Kendra Hub - Media Asset/Content Management, Semantic Syndication/Promotion and Commerce Tool

dahacouk's picture

 

 

 

Proposal

The aim is to create a free open source software tool, available from drupal.org, to simplify the management of digital media catalogues for syndicating to distribution platforms, promoting to new advanced semantic search engines (like Yahoo!'s SearchMonkey) and selling direct to consumers. This project, named Kendra Hub, will create a suite of Drupal modules to enable artists and content owners to manage their media assets and associated metadata from within one system that remains in their total control. Initially, Kendra Hub will promote/syndicate to advanced semantic search engines using semantic web technologies being developed within Drupal. Next stages will involve creating plugins for other distribution platforms. Initially, we will concentrate on making Kendra Hub fit for the music industry as this is the industry most familiar to the development team and Kendra Initiative members. But we want to rapidly roll this out to include video and images. This proposal is to create a software tool and not a hosted service. We hope that owners and producers will find this software useful to their businesses and assist them to monetize their content.

  • Mike Large, chief operating officer at Peter Gabriel's Real World, says "Many thanks for this - I think your one paragraph summary of Kendra Hub is great."

Problem

There are many talented artists creating music, films, photos and images who are struggling to get their content in front of consumers. Digital media creators (musicians, filmmakers, photographers and artists in general) get stuck when it comes to distribution of their content. The distribution options are bewildering: Songbird, MySpace, Facebook, iTunes, YouTube, Napster, etc. Creating your own website is trivial but getting that website to host and sell your own music, films or photos can be quite difficult. Setting up a useful website for a content owner can be quite a daunting task. There are many web services that will host an artist's content but always with a loss of control to the artist. Artists and record labels are tied into distribution deals but want to experiment outside of those deals with different ways of working.

Within the music industry alone there are expensive software solutions that are out of reach to artists and bands going it alone. Then there are many, many good hosted distribution platforms (such as Music Glue, Open Music Source, The Orchard, IODA, AWAL, TuneCore...) but these intermediaries take some control out of the hands of the content creators. There are no free open source software solutions available that we know of.

Motivation

We are submitting this project before the Knight Drupal Initiative officially reopens at DrupalCon because we'd like feedback from the Drupal community as developers, administrators and artists (musicians, photographers, image makers and filmmakers) in advance...

The idea for this project arrived in Drupalcon Szeged. This proposal has been a collaborative effort between Stéphane Corlosquet (of Digital Enterprise Research Institute), Daniel Harris (of Kendra Initiative), Darren Mothersele (of Darren Mothersele), Ryan Szrama (of Ubercart) and Axel Polleres (of Digital Enterprise Research Institute).

Rather than just submitting an "enhance semantic web technologies within Drupal" project we want to create a system that will "show off" a particular Drupal use case for content creators and media industries, create a useful tool for this sector and also enhance semantic web technologies within Drupal. One of our aims is not to create a monolithic module but to use and enhance existing modules and so bring maximum benefit to the Drupal community at large beyond the specific scope of this proposal.

Kendra Initiative is campaigning for a flatter and fairer content distribution system enabling artists to make a living from their art so that they can continue to create their art. Drupal is making good inroads into the media industry. We want to see that trend continue and lower the technical barriers and financial barriers for artists to get a useful site up and running. So, they can devote more of their time to making and promoting their art.

Functionality

Key features for content owners (artists and labels):

  • Simple installation - either by install profile or by enabling an 'install module' that takes care of configuration.
  • Simple customization of base theme with basic HTML/CSS preferably configured as theme options.
  • Publish and manage catalogue information, either by data import or by filling in forms.
  • Aggregate activity from the artist's network, for example, feed from their MySpace, blog, iLike, etc.
  • Upload media assets.
  • Import/export media asset catalogue metadata from/to:
  • Option to have Kendra Hub automatically create correct content type to match catalogue to import.
  • Option to create content types by hand and fill metadata by hand.
  • Add data (like price) and other business rules (like region) for selling media assets via subscription (all you can eat) or pay per media asset.
  • Produce aggregated reports from various syndication portals.
  • Publish the content metadata feed.
  • Inform other search engines and aggregator sites of there content metadata feed.
  • Much of the work will be making the workflow relevant to content owner by running extensive field trials.

Goals

  • Make Drupal more accessible for artists and content owners (such as record labels).
  • Make Drupal be the web framework of choice for record labels, artists, musicians, photographers, filmmakers and anyone creating, managing and selling their art.
  • Kendra Hub will be delivered as a set of modules to install and configure other Drupal modules. It will contain configuration, functionality and templates. This will enable Kendra Hub to be integrated within existing sites. For new sites we will also provide an install profile.
  • Dependencies on other modules required functionality (CCK, Views, RDF, Ubercart). It will provide sensible defaults, useful pre-defined content types and views for artists - like a discography in the case of a musician.
  • Through the semantic web framework Kendra Hub will promote free exchange of information about content, enabling artists to publish their metadata and incorporate community tagging and 'semantic trackbacks' into the 'official' metadata repository for a given media asset.
  • Work with artists and industry to create a system that is born out of their own requirements.
  • Built with multilingual localization functionality from the outset. But only with English for this proposal.

Future Work

The scope of this project will limit Kendra Hub to working with music/audio and by promoting content to search engines using semantic web technologies. Future work will include:

  • Expanding the content types to films/video and photos/image.
  • Develop syndicating content to online stores. Many stores have different formats for metadata and content files. For content owners that wish to syndicate their content to a number of different stores this creates quite a burden of work having to repurpose the content and metadata for each store. When this repurposing is done automatically then artists can get on with creating their art. Plugins will be built that will provide the correct input (transcoding of metadata and media formats) to the specifications of the destination portal.
  • Enable artists and bands to manage their online profiles for other social networks (MySpace and Facebook) from Kendra Hub. For example injecting gig lists into MySpace and Facebook via their respective APIs.
  • Build out languages to FIGS (French, Italian, German and Spanish) initially.

Collaboration

There is much work taking place on a vast number of media related modules and Drupal media initiatives. The Kendra Hub project team will collaborate with as many of these module maintainers and initiatives as possible. Just a few links:

How does your proposal meet the stated goals of the Knight Drupal Initiative program?: 

Artists are currently caught in a net of distribution gateways. They need to be able to take back control of their work. Kendra Initiative has been campaigning for a flatter and fairer distribution system for over 10 years now. We believe that these goals align very closely with Knight Drupal Initiative.

  • This project will lower the barriers to entry for artists wanting to distribute their media content.
  • All code will be available from http://drupal.org licensed under the GPL license.

Drupal Community Benefits

  • We will tie together existing Drupal modules (and not reinvent the wheel). We will work with module maintainers to enhance their modules with features that fit our requirements. So, all changes will be contributed back to the modules - this will benefit the Drupal community at large.
  • Work to enhance semantic web technologies within Drupal.
  • Increase the reach of Drupal by making it more accessible for artists and labels.

Media Sector Benefits

Kendra Hub will enable digital media creators to manage their content and make it directly available to the general public. This project will enable digital media owners to:

  • Easily set up a website to manage, sell and promote their content.
  • Promote their content direct to search engines via semantic web technologies.
  • Describe their content with details beyond the conventional track/video/image metadata.
  • Go through distribution gateways.
  • Enable artists to experiment with different ways to distribute their content: the option to use (or not use) DRM on individual media assets or collections of media assets; and being able to choose to charge for some media assets and not for others.
  • Empower the fans with metadata feedback. Kendra Hub will also enable fans to add information about each media asset either direct on the owners website or on the fans website. The fan would then do a "semantic trackback" to the content owners website. The content owner could then rate the validity of the new information creating a semantic web of trust.
How long will your project take to complete?: 

Development and Delivery Schedule

Development will take place along side feedback from participants in the media industry and from the Drupal community. We'll have access to many willing media beta testers through Kendra Initiative members. We want to give time for feedback from industry to be gathered and analyzed so we'll aim to work roughly one day per week to give lots of space for this process. At 250 hours per person. 36 days per person. Working one day per week. 36 weeks is 8.3 months. So, there will be enough space for feedback trials. Estimate 10 months in total.

  • Milestone 1
    • Artist (or label) site with catalogue import (media assets). It makes sense to have this ready first so that artists and labels can start working with the Kendra Hub trial and we can add/configure extra functionality as it becomes available.
  • Milestone 2
    • Framework for install profiles/install modules for deploying configuration.
  • Milestone 3
    • Integration of RDF and commerce using appropriate vocabularies like SIOC and GoodRelations. Integration with revyu.com for reviews.
  • Milestone 4
    • Integration with third-party distribution services.
How will you implement and distribute your project?: 

Deliverable Distribution

  • Distributed by hosting modules as projects on http://drupal.org
  • Implemented as a set of Drupal Modules. An install profile for new sites and a configuration module for existing sites will be provided.

Modules to be created

  • Kendra Hub

    • Master installation and configuration module for the whole project.
  • Kendra Hub Manage

    • Discography functionality
      • Provides default CCK content type with fields for artist/title/imagefield/isrc/composers/publisher/(p) and (c) copyright holders/other standard metadata fields. Default views: discography page, latest releases block. RDF publishing of metadata.
      • Enables add and modification of media asset files metadata.
  • Kendra Hub RDF

    • Provides configuration for RDF modules to publish metadata for search engines / other uses.
  • Kendra Hub Theme

    • Provides theme settings - easy way for artists to manage the look and feel of their site. Based on something like Zen that allows configuration by CSS with additional configuration tools via web GUI using the theme settings api.
  • Kendra Hub Syndicate

    • Interfaces with external sources to provide metadata and media assets.
  • Kendra Hub Commerce

    • Interfaces with Ubercart.
  • Kendra Hub Report

    • Receives reports and statistics from external sources and repurposes them for use with Kendra Hub.

