Collaborating on a deck

[hssm]

Originally reported on Google Code with ID 322

I have recently begun to use a number of shared decks containing foreign language vocabulary.
Where I have spotted mistakes I have found it invaluable to be able to edit the cards.
It would be nice to have a mechanism of feeding this change back to the deck maintainer.

As a very simple resolution, the author of a deck could have the option to tag a deck
with the maintainer's email address. The application could mail back updates on the
user's behalf to be reviewed and incorporated by the maintainer at his discretion.

A more advanced resolution would feature some kind of wiki hosting service which allows
users to post edits back to a centrally hosted repository.

Reported by reetep on 2011-01-17 14:22:04

[hssm]

I like this idea.
I think it might be best to have a forum though, as some words can have a different
translation and might need a discussion on how to display the cards etc.

I don't know if setting up a forum page for each deck is feasible though. Maybe a blog
that the deck creator has to make, they seem easy to set up.

I'm sure there's a perfect solution and someone with more knowledge about forums/blogs.

Reported by sslater11 on 2011-01-18 06:31:51

[hssm]

Reported by NorbertNagold on 2011-01-18 08:13:22

• Labels added: Type-Enhancement
• Labels removed: Type-Defect

[hssm]

Damien recently mentioned that some features of AnkiWeb might be used.
I think a forum is not sufficient, we probably need a specific tool, maybe something
like wiki+stackexchange+sync

It is time to brainstorm the idea. Please say what would be the goals of the system,
and list the various scenarios that a user could perform.

If we have great shared decks, 10 times more people will be able to use the Anki* tools
:-)
Thanks for your ideas!
Nicolas Raoul

Reported by nicolas.raoul on 2011-01-20 09:06:52

• Status changed: Accepted

[hssm]

I think that this might be a whole other feature request, but I imagine the use of memory
techniques and mnemonics that are rated on their effectiveness.

If there was a rating system, people could share their mnemonics and the best ones
would reach the top and it would help everyone learn a language quicker.

For example: A bear in japanese = Kuma. I see a bear in a leather top and sun glasses
saying "Cool Man" which sounds like KuMa.
That mnemonic could win over hearing someone singing the song Karma Chameleon "Kuma
kuma kuma kuma kuma chameleon, you come and go"

The downside is when you get to a long word with almost no English words that sound
like it. This would allow the best mnemonic to get to the top and remind others.

Reported by sslater11 on 2011-01-26 06:40:29

[hssm]

Hi Sslate,
Very interesting idea, and it is another form of collaboration on decks.
That supposes that the mnemonic is a field of the fact, right? So 2 decks could be
similar, but have different mnemonics.
If you have time, feel free to start thinking about this (and the about the fact-editing-ranking
mechanism, which has a higher priority because not everybody uses mnemonics), and uploading
a drawing showing the various screens.
Thanks!

Reported by nicolas.raoul on 2011-01-26 07:12:01

[hssm]

Hi Nicolas,
I thought the facts could all be a part of the same deck, so it would be like an extra
field that stores all of them, but keeps them separate within the group.
Then when showing the mnemonic, there would be a list to the left of the text box and
it would allow the user to view different mnemonics and to vote for their favourite.

I think automatically updating a deck would be the next step after collaborating on
one, so the mnemonics field could be updated with it.

Reported by sslater11 on 2011-01-26 07:33:12

[hssm]

I'm not sure it's the best idea to let users edit lists O: Sure, some lists can work
with constantly adding items to them and such... But a lot of other lists may not require
it.... say someone created a list to study specific vocabulary he doesn't know, if
people can add items to his lists, he might not want to share them, as he doesn't need
the extra words.... Tho you can argue he doesn't have to download the updates to their
list....but it would be like...a downer. 
Maybe people should be able to improve item quality, like add sentences or make things
more accurate, but if they think another item should be added they can suggest it to
the creator, or fill in a form for it in the site....

I think it would be better to have a wiki of ITEMS. Like, words and sentences...
Say an item is a word in language X. It can have discussions of how to use it in context,
translations to several languages, mnemonics, example sentences, recordings for words
etc'. The site can offer a list building tool, say for Anki, and they can choose to
what language the items would be translated (Would it be language X to Y list or X
to Z, as they will have  different choices... if the word they want to add doesn't
have a translation to their native tongue, they can add it, etc. ), what sentences
they want to add, what mnemonic, if any.
Those lists can be made public for download and stuff, with a page to suggest improvements,
alternatives, more items to add, etc. and the list creator can either keep the list
for his own maintenance, and keep track of those pages, or give it to public maintenance
in which case other people can freely edit those lists.... 
Maybe there can be a "related words" sections too... and another one for words that
are easily confused with the item word.... As well as alternative spellings and stuff...
(Say if the item is in English, it can have alternative English and American spellings
listed, like theater and theatre, grey and gray, etc'. )

Wow this was a long comment... but I feel I had to offer my idea to the community as
I have like 0 ability to execute it myself and I really think it could be a good idea.

Reported by MIDOxRI on 2011-02-20 12:02:59

[hssm]

I have been thinking this weekend about how great it would be if I could get together
with several of my classmates in a class I'm taking now and set up a sort of group
card-hopper. With all of us taking notes and creating cards about things we think are
important, it would be extremely unlikely that any truly important facts would slip
through the cracks. Each user would, of course, have their own copy of the deck with
their own personal timing data for repetition.

