Drumbeat/Attribution generator

OpenAttribute: The world's simplest open attribution tool

Get Involved

• Mailing list (Google Group)
• Tweet #Drumbeat #OpenAttribute
• Group chat (IRC)
• Weekly Drumbeat community calls (open to all)

Functional Specification

OpenAttribute
Functional Specification
Authors: Laura F. Hilliger (laura@bigfunarts.com)
Creation Date:11/22/2010
Last Updated: 11/22/2010
Version: 0.1

Authors

Date	Author	Version	Revision
11/22/2010	Laura F. Hilliger	0.1	Initial Draft

Reviewers

Date	Reviewer	Comments

Approvers

Date	Approver	Comments

Table of Contents

1 Project Overview
2 Models

• 2.1 Add On
• 2.2 Plugin

3 Use Cases

• 3.1 Image to Personal Document
• 3.1.1 Copying Images
• 3.1.2 Downloading Images
• 3.2 Text to Personal Document
• 3.3 Video to Personal Document
• 3.3.1 Copying Video
• 3.3.2 Downloading Video
• 3.4 Manual Entry to Web (Add On)
• 3.5 Image to Web
• 3.6 Text to Web
• 3.7 Video to Web
• 3.8 Manual Entry to Web (Plugin)

4 Interfaces

• 4.1 Wikimedia
• 4.2 Flickr
• 4.3 Manual Entry

5 Output

• 5.1 Add on
• 5.2 Plugin

6 Existing Technologies

• 6.1 Citation Machine: http://citationmachine.net/
• 6.2 Xpert: http://www.nottingham.ac.uk/xpert/attribution/
• 6.3 CC Metadata Humanizer: http://code.creativecommons.org/~john/
• 6.4 OCW cite/attribute: http://ocw.korea.edu/ocw
• 6.5 Wikimedia Commons: http://commons.wikimedia.org

7 Technical Considerations

• 7.1 Security
• 7.2 Meta Repository

8 Recommended Approaches
9 Open Issues

[[|]]1. Project Overview
Open Attribute is a project to build attribution generators for Creative Commons licensed content. The project is to build simple tools for a range of platforms so that re-using and attributing open content is as easy as cut and paste.

The problem: Creative Commons licensed content is awesome, but attributing it properly can be difficult and confusing.

The first rule for re-using openly licensed content is that you have to properly attribute the creator. There are specific requirements for what needs to go into that attribution, and most people don’t know what they are or how to follow them. How-to guides and user education haven't helped much; too many people are still confused about what to put into an attribution and how to format it.

The solution: A simple tool everyone can use to do the right thing with the click of a button.

That's why we're building OpenAttribute, a suite of tools that makes it ridiculously simple for anyone to copy and paste the correct attribution for any CC licensed work. These tools will query the metadata around a CC-licensed object and export it in a properly formatted attribution that users can copy and paste wherever they need to. Citation generators already exist for lots of academic bibliographic formats and many different settings. By building on those tools to produce attributions, we’re creating a clean solution to a messy problem.

The tools include a simple add-on for web browsers, and plug-ins for popular publishing platforms like Flickr, Wordpress, MediaWiki and Drupal.

To make automated attribution generation really useful, we need to enable it in a variety of contexts. We need several tools in several places. With an add-on for browsers, users will be able to pull RDFa from any site where it’s available and produce formatted attributions; plugins and widgets for open content publishing platforms like Wordpress and Drupal, will allow creators to facilitate proper attribution of their work.

The result: OpenAttribute makes it easy for everyday people to use Creative Commons licensed content. And ensures content creators get credit for their work.

There is an enormously rich pool of open content out there, the easier we can make it to use and reuse that content, the better.

[[|]] 2. Models
Two models currently exist for this open attribution tool. Both models are limited by their own technical use cases instigating the creation of two separate code solutions. The models define no special functions for any additional user models.

