Creative Commons for Googlers

Creative Commons for Googlers


>>BROWN: My name is Glenn Brown, Iím a Googler.
Iím a product counsel. I work on image search and video and blogger and Wi-Fi and some other
things. And I have the pleasure today of introducing Creative Commons who’s down from San Francisco.
I worked in Creative Commons for three years before joining Google a couple of years ago
and we’re going to hear a little bit about what Creative Commons has been up to recently
and their plans about the future and what Googlers can do help out Creative Commons
both as part of Google and maybe outside. We’re going to talk mostly about technical
aspects of Creative Commons and some of the technical tools they have that help people
who create online media and different kinds of–different kinds of works online to share
them under a some rights reserved copyright, under a less restrictive copyright than the
general default rules of copyright allow. Just a quick sense of how many people here
already knew of Creative Commons before coming to the Tech Talk or know something about it?
Okay, good. So, we’re going to assume–we’re going to assume a good level of knowledge
about Creative Commons. So, again this talk will be, I think, mostly focused on technical
stuff and tools. Then we’re going to have a little movie that Creative Commons is going
to show and then there will be time for some Q and A and weíve got a whole–the whole
Creative Commons staff here from lawyers to designers to business people, so if you have
questions during the presentation or at the end there will be time for that. Right now
I’ll introduce you to the main speaker who is Michael Linksvayer. He’s the CTO of Creative
Commons. And I had the pleasure to work with Mike for two and a half of the years I was
with Creative Commons and heís now the longest serving staff member of Creative Commons.
He’s been around since April of 2003, I think, and his going to tell a little bit more about
what theyíre doing now and different ways that you all can help out. So, hand it over
to Mike now.>>LINKSVAYER: Thank you. Everybody. Hello?
Okay. That’s good. So, from Blogger to YouTube or from analytics to Zeitgeist, I couldn’t
think of any relevant Google products that span A to Z. And analytics and Zeitgeist are,
kind of meta–and then Google, of course, famous for Web search. They’re really expanding
the whole–all of the companyís businesses. Overall really restrictive copyright potentially
impacts every single one of those businesses and Creative Commons is part of the solution
if you want to work together. So, purpose of this talk, as Glen said, is mainly going
to be technical and I want to stimulate of thought and innovation, feedback and its implementation
and there’s some technical details I’m just going to be straight up ask for feedback and
opinion on. Inspiration will be a helpful happy coincidence but I’m not promising it
because the inspirer is in Berlin and this is my terribly cheesy attempt to imitate his
slides because he sometimes uses the opening and closing bracket. That’s Lawrence Lessig
our founder. He spoke here to a much bigger audience; I think last year and heís very–will
get you worked up about free culture. So, do–in order to work together we can get creative.
This is our original movie, released in 2002 that describe Creative Commons in a fair amount
of detail in the background of howÖ Okay. Well, that’s enough of that. That’s a–that
cartoon, I highly recommend that you watch it and show it to your friends who don’t know
anything about copyright and how copyright works on the Internet or doesnít–or fails
to work. We have a new cartoon that I’m going to show you and this first semi public showing
of it, period. It’s a little more succinct and a little more fun. Itís a few minutes long.
So, I think I blew a segue maybe but that video is called ìWanna Work Togetherî and
there is one joke that I wanted sell before, sorry. The inspirer–itís better you have
an inspirer than a decider, I figure. So, first I’m going to just throw out some what
I call chum and chum is of course blood and body parts of fish that’s used to feed sharks.
This is going to be non-technical. Weíll do some fun things to think about so you can
tell your non-technical friends or argue with them about something. So, what are some ways
to think about Creative Commons? Well, one is the way we usually frame it is reasonable,
flexible copyright. Another way is lowering transaction cost, you can think of this in
terms of free speech, cultural environmentalism and if you Google any of these terms you’ll
find some interesting papers and then you can also think of it in terms of free, libre
or open source software and I’ll just say open source from now on but in case anybody
prefers the other terms you can use that. For this audience I’m going to do a couple
of quick illustrations of the relation to open source. So open source, you can think
of as being the green circle in the middle and some Creative Commons licenses fit within
those principles. So, and just be kind of recycling like symbol or share a like, so
that’s our version copy love and then to buy its attribution you have to give credit. And
then there are other things that fall–there are other license conditions that fall outside
of the scope of what would be considered open source such as no derivatives and noncommercial
or commercial uses prohibited. And just a thought, maybe media is 20 years behind software.
