Plans for Expanding Workshop & Offering "Instant Upload/Feedback" Functionality to OC ReMix

[The Derrit]

if a casual listener just wants to give a 1-10 rating based on their preference and nothing else, they should be able to do that with a simple click of their mouse

that is actually *exactly* what you said don't talk down to me like it wasn't

the point of this thread is to find a way to improve the workshop forum to be more accessible to commentary and possibly provide an incentive for doing something

[timaeus222]

Because "Likes" as on Facebook or YouTube are optional, I inferred that ratings and reviews are also optional and neither requires the other. To me, without other details, the proposed rating system most fundamentally sounded like some sort of scale of preference. I didn't see a proposal about the qualitative rating system I described, so I was suggesting that both be an option.

I'd also note that "Likes" have nothing to do with helping someone improve their product either, so if you have a problem with the preference rating system, you might want to voice your critique of that as well. That is, unless OCR plans to give "Likes" an additional purpose as on YouTube, where they are linked to the user's account to give people easy access to items they've marked.

No Likes is much less degrading than a bunch of (1-6)/10 ratings. Depends on your standards, but I consider a 50% failing. Either a track is liked a little, some, or a lot, but since you don't know the extent of the "Like" by itself, you feel obligated to explain it. djp already talked about his preference for Likes without Dislikes, and I have been and still am very much agreeing that Likes is a good choice.

I'm pretty strongly against dislikes. What I'm currently thinking is this:

• Likes (only - no dislikes) enabled for ALL workshop mixes, period
  
• 5-star ratings optional on a per-mix basis [presumably by the remixer]
  

The major issue with numerical ratings is pretty clear. If people only give numerical ratings, it may give them an excuse to not give a review, because let's face it: many people are lazy when given the opportunity unless it's a topic they're passionate about.

You implied that yourself, right here:

if a casual listener just wants to give a 1-10 rating based on their preference and nothing else, they should be able to do that with a simple click of their mouse.

We don't know exactly what people are thinking or doing while they're listening to something, so for all we know, listeners could just give their numerical rating and go to sleep right afterwards without giving a review, thinking a number is enough of a "review". A numerical rating is a sparse opinion by itself, and not extremely helpful towards the improvement of the OCR community's musical skill. We want to know why something sounds good or bad, not that it simply sounds good or bad (that's the whole point of judging threads and mod reviews; to give the why). We want to keep improving and dishing out wonderful music for everyone to hear. Regardless of whether numerical ratings are optional or required, if they're there, then they'll still be there for some of the lazy to pick and some of the passionate to pick less. Out of all the people you know, what percentage is passionate enough about music to compose it? What percentage are they with respect to the entire world? Pretty small, hence it's a thing to pay musicians to write music.

Saying you "Like" something is just like saying "This is cool to a certain extent because...", and thus it makes sense to provide a "because". Sometimes the "because" is "Just because", and sometimes it's a real reason. Regardless, "Likes" imply an ultimate reason, hidden to either the artist or the listener himself/herself (or both), while numerical ratings are concrete, concise, sparse opinions with some sort of reason to match the number. Human reason may or may not suggest that the number itself is self-sufficient. Likes are less hierarchical, and less impactful to any extremes.

Edited March 26, 2014 by timaeus222

[Ab56 v2 aka Ash]

that is actually *exactly* what you said don't talk down to me like it wasn't

the point of this thread is to find a way to improve the workshop forum to be more accessible to commentary and possibly provide an incentive for doing something

I wasn't talking down to you. I said what you quoted me on under the assumption that a 1-10 preference rating scheme was already the plan, because that is what I thought.

Anyway, I'm clearly not at all as passionate about this topic as any of you. I only wanted to offer some thoughts for consideration to be constructive. Now that I've offered them, I'll see myself out.

Edit: And by the way, the reason why a preference rating system made sense to me is because the entire thread was premised not only on the idea of expanding the workshop, but also on the idea that OCReMix could be a suitable replacement for VGMix and potentially other video game remixing sites. If there is no room for a preference-based rating system on OCReMix, then it would make sense to have a distinct website that's purely about about what listeners like--if people wanted it, that is. I'm not sure how successful it would be, but I'm not seeing why a completely different scheme of showcasing content couldn't or shouldn't exist if OCReMix wouldn't do it. It's not a pressing concern, but it does raise a question about whether this site truly can be the community hub that djp is hoping for, or maybe whether the overarching mission needs to be clearer.

Edited March 26, 2014 by Ab56 v2 aka Ash

[Mirby]

The workshop has, to me, always primarily been a place to post work in progress mixes and get feedback on how to improve those mixes. Maybe the balance is off, maybe the soundscape is too sparse, whatever. If there was an issue with a track, the workshop was a prime place to find out what that issue was.