[[|]]2.1 Add On
The Add On is defined as a browser extension, which, when installed, allows the user to pull available RDFa data and format it into the proper attribution. When no RDFa data is available, metadata should be generated from other sources NEEDS TO BE INVESTIGATED. When neither RDFa or alternative metadata exist, the add on should pull appropriate data (alt tags, HTML) using class or id names. The method of pulling non metadata must be specific for each interface (see Ch. 3), meaning that an interface detection method will be necessary. The Add On will also allow for manual entry of attribution data when RDFa or metadata does not exist. The manual entry option should also exist when pulling non metadata, as the HTML may not be properly formatted for attribution.

The Add On's use scenario will be specifically click a button in the browser. A sidebar appears with the formatted attribution in plain text and HTML. The Add On's use case will be pulling content from the web to a personal document (see Ch. 4).

[[|]]2.2 Plugin
The Plugin is defined as an external code snippet created for use with specific PHP based, open source platforms. These platforms will be: Wordpress, Drupal, Joomla, Zotero, dSpace, ccMixter. The plugin will be created and tested using Wordpress. Iterations of the plugin for the other listed platforms will be forthcoming after successful Wordpress trials. The plugin will also allow for manual entry of attribution data when RDFa or metadata does not exist. The manual entry option should also exist when pulling non metadata, as the HTML may not be properly formatted for attribution.

The Plugin's use scenario will be specifically click a button that appears on the content landing page or the download page. A pop-up (e.g., using ajax or javascript) offers plain text and HTML options for copying and pasting the attribution. The Plugin's use case will be pulling content from the web to the web (see Ch. 4).

[[|]]3. Use Cases
[[|]]3.1 Image to Personal Document
[[|]]3.1.1 Copying Images
The user finds an image that he/she would like to use in a personal document. The user copies the image to the clipboard. The attribution information is copied with the image. Upon pasting the image, attribution information is printed below the image with the same width as the image and using left alignment.
[[|]]3.1.2 Downloading Images
The user finds an image that he/she would like to use in a personal document. The user downloads the image. The attribution information is stored in the image file as metadata. (this works for sure with JPG, investigate GIF, PNG file formats)

3.2 Text to Personal Document
The user finds a piece of text that he/she would like to use in a personal document. The user copies the text and upon paste the text is formatted into quotation style with attribution information appearing in italics two points smaller and aligned right of the quotation.

3.3 Video to Personal Document
3.3.1 Copying Video
The user finds a video that he/she would like to use in a personal document. The user tries to copy the video to the clipboard. The attribution information is copied with the video location information. Upon pasting, attribution and a link will be printed to the document.
3.3.2 Downloading Video
The user finds a video that he/she would like to use in a personal document. The user downloads the video. The attribution information is stored in the video file as metadata.

3.4 Manual Entry to Web (Add On)
The user clicks the button to begin the attribution process, when no RDFa, metadata or suitable HTML is found, the Add On opens a form. The form includes entry for the title of the work being attributed, attribution name (e.g., author, company, username), source URL for the attributed work, CC license name, CC license URL, existing copyright notices. Optional attribution elements are the URL for the author site and contact information for additional permissions.

3.5 Image to Web
The user finds an image that he/she would like to use on the web. The user copies the image to the clipboard. The attribution information is copied with the image. Upon pasting the image, attribution information is printed below the image with the same width as the image and using left alignment. Additionally, the image itself is linked to the original source.

3.6 Text to Web
The user finds a piece of text that he/she would like to use on the web. The user copies the text and upon paste the text is formatted into quotation style with attribution information appearing in italics two points smaller and aligned right of the quotation.

3.7 Video to Web
The user finds a video that he/she would like to use on the web. The user tries to copy the video to the clipboard. The attribution information is copied with the video location information. Upon pasting, attribution and a link will be printed to the webpage.

3.8 Manual Entry to Web (Plugin)
The user clicks the button to begin the attribution process, when no RDFa, metadata or suitable HTML is found, the plugin opens a form in the popup window. The form includes entry for the title of the work being attributed, attribution name (e.g., author, company, username), source URL for the attributed work, CC license name, CC license URL, existing copyright notices. Optional attribution elements are the URL for the author site and contact information for additional permissions.