And that’s one of the reasons why we need those options. And then another interesting
thing to think about as well, you’ve got code and media licenses and maybe there’s a tiny
bit of overlap now and there’s a bunch of stuff thatís not copyrightable at all. And
maybe in the future the overlap will get better–bigger. And in the future maybe nothing will be uncopyrightable
because, you know, everything will be fixed when you think it. So, interesting thing about
reuse, I think a lot of programmers, when they think about code and content, they think,
“Well, there’s an obvious pragmatic case for code reuse.” But interesting thing about it
is itís applicable in very specific cases. If you have a Python library and you’re writing,
you know, a program in Java chances are you have to re-implement, you’re not going to
be on the directly reuse although, maybe with Trython that’s not the case but, anyway. But
in ours, it’s not invented here it’s not a bad thing. So, there’s not a pragmatic case
for reuse always. But, on the other hand, any arts can conceivably be mixed with any
other art. So, content license interoperability is as important as code license interoperability
even though there are different use cases. So, theyíre a bunch of different kinds of
interoperability that Creative Commons is interested in. I have no idea what interoperability
between code and media really means, that’s–you guys are smart. Maybe you can daydream about
that someday. And there’s media licenses, there’s oral licenses and the licenses used
by Wikipedia. I have–there’s no reason there should be separate silos of contents. I don’t
know if we have any reason–any way to solve that problem but itís something that’s Larry
Lessig and [INDISTINCT] have talked about. If you’re really interested in this stuff,
you can listen to their talk at Wiki–there was that Wiki Mania at that link. And then
across jurisdictions, this is something that Creative Commons has been doing from the beginning.
It’s for real. Creative Commons has been ported to 30 some jurisdictions. So, that’s the end
of this stuff that–I have no idea what I’m talking about and just things for you to talk
about with your non-technical friends almost, except this is all real. So, I just want to
give you a quick update of some legal stuff we’re doing or sort of legal stuff that’s
[INDISTINCT] in front of it. We’ve got the podcasting legal guide which is of course
a guide to podcasters and what kind of roles you have to follow. Both are incorporating
Creative Commons Contents and how to deal with copyright generally. What does noncommercial
mean? Which is probably the biggest controversy because we have a noncommercial license and
it’s not always–it’s not always cut and dry what exactly a commercial use is. We have
a page of critiques and controversies. Weíre all about being open and you can check out
a bunch of articles on the Wiki page mentioned there. We’ve actually had a couple of cases
where a Creative Commons License was involved. Theyíre–both in Europe but they’re kind
of interest–and you can read all know about them at the URL here. Kind of interesting
because one was a case of a Creative Commons license content being misused or outside of
the scope and license and that was upheld. And an otherwise Creative Commons Contents
was being properly used and someone tried to prevent that use and it was upheld also.
And then we are working on version 3.0 for our licenses. You can find some information
about that at the link there. And then Glenn said you can’t just talk about cool stuff,
you have to talk about some really cool stuff that’s being done with Creative Commons, so
I’m going to mention a few of those. One is we have cycle ccMixter which I’ll show you
very briefly. There was–you’ve probably all heard about lonelygirl15 on YouTube. Some
mixes by a user on a ccMixter where used in that–those videos and got seen by hundreds
of thousands of people and Pat Chilla, the artist or re-mixer has since signed a commercial
deal with that operation. Something cool according to Glenn states, I think, we originally–recently
had a benefit concert in New York City with [INDISTINCT] girl talking peeping Tom–peeping
Tom. Something cool according to my taste, Bob Ostertag, a very important electronic
composer released all of his, you know, 20 years of audio work under CC License earlier
this year. So, what about yours as well–and the problem is, you know, there’s no counting
for taste. And there’s a huge amount of stuff that Creative Commons licensed out there.
You know, much of it very awesome. So, that’s one reason one of my kind of peeves or goals,
I guess, is that discovery is a really important thing from clever to filtering to just conversation.