Modules to be enhanced

  • Ubercart

    • Current functionality
      • Ubercart currently allows users to sell expirable user role promotions and file downloads. This means a base install is already usable by someone wishing to sell content as an "all you can eat" site membership or on a per-asset basis.
      • Currently, Ubercart products must be of a specific node type, meaning any new media that a site owner wishes to sell would have to be made as a product node (of either the base product class or custom product class dependent on the type of media).
      • Ubercart is dependent on the Token to expose various order and product fields using the Token module. Users wanting to take advantage of the core image support must also have CCK, FileField, ImageAPI, ImageCache, ImageField, and Lightbox2 or Thickbox. These modules are all well-known and currently distributed with Acquia Drupal. These modules will be useful for other parts of the project, though if image support is not necessary for an artist's product, these may all be left uninstalled.
    • Proposed enhancements and benefits
      • Ubercart's core import module was removed in the D6 migration. Either a core import module will need to be re-established, or some sort of Kendra Hub specific importer will need to be created that is capable of reading data from the proposed formats and creating new product classes and products on the site.
      • It will also be beneficial for Ubercart to move toward allowing product information to be assigned to existing content on the site. An artist who decides to list assets on their site and sell them later would have a hard time converting that content if it did not start out as a product node. If Ubercart were more flexible in this area, which is a goal of the project in general, it would benefit artists and a lot of users in general. On Drupal 6, the plan is for the 3.x branch to introduce a dependency on CCK in preparation for migrating to Drupal 7 where fields are already in core. Basically, any node with a populated product field would be purchasable through Ubercart.
      • There is huge potential for artists and Ubercart users in general in exposing Ubercart catalog information to the semantic web through RDF. This should be possible either through core improvements or an RDF helper module using Drupal's hook_nodeapi() to add the necessary information to product and catalog displays. Features exposing Ubercart catalog and product data to the semantic web would benefit all stores and have specific applications in affiliate marketing, marketplace applications, cross-site selling, and more.
      • Work on Ubercart installation profiles is underway but lagging from sufficient dedicated time and effort. A major goal for Ubercart as part of Kendra Hub is to produce a base installation profile that contains reusable code (similar to the Installation Profile API) for other profiles to use to simplify basic store setup functionality and add steps to the installation process specific to the e-commerce side of the site. The code should also be usable for non-installation applications where e-commerce functionality is being added to an existing installation.
      • Ubercart's core systems make it difficult for international sellers to correctly display and collect taxes. This needs to be continually worked on, and as this will likely come up during the testing and review stage of development, any necessary changes will be specced out and developed for core inclusion.
    • Kendra Hub interaction
      • Ubercart's basic reports module is likely not going to be sufficient for the hub's needs. As such, it may be necessary to expose additional or more granular data to the hub. This will be specced as necessary.
  • RDF vocabulary importer

    • The RDF vocabulary importer takes care of the management of external vocabularies which are locally cached. The user can specify the URL of an existing vocabulary such as http://xmlns.com/foaf/spec/ and the classes and properties of this vocabulary will be replicated locally. These RDF terms are then made available in the RDF CCK mapper for the user to map the content types and fields to popular RDF classes and properties.
    • New features will be added to this module in order to provide a better user experience:
      • Automatic import of vocabularies upon installation as part of the Kendra Hub module.
      • Consistency checking with regards to restriction of the external vocabularies (subclass, domain, range) in order to avoid inconsistent RDF data.
  • RDF CCK mapper

    • The Content Construction Kit (CCK) is a module which allows site administrators to structure their data and add fields to their content types. Here is an example: say you create a new content type called 'artist'. It is then possible to add several fields to it, such as 'name', 'country', 'homepage', 'album'. When you then create a page of this content type 'artist', you will be asked to fill these information in a structured manner: 'name' and 'country' are text fields, 'homepage' will be a link to the artist homepage. In the case of 'album', you will be able to choose the albums this artist authored in a select list, the content of this list being the albums featured on your site.
    • At present, the structure of the content that CCK constructs is only known by the site itself. The semantic Web and the RDF technology in particular can describe this structure in a standardized manner so that it can be understood by other external systems in the Semantic Web cloud. Each piece of data can be given the semantic it deserves so that other applications can understand its meaning and reuse it seamlessly.
    • The RDF CCK mapper allows to export Drupal site data into RDF following predefined mappings. Moreover, these mappings can be altered by advanced users if necessary, and imported/exported in a similar way as CCK and Views do. These mappings definition will comply to existing CCK definitions so that they can easily be integrated within CCK definitions in other modules.
    • The same mapping logic will be reused in other parts of Drupal in particular in the case of separate data that the Ubercart module stores in specific tables. Further extensions of the CCK RDF mapper include a more general mapping system allowing any type of instance in Drupal to be mapped to RDF terms: users, user profile fields, files, taxonomies and their terms.
  • Neologism vocabulary builder

    • RDF is the underlying key framework that enables interoperability on the semantic web. When publishing RDF content, it is important to make sure the RDF vocabulary used to describe this content is properly defined and available online. To this end, the RDF Schema language was created which allows the precise description of classes and properties used in RDF.
    • Neologism is a lightweight web-based vocabulary editor and publishing tool built with Drupal. It makes vocabulary authoring easy: just create a vocabulary, add classes and properties to it, and your vocabulary gets instantly published and available online! Several formats are supported via content negotiation: HTML, RDF/XML and N3. All the term URIs are dereferenceable and point to their human readable description.
    • Neologism was ported to Drupal 6 recently (v 0.3) but still lacks some critical features. In this project we will extend and refine Neologism, including the following features:
      • good upgrade path from previous Drupal version, namely from Drupal 5
      • XSD datatypes for RDF literals
      • linking to external vocabularies via the RDF terms importer module, opening the possibility to specialize existing vocabularies which might not fit 100% of your needs otherwise
      • import vocabularies to be edited and improved with Neologism, typical case of migration from an old system to Neologism
      • porting the multiselect module to Drupal 6, which is used by neologism to improve the user experience in some places
      • various usability improvements
    • We will also conduct some usability studies in order to facilitate the adoption of Neologism.

Modules to be utilized

  • RDF API will be the main module on which the other RDF modules developed or improved throughout this project will rely on.
  • Token is used to make various bits of data from Ubercart objects available for users to place in text messages.

Development Team

  • Stéphane Corlosquet (scor), Digital Enterprise Research Institute, Semantic Web Lead, Drupal Profile, Groups Profile
    • Role: development, Semantic Web and documentation.
    • Biography: Stéphane has been a Drupal user and contributor since 2006. He is the maintainer of RDF CCK, Neologism and Evoc. He talked at several DrupalCons and BarCamps promoting the Semantic Web and Drupal.
  • Daniel Harris (dahacouk), Kendra Initiative, Project Lead, Drupal Profile, Groups Profile
    • Role: team lead, project management, media industry liaising, trial coordinating, concept design, marketing and documentation.
    • Biography: Runs (1999 to present): Kendra Initiative. Ran (1995 to 2001) Cerbernet - business ISP based in London, UK. Ran (1994 to 1996) Cerberus Sound & Vision - created Cerberus Digital Jukebox - the first system for copyright protected Internet-based music distribution. Runs (1999 to present) Freewheelers - travel sharing/matching website.
  • Darren Mothersele (darrenlondon), Darren Mothersele, Development Lead, Drupal Profile, Groups Profile
    • Role: development, concept design and documentation.
    • Biography: Founding member of digital music distribution company Uploader. Worked with music industry and built systems for B2B distribution of music assets. Headed up technical operations until its acquisition by IODA in 2007. A user of Drupal since 2003 and full time Drupal developer since 2007.
  • Ryan Szrama (rszrama), Ubercart, Commerce Lead, Drupal Profile, Groups Profile
    • Role: development, Ubercart configuration and documentation.
    • Biography: Ryan has been the lead developer of the code and community driving Ubercart since 2006. He's been a part of Drupal conferences and training events since 2007, promoting and instructing people on using and developing for Ubercart.

Marketing and Promotion

  • Industry feedback will be sought throughout the development of this project. Members of the development team will run trials with artists and managers at music conferences and continuously online. Attending these events will enable us to also promote Kendra Hub throughout development cycle. Over the last 10 years Daniel Harris has spoken at and hosted Kendra Initiative workshops at many media industry events. Being invited to run workshops will happen as a matter of course and these can be used to obtain feedback for Kendra Hub.

  • Some ideas for music events include: Canadian Music Week, Womex, In The City, Midem, SXSW and Popkomm.

Industry Feedback

  • Kendra Initiative will host a trial for Kendra Hub and engage with content owners to provide usability feedback.

  • The Kendra Hub trial will be continuous throughout the life of the development of the project.

What is your total budget estimate and how much funding are you requesting: 

Budget

  • 65,000 USD for 3 programmers for 650 hours at 100 USD / hour.
  • 10,000 USD for travel and accommodation to 5 conferences for 2 people.
  • 25,000 USD for project manager for 250 hours at 100 USD / hour.

Total Budget: 100,000 USD

Funding Sources

  • Kendra Initiative will go out to industry for additional sources of funding to extend the scope of this project along the lines detailed in the future work section.
  • Kendra Initiative is receiving some funding from the European Commission via P2P-Next, a 4 year FP7 project in conjunction with the BBC, Pioneer and many others. P2P-Next is not funding this KDI proposal for Kendra Hub. However, there are some parallels between the work Kendra Initiative is doing for P2P-Next and Kendra Hub. This can only be beneficial. The contacts that Kendra Initiative has will be useful to disseminate and get feedback on Kendra Hub and also push out the Drupal project as a whole.

Comments

Avoid "One-Stop-Shop"

Fr0s7's picture

I agree with the idea behind this initiative, and I'm excited to see some development in this area (since I run a band website myself and music distribution is indeed a big challenge).

However, I believe it is important to avoid a "complete package" or "one-stop-shop" mentality. Let's face it; when orchestrating all of the different features into a system like you're proposing, there is no simple turn-key solution. It would make more sense to deploy this system as a suite of modules, each of which manages one feature. This presents the opportunity for designers to use only the features they want/need. It also lays the groundwork for robust APIs so that developers can extend the features they need in a more granular fashion. In my opinion, a good model for deployment would be Media Mover which is broken down into a series of modules for the same reasons.

  • Frost Simula

Agreed...

dahacouk's picture

Hi fr0st,

Thanks for the feedback...

That's exactly how we envision this project. The Kendra Hub suite of modules, that we've specified in the proposal, are lightweight and used, in the main, for configuring and managing the heavy lifter modules: CCK, Views, RDF, Ubercart and many more. We've divided the Kendra Hub suite into the functions as you suggest. Are you saying we should further subdivide the Kendra Hub suite into more granular functionality? If so, what would satisfy your requirements?

