GALILEO Masters 2004/Questionaire

I just submitted our proposal to the Galileo Masters 2004. Thanks to all! Stefan Richter 17:22, 30 Jul 2004 (UTC)

There is a screenshot of the original questionaire at the talk page.

Idea-Proposal (English)

We have to submit a questionnaire and up to two documents. (We do not have to submit documents. We can, if we want/need to).

This is copied from the original forms of the Galileo-Masters Contest.

I wrote a short and a long description. The main story line is: "The core problem of location-based services is the (missing) large information base. We have this information base already in place and it is easy to integrate this with galileo/location-based applications". Always check the accordance to a main story line, if you change the proposal. Of course you can also propose a different story line. And: It would be nice, if a native speaker could check my english... :-) Stefan Richter 12:42, 19 Jul 2004 (UTC)

Description of the idea (1 line)

Delivering location-based knowledge and information with Wikipedia, collaboratively developed by its users.

Specification (textarea)

The idea of providing location-based services and information is a really obvious one. Everybody who owns a mobile phone and knows about GPS technology has had this idea before:

"When I walk through a city, it would be nice if I could get services and information according to my current GPS position. In Rome I could get ancient history information or information about the best restaurants around. When I am standing under the hole of the Pantheon ceiling, I could get more information on the architecture and the builders."

It is easy to think about this. But it is very difficult to implement.

The main problem is the gigantic information-base that is needed for such a project. Until now, no commercial enterprise was able to provide the kind of content base to deliver an information service like this. It is the chicken-and-egg problem: first you have to provide a lot of content, before you can charge for it. So you really need a lot of investment money, but you cannot be sure that it will pay off in the end: will the customer happily pay for this kind of service?

We have a solution for this: we want to make all the free content and information in Wikipedia available for location-based services.

What is Wikipedia?

Wikipedia (www.wikipedia.org) is an encyclopedia that is collaboratively developed using wiki software. Wikipedia is managed and operated by the non-profit Wikimedia Foundation. In addition to standard encyclopedic knowledge, Wikipedia includes information more often associated with almanacs and gazetteers, as well as coverage of current events.

All original material contributed to Wikipedia is deemed to be free content under the GNU Free Documentation License, meaning that it may be freely used, freely edited, freely copied and freely redistributed. We call this "copyleft", in contrast to the idea of "copyright" used by the commercial world.

Wikipedia's content is created by its users. Because pages are continually being edited, no article is ever finished. As such, Wikipedia is subject to some unique hardships [1] (http://en.wikipedia.org/wiki/Wikipedia:Why_Wikipedia_is_not_so_great) that do not affect traditional encyclopedias. It has self-healing systems in place to deal with these challenges, and even a page designed to explain them [2] (http://en.wikipedia.org/wiki/Wikipedia:Replies_to_common_objections).

Wikipedia began as an English language project on January 15, 2001, and soon gained its second language, French, on March 23, 2001. There has since been a great deal of effort devoted to making it multilingual, and it currently contains over 300,000 articles in English and over 400,000 in other languages (as of June 2004 [3] (http://en.wikipedia.org/wiki/Wikipedia:Multilingual_coordination)).

This means: we already have the content and a working mechanism to increase this large information-base in a world-wide, non-profit organisation.

We have to complete five small sub-projects before we can implement services based on Wikipedia content:

• Develop a way to attach geo-information to knowledge artefacts (text, images etc.)
• Implement this concept by adding geo-information to existing content and encourage this for new content.
• Develop a protocol for requesting and delivering location-based content, and for adding geo-information content that is not (precisely) located.
• Implement a protocol server application integrated with Wikipedia.
• Develop proof-of-concept client software to implement this protocol.

Primary advantage (textarea)

The primary advantage of this extension to Wikipedia is that large-scale location-based information services are possible at all! On a large scale, it seems impossible for any other enterprise to provide a gigantic world-wide information-service like this.

Additional advantages (textarea)

The knowledge is freely available. The information is not owned by a corporation, so it is not possible to build monopolies. Everybody can now enter this market. Everybody can provide this information. Enterprises can only differentiate by price and quality. This is good for the customers and good for society. Also, the information is controlled not by companies, but by the public.

What problems does your idea solve? (textarea)

We solve these problems:

• How can we build and maintain a large, world-wide information-base for location-based services?
• How can we attach geo-information to knowledge artefacts?
• How can we request and deliver location-based content?

This first problem is by far the most difficult one, but in Wikipedia it is already solved! The second and the third are quite simple. We are in the specification phase for these.