And a few other–bunch of other cool stuff. Second Live Concert with Jonathan Coulton
of Board Game, audio book remix contests, remix, a museum in Boston releasing classical
music under a CC License. A Brazilian movie released simultaneously in theaters and online
under CC, Diggnation and other podcast. We have salons in San Francisco and a bunch of
other cities around the world. And there’s just so much more going on, as we’ll come
up when I talk about some of the technical stuff. Creative Commons both the technical
and the community infrastructures really decentralized, so we donít–there’s a lot of cool stuff
happening that we don’t find out about until six months later or, you know, years later.
I mean just like in real culture. So, now I’m going to segue into the technical part
of this–of this talk. So, in December 2002 we launched and one of the–one of the innovations
was that we had three separate representations of the licenses. If you think of a code license
like the GPL, there’s just the GPL and you can read it if you want to. Very few people
will–people have under–people now understand the GPL, I think kind of through osmosis.
Very few people have actually read it. So, we have three representations, one is human
readable, so we have icons that describe the conditions associated with the license. So,
this one, we have to give attribution. Can’t use commercial use and if you make it really,
really work you have to share a like. Then we also have the actual lawyer-readable code.
And then we have–more interestingly here, we have machine-readable code which describes
both the worked being licensed and the license itself. So, you get–so a machine can discover
is this work licensed and then what conditions are associated with the license. Oh, and we
thought–or I hadn’t joined until April 2003, but basically the thought was this is a really
cool feature to be added to a search engine. Somebody’s going to be building it, right?
Well, it took a little while. In early 2004, we got tired of waiting, I guess, and built
a prototype using Postgres tsearch2 which is a full text indexer for Postgres and Python.
And boy was it ever slow but it was a good prototype. Then in late 2004, we built a search
engine for CC material based on an open source search engine called Nutch which uses Lucene.
If some of you are familiar with Java full text indexing, you might be familiar with
that code base which was an even better prototype and somewhat usable. And then finally early
in 2005, Yahoo! built Yahoo! Search for CC in which basically they filter their web index
according to materials they think are CC licensed. And then in late 2005, you guys did it and
I hope you don’t want them beat you at anything ever again. So, then this is the interface
we have now. If you guys search.creativecommons.org. So, we have a series of tabs that enable you
to access search engines that are CC-enabled in some fashion. And you can get there either
by going to search.creativecommons.org or using the Firefox search drop down which most
people think is just a Google Search bar, I think, but credit to you guys for that.
So, this is an example of search where I’m searching for a clipart and I’m using the
Google CC Search basically, so it’s only searching the portion of the Google index that is thought
to be licensed under Creative Commons License. And you can see the first hint that comes
up is the open clipart library that’s a pretty good–a pretty good hit because that’s a public
domain–that’s a library of public domain clipart. It doesn’t work all that well for
some other terms though. So, I’m going to–before we get to–get to other uses of Search I’m
going to dive a little bit and technically and talk about our metadata infrastructure
that enables this kind of search. So, we started out with RDF/XML, and HTML comments and what
that means is–I’ll describe RDF very briefly later but basically it’s a way to put metadata
in a web page such as invisible to a user and can be copy and pasted into that webpage,
that’s why it has in comments otherwise it would show up as random characters in the
webpage. And I can show you really quick why that was a requirement for us. If you go to
creativecommons.org and click publish and choose a license, let’s say I want to make
other people share a like. Let’s select the license and I get this lock of HTML which
I can put in my templates or my webpage or anything that’s HTML or even word processor
is accepted. Now, we had to be–had to be super generic because otherwise in order to
use CC License you hade to integrate with every single software package. So, that’s
the copy and paste requirement. But–and so, why use–we could have just made up our own
XML format and still done that but the Semantic Web seems like the way to go. There’s a lot
of hype around it. However RDF/XML and HTML comments are super ugly and it may have even
be one of the reasons why it took Google a while to build a CC search engine. Theyíre
heading to humans which means that the–it’s amenable to both spamming and also this ill
maintenance as if a web master changes their license or changes other metadata they might
forget to change the hidden metadata. It’s hidden to XML parsers in a sense because an
XML parser can throw out comments. RDF/XML is confusing to many people if–to be honest
I started–I started using it at another company in 2000–late 2000, I guess, and it literally
took me a year sort of to really crack it. It’s easy to mess up because it’s embedded
in comments. CMS might escape it. It’s more of [INDISTINCT] they needed credit information
were conveying. And this shows you–if you look in this copy and paste little area here,
I think you can see, here’s the beginning of an HTML comment and it’s probably making
you cringe. So, another metadata format that’s new to the scene and I meant to say something
before I started with metadata. This is an area that’s really still developing and there
isnít an ideal solution because Google–I think one of the ideal things for Creative
Commons to let Google is because you guys both facilitate publishing but then also search
is kind of the ultimate consumer so, I’ll come back to that point later but–so, I want
both–I want to arrive at standards that Google wants to implement. Here’s a short version
of it. So, I want an honest feedback. So, microformats have taken up–taken off in the
last year or so. They’re initially called Semantic XHTML basically just leveraging semantic
class names that CSS designers already use. And rel=”license” is one of the simplest microformats.