One of our aims is to provide a coherent project that can be used "out of the box" by artists, bands and record labels (initially). So, we'll define a common feature set that satisfies their base (plus a little bit more) requirements and provides configuration options for the features they want. We don't want a "one-stop-shop" solution either, as we said "one of our aims is not to create a monolithic module..." We want to have functionality easily extendable using common Drupal methodologies and the use of APIs.

We'll essentially be configuring modules that are already available at large within Drupal. But, if we have a requirement for significant functionality that can't be fulfilled by an existing module or even fulfilled by enhancing an existing module then we'll create a new standalone module project. Then configure and invoke this new module from one of the Kendra Hub suite modules. Hence, there should not be any real juicy functionality in the Kendra Hub suite - only configuration and process control. We want any code we create to be as reusable as possible beyond the scope of this specific project.

On robust APIs and discrete features, we agree. How can we better express that in the proposal? How would you like to see that implemented as a band website manager? What would work for you?

We may well end up extending the existing semantic web APIs but these extension won't be specific to this project. With this project we are proposing a use case for catalysing the evolution of semantic web technologies in Drupal - albeit a very useful use case.

I hope that clarifies. Let us know...

Cheers Daniel

Open Media Project

deproduction's picture

If you look at the evolving Open-Media Project module-set (introduced here: http://groups.drupal.org/node/12876), you'll find a lot of overlap and opportunity for cooperative development. We'll work on revising these modules and our use/tweaking of existing modules and if this Kendra project moves forward, it would be good to leverage the work we're doing.

Tony Shawcross

Whatever your first issue of concern, media had better be your second, because without change in the media, the chances of progress in your primary area are far less likely. http://denveropenmedia.org

Whatever your first issue of concern, media had better be your second, because without change in the media, the chances of progress in your primary area are far less likely. http://denveropenmedia.org

Areas of Overlap

darrenmothersele's picture

Hi Tony, I'll keep an eye on the Open-Media Project as it evolves.

What are the areas of overlap you are expecting?

Complimentary Drupal music/media projects

gusaus's picture

After spending a good deal of time lately looking at the Open Media project, I can attest to the fact there's a good deal of compliment. Even moreso are the potential compliments to the current effort to 'finally' build a music distribution for Drupal. When you look under the hood, you'll see a lot common features/functionality as well as need for qualified people to do the work.

Funding, on the other hand, really isn't the issue. A lot of the media projects, including Open Media, have received funding and there are a good deal of stakeholders in addition to Google and Knight who would benefit from sponsoring these projects.

Based on this, I think might be a good time combine resources and collaboratively build out common pieces of functionality. From that, we can leverage something like Patterns to create custom setups and branded solutions.

If it seems like a good idea, possibly we could start by looking for overlap and ways to collaborate on the media projects listed here.

Gus Austin
PepperAlley Productions

Gus Austin

Where are these programmers

Garrett Albright's picture

Where are these programmers who are working for $100 an hour, and who do I have to kill to become one?!

That's the average rate in

scor's picture

That's the average rate in the KDI proposals. As doublejosh said, it includes various taxes and insurance fees.

Keep this as an Install Profile or a module group

doublejosh's picture

Have to agree about overdoing a feature set. Keep this as an Install Profile or a module group, not a full platform solution-- sounds a little like a whole open-source statup model. The project would be great, I would use it, but gotta float before you sail. Subscribing to the project.

Garrett, freelancers have to change these kinds of rates due to self-employment tax, health insurance, etc. In the end it's still best for the project initiator who doesn't have to employ talent.

I would use it too...

dahacouk's picture

Hi doublejosh,

Thanks for your input. This project will be supplied as an install profile and a module group. The Kendra Hub suite are merely lightweight modules for configuration and process control.

I don't think fr0st was suggesting that we were "overdoing a feature set", was he? I think he just wanted the feature set to be implemented as a set of discrete modules linked by robust APIs rather than a mass of tightly coupled and difficult to extend modules. And we agree. Make sense?

A significant part of this project is being spent on requirements gathering and usability testing with those in the media industry. It's got to work for them too and not just in the lab. There's mileage to be gained by bringing them into the development process. When they become part of the project they are more likely use it and promote it to others. In order to trial Kendra Hub and get meaningful feedback as we proceed we do need to have a fully working system that is configured to give a base set of functions.

I hope that makes sense...

Cheers Daniel

Wow, apologies.

doublejosh's picture

Hadn't seen the momentum... http://www.kendra.org.uk/ ...awesome.

Momentum...

dahacouk's picture

No need to apologise - we're all here learning together from each other. Anyway, "awesome" is gratefully received. Do you reckon we've done with learning to float then? ;-)

Though Kendra Initiative has been going for around ten years it still feels like nascent days. It's been a long haul but lots of fun. Only from the start of 2008 have we had any serious funding. We can now actually employ staff, which feels great. Being funded by the European Union for 4 years with http://www.p2p-next.org has given the project real stability and great kudos. Doors are now starting to open for collaboration on other EU and globally funded projects.

Cheers Daniel

I watched the video on your

zirafa's picture

I watched the video on your website and read through this proposal, but unfortunately the goal of the project still feels vague to me. On the one hand, you are promoting decentralization through RDF and aggregation by the content creator. On the other hand, the project aims to make a single point of entry for a consumer to buy any content from any vendor. I understand the benefits from a content creator point of view - I can publish some media and then it gets picked up through aggregation and metadata to everyone else in the network. But from the point of view of a consumer, how will this work and more importantly how will this look to the end user? A link? A website? A search engine?

That being said, I support any initiative that promotes semantic web publishing because it makes the web a more organized place. I just want to understand the project further and what makes it distinctly unique or innovative from other semantic web projects out there.

To clarify...

dahacouk's picture

Hi zirafa,

Thanks for your questions and support...

It's important to keep in mind that Kendra Initiative and Kendra Hub are not the same project - albeit they share the name. I'm a bit confused about which project you are discussing. It may help to specify which projects we are talking about. Kendra Hub is only what you see in the proposal at http://groups.drupal.org/node/18936

We chose to associate this project with Kendra Initiative so that there'll be recognition/acceptance with Kendra Initiative members and onlookers in the media industry when it comes to trials. Kendra Initiative is fostering a number of open source/architecture projects that align with our goals of a flatter, fairer and more interoperable content distribution system. Kendra Hub is a very recent addition to the fold.

So, there's understandably some confusion about the specific functions that Kendra Hub is being built to provide and the sometimes enormous looking goals of Kendra Initiative. They are not the same. Kendra Hub wont fully satisfy the goals of Kendra Initiative. It's not being proposed for that reason. The reasons are in the proposal. We want to support the Drupal project and especially semantic web technologies.

I'm guessing that you're really asking questions about Kendra Initiative. I'll keep it short because I don't want to get into a discussion about Kendra Initiative here as we should keep to the thread topic of the Kendra Hub proposal. Kendra Initiative is not about defining how this will "look to the end user". It's about defining interoperability within and between all the links in the content value chain. It's about providing glue between all of the services providers and content owners. In some ways this may mean that there will be very little change to what the consumer sees. But we reckon (and hope) that by standardising the protocols and APIs between all the different services that new ways of doing things will be created and artistic/creative work will flourish.

Kendra Hub is, in part, providing content owners with a shop front using Ubercart but I would reckon that they'll still want to put their music up on iTunes, Napster, Songbird, etc because that's where the consumers hang out. With Kendra Hub we want to make it easy for them to do both, experiment and see what works best for them.

And leverage new semantic web search engines. Let's be honest, we don't know how the semantic web is going to pan out. Or what it's going to look like to the consumer. But it's sure going to make things interesting. And being there is so important because it does seem like it's the way forward. I think we all agree with that.

So, how will Kendra Hub "look to the end user"? A shopping cart? Or perhaps not at all directly. It's a tool to make it easier for content owners to manage, promote and sell (if they wish) their content using semantic web technologies.

I hope that helps...

Cheers Daniel

Ah, you are right that I was

zirafa's picture

Ah, you are right that I was confusing Kendra Hub with the Kendra Initiative. Thanks for the clarification.

I think overall it is a very interesting project and I will definitely be following the work.

As far as the proposal I think the main issue surrounding media publishing with Drupal is dealing with the file upload process. I think a CCK solution is by far the most extendable, perhaps you could create a way to generically connect metadata to a CCK file field which could then be used to create XML feeds, display method (i.e. show this file in flash player), or external links to more info, etc...perhaps you should contact drewish who has done a lot of work on the file system and is also the maintainer of image and audio modules and he might have some ideas.

Naah. If I'm selling tracks

Garrett Albright's picture

Naah. If I'm selling tracks from my band's eleven-track CD, am I going to want to upload them one-by-one inside a web browser? And who ever heard of a module which added CCK fields to content, anyway?

CCK is great for end users, but when you're writing code, it's saner and more flexible to just implement the fields you want to use in that code.

Hi, CCK File Fields allows

darrenmothersele's picture

Hi,

CCK File Fields allows us to attached multiple files to node, but I get your point about not wanting to upload multiple tracks through a browser. I think we should look at some kind of import module. I recently had to import 1000s of tracks of audio to a site, and uploading each one via the browser was not an option! We can still use CCK File Field to attach the files as it can be populated from import code, and then this provides the interface for adding and manipulating files through the browser later.

XSPF For playback

arthurf's picture

XSPF playlist can generate a playlist from an arbitrary number of items attached to a node... just FYI


http://24b6.net

Agreed, file fields are

darrenmothersele's picture

Agreed, file fields are probably the most extensible way to deal with files and it is certainly possible to use metadata from files for populating CCK fields. We also have to think about how this will work with the commerce part of the tool, as Ubercart offers another alternative for dealing with files.

looks interesting..2 quick points

Dublin Drupaller's picture
  1. Drupal eCommerce API

I think it's worth suggesting that you look at the new Drupal eCommerce API as an alternative to Ubercart for this project.

