Can we please get multiple titles again?

[Spud]

we seem to be at the mercy of Spud and JB...

[fluffy]

Frankie: The charts and graphs are there purely to quantify the only quantifiable part of this discussion. Your little missive did point out one missing graph which actually would address one part of what you did say, though, and that's votes over time (as opposed to votes per fight size):

<img src="http://beesbuzz.biz/crap/sfstats/votespertime.png">

Obviously the votes per song count has been more or less steadily declining, even when there were multiple titles, BUT I don't think that vote count is necessarily a symptom of the problems.

The reason I even bring this up is that I feel that fight size IS the problem which leads people to go away. Large fights make the fights very uninviting, because entrants' songs are just a single drop in a flood of music, which doesn't seem to get any attention. That is why I think the multiple-fight format is goo; I don't necessarily think that the review quality is the goal so much as part of what makes this place compelling to begin with.

I don't think you've considered that maybe I've BEEN THINKING about the very issues that you imply that I'm not thinking about, but if I hadn't been thinking about them, I wouldn't have even started this thread to begin with.

[Ross]

One anecdotal note about votes and listeners.

I only vote if I listen to all of the songs - I tend to do this for fights I'm in, but with 30 songs in a fight I end up not listening to all of the songs if I'm not in it.

My wife has commented that she used to listen to all of the songs in my fight (she didn't always get through all of them, but she starting listening to each one) but doesn't bother with the list so long.

[frankie big face]

I don't think you've considered that maybe I've BEEN THINKING about the very issues that you imply that I'm not thinking about, but if I hadn't been thinking about them, I wouldn't have even started this thread to begin with.

I didn't imply anything about you at all, but if you have BEEN THINKING about these issues, then why didn't you write a post about them instead of bloating the thread with graphs? Is this like that time I covered one of your songs and you said my arrangement was exactly what you would have done if you had only had the time?

[fluffy]

I have never fucking said that. And the graphs only came later when people were asking for data, after around five pages of discussion.

[jb]

Votes, schmotes. Here are some Analog stats for November for songfight.org (does not include .net). Assuming 4 MB per mp3 file, the numbers below work out to about 66,081 songs streamed/downloaded in the month of November.

General Summary:
Successful requests: 1,276,702
Average successful requests per day: 42,520
Successful requests for pages: 930,598
Distinct files requested: 24,521
Distinct hosts served: 166,880
Data transferred: 269.39 gigabytes
Average data transferred per day: 8.97 gigabytes


Listing extensions, sorted by the number of requests.
#reqs %bytes extension
564796 98.12% .mp3 [MP3 sound files]
186760 0.36% .php [PHP]

Listing directories, sorted by the number of requests.
#reqs %bytes directory
594757 98.05% /music/
301357 0.82% /pix/
250861 0.43% [root directory]
46393 0.06% /podcast/

[roymond]

I don't know about you guys but I'm just trying to write some useful reviews for a change. You know, doing my thing to keep KA in the hood.

[fluffy]

To me, the votes are more or less unimportant (I only provided graphs because I happened to have the data anyway), but also the download count doesn't reflect much on the community aspects.

Let me back up a bit and try restating things more simply rather than have a shouting match based on what people are reading into what everyone else is saying:

Assumptions

1. Large fights are much more difficult to set aside time to listen to and review

2. Listening to and reviewing songs is an important part of the community, and why people come here (both reviewing and being reviewed, specifically)

3. Large fights make it so that even the people who DO want to review songs end up having to write short, often dismissive reviews

Observations

1. Multiple-title fights allow there to be more participants and entries while keeping the individual song collections smaller (see the graphs above)

2. When a fight is between 10 and 20 songs big, the number and quality of reviews goes up (I have no specific data to back this up)

The conclusions I draw from the above

1. Multiple-title fights keep the community alive

[erik]

I think that when having alot of good songs in a fight becomes less important than having alot of good reviews in a fight, then something is wrong. I think it would be good for the songfight community if more people did reviews, sure, but I also think that if like 18 out of 30 songs were badass, then you'd have to beat people away from the review thread with a stick even though it was a superbig fight.

[roymond]

if like 18 out of 30 songs were badass, then you'd have to beat people away from the review thread with a stick even though it was a superbig fight.

But there's no way to do this without filtering, and then what (whose) filter do you use?

[fluffy]

Absolutely, song quality is the most important thing, but I see that as a graffiti-abatement issue. The lack of deep reviews, IMO, leads people to make more disposable "look ma, I'm on the Internet" tracks, and the presence of those leads to more people making those things. Good, critical reviews give people an incentive to do better, especially the people who actually WANT to do better to begin with.

Good songs -> good reviews -> better songs -> better reviews. This is the sort of thing that JeffB @ Amazon refers to as a "flywheel." The problem is getting momentum into the flywheel to begin with, or in this case, getting the momentum back after a couple of bumps on the road. I feel that smaller fights are the simplest way to do this (short of completely redesigning the whole site from the ground up), and the easiest and fairest way to do that is multiple titles.

[Reist]

Where have all the listeners gone and why? That's what you should be asking yourselves.

People have made this pretty clear ... almost nobody has time to listen to so many songs.

Almost every time I click on a band I haven't heard of, I get pummelled by senseless noise and bad sampling ... I just can't stomach half the entries. It sounds a bit nasty to say that, especially due to the garbage I have submitted, but it's true.

[erik]

Absolutely, song quality is the most important thing, but I see that as a graffiti-abatement issue. The lack of deep reviews, IMO, leads people to make more disposable "look ma, I'm on the Internet" tracks, and the presence of those leads to more people making those things. Good, critical reviews give people an incentive to do better, especially the people who actually WANT to do better to begin with.

I agree that crapsongs breed crapsongs, but I think that holding deep reviews as an ideal leads to (some) people sending in half-baked, poorly executed C- type of work because they know that people who are willing to give out free advice will probably expect to have to hear a high amount of crap. Those kinds of songs are arguably better than LM,IotI tracks, but still not the kind of thing most people want to have to wade through. It's like you end up listening to songs to be nice and give people ideas on how to make a song that someone won't want to FF through instead of listening to songs because you're excited to hear really good independent music.

Do people appreciate good critical reviews? I dunno, I guess some people do. But it really seems that people appreciate someone who can write a longwinded review that has some minor ignorable critique combined with a lot of unexplained praise so that they can feel that the person who is writing the review is being honest and not fanboyish.

[frankie big face]

Votes, schmotes. Here are some Analog stats for November for songfight.org (does not include .net).

This is worse than the graphs.

If you don't care at all about votes, then why not just do away with the voting system altogether and just post the songs?

[frankie big face]

I have never fucking said that.

Actually, you did, but I'm not going to waste time trying to find it. I summarized your statement in my somesongs song notes as "(his) cover version is truer to the original vision of the song than (my) own recording" and never heard any complaints from you, but it's easy to mis-remember things and I could be wrong.

BTW, there's only one person shouting here and that's you. F-word and everything, oh my!

[Caravan Ray]

FUCK

[jb]

Votes, schmotes. Here are some Analog stats for November for songfight.org (does not include .net).

This is worse than the graphs.

If you don't care at all about votes, then why not just do away with the voting system altogether and just post the songs?

Sigh. I merely intended to mitigate some of the "the community is dying!" histrionics with some stats that show that people are listening to lots of songs each month. And to (once again) comment somehow that "winning isn't everything".

Votes are fun. It's fun to win, it sucks to lose, it's a motivating factor. I like the votes. Without the votes, it's just SongPost, not SongFight.

But lots of people hear your song if you enter it into a fight. Isn't that great?

Reviews schmeviews. I can't remember the last review that told me something I didn't already know. Consequently, I only review if I just flat out like a song enough to say something, or if I have some kind of insight that I think might be worth the songwriter's time. But that's just the way I roll.