I’ll show you what that looks like in a second but the benefits of microformats are basically
that theyíre collocated with human visual markup which donít have the downsides of
hidden metadata. They’re also very concise, so they’re not verbose like all of that XML
you saw a second ago. They have a couple of problems. The first problem is the problem
with rel-license by itself really. It’s at the page level. So you can know that a page
is licensed but you canít know if an image inside the page is licensed without using
some kind of heuristics. And then also, what about saying that an mp3 or mp2 is licensed,
what about saying an image in the page is licensed or a video in the page is licensed
et cetera. And there are some skill building interoperability questions about microformats.
It’s a centralized process. You donít need to use names or spaces, so it canít really
be decentralized and that’s depending on the prospective you are coming from, that’s either
a plus or a minus. And this is what the rel-license, rel=”license” microformat actually looks like.
We have a bit of HTML test. This work is licensed under Creative Commons License. And it links
to the license. We just add a rel=”license” attribute to that HTML and that indicates
that we’re not just linking to the license because it’s a cool link or weíre talking
about it. We’re saying that there’s some content here that’s licensed. And you can see that
documented at–on the microformats Wiki. The text of the wrong link–wrong link, it is
actually microformats.org/wiki/rel-license but, anyway. Yeah?
>>Is this out there in the [INDISTINCT] already orÖ?
>>LINKSVAYER: Yes.>>Or itÖ?
>>LINKSVAYER: Yes. This is out–this is out there in a while since 2004. Yeah.
>>Thank you.>>LINKSVAYER: Yes and–yeah. So, what we–what
we want is a solution that both can–you can both–you can make statements about things
whether it’s–whether it’s at a whole page level or fragments or a link in that page
or an embedded object, be it an image, a movie, whatever. And that has a mic–it has the advantages
of microfromat. It’s collocated with–or one of the advantages of microformats. It’s collocated
with human visible markup, so again under premises of hidden metadata. It’s publisher
and consumer friendly, basically. I mean, that it’s low impact for publishers and can
be easily be exposed to consumers or browsers and then it’s flexible and interoperable.
So, one solution that’s–some people associate with Creative Commons and to the v3c Iíve
been working on is color DFA. I’m going to briefly run through that–what that is. It’s
an RDF serialization. The A stands for attributes. It’s the mythical RDF in HTML that many people
wanted for years, including Creative Commons that’s how we arrive at determining RDF in
HTML solution. It was developed for–there are some problems with it. It was developed
for XHTML 2 originally which nobody uses and nobody probably ever will. It uses the [INDISTINCT]
and property attributes which are defined for XHTML 2 but are invalid in XHTML 1. And
some people just hate RDF. The solutions are probably going to be a new revision of XHTML
1 that’s upward compatible. And many of the people–my conjecture is that many people
who hate–say they hate RDF, actually hate RDF/XML because that’s all theyíve been exposed
too. And you can see ongoing developments about this at RDFa.info and I’m going to show
you a couple of examples of what that actually looks like. And this is not actually deployed
except experimentally but I’d be keen to hear feedback on it. So, we got a–this is a simple
RDFa example. It looks exactly like the microformat example but it produces–and in fact it is
exactly the same as Microsoft–microformat example. But you can think of it as producing
what’s called a triple or a sentence in RDF. And that code at the bottom basically is–the
two square brackets right next to each other represents the current URLs in the current
page and then we’re saying that page has a license and the license is identified by the
URI you see there. So what–so a bigger–and I’m going to show a bigger RDFa example that
attacks some of the issues I described before. How do you precisely identify things other
than web pages that are licensed. So images, here–this will be an example for an image
and also incorporating other metadata about the damage besides license. So the stuff in
red as with the previous microformat and RDFa examples are the additional metadata added
to the HTML and here we have a span in the about attribute, identifies the subject so
this is the URL we’re going to be making statements about. And then the rel and property attributes
are verbs or predicates so these are the type of statements we’re going to be making about
that URL. And then either the contents–or I guess either the contents or the HREF is
the object. And so this HTML produces the following triples or the following statements,
so we’re going to–this is using an easy to read RDF serialization called M3 but it’s–I
think it’s pretty self-explanatory. We have–we have an image of food.jpeg. And we’re saying–the
first lines says that it’s an image. The second says that it’s called food captured. The third
says you’re supposed–you’re supposed to attribute that image with the link to bobhome.example.org.
