Semantic Web languages allow to express ontologies and knowledge bases in a way meant to be particularly amenable to the Web. Ontologies formalize the shared understanding of a domain. But the most expressive and widespread languages that we know of are human natural languages, and the largest knowledge base we have is the wealth of text written in human languages. We looks for a path to bridge the gap between knowledge representation languages such as OWL and human natural languages such as English. We propose a project to simultaneously expose that gap, allow to collaborate on closing it, make progress widely visible, and is highly attractive and valuable in its own right: a Wikipedia written in an abstract language to be rendered into any natural language on request. This would make current Wikipedia editors about 100x more productive, and increase the content of Wikipedia by 10x. For billions of users this will unlock knowledge they currently do not have access to. My first talk on this topic will be on October 10, 2018, 16:45-17:00, at the Asilomar in Monterey, CA during the Blue Sky track of ISWC. My second, longer talk on the topic will be at the DL workshop in Tempe, AZ, October 27-29. Comments are very welcome as I prepare the slides and the talk. Link to the paper: http://simia.net/download/abstractwikipedia.pdf Cheers, Denny

Denny, thanks for writing and rewriting this piece. I finally got a chance to go through it end-to-end. Challenge accepted! :) Here are a few early thoughts, and I look forward to discussing it with you and others further. * I tend to agree with you that the challenges of artificial intelligence are a superset of the challenges of bringing to life the abstract Wikipedia. Quite a few items you list in "Unique advantages" section make the abstract-Wikipedia space more easily approachable. * I agree with you that if we are to take the content of Wikipedia to many of the languages spoken in the world today, and engage their speakers to share in, the current model won't work/scale (at least soon enough). * You've raised a great point about "Graceful degradation". A very nice challenge. * In "Unique advantages" you talk about "a single genre of text, encyclopedias" and I wonder what it takes to expand our thinking to include images as well. Will we need to rethink your current construct? Including images is attractive for at least two reasons: Because in terms of learning people have different needs and we will likely need to (continue to) include images as we create the abstractions, but also because one can potentially think of images as representations that are already abstract. Best, Leila -- Leila Zia Senior Research Scientist, Lead Wikimedia Foundation On Tue, Nov 13, 2018 at 10:13 AM Dariusz Jemielniak &lt;darekj(a)alk.edu.pl&gt; wrote: in formalize the collaborate and to current of at October 10. Limits Journal Contributor’s http://journals.plos.org/plosone/article/asset?id=10.1371%2Fjournal.pone.01… _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>

Hi Ariel, thanks for the very thoughtful question. I got asked this question every time I present it, and during the Blue Sky presentation this question - or a variation of it - was asked three times. It really is on top of people's mind! My answer is half inconsistent, I am afraid, because I have by now come up with three ways to answer this question, and they contradict each other. So I am glad to hear more thoughts on it. Here are the three answers: 1) I think that language is a pretty bad delimiter to keep points of view apart. Yes, sure, it allows the Japanese Wikipedia to offer a different description of World War 2 than the Korean Wikipedia has, but I am not sure that is entirely a good thing. We don't have two Wikipedias for Portugal and Brazil, they have to agree and what they say, but we have Wikipedias in Croatian, Serbian, Bosnian, Serbocroatian... and I am not sure that the outcome of this decision is fully positive. So, my main point is, if we really want to capture cultural differences, let's align the borders of the editions of Wikipedia along these cultures. But aligning the cultural borders solely along language borders is badly imperfect. 2) But in general, I think that accepting that different Wikipedias should have different contents are incompatible with our NPOV policy. Now we could have a lengthy discussion whether NPOV is a good policy or not. But in general, I'd really prefer to have all points of views being presented with their due weight in all languages, instead of using languages to represent a point of view only in one language, and have a different point of view in another language. I would love to be able to read both the Japanese and the Korean point of view on contentious issues between these two countries - as I can in Serbian and Croatian, because I can read both languages just fine - but I think it is rather problematic that language barriers dictate the point of view I have access to. In fact, in many cases, we can see in the English speaking Wikipedia how the very same editors from the say Croatian and Serbian Wikipedia come to a more balanced result in the English Wikipedia, which they wouldn't accept in their 'home' Wikipedias. Funny, isn't it? 3) More importantly, and entirely disagreeing with #1 and #2, is that the Abstract Wikipedia never suggests to replace the current language editions, but to fill up the gaps in any given language edition. So, if the Croatian Wikipedia really wants to go into details on Croatian folk songs and Croatian food items, they should be totally able to do so without having to feel bad that they might be missing basic information about South American countries and Australian Aboriginal cultures. In fact, what I hope is that each language edition can choose to display the renderings from the Abstract Wikipedia for most articles, and then they can concentrate on creating in-depth articles on the topics they really care about - local cities, cultures, traditions. I remember in the beginning when working on the Croatian Wikipedia - it feels weird to work on the article about a local dish if you're still missing articles on all chemical elements. How can I write an article about the town my mom lives in (pop. 148) if there is no article yet about the country of Mexico? The abstract Wikipedia has the ability to lessen that pressure and allow the local communities to focus on their interests more. I assume that the solution that combines the three answers is that we will display the local articles whenever available, but be able to take a look at the article rendered from the abstract version (for contrast and comparison and maintenance). And if there is no local article, we would treat the article rendered from the abstract version like a proper article. So, as said, the actual answer to your question is still work in progress, but I wanted to write down a first sketch towards the answer. I am also very happy to hear other people thoughts on this question. But all in all I think that going toward an Abstract Wikipedia will improve along all the dimensions discussed. But I am sure I missed important dimensions on this question. Cheers, Denny On Thu, Nov 15, 2018 at 11:15 PM Ariel Glenn WMF &lt;ariel(a)wikimedia.org&gt; wrote: of a one intelligence make to Including abstract. and human We attractive 16:45-17:00, https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, Societies) Diversity Between 1773-1776. http://journals.plos.org/plosone/article/asset?id=10.1371%2Fjournal.pone.01… https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l New messages to: Wikimedia-l(a)lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe>