Science

Science: journal articles in freebase

2008-05-27T17:40:01.0000Z

@vcs: Freebase has a model that supports scientific and scholarly articles, but that's about it right now, as far as I know. There have been a few projects that have come up that might get us some of this kind of data, but I can't really guess when (or if) they would be done.

@spatialed: using a CVT for journal articles is an interesting idea; you may be right that the issue is not as important as its contents. The contents model, however, requires the Publication (the place the article was published) to be a standard type, since it has to be co-typed with Publication. Trying to co-type a CVT to a standard type is even more cumbersome than entering denormalized data. It could be done, if we stuck an extra property ("journals published in") somewhere. The next question is, how much of this data do we expect to be entered by hand, and how much done by bulk imports? A good bulk-loading tool could auto-generate the name of the issue from the journal title, volume, and issue. (This is, of course, highly speculative, since no such tool exists.)

Science: journal articles in freebase

2008-05-25T18:58:59.0007Z

I'd love to be involved with the effort of getting scientific articles into freebase. I'm very interested in how freebase is approaching this - how can I learn more?

Victoria

Science: journal articles in freebase

2008-05-13T20:44:38.0000Z

We've made quite a bit of progress to get to the Scholarly work type but there is still a ways to go. For example, see he redundant name and properties of Journal issue. That one really irks me. It works but it's not very efficient or user friendly to type a topic name that is defined by its properties and then add those properties to the type. I'm really leaning towards using a CVT for describing a journal issue instead. It ignores the properties of the issue but most users generally don't care about those properties.

I think we are nearly ready to construct a "Biblography import" type via existing co-types that could be used to import/export databases of journal articles. Please post in the Publishing domain if you want to discuss ways to get there. This is the primary reason I became involved with Freebase and I would love to get to the point where I could upload LitCentral.

One of the hurdles has been working on ways to make the publishing domain relevant to all publications. I think we are mostly there now and can start refining the journal article use case.

Science: journal articles in freebase

2008-05-13T19:53:48.0007Z

We recently added types to support scholarly articles and journals in the publishing domain. I'm in the middle of writing documentation for it, but I can point you to the types in question.

Journal articles should use the type Scholarly work; this is a general type that can include journal articles, papers, proceedings, etc. Scholarly work includes two other types -- written work, which allows you to enter the authors, subject, etc., and published work, which allows you to enter the places it was published. (For details on how published work is used, see Entering the Contents of a Book or Periodical.)

There are also types for jounal and journal issue.

Take a look at these, and let me know if you have more questions or if you think there are things we've missed. If you want to play around with the types without worrying about messing up the data, you can practice on sandbox.freebase.com.

Science: journal articles in freebase

2008-05-13T19:15:31.0005Z

I am interested in building a database of science journal articles in freebase and want some help brainstorming about it. I am new to the "topic" "type" "property" schema and any help would be appreciated. For example would the journal title be the "topic" or a "type" of article by the author, or both. Any other ideas about puting contents of journal articles into bins.

Science: Huh?

2008-03-05T19:31:07.0000Z

I think that we really need a "Physics" domain, like Chemistry, and this one is should just become General Science

Science: Philosophy

2007-09-21T10:19:13.0007Z

I think it's a little tricky to start categorizing things as being "science" or not. You'll soon have the scientology people running after you, cause they'll want to be in the "religion" and "science" categories simultaneously (plus, i'll run after you cause I want'em in the "bogus" category too).
More seriously, nice to meet you guys, great enterprise what you're trying to do here. I've been involved with these sort of categorizations for a little while during my phd - so I'll be happy to contribute here. However, I need look more into the freebase "mission statement" first (categorize the world? or let users categorize it?).
I've been looking at how to categorize the philosophical world, and I have a recent publication here:
http://kmi.open.ac.uk/people/mikele/researchStuff/papers/kcap-07-pasin.pdf

More generally, a VERY interesting publication which explores the categorization of topics and defined a so called "topological approach" (briefly - topics are not elements in a hierarchical structure, but are more places in a multi-dimensional space) is here: http://www.cs.vassar.edu/faculty/welty/papers/subjects/subject.html

mmm hope it helps!!!