The fourth says that you should attribute the creator’s name as Bob and then finally
we say that the image is licensed under a particular license. And within a single page
you could annotate any number of images or links to mp3s or whatever with his style of
markup. So, it’s–now, I didnít want to try ex–for anybody who doesnít know what RDF
is, I donít want to explain it. Iím kind of cold. It’s–I think it’s easier to understand
after you’ve some examples. But a super short explanation is that you’re making sentences.
You’ve got a subject. That’s always a URL or URI. Got a predicate that’s also always
the URI and youíve got an object and thatís either a URI or text. So example of an object
is the license if your–if your predicate is licensed. And you can say a lot of things
with those primitives and you can do a lot of things with what is like influencing. The
super short explanation of RDFa which I’m showed the demo of, you’ve got the subject
denoted by the about attribute and a differed–if the about is not there then it defaults to
the current document. The predicate is denoted by rel or rev if you want to flip your subject
and object around and its URI and the predicate–whoops, typo. I meant–no. No, it’s not a typo. So
if your–if you’re using a rel then the object is URI. If youíre using a property attribute
then the object is a literal or just some text. Example of that would be Bob is just
some text, so we use the property attribute to denote that. So, Ben Adida, the–who’s
on–who’s the first Creative Commons CTO and went on to do a PhD at MIT and is a Creative
Commons representative under W3C has a briefcase for RDFa and if you’re interested, I encourage
you to go to his page, benadita–ben.adita.net/presentations. But he lays out a case comparing RDFa to microformats
XML and a couple of other metadata technologies and the benefits and summary are publisher
independents. And it’s the creation process is decentralized so any publisher can come
up with their own schema. Data were used in containments are related. Basically, one end
user feature you can get with them is copy and paste because the metadata isnít in a
separate file and is collocated with the human readable HTML. Schema modularity and schema
evolvability are really outcomes of decentralization in part. I–the schemas arenít rigid. You
can–you can have statements from different schemas in the same page, for example. And
they donít have to know about–you donít have to know about how to compose them. You
can be saying that if you had a schema for describing a museum work of art and then you
also had a way to say that it was licensed, you donít have to worry about how they play
together because you’re just making statements about URLs. Then, I want to throw in a fun
Semantic Web analogy or at least I think itís fun. Java outputs are to Java as Cobol as
the Semantic Web vision in hype in 2001 are to Semantic Web technologies now being used
to integrate heterogeneous enterprise data and if you get that analogy then it’s great,
I hope you’ll think it’s funny if not, oh well. But I think the difference might be
that while Java outputs may never come back on the public web, I think Semantic Web technology
as well. So, microformats are worse. And Iím kind of belaboring this sort of pseudo comparison
because it’s something that I really want engineers to feedback on. So, one of the bad
thing–one of the bad things about microformats or might be the worse thing is the S, thereís
no consistent model. Every microformat is a new thing and if you have a parser or like
the details a plug in for Firefox, you basically have a bunch of different statements and you
need to parse out each individual microformat. So, the parsing and technology what the microformat
does or–is tightly coupled as opposed to RDFa, you’ve got a model, so you can go over–a
parser can just extract all the triples and then other software is tasked with deciding
what to do with that. There’s no detours or research. I’ve already talked about decentralization.
The detours in research is really that you’ve seen the Semantic Web layer cake with inferencing
roles and traps and stuff like that. None of that applies directly to microformats.