If there's a simple 1-10 system applied to all songs, then the workshop loses that purpose.

"Oh I got all 1s. I know it was bad and a very very rough WIP, but I'd like to know what could've been done to improve it."

"Not bad. A mix of 7s and 6s with an 8 or two thrown in. I wonder what they liked the most."

"Ah sweet, 9s and 10s. I sure hope that's not just because and there's actually a reason for it."

Those are three completely hypothetical thoughts that could be, well, thought if we relied on the 1-10 rating system and nobody ever explained their ratings.

One of the great things about the compos we've had in recent years is that even if there's separate places to vote and to post about the mixes, there's been a lot of good reviews posted about the songs in each round. And those often point out little issues that could be done to improve each track. Which causes the songs to wind up in the workshop when people want to improve them. And if they went from "Cool I know exactly what to do" and then they did it, posting it in the workshop for reviews on the improvements... a 1-10 system would be the complete opposite of that. No feedback, just a number. A number that can mean many different things to different people, depending on how they interpret the score, but generally a lower number is worse. Unless you're playing golf, which we aren't.

I think the Like system would be much more beneficial, as it prevents any negative reactions from a low score in addition to also showing how good it is. It might be even better if the Likes were hidden (as timaeus suggested above), as those would prevent people from liking on principle because the song has a lot of likes and instead would require them to listen to the song and decide for themselves.

And really, what use is the workshop if people are just liking without listening?

[Mak Eightman]

I'm sorry I didn't read all pages.

Probably sounds stupid, but.. Work shop made for work.

I want to improve. I need feedback. I want to know why my mix sounds bad. I don't want likes/ratings/whatever.

Look around. Youtube and other places filled with tones of crap. And this crap has billions of likes and views..

In other words, I have a bad feeling about such "upgrade" for work shop. Maybe adding some kind of check boxes will be useful.

ReMixer/listener make a small list with his/her problems. Something like:

Rhythm Guitar +/-

Lead GTR +/-

drums +/-

ETC.

This could help to lazy people.

Also I'd like to say. When ReMixer chose his/her source and genre, he/she already made his/her "rating"(amount of likes). Popular game? Genre? Cool. Now you have your likes/rating.

Sorry for my english.

[MindWanderer]

I agree with Timaeus: a "Like" is moderately useful without any context. It can be applied without a great deal of thought, without any negative impact. I like the idea of this being the primary way remixes can be sorted by popularity, because this is the easiest way of getting "scores" for as many mixes as possible.

A numerical score is just not helpful without an accompanying review. I don't believe it's a good idea for people to be able to give a mix 1/5 stars without justification, for all the reasons Timaeus said. Something like the Amazon method might be a good idea: If you want to leave a numerical rating, you have to leave the review. And reviews can themselves be "liked."

(BTW, if anyone wants to get into an elaborate discussion with me regarding the merits of a 1/5, 1-7, or 1-10 scale, we can take that to PM or a separate thread. It's a complex psychometric topic.)

[Mak Eightman]

I like the idea of this being the primary way remixes can be sorted by popularity, because this is the easiest way of getting "scores" for as many mixes as possible.

Aaand.. Guys like me will be buried under celebrities)

[Nase]

Aaand.. Guys like me will be buried under celebrities)

Again, that's why likes + simple ratings make sense, to offer 2 different filters.

Personally, i've arrived at the opinion that the rating should not be optional from the artist's side, but from the reviewer's.

The ratings, much more than the likes (due to the hype/critical mass issue), should be used as a rough means of filtering. That's the whole point innit? Just allowing that for some mixes seems to split up things unnecessarily. Talk about "bifurcation".

And yes, mandatory reviews coupled with ratings make a lot of sense to me. So when mixers who hate ratings get rated, they atleast get something with it they probably enjoy, a review.

Mindwanderer: yeah the psychometric thingumajick is an interesting topic. Just briefly, what'd be your fav?

To me it's 5 stars/shoes/golden dildos all the way. It's just very simple and a rating standard for anything art related for good reason. For that purpose, low res ends up being more imo.

[Brandon Strader]

I think improving things all around is a better move at the moment.. those new symbols and the fixed way that comments show on mixpost pages is a huge step towards making the site look just that little bit moar professional Very good stuff.

[avaris]

Aaand.. Guys like me will be buried under celebrities)

Would be great if there was a level of anonymity when songs were posted. Or even if that was a search option for displaying songs. That way listeners just listen to the song for what it is, and not for who wrote it.

[Gollgagh]

It would be pretty neat to extend (or replicate) the ReMix Roulette function to the new workshop to achieve something like that.