[[|]]4. Interfaces
RESEARCH NEEDED: This section needs to contain the information for how these sites store their metadata
Wikimedia
The Wikimedia Commons offers an API for pulling an XML of the metadata included in a document. This API must be implemented and queried to pull the attribution information. http://toolserver.org/~magnus/commonsapi_test.php
Flickr
Manual Entry

[[|]]5. Output
Add on
The attribution will be created based on the type of content is being pulled. If the content is an image, attribution information is printed below the image with the same width as the image and using left alignment. In this instance, the attribution information should also be embedded in the image metadata. Text should be formatted as a quote with 3px spacing on all sides with attribution information on the new line (e.g.<br/>), 2px smaller and in italics. Text output can use a standard web font at 12 pixels, making the attribution information 10px. Videos will be attributed with the attribution information followed by a <br/> tag and subsequently, the link information. When a video is downloaded, attribution information will be embedded in the file.
Plugin
The attribution will be created based on the type of content is being pulled. If the content is an image, attribution information is printed below the image with the same width as the image and using left alignment. In this instance, the image will also link back to the original source. Text should be formatted as a quote with 3px spacing on all sides with attribution information on the new line (e.g.<br/>), 2px smaller and in italics. Text output can use a standard web font at 12 pixels, making the attribution information 10px. Videos will be linked or embedded with the attribution information on the next line.

[[|]]6. Existing Technologies
Citation Machine: http://citationmachine.net/
Xpert: http://www.nottingham.ac.uk/xpert/attribution/
CC Metadata Humanizer: http://code.creativecommons.org/~john/
OCW cite/attribute: http://ocw.korea.edu/ocw
Wikimedia Commons: http://commons.wikimedia.org

[[|]]7. Technical Considerations
This functional spec assumes that RDFa and metadata will be accessible. The persons responsible for the implementation of this functional spec will need to determine which interfaces allow the usage of stored RDFa or metadata and how the API must be implemented to pull the data. For purposes of the functional specification, the authors have assumed all interfaces have open metadata. 7.1 Security
This section needs to detail whether or not the attribution can be deleted. If the add on or plugin pulls the metadata and prints it with text or HTML, the user has the ability to delete the data. Does it matter?

7.2 Meta Repository
Where and how will we store manual entry metadata?

[[|]]8. Recommended Approaches
[[|]]9. Open Issues

1 pager

OpenAttribute: building the world's simplest open attribution tool Project Description

The project in five sentences:

•Creative Commons content is awesome, but attributing it properly can be difficult and confusing.

•How-to guides and user education haven't helped much. We need a simple tool everyone can use to do the right thing with the click of a button.

•That's why we're building OpenAttribute, a suite of tools that makes it ridiculously simple for anyone to copy and paste the correct attribution for any CC licensed work.

•The tools include a simple add-on for web browsers, and plug-ins for popular publishing platforms like Flickr, Wordpress, MediaWiki and Drupal.

•The result: OpenAttribute makes it easy for everyday people to use open and Creative Commons content. And ensures content creators get credit for their work.

The project in one page:

Open Attribute: the world's simplest open attribution tools Open Attribute is a project to build attribution generators for Creative Commons licensed content. We’re building simple tools for a range of platforms so that re-using and attributing open content is as easy as cut and paste.

The problem: Creative Commons licensed content is awesome, but attributing it properly can be difficult and confusing. The first rule for re-using openly licensed content is that you have to properly attribute the creator. There are specific requirements for what needs to go into that attribution, and most people don’t know what they are or how to follow them. How-to guides and user education haven't helped much; too many people are still confused about what to put into an attribution and how to format it.

The solution: A simple tool everyone can use to do the right thing with the click of a button. That's why we're building OpenAttribute, a suite of tools that makes it ridiculously simple for anyone to copy and paste the correct attribution for any CC licensed work. These tools will query the metadata around a CC-licensed object and export it in a properly formatted attribution that users can copy and paste wherever they need to. Citation generators already exist for lots of academic bibliographic formats and many different settings. By building on those tools to produce attributions, we’re creating a clean solution to a messy problem.