However, if you havenít seen [INDISTINCT] works is better, I strongly suggest you read
it. Basically elegant solutions often lose to solutions that do something good right
now and that might be where microformats is. A lot of people are publishing microformats
for calendar data and contact data for instance. And their tools being written and however
cool j they are to actually do interesting stuff with microformats. And there’s a potential
media info microformat which would be yet another microformat that, you know, if you
wanted to have a universal microformat parser would be, you know, you had another set of
event statements or event handlers or whatever. But, it could be very useful to Creative Commons
to be able to describe chunks of media as supposed to just web pages using microformats.
And it doesnít really exist yet but it’s sort of in discussion stages. And there are
a couple of URLs there if you want to–if you go to a Microformats Wiki and search on
media, you’ll find them. This is a cooked up media info microformat example that incorporates
another microformat as well but basically how microformats work is you’ve got a class
which designers are familiar with and you denote that, “Okay, I’m starting a microformats
by using a class name everybody has agreed on, so letís pretend media info.î And the
stuff in purple is completely made up. There’s no existing microformat like this. And then,
you might–then youíve got an image here. The famous food.jpeg and weíre saying that’s
with class equals media or I’m imagining this is how it might work. That this is the media
that we’re describing within this microformat and then we’re using vCard to say that the
creator–the creator’s name is Bob and we can use other vCard components to give the
URL the creator and things like that. Now, were–we also reusing the rel-license microformat
to say that the media is licensed. And I just want to stress this is not–thereís no spec
like this right now. There is a case we made for microformats and other kind of non-native
RDF metadata to be made interoperable. Thereís something called GRDDL which stands for Gleaning
Resources–Resource Descriptions from Dialects of Language. It’s basically you associate
some XSLT with some metadata and given knowledge of that metadata, you can get RDF statements
out of it so that way you can take advantage of all the tools that Semantic Web researchers
are building. But, fundamentally, what Creative Commons wants is implementation. We want audio,
image, videos, search et cetera. It’s bcc–CC enabled. And the lack of precise metadata
for objects other than web pages thatís been holding that back so we’re very eager whether
it’s RDFa microformats or something else to see something adapted and obviously Google
is the key adapter. And now, I see you have people doing CMS and another software although
Google does that as well with blogger and other properties. And as I just alluded to,
Google has chickens and eggs. You both have publishing tools and search. So you can really
motivate the adoption of better metadata. So, as I’ve been saying, I want feedback on
this if you want to jump in and contribute, there’s discussions for both microformats
and RDFa. The one for RDFa is a little bit harder to find but, if you search for RDFa,
you can find–work your way towards the W3C discussion list and just start–if either
of these technologies you find workable, just start implementing them in products and even
in your own software products–projects. Then, I want to very quickly run through some other
technologies of interest through Creative Commons. Embedded metadata, it sucks. This
is the metadata that you put inside of a file like an MP3 or pdf or whatever and it almost
always sucks. They’re crappy standards that are not extensible like, id3 for MP3. Everybody
you–programs against it, absolutely hates it. It’s almost never exposed in UIs except
for, maybe in an MP3 player, you might see the title of the song, it’s mostly unused
and there’s absolutely no reason to trust it. So thereís a none–I’ll explain why Creative
Commons is interested in this technology that sucks in a second because there is a–there’s
a somewhat non-crappy standard being pushed by Adobe called XMP. It’s accessible. It’s
basically constrained RDF that can be injected into many different file types including PDF,
JPEG, Flash and pretty much anything that you can–that have an area in the file format
to stick junk in. It’s from Adobe being built into their products line. It’s kind of becoming
a standard in high-end visual asset management systems where an ad agency, for example, needs
to attract its assets but itís being pushed down to consumer level. Microsoft is using
it in some of its Vista applications. So weíre kind of very interested in seeing web applications
also support that. Web applications that deal with media that arenít–that arenít web
pages. And again Search is an obvious driver for that. And then in–another reason that
embedded metadata sucks is because thereís no reason to trust it. Anybody can take a
Madonna MP3 and, you know, stick a comment in there that says this is a CC licensed.
