Reworking assets

Discussions that concern the community rather than the makehuman software

Reworking assets

Postby joepal » Sat Dec 26, 2015 9:35 am

Fred, in other thread:

If you can provide a 5 star rating system once, then why can't you provide more than one? Why can't you have a rating on the quality of the artwork of the piece AND the quality of the download? Perhaps one also for the stability across platforms such as the stable versions and the nightly? I buy stuff from Ebay and in their rating system, which is pretty basic, they have their feedback set up so that you can give a 5 star rating on the speed of delivery, how well the seller kept in touch, how accurate the description of the item was, etc... and at the bottom, there is a box where you get to write any further comments related to rating the seller. Could we not have something set up like that? It would be separate from the comments section that is currently there. I think that something like that would indeed keep the area self regulating because you could set it up so that only registered members could vote and also only one vote per member. When a threshold number of votes saying the item doesn't work is reached, then the item is either automatically purged, or a notification is sent automatically to the author of the item telling them that it has been flagged. You'd have to have a threshold in order to weed out those who simply didn't know how to use it, or downloaded it wrong.


This makes a lot of sense. In order for the repos to self-regulate, it must be possible for users to provide more fine-grained feedback.

Further, I think I have now figured out how to be able to allow for more than one material per asset post, but I don't want to experiment via 4G on tablet. :-)

Downside of this is that the changes aren't compatible with the current repo structure. I'll have to port stuff once there is a new structure. So I'd like to be sure the new structure is sound before making the switch.

So, what fields should there be for assets (I'm primarily talking clothes/hair now) ?

Uploader editable:

Name
Description
License
Category
(Material fields)

Community editable (any logged in user can add/edit)

Is broken (yes/no) + one line reason
Mesh quality
Mateial quality
Overall usefulness
Keywords (possible to add new ones or select fom list)

Any other suggestions?
Joel Palmius (LinkedIn)
MakeHuman Infrastructure Manager
http://www.palmius.com/joel
joepal
 
Posts: 4465
Joined: Wed Jun 04, 2008 11:20 am

Re: Reworking assets

Postby MTKnife » Sat Dec 26, 2015 7:47 pm

I like the idea of being able to flag an asset as broken. However, as someone who works professionally with website ratings, I should caution that rating systems only work when people are willing to give ratings, and the harder it is to give them, the less likely they are to do so. As it is, assets don't have more than one or two ratings each, and asking people to rate three or four aspects for each item will just drive the number of ratings down.
MTKnife
 
Posts: 311
Joined: Tue Sep 01, 2015 5:22 am

Re: Reworking assets

Postby brkurt » Sun Dec 27, 2015 5:19 am

MTKnife wrote:I like the idea of being able to flag an asset as broken. However, as someone who works professionally with website ratings, I should caution that rating systems only work when people are willing to give ratings, and the harder it is to give them, the less likely they are to do so. As it is, assets don't have more than one or two ratings each, and asking people to rate three or four aspects for each item will just drive the number of ratings down.


As someone who is a professional instructor, I'm interested in the whole rating thing. My goal has been to create templates so that my students can redesign and refine them. Exactly who is it that needs professional-level assets, and who is to judge what is professional?

If I remember correctly, one of the intentions of Makehuman was to create models that would work in photo-realistic still renders, and also in animations (cartoons and anime). Those are two very different goals, yet the team wanted both possibilities.

I am the first to contribute the hair models, and it is always been my intention to improve them, and I have done that, but I don't see any new ratings.
Sorry, guys, but that is not how education works.

I have the same goal as always; create submissions that move the whole project forward, not simply please a few users who won't even explain what they want.
brkurt
 
Posts: 1100
Joined: Sun Feb 17, 2008 8:49 pm

Re: Reworking assets

Postby learning » Sun Dec 27, 2015 10:53 am

I've proposed the editable tag system, as in art sites like DA or Japanese site pixiv. I still think it's the best solution, you can add whatever tags you like, including type of clothing, material, state of development etc, so like, Aethelred can tag his viking dress as, say, "dress", "historical", "medieval", "detailed texture", and brkurt can tag his assets as "template", "unfinished", "wip" or whatever, and if some asset is bugged someone can add "broken" tag etc. This will add almost unbounded flexibility for next to no cost, as opposed to any formalized rating system. Personally, I can't see any downside to it.
learning
 
Posts: 317
Joined: Mon Feb 24, 2014 9:36 am

Re: Reworking assets

Postby joepal » Sun Dec 27, 2015 5:28 pm

Well, some kind of free keywords assignment is in the pipeline. Concerning other free-form or static classification, I hear the arguments above.

I'm going to think a bit before doing anything about the issue, so suggestions and opinions are still very welcome.
Joel Palmius (LinkedIn)
MakeHuman Infrastructure Manager
http://www.palmius.com/joel
joepal
 
Posts: 4465
Joined: Wed Jun 04, 2008 11:20 am


Return to Community discussions

Who is online

Users browsing this forum: No registered users and 1 guest

cron