The tools include a simple add-on for web browsers, and plug-ins for popular publishing platforms like Flickr, Wordpress, MediaWiki and Drupal. To make automated attribution generation really useful, we need to enable it in a variety of contexts. We need several tools in several places. With an add-on for browsers, users will be able to pull RDFa from any site where it’s available and produce formatted attributions; plugins and widgets for open content publishing platforms like Wordpress and Drupal, will allow creators to facilitate proper attribution of their work.

The result: OpenAttribute makes it easy for everyday people to use Creative Commons licensed content. And ensures content creators get credit for their work. There is an enormously rich pool of open content out there, the easier we can make it to use and reuse that content, the better.

You can get involved

•Collaborate on a functional spec for the tools

•Help with browser add-on, Wordpress and Drupal plugin development

•Become an alpha tester

POTENTIAL PROJECT NAMES:

•Attaboy +3

•(I don't love using a gendered name - Molly) +1

•CAT (Commons Attribution Tool)

•ChalkUp +2

•(Given Molly's note about gender, this is my new favorite as a temporary working title. Cuz it's so short. e.g., "#drumbeat #chalkup")

•Copy Credit

•Creative Credit

•CreditDue

•OpenAttribute +3

GET INVOLVED. How can people help the project right now?

•Join the project mailing list: http://groups.google.com/group/attrib-generator

•Developers who can help advise & build. (Before they start building, they probably need to help spec it and advise on technical strategy.)

•The first step here is to develop a functional spec. A lot of developers will only get onboard if everything is spelled out and formatted specifically for their techie brains. This document is on the way and could be utilized for the FS.

•Basic FS points (these will, of course, be modified for this specific tool)

•Project Overview with User Model

•Concepts

•Public Features (if separate from Admin)

•Admin Features (if separate from Public)

   • Technical Considerations (Security & Metatagging)

   • Recommended Softwares or Implementations/Examples and Review of preexisting tools

   • Open Issues

   • Turn this etherpad into a crisp one-page overview of the project.

   • Provide  feedback on potential names, or suggest others. Don't have to be perfect or final -- just need a working title we can use for now.

   • We  need to ensure that what we're proposing to build doesn't exist  already. What's the competitive / co-opetive landscape or ecology?

WHAT ALREADY EXISTS?

   • What other solutions already exist that we might collaborate with or learn from?

   •  What can we learn from Citation Machine/Son of Citation Machine? (http://citationmachine.net/)

   •  How much of this functionality can be built on top of existing free citation tools like the Zotero plugin?

   • We    have a specific kind of data that is being cited (media with CC RDF    information) and we have a specific kind of citation we want  (something   capable of being inserted into other web tools) but the  general  problem  of citation creation, formating, and sharing has some  very  capable tools  already.

   • The attribution tool from Xpert: http://bit.ly/9z7GBX by @patlockley. Can we collaborate with them? Can we build off of what they've already built?

   • @msbonn   Bryan Smith et  al over in MPub are noodling on citation generators  for  our content  (open and not). Might be worth touching base.

   • CC Metadata Humanizer: (http://code.creativecommons.org/~john/?url=http://joi.ito.com/) (courtesy of John Doig! - I think he's on the mailing list now..)

   • example, in terms of simple user interaction (this tool is NOT ideal as according to piet it is nearly impossible to format correctly): OCW cite/attribute, ie. http://ocw.korea.edu/ocw

   • Flickr already has a "Share this" drop-down that points to "Grab the html/bbc code". Now to add RDFa, license info to that... +1

   • on Wikimedia Commons:

   • Choose a random file, for instance http://commons.wikimedia.org/wiki/File:Zebras_Ngorongoro_Crater.jpg

   • Click "Use this file on the Web"

   • Then you can copy/paste the attribution (text or HTML format, but without RDFa)

   • (Please list others here)

FAQ

FREQUENTLY ASKED QUESTIONS

   • What does "attribute" and "attribution" mean? Why should I care about it?

   • Simple, one or two-sentence anwswer in a nutshell, anyone? :)

   • To define "attribute" let's just break the word into two words, as in "a" "tribute". Basically, attribution is giving "a" "tribute" to the person that created the awesome content you're using. Why should you care? Because content creators using creative commons licenses care about giving you access to their creations...

