[PageOneX] [dev] Should we simplify the data model?
Edward Platt
elplatt at media.mit.edu
Thu Feb 14 10:54:58 EST 2013
You're absolutely right about needing an association between user and
highlighted area. We also need to add a way to keep track of whether a
user has coded a certain image within a certain thread. There are at least
a few possible approaches, and I'll be brainstorming. Maybe we can discuss
Monday.
-Ed
On Thu, Feb 14, 2013 at 9:24 AM, pablo rey <pablo at basurama.org> wrote:
> Hi,
> Good idea to clean first the data model.
>
> On Wed, Feb 13, 2013 at 5:32 PM, Edward Platt <elplatt at media.mit.edu>wrote:
>
>> Hi All,
>>
>> I've mapped out the data model of PageOneX. In the attached png, an
>> arrow represents a "has_one / belongs_to" relationship and a double arrow
>> represents a "has_many / belongs_to" relationship.
>>
> Great to understand the structure. Thanks!
>
>>
>> It looks like threadx_image and image_codes are redundant, as well as the
>> direct association between threadx and highlighted_area.
>>
> I agree.
>
> I think there is at least one link missing: the association between
> highlighted_area and user. The user has_many :highlighted_areas. (maybe
> with area directly). I guess it is not redundant, as it is important to
> know who has drawn what in a multi-user thread.
>
>
>>
>> There seems to be a danger of things getting out of sync. For instance,
>> if I understand things correctly, the following two commands should return
>> the same thing:
>>
>> HighlightedArea.joins(:code => :threadx).where('threadxes.id = ?',
>> thread.id).length
>> # => 6
>> thread.highlighted_areas.length
>> # => 237
>>
>> Mmm, I do not fully understand the expression to evaluate it :(
>
>
>> Pablo, do you know if there was a reason for creating these redundant
>> associations and models? Should we remove them?
>>
>> As far as I understand, it is fine to remove those associations and
> models.
>
> I was looking at the post that documented the data model
> http://montera34.org/pageonex/2012/06/22/pageonex-initial-data-model/ but
> it is not clear why those associations were needed.
>
> p
>
>
>> -Ed
>> --
>> Edward L. Platt
>> Civic Technology Programmer
>> MIT Center for Civic Media
>> E15-348
>> http://civic.mit.edu
>> http://elplatt.com
>> @EdwardLPlatt <http://twitter.com/edwardlplatt>
>>
>> _______________________________________________
>> Pageonexdev mailing list
>> Pageonexdev at mit.edu
>> http://mailman.mit.edu/mailman/listinfo/pageonexdev
>>
>>
>
--
Edward L. Platt
Civic Technology Programmer
MIT Center for Civic Media
E15-348
http://civic.mit.edu
http://elplatt.com
@EdwardLPlatt <http://twitter.com/edwardlplatt>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/pageonexdev/attachments/20130214/2e028f24/attachment.htm
More information about the Pageonexdev
mailing list