Things may change, they may not. It depends on some things I won't go into. We're listening, but obviously the community influences, but doesn't dictate the path of the site.

[king_arthur]

Okay, so I just went through the current 30-song fight (501) and everybody got 30 seconds, and I'm now down to a list of 10 songs I want to hear more of. That's a manageable number of songs to review.

Basically, I'm looking through the fight for "keepers," songs I want to save for my own future listening. Some songs got dropped for "genre bias," some for poor performance (tuning, rhythm, playing), some for poor recording quality, some for poor singing. I may well have missed a brilliant song or two, but in 15 minutes, I reduced the task to something more manageable.

It does mean that if I cut your song because of one of those reasons, you're not going to get a detailed review from me. But if you were one of the 10 artists who made me want to listen to more, then I'll be able to put more time into commenting on your song. And if you want better reviews from me in the future, you have 30 seconds to grab me.

I dunno. I'm trying that this week.

Charles (KA)

[Hoblit]

There is so much to say, I have so much to say. But this thread is just too cluttered for everything in my head at this point.

Whats funny is that signal to noise ratio was brought back up.

[fluffy]

Well, could you start a new thread then? I'd definitely be interested to hear your ideas, and yeah, this thread has turned into a complete trainwreck.

[obscurity]

Do people appreciate good critical reviews?

Absolutely. I have had some critical reviews that I have found enormously helpful (FBF and Mostess spring to mind here, thanks both!). As much as I tire of the 'sing higher, damnit!' reviews, I'd gladly wade through a thousand of them to get one that points out a flaw I was unaware of and can easily fix.

[Spud]

I am going to try to boil down some issues here, based on both what's been posted here and what I have talked to people about privately on line. I will address some of the details discussed, but in the end, I am going to try to get at the BIG PICTURE.

<B>Current Situation</B>

Single titles, posted on an irregular basis, with a variable amount of time between deadline, posting of songs, and issuance of a new title. Currently, we are getting around 30 entries per title.

<B>Percieved Problem</B>

Number of songs per fight is daunting for potential reviewers. With more time commitment required to review the entire fight, many reviewers are bagging reviews altogether, reviewing selectively, or reviewing poorly, due to skimming, time constraints, etc.

This has a deleterious effect on the song submissions for several potential reasons:

1. People get frustrated by lack of quality reviews.
2. Not-so-serious feedback leads to not-so-serious submissions (cyclical).
3. Serious musicians are frustrated by the lack of seriousness.
4. Unpredictable schedule makes participation either difficult or frustrating for some.
5. People fail to improve due to lack of proper guidance.

O.K I made that last one up.

<B>Goals (Desirable Outcomes of Potential Changes)</B>

1. Better song submissions (subjective)
2. Better reviews (somewhat subjective)
3. Less non-serious crap in both categories (very subjective).

<B>Possible Solutions (either proposed or implied)</B>

by fluffy: multiple titles.
<i>argument for:</i>
The thought is that with multiple titles, there would be less submissions per title, leading to smaller individual fights, which would be less daunting to review. Reviewers would theoretically spend more quality time on each song, rather than dashing through 30 of them or skipping it altogether.
<i>argument against:</i>
Tradition. More work for the fightmasters. The fact that increasing the number of fights seemed to increase the number of entries, which could be counterproductive to the overall goal of making the fights more listenable. Some would argue that the signal to noise ratio degrades under this scenario.

by king arthur: filtering.
The King has devised a system for his own use, by which he "previews" the songs and makes a short list for future listening and reviewing. A general solution is not proposed, but perhaps something could be implemented which makes this process easier.
<i>argument for:</i>
Filtering the entries somehow allows the listener to get to the meat faster, and review only those songs that are of interest. Good material gets the attention it deserves, and the crap is flushed, as it should be.
<i>argument against:</i>
What is the filter? In the King's method, a 30-second preview does the trick. What if the beauty of the song kicks in at 0:31? Any other filter might be just a subjective or innacurate.