I've been playing around with it recently and the way it's structured might suit an initiative like yours more than Ubercart. Ubercart is very good, but, it's striving to be a turnkey shop solution like Magento or previous versions of Drupal eCommerce. In other words, the new Drupal eCommerce API is a complete re-write towards a more native and expansive eCommerce solution for Drupal.

That said..the new Drupal eCommerce API requires a much higher level of Drupal savvy than perhaps Ubercart/Magento/Oscommerce and others. On the flip-side, it's much much lighter and efficient (code-wise)...which may make it a better candidate for building a Kendra hub in the way you described. That's not to say it's better than Ubercart...I'm just suggesting that that Ubercart is a good turnkey solution for beginners and intermediate drupal developers whereas the new Drupal eCommerce API is designed for more sophisticated and enterprise-level solutions.

  1. Localisation

It's easier to apply localisation to a project if it's planned at the outset, rather than added in later. So, while the lead language for the project is obvioiusly English and it will make sense to develop betas in English, it maybe worth adding in provisions for localisation at the outset..even if it just covers FIGS (French, Italian, German and Spanish). That's not to say you introduce extra costs of translation at the development stage, rather including the foundation stones for localisation..e.g. all the developers understand that they are building towards a multilingual solution and code with that in mind.

cheers

Dub

Ubercart has an API as well.

Garrett Albright's picture

Ubercart has an API as well. It has a couple challenging holes, but it's at least competent enough that I'm able to write something like Ubercart Auction. I'm interested in comparing the two, but the eCommerce API documentation seems to be lacking a bit…

I honestly don't recall the

rszrama's picture

I honestly don't recall the initial contact and brainstorming around the decision for the underlying e-commerce component, but that said, even given your description of it, I'm not sure it's right for this project. When you consider the first goal of Kendra Initiative, that it wants to simplify and streamline buying and selling digital content, I think it makes more sense to work with a system that is easier to use out of the box and requires less "Drupal savvy." Otherwise, Kendra Hub would have to devote developer time to the task of providing a simple UI for administrators and customers in addition to all the other tasks required for the project.

I'll be the first to say that Ubercart needs continual improvement, but we're continuing to make strides in Ubercart and in Drupal core to simplify the selling process. While I, too, am impressed by the latest development in the e-Commerce API, I will agree with Garrett that while imperfect, the underlying Ubercart APIs are sufficient for any task mentioned in this proposal. Where it needs to be improved, well, that's what bringing more projects into the Ubersphere is all about. If a patch makes sense to push upstream, we'll get it in the Ubercart issue tracker for review and inclusion. Being the maintainer of the project will certainly help streamline the process. ; )

What's exciting to me is the way this project's goals, including simplifying selling non-physical products and bringing e-commerce to the semantic web, absolutely line up with two of my major goals for Ubercart innovation on Drupal 7. I absolutely want to move in these directions, and there is already work in the form of a core proposal and Ubercart installation profile development. I see Ubercart and Kendra Hub feeding off of each other's momentum and producing solid contributions that everyone will benefit from, including future Ubercart users who haven't heard of Kendra Hub. Furthermore, there's just a lot of momentum in the Ubercart user and developer communities in general that will benefit anyone who decides on Kendra Hub before they know what Ubercart is.

I also think it's worth pointing out that Ubercart already has adoption and support from the likes of Warner Bros. Records, indie record labels, media distribution sites like Riff Trax, and various and sundry musicians, artists, photographers, and others who will benefit from the work of Kendra Initiative.

just spotted this

Dublin Drupaller's picture

hi Ryan,

I'm not sure [Drupal ecommerce] is right for this project. When you consider the first goal of Kendra Initiative, that it wants to simplify and streamline buying and selling digital content, I think it makes more sense to work with a system that is easier to use out of the box and requires less "Drupal savvy." Otherwise, Kendra Hub would have to devote developer time to the task of providing a simple UI for administrators and customers in addition to all the other tasks required for the project.

I'm not sure it's right for the project either..but having looked at the Kendra hub spec...and having looked at the most recent dev version of Drupal eCommerce...it struck me as a much better fit than Ubercart.

I recommend you have a look at the new Drupal eCommerce API....it's designed as an API rather than a package. In other words, instead of downloading 20 or 30 modules, include files and css, you're just downloading about 9 modules that work with other Drupal modules. So you're actually reducing the amount of UI development required. Or to put it another way, because the new Drupal eCommerce API is designed to be more Drupal "native" - less work is involved, particularly theming-wise, because it's essentially plugging into the other modules and the UI created for those modules.

Another bonus is that the new Drupal eCommerce API is less "product" led and more "content" led..which essentially means any content on a drupal site can be sold...text, graphics, photos, articles, video, audio etc. which fits with the core Kendra Initiative.

Ubercart already has adoption and support from the likes of Warner Bros. Records, indie record labels, media distribution sites like Riff Trax, and various and sundry musicians, artists, photographers, and others who will benefit from the work of Kendra Initiative.

I appreciate the enthusiasm for your project, but, I think it's a tad ambiguous and grandiloquent to suggest that Warner Bros and indie labels are adopting and supporting Ubercart. A more objective viewpoint would be that they are adopting and supporting Drupal....and Ubercart is just one of the many modules that came with their Drupal sites. In the same way The Fast Company didn't choose to adopt organic groups first and then pick Drupal, or how Just Racing UK didn't adopt the views suite of modules first and then pick Drupal.

It could also be argued that just because many companies are using Ubercart, doesn't automatically qualify it as the best ecommerce solution. I don't think there is a one-size-fits-all ecommerce solution. Ubercart is a great turnkey shop solution foir newbies while the new Drupal eCommerce API, to me, is a more involved set of APIs that will work with other modules.

In other words, if you are going for a jog, you would tend to wear a pair of trainers but that doesn't mean you should wear trainers if you want to go for a cross-country hike.

If a Drupal newbie asked me for advice on setting up a drupal shop, I would point them straight at Ubercart..but, if someone approached me, in the same way Scor has approached the Drupal community with this thread, asking what would be a smart choice for including ecommerce in a drupal distribution for record labels and artists..I would definitely opt for the new Drupal eCommerce. I don't think Ubercart is a bad choice for what you're planning...but, I would definitely look more closely at the options.

Like I said earlier..there's not much docs available yet...it's been in heavy development recently so I recommend Scor has a chat with Gordon or maybe take a look at the latest dev version.

cheers

Dub

Ubercart decision and localisation (a great idea)...

dahacouk's picture

Hi Dub,

Thanks for your points...

  1. Ubercart decision

In Szeged, after seeing Ken's KDI presentation and having a chat with him I was racking my brains for a project idea. I had attended presentations from Stephane, Darren and Ryan. Things started to click and the concept for Kendra Hub was born. I knew a little about the "which e-commerce system to use" issue. But I have a visual kind of mind and was impressed by the Ubercart project's attention to detail, look and feel - albeit, admittedly, on a superficial level. Decisions get made on the spur of the moment. It's serendipity - right place and right time. I got a good gut feeling from chatting with these guys. They are passionate about what they do. I trust my gut.

All that said, I'm really pleased with the way fate/chance has worked it all out and Ubercart are on the team. I'm really excited about the potential for innovation, reading what Ryan says. I especially like the ideas of Ubercart integrating more with CCK and selling already existing content.

  1. Localisation

Great idea, thanks. I've added it into our proposal and it should appear here the next time Stephane updates. Kendra Hub will be multilingual capable from the start. Cool.

Cheers Daniel

understood..

Dublin Drupaller's picture

hi Daniel,

Points taken. I agree with you re: Ubercart, it is exciting what the UC team are doing but I don't really see the "which ecommerce system to use" question as an issue..I think it's really simple: you use whatever works best for the particular situation. When I saw the Kendra initiative website and was reading through it's goals and objectives, I immediately thought that Drupal eCommerce is a perfect fit for this.

The main reason is because the new Drupal eCommerce API doesn't just interact with CCK, it interacts with everything i.e. it's less Product led and more Content led...which, for me, fits with the core concept of what you're planning.

The Drupal EC Team are deep in development at the moment, so there's not much documentation available I can point you to, but, I recommend you have a chat with Gordon the EC Team leader or download and have a look at the latest dev version, which I did recently. It's obviously your call, but, when you're looking for funding I think it will look better if you can display a S.W.O.T. on both and the logic behind choosing a particular building block...attention to detail and all that sort of thing.

For others reading this (I cringe when I see "which ecommerce system to use" discussions on drupal.org) - [b]I'm not saying the new Drupal eCommerce API is the best ecommerce solution for Drupal site admins. It's not...because there is no such thing[/b]. Both Ubercart and Drupal eCommerce have their merits and foibles, but, neither is an one-size-fits-all solution. If you're looking for a quick shop-in-a-box, I would tend to swing towards Ubercart or Magento. If you are an experienced Drupal developer and are looking for something more adaptable or building an enterprise level solution I would tend to swing towards Drupal eCommerce.

re: localisation

You're welcome. Glad to be able to help and am looking forward to seeing how it progresses.

My background is in localisation and more recently I have been working with record labels and artist management companies. That industry is slowly shaking off the legacy (or baggage) of a licensing-led business - where recording and publishing licenses were usually territorial - for example, the single european license is almost a reality now. An Irish guy called Charlie McCreevy is the driving force behind that new single european content license...and it's going to be based, pretty much, on the French system. As a result, labels/publishers/content companies are realising that markets are not seperated by customs or borders anymore, they are seperated by language. Which is stating the bleeding obvious, I know, but, it is remarkable how many projects overlook it.

cheers

Dub

S.W.O.T.

dahacouk's picture

Yes, you're right about S.W.O.T. - funders will want to see analysis rather than gut feelings.

Cheers Daniel

Hello everybody, I applaud

chrisroditis's picture

Hello everybody, I applaud the Kendra Hub effort but I am interested to hear about how are you planning to address the issue of copyright? You mention that:

Kendra Hub is, in part, providing content owners with a shop front using Ubercart but I would reckon that they'll still want to put their music up on iTunes, Napster, Songbird, etc because that's where the consumers hang out. With Kendra Hub we want to make it easy for them to do both, experiment and see what works best for them.

What kind of copyright will you allow to the shop? Will it consist of mostly label music or unsigned?