What industrial sector(s) can profit from your idea?

Select one of the following:

• Plant construction
• Automobiles and automotive supply
• Construction
• Chemicals and pharmaceuticals
• Electronics and electrical engineering
• Power and water supply, waste technology, recycling
• Precision mechanics and optics
• Commerce, transport and logistics
• Wood and paper manufacturing
• Information and communications
• Plastics and rubber processing
• Measuring and control technology
• Textiles and clothing
• Metal processing
• Software
• Coal, iron and steel
• Raw materials
• Oil and natural gas
• Food and tobacco
• Publishing and printing
• Medical technology
• Arts and crafts, musical instruments, furniture
• Sports equipment/games/toys, office supplies/equipment
• Hospitality
• Banking and insurance
• Aerospace
• Education and training
• Healthcare and social services
• Glass and ceramics processing
• Mechanical engineering
• Other

Other industrial sectors? (1 line)

Hardware (e.g. personal electronics, GPS), cars, real estate sales, tourism...

Please describe how your idea can be used in the specified industry (textarea)

Car manufactures can implement our data in their navigation systems to lead people to interesting places. Anybody can use our data and protocol and implement it in their systems.

Who will be the user/operator/customer of your idea? (textarea)

• First, everybody can access this information, so our main user will be the general public. There will be free implementations of software clients that can access location-based Wikipedia content.
• Second, there will be service providers from the telco industry, that want to deliver this information to their customers using our software protocol to query/request location-based information. They will not charge for the information, but they will charge for the bandwidth usage.

Maybe Wikipedia has to charge for large amounts of commercial requests from telcos to the Wikipedia servers, so Wikipedia e.V. can buy new hardware and provide services and maintenance and fund new projects.Stefan Richter 13:07, 19 Jul 2004 (UTC)

I do not understand how you want to make anybody pay for wikimedia's band-width. -- Nichtich 23:35, 20 Jul 2004 (UTC)

I dont want make to anybody pay for it. But think about this hypothetical case: Vodafone likes this idea. They want the Wikipedia-Content, write a software and build a nice PDA-device. Then you have millions (!) of new Wikipedia-Users requesting geo-data. And: You will have much more contributers. So why shouldn't Vodafone (in general: large commercial coorporations) pay for access, because there is so much traffic from them. But if you have a better idea, on how to get money for hardware/bandwidth...Stefan Richter 07:53, 21 Jul 2004 (UTC)

Has your idea already been tested?

• Yes
• No

The idea of collaborative gathering of data/knowledge has been tested for more than three years - Wikipedia has now more than 800.000 articles. This is proof that the idea works.

Wikipedia aricles already contain special markup and metadata such as math and what we call "categories". To add geo-information is a comparable effort: we know how, and that we can make it happen.

Does a prototype already exist?

• Yes
• No

A prototype version has been checked into the source code by Magnus Manske (cf. Wikitech-l). Currently it works the way we previously handled ISBN data. Something like GEO 1.22.33:-4.55.66 will be automatically converted into a link to a specialpage, which currently links to a MapQuest page of that area, and can include other commentary about that lat/long.

Well you need some fantasy to name it a prototype...--Nichtich 23:19, 20 Jul 2004 (UTC)

OK, so Wikipedia is our Prototype...

What applications and products would be improved by your idea? (1 line)

This idea is a revolutionary new market opportunity for telcos. It also helps to deliver free information to a wider public.

What is the difference between/benefit (profit, costs) of your idea compared to existing products? (1 line)

There is no comparable product on the horizon. On account of our large and dedicated community we have almost no costs and it is likely that our project will achieve greater success than other (commercial) solutions.

What qualifications must users have? (1 line)

We aim to keep it as simple as possible. Everybody able to use a remote control should able to use it. With voice output, the user does not even need to be able to read.

What do you need to realize your idea? (textarea)

Because the complicated part of our idea is already running, we need the support of free software developers willing to contribute their work and knowledge under the GPL and/or LGPL license. There is a group already specifying the requirements and we are sure that we will not lack technical expertise and dedication to our project.

Of course we need to get the word out to the Wikipedia community, to effectively advertise this new feature and its advantages to our 50,000 users. Once we have the support of the existing Wikipedia Community, they will have to attach geo-information to their work. Again, because this is a good idea, we are convinced that the community will follow with the same enthusiasm they have shown when other similarly-complex markup was introduced (cf. Table, TeX and Timeline markup).

