Re: Can we create better links by playing games?

I can only add to Elena's statement - in fact, it is rather the exception than the rule that a Semantic Web task can be turned into a good game that attracts large, non-nerd audiences. Over the years since our first experiments in 2007, I have come to the conclusion that it is way more rewarding to turn such tasks into Amazon Mechanical Turk tasks (HITs) than to develop games. If we are honest to ourselves, then all of the existing SW games fall short in a terribly in terms of gaming fun and understandability.

The difference between Luis van Ahn's successful games and our attempts of using this for the SW is that Luis used challenges where the processing of visual data and applying linguistic competence are the core intelligence task, two areas that are suited for broad audiences and easily link to entertaining game scenarios.

But validating mapping axioms between bio ontologies and even open street map data is terribly boring in comparison. 

Plus, the level of competence needed for cracking the interesting nuts in our data (e.g. subtle forms of polysemy like the city of Munich vs. the district of Munich) restricts the target audience significantly.

To be frank, I consider GWAPs for the Semantic Web a dead end and would not invest additional lifetime into it. It was a promising field back then, and has a lot of appeal at first sight, but it will not solve any of our big challenges.

Martin

On Jun 20, 2012, at 10:59 PM, Elena Simperl wrote:

> Am 20.06.2012 17:52, schrieb Melvin Carvalho:
>> 
>> 
>> On 20 June 2012 17:44, Elena Simperl <elena.simperl@aifb.uni-karlsruhe.de> wrote:
>> Am 20.06.2012 15:19, schrieb Melvin Carvalho:
>>> 
>>> 
>>> On 20 June 2012 15:11, Kingsley Idehen <kidehen@openlinksw.com> wrote:
>>> On 6/19/12 3:23 PM, Martin Hepp wrote:
>>> [1] Games with a Purpose for the Semantic Web, IEEE Intelligent Systems, Vol. 23, No. 3, pp. 50-60, May/June 2008.
>>> 
>>> Do the games at: http://ontogame.sti2.at/games/, still work? The more data quality oriented games the better re. LOD and the Semantic Web in general.
>> Hey,
>> 
>> Most of the OntoGame games still work, and a more comprehensive list of related games is available at http://semanticgames.org/. One of the problems I see, however, is that all data collected through such games is not accessible or reusable by applications (or in other games, as a matter of fact).
>> 
>> Yes this is a really important point.
>> 
>> If you get the high score it should be part of linked data to your identity (eg like a badge).  This makes the game 100 times more worthwhile to play!
> In fairness, you want the games to be played by a very large user base, and most of these players will have nothing to do with Linked Data. They will need other incentives to engage with the game :-) But the results would be more useful, indeed. 
> 
> A second problem that I've seen with the increasing number of games being released over the past years (including ours) is that they produce very similar data sets, mostly in general-purpose domains, for which there are actually knowledge bases available containing that knowledge (as RDF).  Having a standard means to reuse such crowdsourced data sets would make the games definitely more valuable.
>>  
>> 
>> Elena
>> 
>>> 
>>> Others: Are there any other games out there?
>>> 
>>> iand is working on a game:
>>> 
>>> http://blog.iandavis.com/2012/05/21/wolfie/
>>>  
>>> 
>>> -- 
>>> 
>>> Regards,
>>> 
>>> Kingsley Idehen 
>>> Founder & CEO
>>> OpenLink Software
>>> Company Web: http://www.openlinksw.com
>>> Personal Weblog: http://www.openlinksw.com/blog/~kidehen
>>> Twitter/Identi.ca handle: @kidehen
>>> Google+ Profile: https://plus.google.com/112399767740508618350/about
>>> LinkedIn Profile: http://www.linkedin.com/in/kidehen
>>> 
>>> 
>>> 
>>> 
>>> 
>>> 
>> 
>> 
>>  
>> -- 
>> Dr. Elena Simperl
>> Assistant Professor
>> Karlsruhe Institute of Technology
>> t: 
>> +49 721 608 45778
>> 
>> m: 
>> +49 1520 1600994
>> 
>> e: 
>> elena.simperl@kit.edu
>> 
> 
> 
> -- 
> Dr. Elena Simperl
> Assistant Professor
> Karlsruhe Institute of Technology
> t: +49 721 608 45778
> m: +49 1520 1600994
> e: 
> elena.simperl@kit.edu

--------------------------------------------------------
martin hepp
e-business & web science research group
universitaet der bundeswehr muenchen

e-mail:  hepp@ebusiness-unibw.org
phone:   +49-(0)89-6004-4217
fax:     +49-(0)89-6004-4620
www:     http://www.unibw.de/ebusiness/ (group)
         http://www.heppnetz.de/ (personal)
skype:   mfhepp 
twitter: mfhepp

Check out GoodRelations for E-Commerce on the Web of Linked Data!
=================================================================
* Project Main Page: http://purl.org/goodrelations/

Received on Wednesday, 20 June 2012 22:05:05 UTC