wiki_tomos wrote:

Hi. I find it rather odd that nobody so far seems to have responded to this interesting proposal. So I am making a modest attempt. I have to admit that I just ask questions & clarifications mostly, and I am not sure if others have the same questions.

----- Original Message ----- From: Gerard Meijssen Sent: Sat, 04 Jun 2005 08:10:53 +0200 Subject: Re: [Foundation-l] IEEE LOM

...

...

However if 50% of the Dutch data is of an universal nature, it would mean that this 50% does not need to be entered for the articles in other languages. Hosting this metadata on the Wikimedia servers makes sense;

I think I want to learn more about this part. What do you mean by the universal data?

The Dutch implementation of IEEE LOM has Dutch localisation for the international labels. When 50% of the data is the same on every article independent of the localisation used, and when an article on the same topic in different languages shares this metadata, you do not have to enter those again. To make IEEE LOM available, you need to implement the local standard next to the published standard.

Some of the IEEE data are common to all Wikipedia (or Wikimedia) articles/pages so that they do not have to be manually entered for individual articles/pages separately, but could simply be implemented at a different level?

...

...

it allows for the opening up of Free content in a proprietary world. It would make a huge deduction in cost for every second language implementing the IEEE LOM data.

I want to know more about these statements, too.

What are some of the expected benefits of tagging the contents with IEEE LOM? Are you suggesting many commercial content providers will take Wikipedia contents and offer them as part of their contents? That sounds good to me (reuse of contents is good in principle in my opinion).

Yes, that is what I suggest.

I also wonder about how they will comply with GFDL that requires, among other things, that contents should not be technically protected for people to read or copy, and that no other terms than license will be applied. Well, that's their (those commercial content providers) business to figuire things out, so perhaps the question is not very relevant to our discussion.

Making content accessible is what we do. Making people comply with our license is a different matter. I learned that schools themselves often ask organisations to rate info according to IEEE LOM. But once info HAS been categorized according to this standard, our information will be found for use. The issues re licensecompliance are not different for us as it is for the commercial content providers, if anything the commercial boys have more to lose.

...

The questions I put to you are:

• Are we willing to host open standard meta data for the educational world.
• Are we willing to cooperate with organisations that are interested in implementing this data.
• How will we manage such things; funds can be found to pay people doing this kind of work - can we consider this

Thanks, GerardM

Am I correct in assuming that what they want is that somehow we (Dutch Wikipedians, or Wikipedians in general) will tag our contents in the way specified in IEEE LOM? And they will pay for the server and traffic in return? Is that the deal they are proposing?

No, they want to host our content and tag the info. Us hosting the content, will allow for sharing the "universal" IEEE LOM labels and this in turn will lead to less effort in tagging our content for other countries following the paths of interwiki links. This is what I propose what we should do. Finding 100% funding is something that is relatively easy to do certainly when the management of this IEEE LOM stuff is what the rationale is.

Immediate question that I have, in that case, is that: *Do they want to somehow host the Dutch Wikipedia only, on a separate server?

*Did they consider creating a mirror themselves, and do all the tagging by themselves? In that case, for good or not, they perhaps need only a few experienced Wikipedians as advisors/consultants. Not particularly an exciting choice, but could be quite sensible.

This will prevent the economies of scale. So it is cool when they do this, it makes however more sense when this metadata is integrated within our own systems.

Thanks, GerardM

Cheers,

Tomos

Erik Moeller wrote:

Gerard,

first of all, pursuing cooperations like this will always be very slow as long as it's a completely volunteer-driven effort. I think Wikimedia absolutely needs several people who can work on such matters full-time. Of course, you could ask the organization you have spoken to if they would be willing to fund the preparation itself (technical evaluation, research, planning, negotiation).

However, I also think we will have a much better negotiation position once Wikidata is past the prototype stage. Getting international metadata for Wikipedia for free is indeed a fascinating prospect, but I don't see it as an issue we need to address *right now*.

Best,

Erik

Erik, This is the moment when a Dutch organisation asked to have our content in order to add metadata. Therefore this is the time we have to formulate an answer. When we do not, the opportunity to maximise the return of this effort will go past us. You are 100% correct that this will need Wikidata and we do agree that having more people working on Wikidata and Ultimate Wiktionary is a good thing.

This organisation needs to set up its infrastructure to host the Wikipedia content and add the IEEE LOM data, it will not benefit from similar efforts when it goes it alone. It is therefore in its own intrest to cooperate with us to have us host this data. The timeline for this project is not only for us to decide as we are not driving this effort so I disagree, we need to address this now and work out what the optimum timeline is.

Thanks, GerardM