Dub, regarding the single european license, I was aware of commissioner McCreevy's effort to eliminate European Collecting Societies' monopolies but I am not aware of the outcome. Any references to what happened, or what is about to happen would be highly appreciated.

Christopher Skauss

Hi Christopher, Do you mean

dahacouk's picture

Hi Christopher,

Do you mean what kind of license will Kendra Hub enable content owners to use? I think this is totally up to the content owner to decide and not for us to dictate or restrict. Let's also be clear that Kendra Hub is a "Tool" and not a service. It's a set of Drupal modules that content owners will have to install on a server somewhere on the Net. It'll be their own server. They'll decide. Thanks for asking. I'm going to make this real clear in the first paragraph.

The Kendra Hub trial will have very limited participant numbers and we'll make sure that everyone on that trial owns their own content. I would hope that during the course of this project that we could get other artists and record labels to install and run betas on their own server infrastructure. They can put whatever content they want up on their own servers. If the content that they host is questionable in any way then that is for their own conscience/karma/lawyer to deal with.

Does that answer your question? If not then please clarify...

Cheers Daniel

quick one..

Dublin Drupaller's picture

re: Kendra hub

Just to add to what Daniel mentioned..another artist empowering initiative well worth checking out (and joining if you're an artist/label) is the Merlin Music Network - a nonprofit organisation that grew out of the Association of Independent Music.

Part of the merlin music network brief is to aggregate music content, on behalf of members, up onto the big music services. They have so many members now they can negotiate a decent wedge from sites like the excellent Spotify site and many others, which is great for artists, small labels and consumers. Artists/Labels who would normally upload their content for free to websites like that, just to get a bit of exposure, now get paid a fair royalty when their music is used.

re: pan european licensing

It is moving forward. Slowly. The main stumbling block appears to be collection agencies who are trying to fight the idea i.e. the companies in each country that collect performance royalties and live off the comission. They don't want artists to be able to deal with 1 collection agency for all of the EU, because that means their comission will obviously go to the artist instead of themselves.

At http://midem.com a few weeks ago it was mentioned that Nokia had got one of the first pan european licensing deals with all the majors for their Comes With Music idea.

It's difficult to understate the significance of that deal in music industry terms. If it's solid..it opens up a huge amount of possibilities..not just for music, but for video and other content as well.

thanks for the answers

chrisroditis's picture

thanks for the answers

Media Sprint 2009

aaron's picture

You should take a look at some of the work coming out of the Media Sprint 2009, particularly the PHP Stream Wrappers. I think that could go a long way towards some of your distribution ideas.

Aaron Winborn
Drupal Multimedia (book)
AaronWinborn.com (blog)
Advomatic (work)

Aaron Winborn
Drupal Multimedia (my book, available now!)
AaronWinborn.com
Advomatic

Media sprint and third party solutions

gusaus's picture

There's definitely a huge need for tools that make it easy for producers to distribute and monetize their content. Even if the tools exist, it's a challenge to get them in the hands of the artists. Whatever the solution ends up being, it has to be drop dead easy to use, portable (i.e. it's something that can be integrated into any website), and more importantly, the artist needs to know that it exists.

I'm still trying to grasp this proposal, so please bear with my silly questions. First off, is this going to be a site where the artist has to create an account (ala myspace, virb, etc.) or can they integrate into their own website? How would the Kendra Hub be different from third party solutions like Topspin or iLike (related thread here)? And to Aaron's point, there's been some amazing innovation lately in Drupal media which could be very complimentary to this project.

Don't get me wrong - I'm for any solution that's going to make it easy for content creators to make a living. If anything, the sooner we can make it happen, the better.

Gus Austin
PepperAlley Productions

Gus Austin

totally agree..

Dublin Drupaller's picture

There's a good intro video across on the Kendra initiative website that might answer some of your questions.

Totally agree about the point of getting the technology in the hands of artists.

I notice that some labels, the majors in particular, are exploiting technology to encroach on the realm of an artist by building websites that appear to be official artist websites...but are really just shops dressed up as artist sites. In other words, the artist has nothing to do with the site, other than it's their music and image that is being used. More often than not, without their permission and in some cases, without their knowledge.

http://www.bobdylan.com is a good example. It's a very good implementation of Drupal and and led to a few "bob dylan is using Drupal" blog posts around the Drupal Planet, but, I would argue it would be more interesting in the hands of Dylan himself rather than Sony Music. One of the key ideologies behind Drupal is to remove the need for a webmaster (i.e. anyone can update their Drupal site), so there is little reason why the technology can't empower artists...as well as the business end. By the way, Sony Music aren't alone to exploit technology in this way...most of the majors employ the same tactic.

Many artists are familiar with updating their own facebook/myspace/youtube page...i.e. uploading images/audio/video etc. so I think it's only a matter of time before the penny drops that they can build their own destination site using Drupal, rather than promoting myspace/other.

Slightly off topic...but, there are a lot of initiatives in this direction out there...one I came across that I really like is People's music store - which allows anyone, particularly music bloggers to setup their own music shop to sit alongside their blog.

Agreed...

dahacouk's picture

Hi Gus,

Thanks for the "related thread". Lots of good information there...

Point taken on marketing and promoting Kendra Hub. I've elaborated in the proposal but it is part of our plan to demonstrate and promote at various conferences and through as many media industry associations as time permits. Kendra Initiative already has links with a number of these but I'm sure the Drupal community will be able to help too.

To the question, Kendra Hub is not a site or a service, it's a set of Drupal modules that content owners will integrate into their existing websites or use to form the basis of a new Drupal website. Topspin or iLike are hosted solutions run by companies. I don't see why Kendra Hub couldn't, with additional development, be used by a company to offer a hosted solution similar to Topspin or iLike. However, it's not the interest or motivation for this team to set up a business to host content, artists or their fans - nor, I should add, for Kendra Initiative. However, we'd love to see other companies offer out such a service using Kendra Hub as well as less technophobic artists administer their own Drupal sites directly.

And Dublin Drupaller is totally spot on with "artists are familiar with updating..." and it's about time there was a FOSS solution that could let them take back control. Looks like you're deep in the same groove as what we are trying to achieve with Kendra Hub. Hope we can work together somehow as we go forward.

Interestingly, I've just started putting up my artists page for Facebook, MySpace and, following your link today, iLike. I must admit this is starting to get painful as it seems iLike doesn't update the real Facebook page that I created but has another Facebook page under the app server. All this stuff is a mess! And keeps artists away from creating their art. We've got to sort it out.

Cheers Daniel

Many indie artists just use MySpace

gusaus's picture

Check a few links on this local club website and you'll see what artists are using. It's amazing how many indie artists don't even have a website because they're expensive, difficult to manage, and usually don't have any tools for selling, distributing, or often times just playing music. So if you provide something as easy to use as MySpace (which shouldn't be a tall order), the artists will have something they can manage.

Drupal, of course, can be molded into something artist friendly. This is a pretty good example of a site the artist actually updates and there is a good amount of aggregation and distribution to/from third party services and stores. Again, the ongoing innovation in Drupal media, is already producing effective distribution, syndication tools, and publishing.

All in all, there are many people working towards similar goals and probably a lot more who would get involved if they could (or knew what to do). A bit of KDI funding would be great, but there are other potential sources from Summer of Code to other sorts of collective funding. If you think of all the people that may value these tools (artists, labels, managers, consumers, developers, hosting providers, etc.), there are a lot of potential funders who might be able to help kick start some of this development. Drupal Dojo's Media Ninja class should be a great vehicle to collaborate, raise awareness and funds.

Gus Austin
PepperAlley Productions

Gus Austin

Our support..

psipi@drupal.org's picture

Hi,

This is exactly what we need.. ready to go solutions.. with the building blocks also available. Whether or not this gets accepted http://groups.drupal.org/node/19479

I'd still love to discuss offering your ready to go sites free for musicians starting out.

I think we could easily provide place where
- modules exist - available for one click install - working example http://www.screencast.com/t/dElHPEz6
- we could also "template" sites in various states of setup

That way people could get the blocks and built it themselves
or if they just want to plug and play they could..

Would love to give you access so we can move it forward,
i have a friend who's a poet.. and this would be perfect for him.

Industry interest is high!...

dahacouk's picture

Mike Large, chief operating officer at Peter Gabriel's Real World, says "Many thanks for this - I think your one paragraph summary of Kendra Hub is great."

Looks like we're hitting the nail on the head in terms of providing the music industry with what they really want.

Let's hope that Knight see it the same way...

Cheers Daniel

KDI status

agentrickard's picture

The KDI is currently on hiatus and is not passing along proposals for funding. We hope that Knight will re-launch the project within 2009.

If and when we re-open, I will be sure that we circle back to this proposal. But I would continue to pursue other sources of funding right now.

Apologies if this has not been clear.

--
http://ken.therickards.com/

Not workable / wrong approach

privacyPreferred's picture

"Within the music industry alone there are expensive software solutions that are out of reach to artists and bands going it alone. Then there are many, many good hosted distribution platforms (such as Music Glue, Open Music Source, The Orchard, IODA, AWAL, TuneCore...) but these intermediaries take some control out of the hands of the content creators. There are no free open source software solutions available that we know of."

This reveals a fundamental lack of understanding of how the ecosystem works. This simply isn't a software issue.

"Within the music industry alone there are expensive software solutions "
Actually many of them are extremely cheap. You need to research how these companies actually service the market.
This is especially true of the independent sector, which is entirely enabled by this new world order.

"intermediaries take some control out of the hands of the content creators"
The "lack of control" you refer to is actually the opposite - you need people to help you distribute, and give them control in order to achieve that. It's also simply not the case for many of the services who work to increase the value of the whole ecosystem, not just profiteer out of a tiny few.

"This project will lower the barriers to entry for artists wanting to distribute their media content."
It's hard to imagine how online distribution could be made any easier.
Ranging from free services like YouTube, to TuneCore which enables you to keep all your rights and keep all your royalties to other services that do more, and charge more - I'm not sure what the problem is that you're trying to solve. The are also already international standards for metadata, and in many cases this is only entered once for distribution.

The value-add of labels and distributors is a human one - they help you get noticed, provide seed funding, financial management and a host of other services. It's a common error to assume that those roles are redundant - actually they are merely transformed in the digital age.

