version 2.3) does not natively support SVG files, however there are two broad approaches to getting the OCAL images into OpenOffice. The first approach is using vector formats, which are inherently higher quality and scale very well. The second, or fall back, approach is to use bitmap/raster formats, which do not scale well.
Using Vector Formats
While OpenOffice doesn't natively support SVG images, a third-party plugin available from http://haumacher.de/svg-import/ can add this support. This plugin internally converts SVG files into OpenDocument Graphics using the Apache Batik SVG Engine. It's in an early stage but produces reasonable results for simple SVG graphics. There are instructions for installing the plugin on that website. Using this plugin, you must select File -> Open to open the SVG file, rather than Insert -> Picture -> From File. as SVG files are effectively treated like OpenDocument Graphics (ODG) files, and opened in OpenOffice Draw. Thus, to use the graphic in Writer or Impress, you will need to copy and paste the graphic from OpenOffice Draw.
The same people who made the above plugin have also produced a batch processing utility called svg2office to convert SVG files to ODG files in bulk.
Inkscape will also convert SVG files into OpenDocument Graphics (ODG) documents, which is the native format for OpenOffice Draw and several recent office and graphics programs. Again this is at an early stage, but the results are reasonable. Inkscape can also export SVG files to various other vector formats that OpenOffice can import, inluding EMF and WMF (Windows only and inferior to EMF).
Note: All of these vector conversions tend to lose some detail, primarily colour and transparency gradients/blends, plus also SVG effects, such as the blur effect that became popular in Inkscape 0.45. These problems will probably be resolved in the near future. See the OpenOffice Wiki for some user experiences of conversion.
Using Bitmap/Raster Formats
Many of the pictures in the Open Clip Art Library now include a PNG version, which can be directly imported into OpenOffice. Once you have the file on your computer, simply go to Insert-->Picture-->From File, and open the file from your hard drive.
Additionally, images can be converted to bitmap/raster formats using Inkscape or the GIMP, among others. Inkscape will export to PNG, but the GIMP will export to many formats, including PNG, GIF, JPG, BMP. Both these programs are free and work on all major operating systems.
Note: Bitmap/Raster images will produce faithfully at the designated size, however as with any bitmap image they do not scale well, and produce jagged or pixelated images.
For More Information
For more discussion of OpenOffice.org and SVG importing, see Issue 2497 and Issue 49991 in the OpenOffice.org Issue Tracker.
What's the deal with all these different graphic formats? BMP, JPEG, GIF, PNG, SVG, WMF...?
These formats can be broken up into two major categories: raster, and vector. This project uses the vector format, but we will describe raster first.
Popular raster formats include GIF, JPEG, PNG, and TIFF. In these formats, each pixel in the image is represented by a specific bit of data in the file. Storing each pixel of a large image creates a huge file! Many formats try to get around this by compressing the data so that each bit of data can represent several pixels. For example, the GIF format uses a system called Lempel-Ziv-Welch, which detects patterns of pixels. If there is a section in the image with 100 blue pixels, followed by 100 red ones, and another part also has 100 blue followed by 100 red, it will reuse the previously stored segment. This is why GIF works well for images that have large areas of similar colors and high amounts of regularity, like line art. This type of encoding is called "lossless", and something similar is used by BMP, PNG, and TIFF formats. They can be saved and converted without ever losing any quality. (Actually, GIFs only use 256 colors, so they can lose some of the color data the first time you save them.)
JPEG, on the other hand, uses a different approach for achieving compression. The method it uses drops some of the data, so it is called "lossy". It does this using some known tricks to how the human eye works. This works well for photographic images, less so for artwork, and extremely poorly for line art. The problem with JPEG, as it relates to clip art, is that JPEGs show artifacts (strange blocks and shadows) when you zoom in on them. You also can't edit them often. When you save the JPEG again, the computer saves the original artifacts and creates more, so the quality goes down. The higher the compression ratio you use, the faster it'll get bad.
PNG (pronounced "ping") is usually the best raster format for saving clip art, but it too has a few tradeoffs. It is a lossless format so it doesn't have the problems that JPEG has, and it uses more colors and better compression than GIF. However, the file size still gets unreasonable for large images, and it is not supported by some applications. This is much less of an issue today than it was a few years ago, but since most Microsoft programs don't display PNGs correctly, many users are still affected.
This project uses the vector format. Popular vector formats include WMF and SVG. The difference between raster and vector formats is significant. If you draw a circle and save it as a raster image the file will describe a bunch of white pixels and a bunch of black pixel. The fact that it represents a circle is lost. If you zoom in on the picture, the computer can only make each dot bigger, and it looks worse.
In vector formats, what gets stored is data describing the shape. If you draw a circle, it'll store bits of info that mean "circle of this radius at this point". If you want to go back and re-edit it, you can use your application's knowledge of editing circles. Also, if you zoom in on the circle, the program redraws it completely instead of just making each dot bigger. Another advantage is that the files are usually quite small despite their clarity. All this makes vector formats great for clipart.
How do I batch-convert SVG files into PNG files?
You can use Inkscape to do this.