Science: Type Suggestion: Theory / Scientific Theory

2007-07-20T23:12:30.0000Z

First, a digression. When you indicate Type B as an included type of Type A, it is basically a shortcut of saying: any topic assigned to Type A will also be co-typed to be of Type B, and get all of the properties of Type B. That's it. Though it may look like inheritance, it's not. There's no implied strict relationships between the two types, and you can easily remove any of the type indication of any topic, even if that type is included. Example: Film Actor has Person as an included type, because most of the instances will be, and with an included type you don't have to manually co-type each Film Actor as a Person as well. But, for those non-human actors (think Lassie, Babe, etc.), you can easily remove the Person type from their topics: easy because it's not inheritance.

Anyway, thanks for adding the additional properties. Theory is such a loaded word that has different meanings in different contexts. I suggest focusing on Scientific Theory and polish it first. Put all of the properties you need in there. Add some instances and see if the type fits. If a need should arise later for some generic Theory type, we can worry about it then. It's too abstract to try to design schema for some unknown future data -- it's much easier the other way around, starting with data that needs to be structured and categorized. One thing at a time. :)

Science: Type Suggestion: Theory / Scientific Theory

2007-07-19T22:09:45.0007Z

Nevermind the question on subtypes, I must've not added it properly last time. It's there now.

Science: Type Suggestion: Theory / Scientific Theory

2007-07-19T21:56:51.0000Z

Hi again. I'd already added the types "Conceived in", "Formulated by" and "Scientific field" to Theory. I hadn't put any subfields in there though, so perhaps I should add some. A quick question — by making Scientific Theory a subtype of Theory, does it not inherit the properties of Theory automatically? I don't see them there. On keeping "Scientific Theory" a subtype of "Theory": A Theory doesn't necessarily have to be based on/or use the scientific method. For example, many religious beliefs have theories on how the world came to be without any scientific support. But perhaps this is an unnecessary distinction. Not sure if it'd turn out to be useful on Freebase.

Science: Type Suggestion: Theory / Scientific Theory

2007-07-05T19:33:07.0007Z

This is a good start!

One of the advantages of Freebase over Wikipedia is that instead of just having flat data, key data common to a type are called out by way of properties, which can be queried. Here's a question for you to consider: What properties do scientific theories have in common? For starters, I'd suggest adding the following properties (feel free to use better names):

1) "Introduced By", where the expected type is a new type that has Person as an included type. Perhaps "Scientist" or something more specific. This is to give credit to the scientist(s) who first brought the theory to light. For example, the theory Autopoiesis was introduced by Francisco Varela and Humberto Maturana.
2) "Introduction Date", expected type: Date/Time. This would be the date the theory is introduced. For Autopoiesis, it would be 1973.
3) "Scientific Field", expected type: Field of Study. This would categorize the field for which this theory is significant. For autopoiesis, it would be Biology. You could even input the subfield, something more specific, say Cellular Biology (not sure if that actually applies to autopoiesis, which I'm not familiar with).

What do you think?

Also, I'm not sure what distinguishes "Theory" and "Scientific Theory", or how "Theory" is intended to be used. Can you give me an example?

Science: Type Suggestion: Theory / Scientific Theory

2007-06-27T21:12:35.0007Z

Hi Faye. I've created the types, I've added some properties to Theory, and added Scientific Theory as a subtype. It's a bit unclear to me though how I get the type reviewed and promoted.

Science: Huh?

2007-06-25T05:08:37.0000Z

We deffinately need a Physics subtype. I guess we need more physics articles. I'll get crackin' .

Science: Huh?

2007-06-25T05:08:36.0000Z

We deffinately need a Physics subtype. I guess we need more physics articles. I'll get crackin' .