Which means nothing because you have no idea–you know, you might get that through email or
a PDP file sharing network or whatever. You have no reason to trust it. So, we say the
embedded license is meaningless by itself. It needs a reference to a web page that has
the same license information for that media and in that way you can kind of trust the
embedded license as much as you would as a license assertion on a web page because in
fact license assertion is on the web page. And you might not trust license assertion
published on Geocities, so ignore that. You can find a brief write up of this, with some
cartoons actually, at–on our Wiki, the web statements article. So why–so, why do we
care about embedded metadata? Well, for [INDISTINCT] use as they just describe it could harm the
brand or the trust in our licenses and fundamentally, objects or works of art, this sort of stuff
that Creative Commons is interesting for are not only shared via the original URL with,
you know, or web page with metadata. It could be shared by email, PDP file sharing, be it
from a wireless device, a writable DVD, IM, memory stick, hard drive, pretty much anything
and none of these conveyances have facilities for metadata unless it’s in the file itself.
And it’s also just easier for software to assume that metadata comes with the object
in some cases. I’ve talked to many developers of media players for example, even though
they realized that IE3 is terrible, they really–they still have the exotic goal of getting people
to put in correct IE3 text because it’s a hell of a lot easier for them to deal with.
In chicken and egg, again, it’s a big problem. So, we would like companies like Google to
help drive better and better metadata. And then beyond we’re going to do a lot more with
metadata than just licenses. There are a lot complementary things. Attribution is an obvious
one. All of the Creative Commons Licenses require attribution and links are web gold
essentially. So, you can require somebody who went to your website, when they reused
your Creative Commons License content and I think maybe not having a standard way to
attribute with metadata and the link may have been a big–has–may have been CC’s biggest
missed opportunity so far because it’s not–I donít think it’s part of people’s consciousness
when they think of Creative Commons, this is the way I can tell people to link to me
or encourage people to use my content and link back to me. So, commerce are nearby,
they are a bunch of things related to commerce that we would like people to specify with
Creative Commons Licenses or along with Creative Commons Licenses such as where can I buy–where
can I buy related media, where can I get a commercial license, where can I make a donation,
where can I get providence info. And there’s some example–one of the strong things about
the microformats process as an aside is that they really donít want to invent brand new
formats. They do whatís called paving the cow paths. They see how people are already
publishing information and then develop–and then sort of standardize that. So, in the
process of looking at microformats, we’ve started cataloging examples of people doing
all of these things on the Web at this URL you can see on the screen. We want to encourage
decentralized, meaning a [INDISTINCT] and search friendly commerce thatís complementary
to already decentralized public licensing. And also just when people have commercial
success stories and theyíre using a CC license, that reflects well on us. Occasionally, you’ll
run into somebody who comes from the hard–you know, comes from the publishing industry or
just comes from a very strong protectionist background I think, putting Creative Commons
License on your work that’s just a signal to give it away. It’s not true because there
are conditions attached that having commerce linked in in a complementary fashion it reiterates
that message. A couple of examples of commerce, we’ve got Magnatune, a record label using
all CC licenses. And you can see theyíre promoting purchase and commercial licensing.
Jamendo, another great creator of CC licensed music promotes ways to support the artist.
All of this could be marked up with metadata as well. Derivatives, this, I think, is an
interesting idea. A lot of people care about who linked to them but as everybody becomes
creator and re-creators in media besides just web pages, really interest and people arenít
only putting their media in silos. They want to know who on the Web is remixing me. So,
you could imagine a query, like, instead of link call and show me a URL, instead source
call and meaning show me everybody who has used my work at this URL as source material
for their remix. And we kind of have that demonstrated within our own silo. A site called
ccMixter and you can see here, here’s is a track from Pat Chilla of the Beat Gorilla,
the guy who worked with loneygirl15, uses samples from a number of tracks and then has
been resampled. So, all of this could be marked up with metadata to unable that kind of source
search. Syndication, you can stick CC licenses in various indication formats and this is
important for aggregators. And again, Google is just checking in the [INDISTINCT]. You
have both have a blog reader in a number of sites that half feeds or allow people to publish
feeds. And I want–that’s one aside I wanted to make. I’m not really covering here the
technologies to give people the options to license their content because theyíre relatively
simple and straightforward. You could add that to Blogger and we want you to add that
to Blogger. There’s no technology challenges there but when it is added to Blogger, an
interesting thing would be to also add the appropriate license annotation to the feeds
produced by Blogger. So, just do it. There are a huge number of areas that Creative Commons
would like to see Google build Creative Commons into the product. Blog searches now use one
images, there’s a super obvious one video web search is already there blogger docs and
bread spee–dots and spreadsheets, Picasa, SketchUp, YouTube, and I’m sure we missed
some products that either involved content creation or consumption or something in between
or both. Then very briefly, we also want people to write open sourced code that help–that
enables Creative Commons in different applications. I think Google is interested in some of these.