by senza valore, in another thread: registered voting
I throw this in because it might have some impact on reviews, and on song quality.
<i>argument for:</i>
Theoretically, if the voting was more in line with the reviewing, and the winning of a fight was perceived to be more "fair", people would be more inclined to participate all around.
<i>argument against:</i>
Tradition. Makes the community more insular, less populist, more "elitist". Who says that the registered voters have the best musical taste?

by billy's little trip: something for everyone
BLT proposed incorporating just about everything that happens around here into the main body of the site.
<i>argument for:</i>
This would make all of the side fights and other items more visible, more accessible, and more legitimate. There would always be something to do, whether the fightmasters ever got around to posting another title or not.
<i>argument against:</i>
Complexity, dilution, work.

by the weakest suit, transparency:
Exposing who (or just how many) has entered a fight so far as the week goes by is proposed.
<i>argument for:</i>
Coolness factor. Builds excitement and anticipation. Brings people back to the site who might have checked in but not seen the latest song posted.
<i>argument against:</i>
Reduced excitement and anticipation by giving too much away. Might keep some people from entering if they think an unbeatable entity has entered the fight.

by sdurand, clockwork:
<i>argument for</i>
More predictable schedule and timely postings helps people to plan, and eliminates the so-called "songfight gap". Reduces frustration.
<i>argument against</i>
Tradition. Fightmaster's schedules. There is a fun anticipation factor in checking in to see if there is a new title and to see if the songs are posted. Varying schedule for fights allows different schedules to favor different individuals from time to time. Some have more time on weekends, some during the week.

<B>Thoughts</B>

I sense that what is going on here is a gradual change in SongFight! over the years, which includes both the site, its submitting participants, and its audience.

Someone once noted that the rise of Octothorpe paralleled the decline of SongFight! While I am not prepared to state that either SongFight! has declined, nor that Octothorpe either caused or somehow heralded it, the site and the community have changed over the years.

One might say that in the early days, which were also much earlier in the history of both the internet and the world of amateur digital recording capabilities, this was an internet site for musicians. Musicians who were making songs, happened to have computer and internet savvy, and wanted somewhere to hob-nob with other like-minded individuals had an outpost here at SongFight!

Over the years, earlier on due to the connection with dumbrella, but more recently due to broader phenomena, including the ready availability of recording hardware and software (take garageband, for example), the site has evolved into one that attracts internet musicians. I am not talking so much about the "look ma, I'm on the internet" phenomenon, but people who do a lot of other stuff on the internet, not just music. Doing stuff on the internet is no big deal to this audience, as it is where they spend a lot of their time anyway. Putting their music here is as natural as checking their email or chatting in IRC.

This is a subtle change, but an important one. While it doesn't apply equally to everyone, as there were internet musicians here early on, and many traditional musicians still use and appreciate this forum, I would venture to say that the center has shifted.

Of course, many of these people initially attracted by the technology and the community, and less by the music, eventually become more serious about their music (yes, this includes Octothorpe). I am sensing that that is what is happening here. There is a constructive desire to swing back toward the "internet site for musicians" side of the equation as opposed to the "site for internet musicians" side.

<B>Conclusions</B>

This discussion has not gone on in a vacuum. JB and I have been discussing a number of changes to both interface and functionality which, although not specifically designed to address this issue, oddly enough, have that effect. Many of our own discussions preceded this one by several weeks, with some of the same frustrations noted.

Believe me, we are not going to throw the baby out with the bath water. In the big picture, SongFight! will remain the same. However, some fun and subtle changes are in the works. As JB has noted with the stats, listeners are not the big issue. We (you) are getting lots of downloads. Our goal is to make participation in the fights and the review process a more positive experience for those who are truly interested.

We are not yet prepared to launch this new "SongFight! 2.0" nor to provide a date for said launch, but we are also not really looking for specific suggestions, as we have it pretty well underway already. However, you can help indirectly by letting me know if I am reading the tea leaves properly above.

SPUD