Kendra has always been an easy-to-agree-with aim, "The mission is to foster an open distributed marketplace for digital media (including films, music, images, games and text)." - but this has been achieved in spades, it's the internet today and the diverse range of services already on it. The world has moved on...

One final point: the Kendra site lists everyone who's ever expressed an interest (http://www.kendra.org.uk/lists.php#participants-by-list), rather than those actively involved. I believe it is now completely misrepresenting both its need, its activity and its user base. This is one of the reasons I've preferred posting anonymously, it's a delicate issue, but I really believe Kendra's contribution is complete and its time has passed. FYI I have no personal interest in it succeeding or not.

All Drupal needs to do is facilitate the embedding of media UPC/UIDs and associated metadata, to enable traceability, discoverability and trade. Drupal already has most of the functionality to do the rest...

International metadata standards & DIY

allolex's picture

I'm curious about these international standards for metadata you mention. Are any of them open? Do any of them apply to retail delivery with territory pricing etc.? I ask because I've looked and not found anything open that is suitable for digital retail content delivery.

I should also point out that being able to pay someone to do something for you isn't really an argument against doing it yourself for cheaper. I'm surprised this isn't one of the stated goals of the Kendra Initiative!

standards

the full octave's picture

The main open standards are ID3, FreeDB, MusicBrainz, Dublin Core, Music Vocabulary (describes classic musical works and performances in an RDF-ontology), Music Ontology and MPEG-7 which is a multimedia content description standard (MPEG 7 provides standards for describing different aspects of multimedia at different levels of abstraction using an XML-based syntax).

AFAIK most online retailers base their metadata on musicbrainz, freedb or cddb (aka gracenote) standards and add on their own custom fields. So there's no single global standard yet....which was what was attempted by dublin core.

For applying metadata to retail delivery there are many SaaS (software as a solution) options out there already providing that service alongside digital warehousing. Peter Gabriels http://www.od2.com (bought by nokia recently), http://www.muze.com, or http://www.royaltyshare.com are two that spring to mind. just google "digital asset management"+music and you'll find loads. One of the more "open" ones is http://www.bandcamp.mu which is for artists/musicians and bands, more than labels.

Paul

Thanks, but not exactly what I need

allolex's picture

Thanks for your reply.

That's a very interesting list of standards, but as far as I can tell, none of them actually has an implementation that expresses all of the relationships needed to deliver content to a retailer. I guess that was part of the point of my post above.

The digital asset management service providers list you provide might be interesting and valuable to some people, but what I'm looking for is an existing and documented standard that can be used to store metadata that one intends to deliver to digital retailers. The point of asking for this is to help come up with a way avoid using third-party service providers and other middlemen. Dublin Core as RDF/XML could do this, but I haven't found a controlled vocabulary that fits the bill.

[I should probably mention for posterity that RoyaltyShare (above) and CI (Consolidated Independent) offer very similar services. But as far as I know, neither handles legal agreements with retailers--they just do delivery and management. Muze sells music metadata to interested parties like Amazon. Bandcamp is a platform for artists to sell their music digitally with nice bells and whistles, and is similar to some services offered by CDBaby and ReverbNation, as well as SoundCloud.]

Industry thinks Kendra Hub is the right way to go...

dahacouk's picture

Hi privacyPreferred,

You're right to some degree. If I may paraphrase you: This isn't simply a software issue. But software has a large part to play in giving more freedom to content owners.

Totally agree with "you need people to help you distribute". Kendra Hub is just about making it easier for content owners to interface with those distribution points and not having to manually repurpose their content and/or metadata to each and every distribution point/hub.

Totally understand that "services who work to increase the value of the whole ecosystem" are beneficial to the industry. And Kendra Hub is not about replacing those services. Not at all. It's about making it easier to interface with those services. And also it's important to note that we are not creating Kendra Hub as a service we are creating it a piece of software that can reside on a server or even a desktop or laptop. And there'll be plugins to as many content distribution points as have open APIs. So, it's all about streamlining the workflow for content owners. And the fact that the content owner will self host the software means that they really will be in control of the origination content and metadata and they'll be able to decide what goes where.

I don't think it's "hard to imagine how online distribution could be made any easier". Judging from the response we are getting from both Drupal developers/musicians and people from the media industry who all think that things can be improved from the content owners standpoint. Having "international standards for metadata" is great but making systems that make these transparent to the users is another matter.

Totally agree that the "value-add of labels and distributors is a human one". This tool is about assisting those people to manage their content so they can get on with doing the creative work - be that marketing or song writing/producing.

"Kendra's contribution" seems to be more relevant than ever. It seems like industry is at last waking up for the need to work together and produce interoperable systems. And now that Kendra is part funded by the EU it is in a stronger position than ever before.

Yes, the great thing about Drupal is that there are many efforts to produce the components that we need to create such a system as Kendra Hub. However, I reckon there's still a requirement to package it all up in some way that makes it easy and friendly to use and has a consistent interface and workflow. And don't forget we are also incorporating cutting edge semantic web technologies that we'll see come into their own in the not so distant future. Using RDF in the core of Kendra Hub will keep the project future proof in terms of new metadata standards. And, let's face it, they're never going to stop producing new metadata standards. Constantly improving and enhancing our world. Hence, the value of this effort - we've got to not only "keep up" but be the "way forward".

Cheers Daniel

Dublin Drupaller's picture

Was researching for a different, but, not unrelated project this morning and had an idea for the kendra hub team that might be worth suggesting.

An example would be MYCONTENT.COM which is launching in the USA this summer. No idea who is behind it. I just got an email from a pal in the US who heard about it

http://www.mycontent.com/AboutUs.aspx

The following bullet points are copied from the site:

About mycontent.com

  • It is a mass distribution platform for anyone to share their work with everyone.
  • It is an e-commerce infrastructure complete with a pre-established network of advertisers, a platform for easily selling premium content, and payment mechanisms in place so that artists can properly be remunerated from Day 1.
  • It's the ability to upload all content: videos, photos, and audio. Set your price point to sell it, or put it out there for free. The choice is yours, it's YOUR content.
  • Reach millions of potential customers instantly with our extensive social networking platform.

mycontent.com Experience

  • An instant distribution platform for bands, indie filmmakers, networks, labels, and individuals.
  • No longer having to pay thousands of dollars to build & manage your own site.
  • Setting up (upload, arrange, categorize, sell) and design your own web store/social networking page/mass distribution platform in minutes.
  • Communicating with your fans and marketing to them in a way never before thought imaginable.
  • Monetizing all media (video clips, songs, photos) that otherwise were collecting dust through both our extensive network of advertisers, and PPV events.
  • Encouraging new content creation, since e-commerce infrastructure is already in place.
  • Access millions of consumers overnight through an extensive social networking platform.

It's very ambitious, due for launch in the Summer and while the details on the site are a little sketchy (most of the info. is bullet points about what they want to do) it will be interesting to see how it pans out - it's not a million miles from this kendra hub idea - and I was going to suggest the following:

I think there will be a flurry of initiatives like this over the next year or two. The number of Social Networking sites online went past the 1 million mark recently and I think it's only a natural progression for "kendra hub" initiatives to follow....where, instead of people posting content to websites willy-nilly and the domain owners creaming all the ad revenue and profits, creatives will swing more to kendra hub type initiatives so they can get something in return.

The interesting thing about this is that I think creatives will register for a number of hubs...in the same way an artist/band will tend to register on myspace, bebo, imeem, facebook and last.fm to promote their own work.

In other words, would it be a smarter idea to re-jig the kendra-hub proposal to ping these new emerging hub sites with a view of Drupal integration?

As a simple example, whatabout pinging the guys behind mycontent.com (or other similar hub sites, like istockphoto.com etc.) and offering to create an integration module for Drupal?

Thet would provide immediate value to the Drupal community, would require much less funding and may open up an opportunity down the road to build your own kendra hub, based on the experience and knowledge of other succes stories.

You guys have obviously done a lot of work already in this area and it would be a shame to see the project stall and let that effort and energy slip. I know what I'm suggesting it's quite a shift away from the original idea for the kendra hub, but, I think the original idea was a little ambitious and requires a significant amount of funding to get up and running. Approaching it in the way suggested will require less funding...deliver immediate value back to the drupal community and can still act as a stepping stone towards the original goals.

Hope that makes sense...

dub

You're right but this is already in the plan...

dahacouk's picture

You ask "integrating with existing hubs a better idea?" Well, as far as I am concerned that is already well part of the core Kendra Hub idea. So, I totally agree. But it's not anything new for the Kendra Hub concept - as far as I can tell...

Remember, Kendra Hub is about creating a way for content owners to only have to enter information about their content once and once only. This content/information then gets syndicated out to distribution points such as your "existing hubs". The great thing is that the content owners are in control of the content/information because they host the application. That's really important. And the content owners are not having to sign up to any "terms" to use a service hosted by someone else.

What's different about Kendra Hub from your example is that Kendra Hub is just a piece of open source server software written in Drupal. Whereas your example is a service hosted and owned by someone else with strings.

What you are suggesting is not "quite a shift away from the original idea for the kendra hub". It's totally in line with the concept. But in order to integrate with existing hubs we need to have something to integrate with: a content and metadata store, input/output filters/transformers, etc. And I'm not sure if it does "require less funding". Because we still have to do the same thing to get a fully working system.

And I have a strong hunch that if we are going to integrate Drupal into all these "existing hubs" then we are going see real benefit in using semantic technologies (RDF) to transform the metadata to fit all these different hubs. And also enabling the content owners to put their content into Drupal I think it would be a mistake to shoehorn their data into a particular schema so we need to have metadata mapping which I think RDF does really well. So, to produce a powerful, useful and flexible tool will take some work here. Yeah? Open to ideas though!

Sure, creating integration modules for Drupal in a piecemeal way may would work too. But there's still a need to fund those efforts. I think that potential funders would want to see an effort where the fruits yield a substantial working system that hits the ground running.

Listen, we're both right... I think the best we can do is keep on banging on about the general concept of using Drupal as the platform for content owners to syndicate their content to all these other hubs. Keep developing the specifying what functionality we need. And really clearly marking out how the APIs will operate in a clean way. And one day a magic fairy will come along and fund it wholesale or piecemeal. Yeah?