Open Office is a key one. But also, every other content creation package or simply work
to make a free software media tools better and you can go to the developer.creativecommons.org
for this. And then finally, in that vein, we participated in Summer of Code. We had
three successful completions. I just want to give a big thank you to Chris DiBona and
his team which–we didnít participate last year but I think it’s a–get–I think–weíre
really looking forward to Summer of Code next year, let’s put it that way. Because I think
it’s getting bigger and bigger and we’ll have a ton of influence. Then finally I want to
end on a note asking for your support, both not only writing code and integrating into
products, weíre also just started our fall fund–our fall fundraiser and you can go to
creativecommons.org/supports and we–you can buy t-shirts there, you can see letters from
our founder, Larry Lessig and please support us in every fashion because as I said, I think
Creative Commons is part of the solution to a restrictive copyright landscape that impacts
all of Google’s businesses. So, thanks and any questions and answer, feedback I’ll be
happy to take.>>I would say thanks to Mike everybody. Give
him a hand. And now, let’s open up the questions and I just want to remind you that weíre
videotaping this and weíre gong to be putting it on the Web and hopefully under a Creative
Commons License. So justÖ>>LINKSVAYER: Yeah.
>>Ökind of, be careful about asking questions that have confidential information and et
cetera becauseÖ>>LINKSVAYER: Yeah. Also–oh, on that–on
that note, you probably want to ask confidential legal questions because we do have our general
counsel here and our creative director here, so if you have non-technical questions please
ask those as well.>>Great and just for later, just remember
there’s some free pens and stickers and buttons over there too. If you want grab some. AnyÖ
>>So, I [INDISTINCT] see that the creative hard drive and stuff and the–that the licensing
scheme that looks remarkable like your [INDISTINCT] because they have their own freaky things
to it. Are you in general in support this or you would you have want them to come to
you and [INDISTINCT]>>LINKSVAYER: Yeah sure. So, Luke is asking
about the BBC Creative Archive. They used licenses that look, sort of, like Creative
Commons Licenses but they have–if I could quote, “Other freaky stuff attached to them.”
Yes. So, I think that people associated with Creative Commons have been taking to the BBC
for a long time but they–BBC is funded by U.K. taxpayers, so I think one of their requirements
that unfortunately stood in place was that the content can only be free for U.K. taxpayers
and there are–there are other things in there but I’m not–me, I might be able–might want
to say more. That’s about it. Okay.>>Do you think it’s a missed opportunity
or [INDISTINCT]Ö>>LINKSVAYER: Do you want me to give you
this? Itís a little easier toÖ>>Hello? Do you thatís a missed opportunity
and what happens if other organizations start to make their own due of licenses that looks
very much like yours, is it a concern or something you welcome?
>>LINKSVAYER: Okay. It is–it is a concern. I mean, itís a missed opportunity in a sense
but they can only change their mind. Of course, the BBC is a huge organization maybe they
didnít have very much flexibility. But certainly I talked a little bit about interoperability
at the beginning of the talk. Certainly a danger is a bunch–you know, everybody decides
to come up with their own licenses and they’re not interoperable and that reduces the value
of freeing content up a lot because if the licenses are not compatible for some minor
reason, then you have these different silos of content. It’s bad enough now, we have kind
of Wikipedia under the FDL and everything else under Creative Commons Licenses but if
you have BBC, you know, universities try to come up with their own licenses. And fortunately
we’ve been very successful with like open courseware and it’s fall line are using CC
licenses, so. So it’s a danger. Yeah. Any other questions, anybody wants to tell me
how much they hate RDF or love it or anything like that?
>>Well, can I remind you that this is being broadcast.
>>LINKSVAYER: Okay.>>Yes, I hate it. All right. So, thanks once
again to Creative Commons and Mike and the Creative Commons folks will be around if you
have any one on one questions that you want to ask off camera or exchange business cards
or anything like that. And again there’s free pens and stickers and fliers that describe
Creative Commons stuff sitting here. So, grab a handful and thanks again to Mike. Thanks
for coming everybody.

One thought on “Creative Commons for Googlers”

Leave a Reply

Your email address will not be published. Required fields are marked *