In Linux, go to the directory where all SVG files are lying in, and type this command in a console:
- for i in *.svg; do inkscape -f "$i" -e "$i.png" -w 200; done
In Windows, go to the main directory where all SVG files are lying underneath, and type this command in a command window:
- for /R %i in (*.svg) do start /wait C:\inkscape\inkscape.exe -f "%~fi" -e "%~dpni.png" -w 200
Where "C:\inkscape\inkscape.exe" represents the path to where you have installed Inkscape. When you run this command, it recurses through all sub directories from where it is run. If you only want the command to process SVG files in the current directory only then remove the initial "/R". To use this command in a batch file, replace every "%" with "%%".
Note: '-w 200' sets the final pixel width for the resulting PNG images, so change the number to what you want. A value of 80 is useful for generating thumbnails. If you need larger PNG images for OpenOffice.org or something similar, then increase the value to something like 600.
How do I batch-convert inkscape SVG files into plain SVG files?
Same as above but this time use
- inkscape inkskape.svg --export-plain-svg=plain.svg
For more hints: read this howto.
How can I convert my bitmap images to vector graphics?
This is possible using programs that "trace" your image, or for better results that take more time, by hand. Software methods do not work perfectly, or even well in many cases. Photographic images are hard to get good results from, but outline images generally work reasonably well.
Free Software
A free software package that can trace bitmap images into vectors is Inkscape, which uses the Potrace engine. Inkscape will open bitmap images, and then you can select the image and go to the Path menu and select 'Trace Bitmap'. There are many options you can adjust in the tracing process, though a decent tutorial is available on the Inkscape Wiki.
Recently, the Vector Magic website has also started offering a good online tool for tracing bitmaps. This service is free to use, and the resulting file can be emailed in either EPS or SVG format.
Other free, but less straight forward, options include Delineate, AutoTrace and Potrace. These are available for various operating systems, and there are some graphical front ends available for those that are command line based.
Commercial Software
Various commercial offerings exist for tracing, including Corel PowerTRACE (part of CorelDRAW), Adobe Live Trace (part of Illustrator), and many more lesser known packages.
The Vector Magic website offers comparisons to the main commercial offerings, and claims to be much better. Comparisons of Inkscape, such as this one, indicate that Inkscape compares reasonably well to both Vector Magic and commercial offerings.
For Best Results
Many would say that nothing beats hand-tracing a bitmap within a vector editor, but it takes a bit of artistic talent, and plenty of time.
Even if you do use a tracing program, you should use a high resolution source image. If your source image is lower resolution, it is often better to scale it in a bitmap editor such as Gimp. Bitmap editors can also be used to clean up obvious noise or other problems before attempting to trace an image.
Why not add a section for each package with a preview for all images, like www.sodipodi.com?
Thanks to Alberto Simões, we now have a Clip Art Browser!
The link is currently 404'd. I am not sure of the current link, if any.
Is there a MIME type for SVG data?
The SVG 1.1 specification gives the MIME type as image/svg+xml in accordance with RFC 3023. However, this MIME type is not yet registered, and RFC 3023 says that it should not be used until it is registered. In practice it is used anyway.
Yes, there are two commonly in use: image/x-svg and image/svg+xml. The latter is the more "standard" mimetype and should be preferred.
Is there any file(1) magic(5) for SVG images?
This was discussed on the mailing list, and the conclusion that seems to have been reached is that there was no such magic.
Magic is difficult to get right with XML files are identified but it is not impossible. If you can figure out exactly the right incantation please let us know.
Alan Horkan attempted to get the right magic for SVG but did not get a final correct answer.
File is a command line tool for Unix and other Operating Systems which uses known patterns referred to as magic to quickly identify file types.
Is there a CD label for this project?
Not at this time.
Why can't I find my submitted works?
Because they may have been uploaded to our old site, which had buggy software which caused a lot of problems. Please try re-uploading to the new cchost site. Files that were summit successfully to the old site can still be found by downloading from:
http://download.openclipart.org/downloads/ or http://openclipart.org/media/downloads
If you are an author of clipart submitted through the old system, please download form:
http://tinyurl.com/2bu2ue and after re-submitting to cchost, delete your name form:
http://openclipart.org/wiki/Authorlist
Note that files may be deleted form the Open Clipart Library for a number of reasons, most being Trademark or licence issues, or non-clipart files or images using the site as an image hosting service (ie all files submitted without following the guidelines). All of these will be removed.
Can I see excluded trademarked images or names list of those files somehow?
They are in the failed-files archive for release 0.11, in the trademark directory. The SVG filenames are:
divx_logo_none_01.svg harley_davidson_logo_non_01.svg mac.svg playboy_none_01.svg samba.svg windows.svg
Why don't you use UTF-8 character encoding?
The SVG files are supposed to be in UTF-8, and almost all of them are. However, the upload script does not correctly handle non-ASCII characters in the metadata, so we often have to manually fix the files (which will delay their entry into the collection). An effort is underway to address this issue during the spring of 2005.
Why can't I upload my works with Mozilla 1.7.5?
This browser should work at least as well as any other. If you had problems uploading, you need to provide more details.
Can I upload my Blender files here?
Maybe, but there is not much point in doing so because only SVG (and WMF) files are currently being included in the collection.
Sandbox?
On many Wikis, the sandbox is a page where users are free to experiment with codes without destroying or damaging any legitimate content.