   • http://wiki.creativecommons.org/Frequently_Asked_Questions#How_do_I_properly_attribute_a_Creative_Commons_licensed_work.3F

   • What's the specific problem this project will help solve? How does it make the web more open and awesome?

   • Lots of people are using Creative Commons and other open licenses to share their work. We need a tool that can query the metadata around a CC-licensed object and  export it in a properly formatted attribution that users can copy and paste wherever they need to. 

   • What's the difference between "attribution" and "citation?" 

   • Answer: This project is aimed at attribution, not citation. Attribution and citation are different things, even though the tools to generate them will likely end up being technically similar.   

   • Citation = a scholarly practice for tracking the idealogical underpinnings of a work. Usually referencing sources like published books, articles, government documents, primary sources, etc

   • Attribution = crediting a copyright holder according to the terms of a copyright license. Usually (though not always) crediting artistic works like music, fiction, video, and photography. 

   • What's the itch you're trying to scratch?

   • Citation generators already exist for lots of different bibliographic formats  and many different settings. This project should be able to build on that work. 

   • However, as far as we have been able to find, there is still no automated tool that pulls the necessary metadata about an openly licensed work and formats it into a cut and pasteable attribution. 

   •  What's the right technological approach for this project?

   • We need some developers and smart technical minds with some early thoughts on the tech landscape and strategy here. 

   • Please dive in here or on the mailing list.

   • Maybe a simpler technical solution would be to focus on the bookmarklet (aka, "browser add-on").  

   • Most big sites/libraries are loathe to implement third-party features  that aren't in some way "standard." 

   • It's possible that a better use of our energy is to build the bookmarklet and lobby the sites to implement standard RDFa. 

   • When we talked about how to tackle this at Drumbeat, the group agreed that beginning with the bookmarklet / add-on probably made a lot of sense, even  though it won't reach all the people we want to reach. 

   • What specifically will the project build? 

   • Answer: To make automated attribution generation really useful, we need to enable it in a variety of contexts. We need several tools in several places: 

   • An add-on for browsers. A simple tool users can install in their browsers that will pull RDFa from sites where it’s available and produce formatted attributions

   • Plugins and widgets. Built for open content publishing platforms like Wordpress and Drupal, as well as social media sites, that will place a button on web pages where licensed content appears.  Users could click on the button to get a correctly formatted attribution.

   • An attribution generator tool. For closed content platforms (like Flickr) that share lots of openly licensed content. (or just for them to implement RDFa?)

   •  Who are the key AUDIENCES or "customers" for this project?

   • Is this targeted at the end user? Or at a handful of big partners (Creative Commons, Flickr, Wordpress, etc.) who in turn deliver a large number of end users for the tool?

   • Is the tool intended primarily / exclusively for Creative Commons content? Or other use cases as well? 

   • e.g., What's the difference between saying "open content" vs. "CC content"?

   • It  might be interesting to implement this to out of copyright content as well. Just because something is out of copyright doesn't mean it  shouldn't be attributed...just a thought (or even in copyright—does that belie the point?)  - Do you mean attributing fair use cases, and if so how do we implement  something that can judge if the usage is fair use or does it matter?

   • http://creativecommons.org/publicdomain/

   • Is this a "suite" of tools (plural)? Or a "tool" (singular)?

   • (or a "project," if there's a big emphasis on lobbying sites)

   • What format will the attributions use?

   • Answer: Print (plain text) and web (HTML with RDFa)

   • What citation scheme will you use?

   •  MLA, APA? Multiple options?

   • Answer: None, because this is an attribution tool, not a citation tool. Citation generators already exist for all those different bibliographic formats. What we want is an additional output - not a citation, but a proper attribution that conforms to the requirements of a CC (or other open content) license. 

   •  Lots of things that are needed in a "properly formatted" citation won't be included in the creator-provided metadata. Do we need to build a robust citation generator, or how simple will the attribution be?

   • Will this provide a simple attribution based on what's specified in the RDFa? 