[djpretzel]

Would be great if there was a level of anonymity when songs were posted. Or even if that was a search option for displaying songs. That way listeners just listen to the song for what it is, and not for who wrote it.

I think the real celebrities, from a visitor filtering perspective, are the games

I think far more users ignore anything that isn't Chrono Trigger than anything that isn't zircon, for example. Even though I love me some zircon.

Having a dedicated page that offers randomized tracks, and options to NOT show artist OR game might be kinda cool... but we are getting a little outside the original scope, which was just the functionality/integration being added to the workshop, specifically.

To answer Ab's question, my intent with this discussion is to gather both ideas and concerns and help crystallize what I think the best & most pragmatic initial plan for this should be. Step two would be putting together something more concrete, based on actual technical capabilities TBD, and then letting folks poke holes at it - with the understanding that the INITIAL version of this won't be perfect, and that some things will have to be deferred.

[MindWanderer]

yeah the psychometric thingumajick is an interesting topic. Just briefly, what'd be your fav?

To me it's 5 stars/shoes/golden dildos all the way. It's just very simple and a rating standard for anything art related for good reason. For that purpose, low res ends up being more imo.

For a short answer: 1-7, arguably 1-6 or 1-8. For the long answer (which even still I abbreviated considerably to get it under the 1000 character limit), check your message wall. Edited March 26, 2014 by MindWanderer

[Mak Eightman]

I think the real celebrities, from a visitor filtering perspective, are the games

I think far more users ignore anything that isn't Chrono Trigger than anything that isn't zircon, for example. Even though I love me some zircon.

That's what I'm talking about! Most of people search their favorite game. I don't want this community became MegaChronoMetroidPokemon(whatever)Clocked ReMix, because "someone" liked mega/chrono/poke/whatever mix two billions times. Same goes to genre.

The one who searching for metal will totally miss supa awesome funky/jazzy/bluesy/whatever mix. That's sad. Tags are great and useful, but has some bad effect at the end..

And if you let people to rate them.. Pessimism?

Back on topic. I can't see any use for likes or rating or whatever "filter" in WORKshop. Some wips has no feedback or even one reply. Possibly, the artist will not finish it because of ignoring. Possibly. Unfortunately, I have only one thought. Wip should be:

1. Name of the game

2. Mix status(wip/fnshd/mod)

That is all! No more info. No source theme, before mod review. No info about genre or whatever. And remove topic starter name. My thought. Probably not so constructive. Sorry.

[MindWanderer]

That's what I'm talking about! Most of people search their favorite game. I don't want this community became MegaChronoMetroidPokemon(whatever)Clocked ReMix, because "someone" liked mega/chrono/poke/whatever mix two billions times. Same goes to genre.

The one who searching for metal will totally miss supa awesome funky/jazzy/bluesy/whatever mix. That's sad. Tags are great and useful, but has some bad effect at the end.

Well... there has to be some way of searching for and finding stuff you think you might be especially interested in. There are almost 3000 mixes posted on the site right now, and if people can upload things without them being vetted, the number of mixes on the site will explode in no time at all.

Right now, we have a banner with the 20 most recent postings, a list of the next 15, and a completely random list of 6. That doesn't really help the "findability"/"discovery" of mixes someone might not manually search for unless they're obsessive and either check the site weekly or subscribe to the RSS feed.

This is why I was thinking that a banner of "featured" remixes, in the sense most people use the term, might be cool. It would have to be curated, but listing the finals of a compo, a fun cover or joke mix that will never be posted, or an older mix that's relevant to current gaming news would be cool.

And now that the topic of discovery has come up... DJP and team, I know this would be years down the line, but if you want to think about creating an automated recommendation system ("Based on your history, we thought you might like the following..." or "listeners who liked this remix also liked..."), I can totally help. I know a lot about the math and theory behind such systems, and while I've never actually had a reason to make one, it should be something I could cook up the algorithms for.

[YukiKazami]

This thread has some really long posts I couldn't be bothered to read, so sorry if this ideas already been brought up. What if, for sources, we have a system similar to how twitch.tv/hitbox.tv has people pick their games for streaming? It would require some database work, likely outsourced to some active community members, but you would start typing the name in, narrowing the choices down as you approach a games name, then select the game. Once you select the game, a second input form next to the first could be used to pick the track(s) used, and then depending on what the tracks are, the youtube source (also community maintained because soundtrack uploaders randomly get shot down all the time) will be automatically linked, say, at the bottom of the OP/Submission page, depending on the format of how this goes down.

[avaris]

I think the real celebrities, from a visitor filtering perspective, are the games

Haha totally agree. It was just a thought to spark more conversation and brain storming.