I'll sort a BoF this at Paris DC. I do hope you are coming along!

Freedom to content owners! ;-)

Cheers Daniel

dahacouk's picture

I've just set up the Kendra Hub presentation at Drupalcon Paris. If you are attending the conference please show your support and slip it a vote:

http://paris2009.drupalcon.org/session/using-drupal-media-asset-content-...

It's rather late in the day and near the deadline so it needs all the help it can get.

Many thanks!

Cheers Daniel

Media BoF at Paris. Please vote now...

dahacouk's picture

Hi!

I’ve submitted a request to hold a BoF called Drupal for Media.

So, please vote for it as soon as you can if you’re going to DrupalCon Paris. Time is very short!

Many thanks! ;-)

Cheers Daniel

Coming to Kendra

allolex's picture

I help run a very old, very independent label called Southern Records in London. Some of you might know us through our work with bands like Crass, Babes in Toyland, Karate, and William Elliott Whitmore (and hundreds of others) as well as through our partnerships with Subhumans, Nomeansno, and Dischord Records, home of Minor Threat and Fugazi. We've been developing on Unix, Linux and using open source software for a very long time. We've always been a bit heavy on technology because it's allowed us to do things for ourselves that previously only major labels and Very Big "Independents" were able to do.

I've noticed that open source software is getting better and better and much easier to use than ever before, especially in the area of Content Management Systems, which have become complete Content Management Frameworks, suitable for data that goes far beyond displaying stuff on the web. The Kendra Hub proposal looks like step in the right direction. Kendra also looks like exactly the sort of thing that we could use and contribute meaningfully to since we've got a lot of expertise in dealing with various metadata standards, and of course we have a pretty good idea about the kinds of things artists and labels need to get by in the music business.

At Southern, we actually have our own complete digital distribution system, written in Perl--or "congealed" in Perl, as the original developer Dave Coles would put it. The system rips, encodes the audio (as FLAC), stores metadata (as XML), and exports the metadata to various formats used by various retailers and aggregators. The system isn't particularly usable by non-experts though, as it relies on the command line and is dependent on human intervention to do the actual content delivery bit. Still, I can't help but feeling that the Southern Digital Distribution system could be helpful to a lot of other labels.

Kendra appears to be interested in developing a solution for content delivery using Drupal. That's very interesting for us, despite Drupal beinig written in a vile language like PHP. ;)

When I get some time, my plan is to clean up this code and make it available under the GPL. I'm starting to think that, for the sake of getting the code out there before we all go bankrupt, my standards for "clean" might need to be lowered.

One of the problems about being a label and distributing digital content directly to retailers is that most of them cannot be bothered importing one's internal metadata standard. Some will go so far as to accept the data on spreadsheets, but that's no way for two IT systems to be communicating. That said, I also understand why the retailers might not be interested--very few labels have enough content to deliver to them to make this worth their while. However, if we had an open music catalogue metadata standard that actually got used by enough labels and artists, there would be enough of an incentive for the retailers to import this metadata directly. This is what they do for the big digital distributors ("aggregators") like IODA.

It looks like the Kendra Initiative wants to address this problem as well.

We, like a lot of independent labels, don't make much money from individual sales, so each sale is precious to us and to our artist partners. There are a lot of sites out there making a lot of advertising revenue, and advertisers getting lots of clicks, off the fruits of our labour. That just isn't fair, so I've also written some software to help automate the reporting of copyright infringement. This is another aspect of the music business that artists and labels might be interested in developing further.

What I would like to see is an inventory of this project. Have you created any specifications? What modules will the Hub modules require? Do you have any code? I'd like to see a github repository (or similar) for all project files. We've got code we would like to share, albeit in another language, and we would like to see what others are prepared to share.

Let's talk...

dahacouk's picture

Hi Damon,

Thanks for the all the information on your internal digital distribution system. Yes, even if the code can't be directly used in Drupal then the metadata transformations and interfaces to retailers and social networks would be invaluable.

We don't have any code yet but, yes, further work could be done on planning.

Let's talk...

Cheers Daniel

It can certainly help to some extent

allolex's picture

The only problem is that the implementation of the retailer metadata is all protected by non-disclosure agreements. I really think that pushing an open metadata standard to retailers is a way around this, but we might be able to get retailers to allow us to use their specifications. iTunes XML can be reverse engineered from Producer output, but I think it's the only one widely available. Maybe I can ask around.

Barring those possibilities and keeping the status quo, I feel the missing element to make Southern Records' digital distribution system useful is software that allows you to take a sample XML file provided by a retailer and transform it into a template.

A large group of content producers grouped behind something like the Kendra Initiative could deal with the big retailers in terms of accepting metadata using an open standard. The retailers are just interested in saving money and with a free and open metadata standard, everyone wins. I've looked at a few free metadata standards, but so far have not found one with the necessary support for territorial availability and pricing, and similar business rules.

Catalogue import

allolex's picture

Milestone 1

* Artist (or label) site with catalogue import (media assets). It makes sense to have this ready first so that artists and labels can start working with the Kendra Hub trial and we can add/configure extra functionality as it becomes available.

Has someone started work on this?

At first glance, I see a couple of different implementations. A quick and effective solution might be to build something that parses either the iTunes XML format or sales reports. The catalogue reports only show things at the release level, and it's obvious that a track-level database will be needed. The problem with using sales reports is that they only report what you've sold, not your entire catalogue.

I guess the question is actually "in which formats are people most likely to already have their data"?

discography/catalogue module

the full octave's picture

dublin drupaller worked on a Discography module using standard fields compatible with international standards that also allows labels/admins to add in extra custom fields. The code and instructions to implement the basic catalogue is there and can work on a track level. "Import catalogue" functionality isn't included. Work on that led to the suggestion of using the same modules utilised for the discography module for other label/band related functionality i.e. Music distribution for Drupal or seperate modules?

To answer your question: which formats? most follow musicbrainz or freedb for the fields and excel/access/csv/tab delimited are probably the most popular formats.

At least, that's what I've come across.

Paul

Thanks. More thoughts on importing stuff

allolex's picture

Thanks for the reply. I've run across those threads you linked to, but I didn't find any code--maybe I overlooked something?

The reason I was talking about iTunes XML is because almost everyone is on iTunes. Unfortunately, not everyone has access to their XML because they've done delivery via an intermediary of some sort.

I suppose most artists and labels have delivered their content via aggregators, who tend to work with spreadsheets. It would be best to have an importer be able to parse the most common formats in which artists and labels are most likely to already have their data. Most of the labels we have done digital distribution for either provided incomplete data in spreadsheet form or sent us spreadsheet forms that were from one of the big aggregators.

If we could support the most common of these spreadsheet formats, then that would be a big step. Or we could write something that allows the user to do field matching (or mapping) during the import process--which would be the most flexible approach of all, but subject to interpretation and therefore errors.

And now my thoughts turn towards internal representation. I really think we need to have an open metadata standard. I think we need one.

metadata

the full octave's picture

I really think we need to have an open metadata standard. I think we need one.

We already have an open metadata standard - it's called the dublin core. I mentioned that earlier.

Gracenote.com and musicbrainz.org, probably the most well known online music catalogues, base their schemas on dublin core. Although, just because one standard schema exists, doesn't mean everyone will adopt it and use it. The EU are funding a DISMARC portal which disseminates and maps different schemas from various content aggregators, retailers, libraries, record labels etc...incidentally Dismarc is based on dublin core as well.

To be frank, I think you're muddling up schema standards with tech/code and forgetting that the schema is actually a very minor part of syndicating content to retail. The legals (licensing) are the most important and time consuming part. Hence the convenience of a label/band signing one licensing agreement with the orchard, cdbaby, ioda or whoever instead of signing between 10 and 100 licensing agreements to cover all retailers and uploading content 20-100 times to each retailer or music service. Not to mention the convenience of receiving 1 cheque/payment every quarter instead of between 20 and 100.

In theory, calling for an open metadata standard sounds like a great idea. In practice, I don't think it is a massive turn on for labels/retailers. One can only imagine how much money it would cost amazon or itunes to rejig their schemas to comply. Besides, apple are the opposite of an "open" company. They're like a microsoft wanabee, when it comes to "open-ness" and they still refuse labels having the choice to remove the apple-ipod-only-drm from their itunes tracks. Apple also still refuse to license that drm so it will work on other players.

In my opinion, calling for every retailer to use the same standard isn't going to happen any time soon - that would be a bit like asking every payment provider to use the same order submission and payment processing schema as a standard. That aint going to happen. Hence the reason there is a paypal.module, protx.module, worldpay.module for drupal ecommerce and ubercart.

If you follow dublin drupallers approach to the music distribution for drupal - it's very simple and modular based - e.g. for content syndication to other sites you would use a facebook.module, myspace.module, bebo.module that would map drupal (cck) fields to myspace/bebo/etc. schemas. The same simple principles could be applied to mapping retailers schemas as you pointed out, but, like I said earlier, I don't think that's a big requirement for labels.

In theory, calling for an

Garrett Albright's picture

In theory, calling for an open metadata standard sounds like a great idea. In practice, I don't think it is a massive turn on for labels/retailers. One can only imagine how much money it would cost amazon or itunes to rejig their schemas to comply.

It would be a drop in the bucket to these companies, were there a compelling reason to do so.

Besides, apple are the opposite of an "open" company. They're like a microsoft wanabee, when it comes to "open-ness" and they still refuse labels having the choice to remove the apple-ipod-only-drm from their itunes tracks. Apple also still refuse to license that drm so it will work on other players.

Have you visited the iTMS in the last year or so? DRMd tracks are the exception, not the rule, nowadays. It's the labels which were pushing the DRM on Apple, not the other way around; now that the big ones have come around to the idea of selling music without DRM, almost all of the music in the store is DRM-free. I myself listen to most of my iTMS-purchased collection on my SanDisk Sansa player.

apple drm

the full octave's picture

It's a few years to late, Garrett. Besides, it was label pressure that forced apple to drop the apple ipod only drm. it is true that 7 years ago when itunes started, labels wanted something to protect the music, but, they soon realised that the apple drm only protected apple products. it didn't protect the music. It wasn't until they saw the huge success of amazon mp3 and 7digital did apple back down to label pressure.