   • RDFa  (or Resource Description Framework – in – attributes) is a W3C  Recommendation that adds a set of attribute level extensions to XHTML  for embedding rich metadata within Web documents.

   • Are we doing a simple attribution based on what's specified in the RDFa? or formatting that information into a proper citation?

   • I think a simple attribution is best - then others can adapt to their needs if they want to build a citation tool based on it for specific fields

   • What is the "baseline" of data recommended in RDFa? Will it be hard to change that?

   • What's the user experience like? 

   • For people using / attributing Creative Commons content:

   • Browser add-on: Click a button in your browser. A sidebar appears with the formatted attribution in plain text and HTML. 

   • (Then what happens? How does the process end?) (User copies the attribution, closes the sidebar, and pastes it wherever it needs pasting)

   • Platform-based tool: Click a link or button that appears on the content landing page or the download page (if there is one). (ie. Flickr's "Share this" - then "grab the html/bbc code" on any photo)

   • A pop-up (e.g., using ajax or javascript) offers plain text and HTML options for copying and pasting the attribution.

   • For content creators / authors:

   • Platform based tool: The author provides their preferred name for attribution. And has the option to include  CC+ information about contact information for additional permissions. 

   • What metadata will be included in the attribution?

   • Required attribution elements (metadata fields to query):

   • 1) Title of the work being attributed

   • 2) Attribution name (e.g., author, company, username)

   • 3) Source URL for the attributed work

   • 4) CC license name (i.e. CC-BY, Creative Commons Attribution Non Commercial)

   • 5) CC license URL

   • existing copyright notices? see http://wiki.creativecommons.org/FAQ#How_do_I_properly_attribute_a_Creative_Commons_licensed_work.3F

   • Optional attribution elements (CC+)

   • 1) URL for the author site

   • 2) Contact information for additional permissions

   • What "closed" platforms will you support?

   • Flickr

   • Blip.tv

   • Vimeo

   • Soundcloud

   • Jamendo

   • All those other CC-based music sites

   • (in a way, we're just asking them to implement RDFa? -Ben)

   • (In  a fantasty world, they'd do more than that - they would build an "attribute this" button into every page that users could click to get a  properly formatted attribution. Nearly all academic databases include a  "cite this" button - same principle, only for attributions - Molly)

What "open" platforms will you support? (what do we know about these platforms' plug-in systems? can we write once and port easily?)

--I would think that a plugin for Wordpress, Drupal, Joomla and all the rest of the PHP based content management systems would be a once over in one system and then a quick port. If the PHP is clean, it should be easy to write it for multiple platforms.

just sharing— could be a cool test case:

http://ookaboo.com/o/pictures/ http://ookaboo.com/o/pictures/topic/12586624/Auto_rickshaw

   • Wikipedia (MediaWiki)

   • Drupal

   • Wordpress

   • Zotero 

   • DSpace

   • ccMixter

   • (what do we know about these platforms' plug-in systems? can we write once and port easily?)

   •  --I  would think that a plugin for Wordpress, Drupal, Joomla and all the  rest of the PHP based content management systems would be a once over in  one system and then a quick port. If the PHP is clean, it should be  easy to write it for multiple platforms.

   • What's an example of what the plain text attribution would look like?

   • For example:

   • My Drumbeat Experience / Pieter Kleymeer (@bagabot) / Creative Commons Attribution 3.0 Unported License.

   • What's an example of what the HTML attribution looks like?

   • Code for the pasteable HTML attribution, w/ RDFa

   • <a rel="license" href="http://creativecommons.org/licenses/by/3.0/"><img alt="Creative Commons License" style="border-width:0" src="" /></a>
My Drumbeat Experience by <a xmlns:cc="http://creativecommons.org/ns#" href="http://open.umich.edu"  property="cc:attributionName" rel="cc:attributionURL">Pieter  Kleymeer (@bagabot)</a> is licensed under a <a rel="license"  href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution 3.0 Unported License</a>.
Based on a work at <a xmlns:dct="http://purl.org/dc/terms/" href="http://flickr.com" rel="dct:source">flickr.com</a>.