How to fix the SxSW voting problem

Written by

Warning: this content is older than 365 days. It may be out of date and no longer relevant.

South by Southwest (SxSW) is a fairly large conference that offers attendees the opportunity to learn more about movies, music, and interactive (online) content at its weeklong festival. It’s become something of a giant party and simultaneously for a lot of people, a way for them to validate their projects and causes by being selected to speak.

The process for choosing who will speak at SxSW is partly open to the public, where speakers propose panel discussions on the voting web site, then encourage friends, family, and anyone who will stand still long enough to vote for their panel. Panels that receive lots of votes have a greater chance of being selected for public performance at the festival.

Competition is fierce, with over 2,200 proposed panels and about 10% or so will make it through (if I remember correctly).

The consequence of this is that every proposed speaker has been shilling like mad to get their panels selected. Sonny Gill asked if there was a better way than this, than making it effectively a popularity contest, and I think there is.

Make the content stand on its own.

Here’s how I’d approach it. Speakers complete proposed speaking session topics, the same as now, except that they are forbidden from attaching any personally identifiable information to the topic. These are then loaded into the system, validated to ensure speakers followed the rules (anyone who didn’t, obviously, is disqualified outright), and then displayed to voters.

Here’s the catch: voters only see the panel topic and description. No speaker name or bio. No information at all about who’s delivering. The URLs themselves are randomized each time you enter the voting process so a speaker can’t find their panel in the pile and tell people to vote for it or even link to it. Voters vote for 5 at a time, ranking them in order of preference, and at the end, votes are tallied and the schedule is revealed.

This eliminates the popularity contest. This eliminates gender or race bias. This eliminates everything except what the panel is ostensibly about, which is the content, the discussion, the conversation.

Will this ever happen? Not likely. SxSW would never get nearly the same number of voters registering under an anonymous system, which means a smaller database to work with, nor would it create the same kind of buzz that the current system does, so don’t expect it to change. Every potential speaker is SxSW’s marketing department, unpaid. Every potential speaker is generating ridiculous Google Juice for SxSW’s web site. Why would you as a conference organizer ever give that up?

But, that said, if there ever were an opportunity for conference organizers as a professional conference to democratize their voting system, this would be the way to do it.

Did you enjoy this blog post? If so, please subscribe right now!

Enjoyed it? Please share it!

| More

Get this and other great articles from the source at www.ChristopherSPenn.com

Comments

32 responses to “How to fix the SxSW voting problem”

September 1, 2009

Katherine

I like this idea. The problem is that the qualifications of panel members are also really important. It's one thing to say that getting into SXSW is a popularity contest–it is!–but it's another to say that names should be completely eliminated from the voting process. For instance, the presentation “Understanding Lies, Deception, and Truthiness in Social Media,” which is about how people post inaccurate/accurate profile information online, seems like a really awesome speech. I'd love to hear it, but not if there's some high school kid from Kansas (I have nothing against Kansas!) talking about how his friends posted stuff on Xanga in middle school. But, if it's danah boyd, heck yes! That gets my vote. She has a Phd from UC Berkeley and all kinds of fancy titles that I don't understand.

There's more in a name than what you're implying.

Reply
September 1, 2009

stevegarfield

Keep the names. I attend sessions based on the speaker sometimes, more than the subject.

There has to be a better way thatn the popularity contest though, I agree with you on that. I don't know what it is though.

Reply
September 1, 2009

jchutchins

While I've never pitched a presentation to SXSW, I've had identical feelings about the voting process, the shilling, and the uneven playing field due to the sheer number of “built in” voters some presenters have over others. Influencers who have large communities are usually — but far from always — insightful presenters.

Katherine other commenters do raise a valid point about keeping presenters' identities public. Perhaps a “first wave” of voting (in which presenters' names are anonymous) could occur, permitting more resonant presentations to make the cut, and then a “second wave” of voting, in which the identities of the presenters are available?

It's an interesting conundrum, but as you (Chris) say in your post, SXSW won't likely change the model. That's unfortunate. There's a lot of noise out there…

Reply
September 2, 2009

Steve Garfield

Keep the names. I attend sessions based on the speaker sometimes, more than the subject.

There has to be a better way thatn the popularity contest though, I agree with you on that. I don’t know what it is though.

Reply
September 2, 2009

Katherine

I like this idea. The problem is that the qualifications of panel members are also really important. It's one thing to say that getting into SXSW is a popularity contest–it is!–but it's another to say that names should be completely eliminated from the voting process. For instance, the presentation “Understanding Lies, Deception, and Truthiness in Social Media,” which is about how people post inaccurate/accurate profile information online, seems like a really awesome speech. I'd love to hear it, but not if there's some high school kid from Kansas (I have nothing against Kansas!) talking about how his friends posted stuff on Xanga in middle school. But, if it's danah boyd, heck yes! That gets my vote. She has a Phd from UC Berkeley and all kinds of fancy titles that I don't understand.