Of course, apple didn't miss a trick when they backed down to label pressure. Music fans who overpaid for the price of tracks crippled with apple ipod DRM in the first place had to pay once again to have those restrictions removed!

open standard metadata schema for retailers

the full octave's picture

You must be joking about amazon changing their accounting and stock control schema's not costing much. That's a massive job and would require taking the site offline, which would incur a massive loss for every minute it's offline.

Like I said earlier, I can't see a compelling reason for retailers to go through with that expense. Metadata schemas isn't a big problem. Licensing is.

It would be much simpler and logical to develop an amazon.module in the same way you would create a google_checkout.module so your drupal site can "talk" to google checkout.

…What? They don't take the

Garrett Albright's picture

…What?

They don't take the whole site down while developing this. They develop and test it on a dev server. Then, when it's ready to go live, they move the updates to the live servers. It will take a few minutes at most, and surely someone like Amazon already has a mechanism to do something like this. If they do a staged rollout, only taking down and updating one live server at a time while traffic is diverted to the servers which are still live, there may end up being no noticeable downtime at all.

As for the DRM argument, clearly you've already decided that Apple is the bad guy. But I find it odd that you also seem to have decided that the labels are the good guys; that's a new one.

code vs schema / administrative overhead of retail deals

allolex's picture

Thanks for the interesting reply.

To be frank, I think you're muddling up schema standards with tech/code and forgetting that the schema is actually a very minor part of syndicating content to retail. The legals (licensing) are the most important and time consuming part.

I think your perception that I'm conflating code and target schema comes from the way I express these concepts. Since there is so much munging that has to take place to shoe-horn data from one schema into another, I tend to see them as part of the same process.

The legal aspects are a little time-consuming, but not as time-consuming as writing export routines based on a retailer's specification. There is a lot of administrative overhead involved at first, though. In the broadest terms, I think you're overstating just a little bit the difficulty of making a retail deal. Maybe you have some personal experience that has led you to this conclusion?

There's no disputing the advantages of having your content imported, warehoused and delivered by an aggregator, unless of course you've got some system in place that handles that for you. That's one of the things I'd like to help develop.

In theory, calling for an open metadata standard sounds like a great idea. In practice, I don't think it is a massive turn on for labels/retailers. One can only imagine how much money it would cost amazon or itunes to rejig their schemas to comply.

They don't have to change their own schemes to fit with any new standard--they just have to write something that can import it into their existing system. This is what they do for IODA's metadata, for example. Retailers will tend to do things that bring in the most revenue, and helping to cut out middlemen would certainly give them more space for negotiation on prices. If an aggregator takes 15%, the content supplier and the retailer could split that money. Perhaps that's a bit simply stated, but the principal holds.

swings and roundabouts

the full octave's picture

They (retailers) don't have to change their own schemes to fit with any new standard--they just have to write something that can import it into their existing system. This is what they do for IODA's metadata, for example. Retailers will tend to do things that bring in the most revenue, and helping to cut out middlemen would certainly give them more space for negotiation on prices.

You're right about the retailers not having to change their own schemes and writing an importer instead which sorta backs up my point that it's not a big problem (for retailers or labels).

On the non-retailer side, isn't that exactly what the EU funded DISMARC project is all about? i.e. gathering metadata from all the various schemes out there and flowing it into one...based on dublin Core. They are going beyond record labels, publishers & retailers and including libraries and other sources of content.

Either way, I would pick up on the point about cutting out the middlemen. Most of the big music services..such as apple itms, spotify, amazon mp3 or nokia comes with music all prefer to deal through aggregators for the simple reason that instead of signing individual legal agreements with thousands of record labels, they can sign just one with an aggregator.

From a label point of view, aggregators don't get 15% for nothing. They provide a useful service (accounting, tracking, identifying new music services, promoting and pulling content if needs be) that allows a label to get on with making and marking music.

In a nutshell, I'm not convinced that a metadata standard is a priority for this project. In fact, I would call it a hindrance and would tend to just stick with dublin core standards while focussing attention on other, more pressing requirements.

Which standard

allolex's picture

Thanks once again for an interesting reply.

I have to dispute your assertion that Apple, Amazon and Spotify prefer to deal with aggregators. We work directly with them and I have been told that they would prefer to direct deals as long as their technical requirements are met, despite administrative overhead (boilerplate agreements are cheap). The big aggregators apparently have too much bargaining power on both sides of the game, and retailers aren't overly fond of them, either.

I know a little about this business because we also do digital distribution. I know what we do for our fee or percentage and some labels and artists prefer to not have to deal with these things. Others like to have complete control over all aspects of their business and would be working with retailers directly if they could. Margins in the music business are becoming tighter and tighter, particularly for independents. 15% is close to what many physical distributors in the US charge for their services--it's a substantial amount for the services they are able provide on such a large scale. (Digital scales quite nicely, particularly where turnover is concerned.) Suffice it to say that artists and labels are looking for ways to trim the fat and distribution of any kind is going to be looked at much more closely in the months and years to come.

Also, some people just want to do it themselves, which I understand very well. We want to do it ourselves. I'm also interested in broadening the field of people who can do it for themselves.

In a nutshell, I'm not convinced that a metadata standard is a priority for this project. In fact, I would call it a hindrance and would tend to just stick with dublin core standards while focussing attention on other, more pressing requirements.

I'm still confused by your assertion that we should use a standard based on Dublin Core when that standard doesn't exist. Dublin Core is a way of dealing with certain data types and looks like it could be extended for the purpose of digital asset management and distribution, but my research shows that there is no current standard for this. I've asked if anyone knows of one that I've missed, but ones you've cited don't fit the requirements for digital distribution. Since the data stored in any digital asset management/content delivery system should be standardised, this seems quite conclusively to be a prerequisite for doing any other work on such a system.

If there's really a standard that fits these requirements, then please point me to it. Dublin Core as it stands does not by itself have the requisite scope. It requires a vocabulary, and a controlled vocabulary needs to be standardised. Standard attribute sets would also be good.

Maybe I should ask my question differently. How do you store your catalogue metadata? Do you have (part of) a schema you can share?

It might be interesting to hear what you consider to be the most pressing requirements at this point in time. Maybe there's something we could work on together since I'm interested in developing a complete system beyond the scope of digital asset management and content delivery.

Need to define context of standard...

dahacouk's picture

I met Damon (allolex) a couple of days ago in London. We talked about this standards issue. Perhaps we need to define the context of this standards usage. I think Kendra Hub developers have been more thinking of it as a metadata field index rather than an external standard. Some more issues/questions...

  • I tried to explain that we were trying to move away from the monolithic only one standard approach. Damon explained that record labels aren't going to want to map their catalogue field set to each and every syndication target again and again. Fair enough. They'll map their catalogue field set a maximum of once and then we can provide mappings to syndication targets from there on in. But to what are they mapping to in the first place? So, do we need to have some kind of internal Kendra Hub standard?

  • I kind of had it in mind that all of this would come out of usage - if we coded Kendra Hub that way from the start. So that metadata maps would be built and shared across the network of the Kendra Hub install base. But perhaps that comes later. If we make sure that each and every install of Kendra Hub has its own namespace reference then we should be fine. To assist with record labels making initial mappings we'd have a wizard. So, that each field would have a question "when you say A: Performer do you mean B: Artist?". Where B is a field in our internal index/standard and A is the field name from the record labels original catalogue database. But we could (or would even need to) seed Kendra Hub with an standard index to start with. It needs to be flexible and modifiable. However, if a Kendra Hub user changed their "Artist" record link to be anything other than kendra.org.uk/kendra-hub/rdf/artist then all the mappings we provided would be broken, right?

  • Which brings me to another question. When a user installs and starts up Kendra Hub they are going to have to import or synchronise to their existing media catalogue. There will be many instances I'm sure where users will have existing DAMs that they are quite happy with and just want to use some of the functionality of Kendra Hub. In either case we will need to represent their media catalogue in its original form and with original field headings. Damon was adamant that Drupal couldn't programatically create (CCK) field sets on the fly at the time of importing the catalogue. I was pretty sure that we could but not being a developer I bowed out. Can this be done or not?

Cheers Daniel

The way forward for Kendra Hub...

dahacouk's picture

Kendra Hub should talk many languages/protocols. So, the issue of what metadata standard to use is not so important - just as long as we can talk all of them. If a new one pops up then we just build another metadata translator/transformer and hocus pocus we're in business. That's the direction we are taking unless someone has a good reason to take another route. To be honest, I can't think of another possible direction we could take anyway.

So, the best thing we can do is list as many metadata standards, social networks, aggregators and retailers as we can and in order of priority to build connectors. Then we'll have a kind of to do list and a way forward. After all, what we should be building here is a project plan. Cool?

By the way a few days ago the powers that be accepted http://groups.drupal.org/kendra-initiative so please go and sign up and we'll figure out how to build documents. I do prefer documents rather than long forum discussions. In documents one can see what's going on at a glance rather than having to read a whole history.

And we'll get some regular skype/phone calls to discuss those documents because, hey, there's nothing quite like hearing the sound of someone's voice to figure out where they are coming from. And basically, I think we'd all get a lot more done and progress a lot more if we spoke to each other...

Cheers Daniel

Drupal Crash Course

Concise Code's picture

If the project is going to happen, I would like to participate in it. It could be a crash course in Drupal for me.

Vote for Kendra Hub at Knight News Challenge!

dahacouk's picture

Please vote for Kendra Hub at Knight News Challenge. Please provide constructive criticism. Please pass this on to your networks.

Cheers Daniel

Please give me feedback on Kendra Hub...

dahacouk's picture

I've just resubmitted Kendra Hub - http://cli.gs/6qsJN9 - to the competition. It would be great to get some feedback from the Drupal community with any refinements you think I could make.

Thanks!

Cheers Daniel

dahacouk's picture

I see that Tom Erickson (CEO of Acquia no less) talked about "the ability for Drupal Commons to be added to another Drupal installation in a fairly straight forward manner, using basic Drupal constructs.". Well, I think Kendra Hub should be both an install profile and retrofit configuration/modules.

So, how do we offer both and still have a streamlined development process? Anyone have any info/pointers on this?

Cheers Daniel