Science: Type Suggestion: Theory / Scientific Theory

2007-06-14T22:33:52.0006Z

Sounds good to me. Would you like to take a shot at creating these types and appropriate properties? Feel free to experiment in your domain (and try it on sandbox if you're not sure). User-created types can be reviewed and promoted to the public space. Good luck and have fun!

Science: Type Suggestion: Theory / Scientific Theory

2007-06-14T21:41:31.0011Z

Hello everyone, second day of my Freebase experience and my first post.

From what I gather, after having created a type — I should suggest it be added to a certain domain, right?

I discovered yesterday that there's no type yet for "Theory" and "Scientific Theory", so I've created them (discerning the two because a theory can be based on something other than the scientific method.)

Wouldn't it be appropriate to add the type to the Science domain?

Science: Philosophy

2007-06-14T17:47:58.0006Z

Just to be thorough so that people who stumble upon this discussion in the future don't get confused, the Religion type that used to be in the People domain has become part of the newly created Religion domain.

Science: Philosophy

2007-04-11T19:54:58.0000Z

Philosophy and Theology might be better listed as Field of Study under the Education domain, or if not there, I would create a new category and call it Social Sciences to include them. I know Jeff is working on adding schema to the Education domain and you may very well see more stuff there soon. We do have a Religion type currently in the People domain (as it was created as an expected type to one of Person's properties) that will probably be moved into its own in the near future. Again, you're welcome to try modeling anything we don't already have.

Science: More science?

2007-04-11T19:47:08.0006Z

There's definitely a lot to explore in this domain. Don't worry too much about domain names - we usually start with data, and create a domain name to best describe the data we have. I think we are going more fine grain than "natural sciences" vs. "social sciences" in our domain naming - see Chemistry and Biology. Some of the ideas you suggested already exist or are meant to exist under a different category than Science. Space and flight is already a domain, and we have Types for inventions and inventors. Unfortunately we don't have a lot of data on hand for some of the domains you suggested, and that creates difficulties in trying to model them using data we don't have. If you are an expert in an area and would like to suggest ways of modeling it, feel free to create new schema in your domain and point us there. Have fun!

Science: More science?

2007-04-07T18:23:47.0006Z

I'd be interested in doing Acoustic Science, Theoretical Sciences, Space and Time Science, Social Science, Economic Science,
Holistic Science, Weird Science, Scientology, Seance, Alchemy, the History of Science (as in breakthroughs, inventions, world views,
important people,) Modern Science (liquid armor and space program kinds of stuff,) and even a legitimate look at the truth and use of
Science Fiction.
That's mostly domain labels, but the types are in there somewhere. Run with it.

Science: Philosophy

2007-04-07T18:14:15.0012Z

I consider philosophy to be a science, even if it's abstract and without any scientific method. But more so I consider it and a number of
other pursuits to be catagorized as fields of study. In that vein, theology might be considered an academic activity. Is it do-able to put an
"abstract science" or "study" section into the science/tech domain, but also maintain the various fields in their own, more fitting, domains
(like Religion might be considered a social topic to some, but Music has its own science.)

Science: Huh?

2007-04-03T17:48:25.0006Z

The Science domain under Science and Technology is sort of a stub/parking lot right now. Subjects that are more fleshed out get their own domains, and that's why you see domains like Astronomy, Chemistry, Biology, etc. outside of the Science domain. It's possible that we may not need a Science domain after all if we conclude that all of its subjects and related topics are better listed separately or elsewhere. Let us know if you have any ideas.

Science: Huh?

2007-04-03T06:00:52.0006Z

I don't get why there's a "Science" branch of "Science and Tech" which in of itself has various (albeit limited) sciences. Where, for example, should we put studies of space and time, Relativistic Physics, Quantum Physics, etc.? Do they get their own branches or are they put into the "Science" drawer?

Science: Must...have...more...science!

2007-04-03T05:55:46.0006Z

true that

Science: Must...have...more...science!

2007-03-19T22:22:01.0032Z

Scientific-minded people have spoken! We need more Science types! ;)