For this system to work, card changes would need to be synced optionally, with the
master card repository (as it were) keeping track of the revisions; and new cards would
need to be able to be hidden from the deck on a per-user basis. There may be other
necessities I haven't thought of.

The sort of heavyweight collaboration tools discussed above sound good for situations
where the collaborators don't all know each other, e.g. a sort of wiki-deck. For smaller
groups of people who know each other and are in the same class/team, a lot of that
would be unnecessary. Those tools should be optional per collaborative deck—though
I can't say whether I think it would be better to default on or default off.

Reported by voyagerfan5761 on 2013-01-20 17:17:03

[hssm]

I've been a user of anki for around three years. First for European Portuguese, now
for French. I have some ideas on the topic that I would like to share for those with
the tech knowledge to make it happen:

Motivation:

.- There's a lot of free/quality resources in the web to which I'm grateful. I would
like to contribute back.

.- Anki is a great tool for learning, and more important, maintaining knowledge of
a language, with small amounts of time on a daily basis.

.- Currently, anki users are encoureged to create their own cards based on their learning
needs and there are shared decks but with no reference to its content and that each
person would need to analyze. This is duplicating efforts to create cards. 

-. As a user, after having created a collection, it seems a waste of resources not
sharing it. It would require though a little more effort for creating one card as content
would have to comply with licencing restrictions, but if that effort is spread over
a number of users, the total amount of work per user would be highly reduced, and overall
quality of cards would be improved.

.- Cards could be structured in a way that a user would incorporate them into its collection,
chosing which cards fit their needs

.- Current sharing method lacks key desirable features:
 - what the collection covers
 - duplicate management
 - quality
 - updates
 - tags

It seems to me that efficient language learning follows a more or less standard pattern,
and therefore all users could share the same collection (a lot to be argued here, specially
since I am not an expert in the area and as there could be a lot of different views
by specialists. Nonetheless, as learners, we could infer what has worked best for us.
Better to act than not.

At first, those with an advanced level would quickly review a lot of cards and send
them far away into the future, remaining only with those unknow items

So, more or less with what you have pointed, a Vision would be:

.- To have a universal set of cards, a free way to study a language, always improving,
with total coverage of a language (a wikilanguage like system) 

.- It will be used by a lot of people and for a long time, because language learning
is still a need to millions.

A features wish list would be:

.- Organized, well structured, visible, updatable, collection of cards to learn a language.

.- I would look like a table, with a row for each card, with a specific/invariable
record number, and columns for all data related to it. So a row would have a version,
and those who have that record in their deck would update it over time with a sync
function, without affecting their study status, meaning that the card is morphing.
Images or sounds would be in a repository. 

.- So each record/unique card would have: 
- a selection camp (for downloading selection), 
- a unique id
.- order
.- question 
.- answer
.- language (the target language being studied, let's say Japanese)
.- native language (if the record was developed in a language different from that being
studied)
.- word (word being studied with this record)
.- lemma (lemma being studied with record)
.- Wordnet synset (the word net definition)
.- Frequency
.- version
.- rating
.- downloads
.- tags
.- comments
.- update requests

.- To be prepared in the same language of the target language, when possible/desirable,
if not, translation to be included, with replicas for different important translating
languages, like English, etc. for the native or working language of the user.

.- Cards to have simple question/answer format, if reverse required, a new card prepared
(because it is usually not exactly the same information in the reverse card, it may
be desirable to include other information)

.- To include images, sounds, when desirable.

.- To include IPA phonetics when possible/desirable.

.- To include word frequency ranking (to devise a desambiguation for multiple meaning
words), including part of speach, etc.

.- To include author/version.

.- To include licence level, (http://commons.wikimedia.org/wiki/Commons:Copyright_tags)

.- To include source for information (within the question/answer fields)

.- To include average user rating (rate, user, average rating)

.- To include # of downloads

.- Structure for classification as well as tags needed, in order to not duplicate or
keep control of duplicates

.- In the case of tags, some users may tag words as glossary for a certain important
book in that language.

.- Structure for classification should have a order for presentation, from the very
basic to advanced, first 100-1000 cards to be ordered

.- A card should have a word and lemma, to control coverage of both.

.- To include comments, improvement required

.- To include requests for new cards

.- To have a table like interface for chosing which cards to select and add into user
collection

.- A card could be updated, so that an update into the user card keeps its study level.

.-  An update just fine tunes a card/ improves it without affecting the item being
learned.

.- May have a wordnet synset reference

.- In order for users to review updates or new cards available for its deck, a language
would have a unique deck name to be also used by a user.

What would be needed:

.- Support for web page development (crowdsourcing)
Web page, Hosting, Updating system for users, Card creation
Rating, comments, requests

.- Create supporting links for card creators (crowdsourcing):
Public domain:
 dictionaries
 images
 sound files
 phonetic transcription
 special characters

Develop(crowdsourcing):
.- word frequencies
.- language structure tags
. ordering

Funding (crowdfunding)
 - Wiki support - to be determined

Well, I think Anki is great. I recommended a lot. However, is only for those with the
perseverance/patience/time to create its decks.

I think Anki deserves to become a more attractive an popular tool by improving collaborative/shared
decks (in the way wiki does). It would give Anki creators the credit they deserve for
this useful tool

Reported by ajoposor on 2013-11-13 22:18:12

[hssm]

Issue 1108 has been merged into this issue.

Reported by perceptualchaos2 on 2015-06-02 08:35:10

[hssm]

I'm closing this as out-of-scope for the AnkiDroid project. It's a great idea but doesn't belong here as an issue.