There's more in a name than what you're implying.

Reply
September 2, 2009

John Wall

Yes, Katherine has a point – this method doesn't take into account the skill of the presenter. The best selling author that you never want to see on stage is almost cliche. Another problem I have is that I don't know if I'd be interested in a conference built by a popular opinion committee. I would take the popular vote into account along with a smaller panel of judges considering the results and watching video of the presentations to make the final decisions.

Reply
September 2, 2009

jchutchins

While I've never pitched a presentation to SXSW, I've had identical feelings about the voting process, the shilling, and the uneven playing field due to the sheer number of “built in” voters some presenters have over others. Influencers who have large communities are usually — but far from always — insightful presenters.

Katherine and other commenters do raise a valid point about keeping presenters' identities public. Perhaps a “first wave” of voting (in which presenters' names are anonymous) could occur, permitting more resonant presentations to make the cut, and then a “second wave” of voting, in which the identities of the presenters are available?

It's an interesting conundrum, but as you (Chris) say in your post, SXSW won't likely change the model. That's unfortunate. There's a lot of noise out there…

Reply
September 2, 2009

chelpixie

The points you outlined would be helpful to the community but right, not so helpful to their database or traffic.

The question is do they care enough about community to change their voting process in a way that meets their traffic goals while making voting about the content and not the names. Then there's figuring out how that works.

Being a part of the social media community we're focused on being helpful and doing it in a way that is “fair” and perhaps better for community as a whole. SxSW isn't a “social media conference” and they play by different rules. Is it fair as a popularity contest? No. Should they change it? It depends on what their ultimate goals are and if those goals include catering to their community.

Reply
September 2, 2009

Shel Holtz

Good thoughts, but who's going to read through 2,200 session descriptions, names or no names? I'm all for a democratized selection method, but somebody has to winnow the sessions down to a reasonable enough number that people can make informed choices after reviewing all the possibilities.

Reply
September 2, 2009

Mack Collier

Katherine stole my point. SXSW panel picker IS a popularity contest, but the bigger problem IMO is that a lot of people that aren't very good speakers, want to speak at SXSW. I have sat in on only a few sessions that didn't feature a speaker I was familiar with, and those sessions were downright painful to watch. But being able to say you spoke at SXSW is so prestigious that everyone wants to submit an idea.

One thing I would suggest is that everyone can only submit ONE panel idea. That way they'll present only their best topic.

BTW not sure if they changed it for this year, but I've been told that last year they placed a premium more on comments than voting.

Reply
September 2, 2009

JasonFalls

Like the idea, but don't you think it might just then become a headline writing contest? Just a thought. There is no perfect Pepsi.

Reply
September 2, 2009

Tom Martin

Agree that the voting unfairly favors those with large, loyal followings and could quite likely simply add to the echo chamber effect. But more importantly than that — think SXSW needs to simply do a pre-cull on all entries. There are over 2,000 in the picker this year… there is simply not time for anyone that truly wants to find/pick/vote for great panels to do so… at most maybe you'll review a tenth of them.

In the end, seems SXSW has abdicated a conference organizer's primary job (curating really good content) in favor of cheap marketing. Shame really.
@TomMartin

Reply
September 2, 2009

BryanPerson

A huge problem this year comes down to filtering. Why on earth couldn't the SXSW organizers have done a more thorough job of filtering the initial proposals? 2000-plus proposals to choose? Realistically, how many people are ever going to weed through them all to find the best ones (I know I planned to, but …)

Outright pimping for votes ain't my style. But given the alternative of not doing so and seeing the time I spent brainstorming ideas with colleagues, preparing the submission, and reaching out to potential panelists having–let's be honest–ZERO chance of success, I've resorted to pimping.

@Mack: At LiveWorld we submitted 6 proposals (spread across 4 or 5 team members), hoping that 1 or 2 would make the cut for this first round of public voting. Were all 6 brilliant and featuring a so-called “tight focus” that the SXSW praised each of the submissions for? Realistically, probably not. I would expect the SXSW organizers to sort out the stronger proposals in the initial phase.

Reply
September 2, 2009

Jason Baer

Good post. I agree it's a mess, and I like the solution of hiding speaker names. Only shortcoming I see with that approach is that it then becomes a headline writing contest. But they may be the lesser of two evils.

Reply
September 2, 2009

Erica M

I agree with a lot of what people said about the qualifications of panel members. I may be somewhat unique in this, but the so-called popularity of the presenters often works the opposite way for me. There are people who get to present on whatever the hell they want because they've been presenting forever and they know the organizers.

Aside from that, here's where the identity of the presenter really plays in for me: If your panel is about universities/government/libraries and their use of social media, are from a university/government agency/library? Or are you a consultant? I want to hear from the people within these types of institutions who are championing these non-traditional approaches in very traditional environments.

