[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [kEv] Topics for the Summit
You've given us a very sound set of questions with which to use as a basis for the Summit. They bring up many issues - so that's good. See my comments inline.
Bob and Neil, especially, what are your thoughts?
On 02/03/2007 14:07, JG Rae wrote:
I was wondering how people think we should structure the day on the 9th. I
have four business questions and I think if went through them in order, we
could focus people's energy and get something quite productive going. I'm
floating this as a possible idea, would be very interested to know what
others think.
Q1 What kinds of things do we want to do, that better quality metadata could
help us with?
I'd like to start here, because it's a chance to remind ourselves of our
common purpose before we get too geeky. As I said in my post yesterday,
metadata is for answering "have you got..." and "may I have..." type
questions. The invitation to the summit did a good job of flagging up the
business issues that Kendra is trying to solve.
Yes, this is definitely the place to start - at the business level. What are we trying to do? What's the business need? All our answers will have commonality and uniqueness. Part of the reason for the Summit is to show that there are common threads and needs across business sectors. But there are specific needs to some sectors too.
Throughout the day we should be quick to divert ourselves from getting too technical - especially when it comes to solutions. We need to avoid the 'technology comparison spiral'. We must keep coming back to remind ourselves what we want to achieve, followed by how we go about achieving it.
In my opinion I reckon that this question is what the Summit is all about. Just to share our own experiences and desires. The rest of the questions are more about implementation of a standard but that assumes that a 'standard' is what is needed and, at this time, I personally want to hear from the experts in the field before we assume anything. So, perhaps give more space to this question over the others. As with you I'm in to hearing what others think.
Ideally I would like to ask people to think about this before the 9th;
or perhaps we could hand out cards during registration that people could
write an idea on and drop them into a bucket?
Better to give people as much time as possible. I will commit to sending an email to all registered Summit attendees every day of next week with reminders and updates and questions, etc. Please give me some wording and I'll put it in.
Q2 What tags do we need to support those activities?
This is the meatiest topic. There are some good examples on the Kendra wiki,
and I like the way you flag up that Kendra has a more ambitious scope than
Gracenote. The answers to Q1 should help frame the discussion here.
Keep it at a business needs level - so we'd talk about *how* we'd define tags, what kinds/types of tags we want to define. Again, want to avoid getting into a convoluted technical discussion on what to call tags - like should it be "track" or "song" - that should be done on another day and somewhere else.
So, we need to talk about the process by which we come up with the tags. Who should be consulted, etc. I don't reckon we should try to come up with the standard in one day - but more look at the processes by which we'd go about do this.
It may prove helpful to split into smaller groups, but I don't whether it's
good or bad to divide people along industry lines, or to have a content
discovery group, a licensing group, e-commerce group, maybe. What do others
think?
I'd like to encourage people to mix cross-sector/industry - "try something new". But we could suggest both at vertical and horizontal at different parts of the day.
Business Q3 Who is going to populate the tags?
Business Q4 What controls do we need to restrict people changing some tags?
I think maybe we could handle Q3 and Q4 together. I'd like to stimulate a
social media / mashup agenda, in which some tags are completed by fans a la
Wikipedia, some have to be policed by the content owners (e.g. publisher,
collection society, royalty rate) and some can be filled in by software
translating from other metadata schema.
I reckon one the tags have been created then the Wikipedia effect will take over. Policing sounds like we need an centralised authority - eeek. Just my opinions. So, good questions to get the juices flowing for sure.
I was at a conference on web 2.0 and social media last week, and one idea
that came out was that you can use metadata to tag metadata. This really
helps if you want a standard that interoperates well with other industry
initiatives, because it lets you say that "Kendra-title is the equivalent of
CDDB-track-name".
If you take a look at Kendra Base ( http://base.kendra.org.uk ) you will see that we have built a framework that can handle a distributed set of relationships/equivalents between tags of different schema. We have paused that project due to lack of funding - it's a very complex task - but we have proved that it is possible. Aside: if there were people with deep pockets and/or influence at that conference that wanted this schema interoperability then you should *definitely* get them along to the Summit.
Also, the tag equivalent tag stuff is what a large part of the Semantic Web is driving for - see W3C.
If we're not all exhausted by this point we could talk about exposing our
metadata through API's, and getting some dialogue going between software
developers and users:
Q5 What tools will help us leverage the value of metadata?
Q6 What tools will help us populate metadata, improve it, validate it, etc?
Exactly! Without the tools we will not be able to do anything. Over the last few weeks it seemed to be a common complaint that standards bodies don't build enough tools, be they reference or plugins to existing applications/servers. And that has been the failing of many good standards - no tools which leads to no take up.
Add to Q6 What tools will help us build the tags.
I'd love to know what other people think. Please evolve, warp, edit, mutate
and improve it.
Perhaps we should set these questions up as a wiki page and invite all attendees to comment over the coming week. Get the juices going, as it were...
Cheers Daniel