Playing with the Mine

[Richard Marr]

A few notes and questions from using Alec's version of the Mine:

* Platform/app context: How do you guys see this interface evolving?
My head says this should be the admin interface to a platform that's
main purpose is to provide for applications, but my gut doesn't always
feel that way when using it, which makes me wonder what the intention
is.

* Apps: If someone wanted to start putting together an application on
this platform, what form would that application take? Would it be a
Django module hosted in the same instance? If so would those apps
impose any requirements on this interface? Is there a list of app
ideas somewhere that people can contribute to?

* Tags/terminology: I think a lot of people have been trained to think
of tags as an unstructured classification mechanism so might be
surprised to find hierarchies. I guess as long as it's possible to use
the Mine without creating that structure (which in the open web would
only really be something for geeks and collectors) then it wouldn't be
an issue, but I wouldn't be surprised if that feature gets ignored a
lot. It's really cool and useful but I think only certain types of
people would put the time into maintaining it.

Apologies if this stuff has been covered already.




--
Richard Marr
richar...@gmail.com
07976 910 515

[Adriana]

2009/12/7 Richard Marr <richar...@gmail.com>:

> A few notes and questions from using Alec's version of the Mine:
>
> * Platform/app context: How do you guys see this interface evolving?
> My head says this should be the admin interface to a platform that's
> main purpose is to provide for applications, but my gut doesn't always
> feel that way when using it, which makes me wonder what the intention
> is.

Actually, both your head and gut are right. The intention is to use
both the admin interface as well as applications interface... So for
some, Mine! interface proper will be the way they feel comfortable
interacting with their Mine! stuff. And for some it'll be the
applications interfaces. Sort of like some people still use
Twitter.com, though not sure how many :) - and others use tweetdeck or
twidroid etc. That said, I would like to make Mine! proper much better
equipped that Twitter.com is vis-a-vis its clients/applications. Apart
from twitter, which is more API driven (which Mine! will be too),
there is the WP or Gmail model, where you plug in functionality but
still work with it through the 'main' interface. Perhaps something
like Netvibes or iGoogle may be a better model as it uses 'external'
apps but brings them into native interface... so yes, and no. Happy to
discuss further, hope the above makes some sense..

>
> * Apps: If someone wanted to start putting together an application on
> this platform, what form would that application take? Would it be a
> Django module hosted in the same instance? If so would those apps
> impose any requirements on this interface? Is there a list of app
> ideas somewhere that people can contribute to?

Yes, there is a list of ideas that people can contribute to - over to
you Alec - it's on the google code wiki or a page. Re Django module
question - not sure, I don't expect it to be but that was before Alec
fell for Django.. :) As for requirements on the interface, this does
depend on the answer to your first question. For the DB interaction,
there is API. for UI, we got as far as envisaging a 'dashboard module'
for applications where the user goes to use particular app or
functionality. However, this may be only a stage in evolution of Mine!
as there is no reason why need to go to a particular app within Mine!
to enjoy its functionality. That said, the user mental models are such
that we are definitely going to start with that. Again, happy to
discuss, brainstorm further.

>
> * Tags/terminology: I think a lot of people have been trained to think
> of tags as an unstructured classification mechanism so might be
> surprised to find hierarchies.

Well, there are any hierarchies in Mine! as such. There is implied
tags (which we are thinking of renaming categories) which is taxonomy
determined by the user. Mine! provides a tool to create your own
taxonomy but you don't have to use them. Even with implied tags Mine!
does remain an unstructured classification mechanism in traditional
sense. Also, implied tags =! hierarchy, as they are closer to tag
bundles in Delicious, than they are to dewey decimal system. :)

I guess as long as it's possible to use
> the Mine without creating that structure (which in the open web would
> only really be something for geeks and collectors) then it wouldn't be
> an issue, but I wouldn't be surprised if that feature gets ignored a
> lot. It's really cool and useful but I think only certain types of
> people would put the time into maintaining it.

