Hi Phil (and the rest of the list)

Thanks very much for all the comments and questions. 

Your first question - why are we starting from scratch?  I think the answer is simply that it gives us more flexibility to get it right.  However, whatever we develop will be open source, so starting with an existing open source project and building on it or cloning it etc would not be an issue and the first stage to our development would be to do some serious investigation into what was already available.  We work in Python/Django so the number of open source projects available to us isn't as great as if we were developing in Java or PHP (but the number is growing, and I was very interested to head Greg mention his OERC bookmark - I'll be looking at that in a lot more detail).  There isn't yet a demo version of the work we did at the Hackday online.  I'll see what I can do.

We would be very happy to work with the Learning Registry with regard to Paradata.  We see paradata as the big plus point of a centralised system like this because the user base is a ready made community - we know that a bookmark that is recorded 100 times in the one system has a certain value, but if it's recorded once in 100 separate systems how would we ever evaluate that?  I'm not sure if we'd be able to say "yes we'll get it working with the Learning Registry" during this project, BUT we would want input from the community (ie this mailing list) AND we're going to publish an API which will allow you to get paradata out as well as metadata, so data from this system could definitely be used by the Learning Registry.

I think that us calling it a clone of Delicious is a bit of a red herring.  We've specifically mentioned it in the bid because people are familiar with it as a service, and there are definite similarities.  But really it's something that fits into a space away from Delicious and most bookmarking services - it's not designed to just record the basic 'title', 'url', 'description', 'tags' etc data that is captured in a system like that - the added value comes from capturing licence information and allowing communities to generate additional data.

One comment was that bookmark services may be redundant now given the rise of twitter/url shorteners etc.  However, I would disagree with this.  Twitter is immediate - a bookmark service is about storing the data so that you can access it retrospectively, and that's where the idea of using this service to generate 'playlists' would come into it's own.

Another comment was what would happen if the service was discontinued.  We said in the bid that we'd keep a version of it online for three years.  However, if the system is REALLY successful then I couldn't see the University being happy funding it for the agreed three years if we need to put the sort of infrastructure in place that a service with hundreds of thousands of users would need.  On the flip side, if it does have that number of users then I'd be amazed if the community didn't take on responsibility for hosting etc.  We're going to make the project open source so it would be up to JISC to decide whether to set it up as a service or not at the end of the project - I personally like the idea of passing on the code to JISC once the project is complete and leaving it up to them to host it, but that would be up for negotiation.

I like the idea of it being a front end to another bookmarking service (using their API), capturing our own metadata.  We'd have to be pretty careful that the one we go for has a good enough API and that it is a service that is going to be around for a long time.  But we'd be interested in looking at it as an option (at least I would be - I shouldn't speak for the rest of the team!)

It's getting late, so if I've missed anything please accept my apologies.

Thanks

Paul
________________________________________
From: Phil Barker [[log in to unmask]]
Sent: 13 April 2011 17:26
To: Open Educational Resources
Cc: Paul Horner
Subject: Re: Mini Project

Hi Paul, everyone,
I like the aspects of this proposal that go beyond bookmarking as seen
in Delicious (though as we discussed at the hackdays bookmarking
services that don't rely on Yahoo finding the right buyer for Delicious
have their attractions). Anyway, a couple of questions:

1. your starting point: is there a reason for starting from scratch?
Correct me if I am wrong, but I recall from the hack days that you
didn't base your bookmarker on any existing system or portal software.
Even discounting the ideas of using blogging platforms for bookmarking,
there seem to be several existing open source bookmarking software
projects that might make a suitable starting point for what you
propose.  Anyway, is there and instance of the demonstrator available
that people can look at? (on the understanding that it is a rough demo
resulting from a day or so's work)

2. you mentioned paradata, can you see any way of working with the
Learning Registry in order to feed this paradata into their system?

All the best, Phil.



On 08/04/2011 16:05, Paul Horner wrote:
> Dear All
>
> Given that the deadline for mini project proposals is now less than an hour away, please find attached a hastily assembled bid on behalf of Newcastle University, for 'Project 3' - the Open Call.
>
> The document is a bit formal, so as a summary the idea we're putting forward is for an OER Bookmarking tool.  For those of you who were at the Hackdays last week, this idea will sound pretty familiar because what we're proposing is extending the work we demoed at the end of that workshop.  However, this time we'll be providing a fully featured system (well, a fully featured demonstrator and all the source code) rather than the proof of concept we showed last Friday.
>
> We've tried to think from the start about how other systems will be able to utilise/harvest/add to the data that this system will capture, including the social/user-generated metadata/paradata, and we're going to pilot this side of things with an existing system in place at Newcastle (Dynamic Learning Maps).  We've thought a lot about the user experience - making it as easy as possible to get data in and out and providing ways for users to enhance the data (commenting, rating etc).  We're even proposing to build in the 'playlists' idea that was suggested to this group last week (which I think is a fantastic idea in its own right).
>
> One thing I should probably stress is that this isn't just a clone of delicious.  OER (and licensing in particular) is central to this system.  It is an attempt to make it easier for license information about a resource to be discovered and distributed and we hope to do it in such a way that doesn't unduly affect the user experience.
>
> So, over to you.  Please let us know whether you think it's a worthwhile (and achievable) project.  In our opinion we're proposing something that would be extremely useful and would prove to be very good value for money.
>
> Thanks
>
> Paul
>


--

Please note new email address: [log in to unmask]



--
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.