Talk:Knowledge crystal

This article is within the scope of WikiProject Databases, a project which is currently considered to be inactive.DatabasesWikipedia:WikiProject DatabasesTemplate:WikiProject DatabasesDatabases articles

General name for this concept

Lately I am seeing more people make more Wikipedia articles describing "information objects that are used in scientific information production". Wikidata and the Wikimedia Foundation's own proposed Knowledge Engine (Wikimedia Foundation) might be instances of these things, but actually, so many research organizations are trying to configure databases to receive and meaningfully answer questions. I am not sure what the parts to this concept are. Is "knowledge crystal" a general name for the entire concept, or just for part of the whole information production process? Blue Rasberry (talk) 20:19, 20 May 2016 (UTC)[reply]

@Blue Rasberry: "Knowledge crystal" is a general name for an information object that has a specific structure (the main structure is question -- answer -- rationale) and that has been produced according to specific rules (most importantly, following openness and scientific criticism). I am not sure what you mean by "whole information production process". But for example, I vision processes where researchers publish their original data, and then on a web workspace producing knowledge crystals, they analyse, discuss, resolve, and synthesise current best answers to specific research questions. The name knowledge crystal refers to the information object, but it also implies that the appropriate information structures and procedural rules were applied when the object was produced. --Jtuom (talk) 10:51, 24 May 2016 (UTC)[reply]

Jtuom Is my language correct in the following passage?

"I presented a knowledge crystal question to a knowledge crystal generator. The question was, 'How many people in New York City have diabetes?' The knowledge crystal answer was '500,000', then the knowledge crystal rationale was 'because 5% of people in this area have diabetes, and the population is 10,000,000.' The information, 'About 500,000 people in New York City have diabetes because the population is 10,000,000 and the diabetes rate is 5%' is the knowledge crystal given by the knowledge crystal generator."

Are Wolfram Alpha, Watson (computer), and Knowledge Engine (Wikimedia Foundation) knowledge crystal generators? If not, can you say more about how a knowledge crystal is different from what those tools provide? Blue Rasberry (talk) 18:50, 24 May 2016 (UTC)[reply]

Bluerasberry Your terminology in the example case is quite correct. Wolfram Alpha and Watson are NOT producing knowledge crystals, because they are not open and criticizable (I don't know enough about Knowledge Engine to comment on that). This means that if you for some reason have doubts about 5 % of New Yorkers being diabetic, you should be able to dig into the rationale to see where that number came from, and you should be able to argue against that estimate or bring up your own studies not yet considered. If you cannot do that, then the information object is not a knowledge crystal.

Of course, there are practical limits to this ideal requirement. Generally it is thought that scientific articles are credible sources of information and there is no need to dig into the unpublished details of them. So, even with knowledge crystals you will eventually reach a level where the details are no longer openly available and you cannot criticize them.

With such commercial knowledge generators, you very soon reach a point where the rationale of your question is proprietary information and not disclosed to public scrutiny. Having said this, I could argue that if e.g. Wolfram Alpha is widely considered as a very trustworthy source, nobody is interested in challenging its rationales and then it can be said, in practice, to produce knowledge crystals. But personally, I'd like to see a world where there is less and less tolerance to answers whose rationale are not openly available for critique. Even the current scientific articles are poorly open for critique, and therefore we should promote open science. --Jtuom (talk) 11:18, 1 June 2016 (UTC)[reply]

General name for this concept

Categories