Some money is needed mainly to buy geo-specific hardware to distribute to a team of actively interested contributors, and to bring the world-wide developers together to tweak and improve the code prototype. This will be covered by the prize money.

hardware includes handhelds and GPS-devices for developing and demonstration

Evaluation

How will your idea help to resolve the described problem?

• Very good
• Good
• Satisfactory
• Poor

To what extent can you improve the described product/application?

• Very good
• Good
• Satisfactory
• Poor

How would you evaluate the product’s merits?

• Very good
• Good
• Satisfactory
• Poor

How would you assess the technical realization of your idea?

• Easy to realize
• Ordinary to realize
• Difficult to realize
• Extremely complex

In your opinion, what is the estimated time for realizing your idea?

• Less than 6 month
• Up to two years
• Less than three years
• Less than five years
• More than five years

Rights and Patentability

This is one example of what we can expect, when Software Patents become general law in the European Community... Checking every idea and line of code, if you are possibly "infringing IP rights"... (Ideas and Software should't be patentable!).

Has your idea already been published?

• Yes
• No
• Unknown

If so, who has been informed about it? (1 line)

The idea was published on the Wikipedia Mailing List and on the public internet for a general discussion and a joint effort to write this proposal.

If so, in which medium or sector has it been published? (1 line)

The idea underlying our proposal is continuously being developed and published in form of a wiki at the address http://meta.wikimedia.org/wiki/GALILEO_Masters_2004. Further publications in magazines and congress papers are in preparation. Everything is or will be published under the terms of GFDL.

How would you assess the technical head start regarding your idea versus already existing solutions?

• Very large
• Large
• Medium
• Small

Do you think that your idea will be patentable?

Parts of it may be patentable but nothing will be patented because we want everything of the system to be free. Everybody should have the right to freely use, distribute and modify it under the terms of GFDL and/or GPL.

• Yes
• No
• Unknown

Has your idea already been applied for a patent?

• Yes
• No

If so, do you have a provisional application?

• Yes
• No

If so, has your idea already been granted a patent?

• Yes
• No

If so, in which country and with which number is your patent pending?

Country: Number:

Documents

You may now enclose up to two documents. You may send gif, jpg and png files. Each file may be no more than 1 MB in size.

What kind of documents do they expect?

Perhaps we should include a screenshot of a conceptual design geo-page?

How about some kind of chart showing the increasing user base and size of Wikipedia, with an indication of the proportion of geo-specific articles? -- Heron 13:52, 27 Jul 2004 (UTC)

Cool idea! Do you have such charts?! Stefan Richter 11:30, 28 Jul 2004 (UTC)

I will draw such a chart if anybody gives me the GALILEO Masters 2004/proportion of geo-specific articles. You can use the random-page function to get a sample but please tell me the size of the sample (at least >= 100). See also de:Wikipedia:Wikipedistik/Methodik. -- Nichtich 11:51, 28 Jul 2004 (UTC)

I'm sorry that I can't supply the charts, but I have seen charts of Wikipedia size and usage somewhere on the site. I just sampled 55 articles at random, of which 23 were directly geo-specific (not counting articles that might have some indirect relevance to a particular place). I might do another sample later, but I would be grateful if someone else would contribute! -- Heron 21:20, 28 Jul 2004 (UTC)

My values of a tiny 60 articles sample in the German Wikipedia are a little lower. Maybe in the english one there are too many automatically generated articles. See Image:Statistic4GALILEO.png -- Nichtich 11:37, 29 Jul 2004 (UTC)

Beautiful chart! (One small criticism: perhaps you should add text labels to the left and right vertical axes. It only took me a few seconds to work out which axis was which, from your colour-coded red line, but it wouldn't hurt to make it more obvious.)

You're right about the automatically generated articles. A robot called Rambot has added articles for every settlement in the USA (although most have nice maps and geo coordinates, and some have been expanded with useful information).

I just sampled another 45 articles, to bring my total up to 100. The aggregate result for the 100 articles is 36% geo-specific (of which an estimated 13% are by Rambot). Today's random articles seem to be much less geo-specific than yesterday's, I'm afraid. The randomness of the "random article" generator is poor, so I would put a ±50% error on these figures. -- Heron 13:39, 29 Jul 2004 (UTC)

Ideenvorschlag (deutsch)

Das fehlt noch, zumindest ein Diskussionsfeld.

Wir sollten uns auf den englischen Text konzentrieren. Das macht schon genug Arbeit. :-)