Agreed. Just had a long conversation with Carrie on that very topic
and we came to the same conclusion - i.e. most people will not use it
especially not when they start using Mine! first. We'll try to take
that into account when designing the UI etc. Ultimately, we provide
the functionality, but I am not holding my breath for 'mass use'. That
said, I am assuming/hoping that every user has an area that will over
time require more granular approach/structure. This is where implied
tags come in. For example, I expect to be using implied tags on my
wine stuff, travel, restaurants or food but am unlikely to use them
for topics that I cover only occasionally or one-off instances. The
challenge will be providing the best tools and interface for users to
a) tag consistently and b) manage and tidy up tags on an ongoing
basis. Mine! tagging also has to be able to handle complete tagging
newbies as well as people used to tagging in Delicious, diigo and
other bookmarking apps. Tall order but we have some ideas on how to
have a go at this.

>
> Apologies if this stuff has been covered already.

No need to apologise, it's really good to see these questions as they
validate the direction we have been going, also it's cool to see
someone not involved in day to day Mine! stuff thinking about this at
all! :)

I'd love to meet up as soon as you can to show you what we have done
so far and incorporate your questions and thoughts into the project.

Any good dates? :)
>
Adriana

>
> --
> Richard Marr
> richar...@gmail.com
> 07976 910 515
>

> --
>
> You received this message because you are subscribed to the Google Groups "themineproject discussion" group.
> To post to this group, send email to themine...@googlegroups.com.
> To unsubscribe from this group, send email to themineprojec...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/themineproject?hl=en.
>
>
>

[Alec Muffett]

>> * Platform/app context: How do you guys see this interface evolving?
>

> The intention is to use
> both the admin interface as well as applications interface... So for
> some, Mine! interface proper will be the way they feel comfortable
> interacting with their Mine! stuff. And for some it'll be the
> applications interfaces. Sort of like some people still use
> Twitter.com, though not sure how many :)

That's a good example; the pymine software should accept dashboardy plugins a-la Wordpress, but the intent of the API is to permit development of rich applications to drive the mine from other platforms.

>> * Apps: If someone wanted to start putting together an application on
>> this platform, what form would that application take? Would it be a
>> Django module hosted in the same instance?

Well... I have yet to create a Python Mine API remote-management class, but that is definitely on the todo list as it will replace "minectl.pl" which is the current interface to the commandline.

As a kinda one-off sick joke I wrote:

http://code.google.com/p/pymine/source/browse/trunk/toys/last.fm.uploader.sh

...to show a kludge way to upload your scrobbling history into the mine for searching and tagging, etc, but yet it really needs something better.

Maybe now is a good time for me to work on the remote-management class, because then what you could do is import it into any other python program and drive/integrate your mine from that, with a minimum of linkage.

Or: you could talk to the REST API direct from (say) PHP or Java

Or: you could go the AJAX route, write your app in Javascript and install it in the "public_html" folder of your mine.

There's a whole pile of possibilities.

>> If so would those apps
>> impose any requirements on this interface?

The Mine-plugin stuff is still being thrashed out, but would love to talk about that; myself, I would encourage the AJAX route for optimal reuse; the search/select stuff I am working on will help greatly with that.

>> Is there a list of app
>> ideas somewhere that people can contribute to?

http://code.google.com/p/pymine/wiki/MineAppIdeas

>> * Tags/terminology: I think a lot of people have been trained to think
>> of tags as an unstructured classification mechanism so might be
>> surprised to find hierarchies.

Yeah, it's not exactly a hierarchy as a mesh; with the latest updates you no longer have to declare a tag before using it, so the hierarchy element is doubly gone. What remains is the ability to set up one tag as "implying" something more generic, so where "->" reads "implies"

merlot -> red-wine -> wine -> booze

...what happens is:

1) if I tag "winebottle.jpg" as "merlot"
2) then it will end up in the feed of someone who gets "booze"-related stuff,

...but NOT VICE VERSA because something tagged as "booze" does not necessarily imply "merlot".

You can have multiple implications, too:

scrumpy --> cider
+-> somerset

...but it's hard to draw in e-mail.


Ketan came up with the term "reverse taxonomy" last night, which sounds cool. Nobody is forcing you to prefer this over and above the traditional tagging clusterfuck, by the way; you can still do things the old way. :-)


- alec


--
alec.m...@gmail.com
http://www.crypticide.com/dropsafe/