Organization of Non-Language Topics

The non-language topics desperately need to be reorganized. There are many duplicates (geography alone has three separate topics) and there are many courses that are in topics they don't necessarily belong in. And to be honest it's just sloppy. Is there any way all of these topics and the courses within them could be reorganized?

It seems like the best option would be to create major topics and then split those up into subcategories if possible. For example Science could be a main topic and as the amount of let's say physics courses becomes too large, the curators of science could create a physics subcategory, with its own dedicated curators. That way there could be a tiered system of curators that can oversee the system of topics at all levels, with a few at the top that could oversee the creation and management of the main topics. I don't know how difficult this would be to implement as far as the design of the website, or how much control over content the users should be allowed. And perhaps someone else has a much better idea for how these topics should be structured. I would love to hear some input.

I think at least some preliminary structure that allows for the evolution of new categories and at the same time keeps things organized is needed so that the wiki can be maintained as the content expands, not just on the word level but on the course and topic levels as well. And if it can be done soon while the amount of content is still easy to manage, that's all the better. But if the creation of a new system isn't very realistic could the current system at least be reorganized and then overseen to prevent duplicate topics and the correct categorization of each course? I would be eager to help with this task any way I could.

Posted by spacemoose 5/4/12, last update 7/9/12 (2 years ago)
  • And history is split up too - better to be one topic for the moment

    Posted by tonyyyy 5/5/12 (2 years ago)
  • This is an excellent point: the reason for the duplicates is actually that the topics are set up to test differently - the fields (eg word and definition) are named differently, and they are set up to test in different directions. But that is not clearly explained anywhere obvious. Also they should be grouped together under single headings at least.

    I will see what I can do to organise this in the short term.

    Best wishes

    Ben

    Posted by BenWhately Staff   5/5/12 (2 years ago)
  • Current situation is pretty terrible. There's "random facts" with a big mess of presumably everything older than a certain point, and outside of that... A big mess of a load more stuff. Some topics are very broad, and others very specific. Some of them have courses in which have no relevance to the topic. The problem with narrow categories ("SWIFT Message types" for example) is that you get a plant with far too many "branches" each having only one or two "twigs" and a handful of "leaves". The problem with broad categories is you end up with a "dictionary" full of very disparate stuff, as it's put together from "courses" that aren't really related. Your plant is now a whole garden in itself, and identifying the species of any particular part isn't so easy. For languages, this isn't a problem, as the "dictionary" would be just that. Spanish, for instance, has a certain number of words, of which you may be interested in only a subset. But "your" words are just like any other Spanish words, so including words created by other people shouldn't be a problem. So, the system works well for languages, where "courses" are just subsets of the total words of that language (that have been added so far). Plant one shrub next to another of the same species, they grow together and basically form a single large homogeneous plant. It doesn't really work however for the non-language topics. Usually here each "course" is really a whole topic in itself, with little if no connection to the other courses in the actual "topic". Everyone planted their little plants near each other with no real vision of how it would look in the end, resulting in a mess of random colours and no one knowing what is where. I think definitely make the "headings" more hierarchical, so an example may be Science->Chemistry->Elements. Now we come to another problem. An element for instance could have a name, a symbol, a number, a usage, etc. etc... It would be best if all the information about that element was kept in the same place, rather than being separated into a symbol/name list, a name/usage list, a number/symbol list, etc. This is also similar for languages, where you might have a written form (ie, for Japanese, kanji), a decomposed form (ie kana), a transliterated form (ie romaji) and a translation (ie English). The "dictionary" for Spanish should have the same words in whether your native language is English or Dutch, so the translations in those languages could be fields in the same entry. Ideally, the dictionary for each topic would be more like a database, and mems etc would be related to a pair of fields in that database. When setting up a course, you would choose which field-pairs from the database were relevant, as well as which data, so for instance "element symbol<->element name". Some pairs would work both directions, others only one way. A course could include items from more than one topic, so for instance a general chemistry "course" for a particular exam could be created using items from several different subtopics. Maybe this is too ambitious though.

    Just rearranging the current topics into better categories and subcategories would be a start. Maybe move most things out of the old "random facts" topic into more sensible places and rename it to "other" or something, as a place for the things that don't deserve their own category (yet).

    Posted by Moo0 7/7/12 (2 years ago)
  • How about working off of this? - it's worked for a couple of years :)

    The classes are: 000 – Computer science, information and general works 100 – Philosophy and psychology 200 – Religion 300 – Social sciences 400 – Language 500 – Science (including mathematics) 600 – Technology and applied science 700 – Arts and recreation 800 – Literature 900 – History and geography

    Dewey Decimal Classification

    this would be sub-directories of what goes under "000" but we could create our own subcategories as well so that it fits better with how memrise is used.

    000 Computer science, knowledge & systems 000 Computer science, knowledge & general works 001 Knowledge 002 The book (i.e. Meta writings about books) 003 Systems 004 Data processing & computer science 005 Computer programming, programs & data 006 Special computer methods 007 Unassigned 008 Unassigned 009 Unassigned

    Posted by jenniferhunter 7/7/12 (2 years ago)
  • Also, could we maybe have a new section for Fictional Languages? There's already a few like Dothraki and Klingon, and I'm considering adding in another.

    Posted by EtyMed 7/8/12 (2 years ago)
  • @jenniferhunter, we will give this some more thought; very interesting idea.

    @eric. at the moment we treat "normal" and "constructed" languages in the same way, and all are just "languages"; our thinking is that people are likely to search for the language that they want to learn, rather than for wether it is a constructed language or not. For example, lots of people want to learn Esperanto, and those people would put "Esperanto" into the search box rather than "constructed language". Does that make sense? OR do you think that for less well-known constructed languages it would be more useful to be able to search by that?

    Thanks,

    Ben

    Posted by BenWhately Staff   7/9/12 (2 years ago)
  • Oh that's true.. I didn't even think about Esperanto.. So that's fine the way it is then probably.

    Posted by EtyMed 7/9/12 (2 years ago)