On a personal side note, I've always preferred anonymous grading/reviewing. This system's main focus should not be about that. It should always be about the music, which OCR has always done an excellent sticking to that goal. As Gollgagh said, would be cool to copy the ReMix Roulette function in an anonymous fashion.

[Brandon Strader]

I think it'd be awesome to give a better focus to the remixes submitted to the site in general. That spotlight that Bubble Dragon got, all of the mixes deserve that, they're all important. Sometimes it feels like they don't get the respect they deserve. (Also it's uncool that Bubble Dragon is still the front page item considering how many more mixes have been posted since then). I understand that it is a good mix but I don't understand how it is a better mix than any other one submitted to the site.

[Nase]

I think it'd be awesome to give a better focus to the remixes submitted to the site in general. That spotlight that Bubble Dragon got, all of the mixes deserve that, they're all important. Sometimes it feels like they don't get the respect they deserve. (Also it's uncool that Bubble Dragon is still the front page item considering how many more mixes have been posted since then). I understand that it is a good mix but I don't understand how it is a better mix than any other one submitted to the site.

I think it succeeded at turning a (cheesy) game tune into a legit pop song fit for radio play, and that might've been exactly what djp dreamed about when he started the site. So maybe it's not fair to the rest of the pack, but it's an exception where the site creator gets really enthusiastic about something. His choice.

Understand that if every mix got a home spotlight like that, the effect would soon be inflationary. Gotta find other ways.

[Argle]

I think it'd be awesome to give a better focus to the remixes submitted to the site in general. That spotlight that Bubble Dragon got, all of the mixes deserve that, they're all important. Sometimes it feels like they don't get the respect they deserve. (Also it's uncool that Bubble Dragon is still the front page item considering how many more mixes have been posted since then). I understand that it is a good mix but I don't understand how it is a better mix than any other one submitted to the site.

Sour grapes?

[zircon]

I think it'd be awesome to give a better focus to the remixes submitted to the site in general. That spotlight that Bubble Dragon got, all of the mixes deserve that, they're all important. Sometimes it feels like they don't get the respect they deserve. (Also it's uncool that Bubble Dragon is still the front page item considering how many more mixes have been posted since then). I understand that it is a good mix but I don't understand how it is a better mix than any other one submitted to the site.

Bubble Dragon has a really cool music video. That's why it got spotlighted. Do you have any concrete suggestions for giving mixes "more respect"? Every individual mix already gets reviewed by judges, a writeup by djpretzel, a unique page on the site, a YouTube video, Facebook post and tweet to hundreds of thousands of people. Isn't that enough?

[Brandon Strader]

Bubble Dragon has a really cool music video. That's why it got spotlighted. Do you have any concrete suggestions for giving mixes "more respect"? Every individual mix already gets reviewed by judges, a writeup by djpretzel, a unique page on the site, a YouTube video, Facebook post and tweet to hundreds of thousands of people. Isn't that enough?

The answer is obviously a scrolling marquee. Remember those?

But yeah I mean Sir_NutS's mix has origins from like 2007 and it feels like it didn't get much love.. sure it got a facebook and twitter post, and a write-up and all that stuff. I honestly don't know what the answer is to improve visibility or anything.

Edited March 27, 2014 by Brandon Strader

[djpretzel]

The answer is obviously a scrolling marquee. Remember those?

But yeah I mean Sir_NutS's mix has origins from like 2007 and it feels like it didn't get much love.. sure it got a facebook and twitter post, and a write-up and all that stuff. I honestly don't know what the answer is to improve visibility or anything.

We do have some ideas to increase exposure for recently posted mixes; we'll be doing SOMETHING with the next minor site update that should help. As Andy says, though, from time to time we like to highlight something different, and "Bubble Dragon" was certainly that. We feel it is to everyone's advantage that we have a certain degree of flexibility and freedom to promote things outside of regular processes.

[timaeus222]

We do have some ideas to increase exposure for recently posted mixes; we'll be doing SOMETHING with the next minor site update that should help. As Andy says, though, from time to time we like to highlight something different, and "Bubble Dragon" was certainly that. We feel it is to everyone's advantage that we have a certain degree of flexibility and freedom to promote things outside of regular processes.

I can certainly agree that the music video was really well done, so I of course had no issues with that being posted. It's definitely a fun remix, and it deserves that spotlight.

[Eino Keskitalo]

I thought of an use case that would be nice for me: submit piece to evaluation and write/edit ReMixer notes afterwards. I have sometimes delayed a sub because I tend to write essay-length submission notes.. or I'm doing the source breakdown as the last thing, which takes a good chunk of time when you do it after the fact.

Couple of other thoughts: multiple ReMixers submitting might get interesting, and perhaps the same system could be used for project submissions (would be handy for consent).