Reply
September 2, 2009

Erica M

I agree with a lot of what people said about the qualifications of panel members. I may be somewhat unique in this, but the so-called popularity of the presenters often works the opposite way for me. There are people who get to present on whatever the hell they want because they've been presenting forever and they know the organizers.

Aside from that, here's where the identity of the presenter really plays in for me: If your panel is about universities/government/libraries and their use of social media, are from a university/government agency/library? Or are you a consultant? I want to hear from the people within these types of institutions who are championing these non-traditional approaches in very traditional environments.

Reply
September 2, 2009

Rob Blatt

Here's the issue with what you present. The idea of voting for sessions is to be a popularity contest. The organizers want as many people to go to SXSW as possible, and in order to do that, the sessions need to be as popular as possible. They aren't trying to make voting fair. They WANT the proposed panelists to shill and to do the work of getting people excited for the conference early.

The reason they put this out to the public is to gauge interest in the panels, but then it goes back to the board of people who actually decide.

Reply
September 2, 2009

Rob Blatt

Here's the issue with what you present. The idea of voting for sessions is to be a popularity contest. The organizers want as many people to go to SXSW as possible, and in order to do that, the sessions need to be as popular as possible. They aren't trying to make voting fair. They WANT the proposed panelists to shill and to do the work of getting people excited for the conference early.

The reason they put this out to the public is to gauge interest in the panels, but then it goes back to the board of people who actually decide.

Reply
September 2, 2009

Chel Wolverton

The points you outlined would be helpful to the community but right, not so helpful to their database or traffic.

The question is do they care enough about community to change their voting process in a way that meets their traffic goals while making voting about the content and not the names. Then there's figuring out how that works.

Being a part of the social media community we're focused on being helpful and doing it in a way that is “fair” and perhaps better for community as a whole. SxSW isn't a “social media conference” and they play by different rules. Is it fair as a popularity contest? No. Should they change it? It depends on what their ultimate goals are and if those goals include catering to their community.

Reply
September 2, 2009

Chel Wolverton

The points you outlined would be helpful to the community but right, not so helpful to their database or traffic.

The question is do they care enough about community to change their voting process in a way that meets their traffic goals while making voting about the content and not the names. Then there's figuring out how that works.

Being a part of the social media community we're focused on being helpful and doing it in a way that is “fair” and perhaps better for community as a whole. SxSW isn't a “social media conference” and they play by different rules. Is it fair as a popularity contest? No. Should they change it? It depends on what their ultimate goals are and if those goals include catering to their community.

Reply
September 2, 2009

Shel Holtz

Good thoughts, but who's going to read through 2,200 session descriptions, names or no names? I'm all for a democratized selection method, but somebody has to winnow the sessions down to a reasonable enough number that people can make informed choices after reviewing all the possibilities.

Reply
September 2, 2009

Shel Holtz

Good thoughts, but who's going to read through 2,200 session descriptions, names or no names? I'm all for a democratized selection method, but somebody has to winnow the sessions down to a reasonable enough number that people can make informed choices after reviewing all the possibilities.

Reply
September 2, 2009

Mack Collier

Katherine stole my point. SXSW panel picker IS a popularity contest, but the bigger problem IMO is that a lot of people that aren't very good speakers, want to speak at SXSW. I have sat in on only a few sessions that didn't feature a speaker I was familiar with, and those sessions were downright painful to watch. But being able to say you spoke at SXSW is so prestigious that everyone wants to submit an idea.

One thing I would suggest is that everyone can only submit ONE panel idea. That way they'll present only their best topic.

BTW not sure if they changed it for this year, but I've been told that last year they placed a premium more on comments than voting.

Reply
September 2, 2009

Mack Collier

Katherine stole my point. SXSW panel picker IS a popularity contest, but the bigger problem IMO is that a lot of people that aren't very good speakers, want to speak at SXSW. I have sat in on only a few sessions that didn't feature a speaker I was familiar with, and those sessions were downright painful to watch. But being able to say you spoke at SXSW is so prestigious that everyone wants to submit an idea.

One thing I would suggest is that everyone can only submit ONE panel idea. That way they'll present only their best topic.

BTW not sure if they changed it for this year, but I've been told that last year they placed a premium more on comments than voting.

Reply
September 2, 2009

JasonFalls

Like the idea, but don’t you think it might just then become a headline writing contest? Just a thought. There is no perfect Pepsi.

Reply
September 2, 2009

JasonFalls

Like the idea, but don't you think it might just then become a headline writing contest? Just a thought. There is no perfect Pepsi.

Reply
September 2, 2009

Tom Martin

Agree that the voting unfairly favors those with large, loyal followings and could quite likely simply add to the echo chamber effect. But more importantly than that — think SXSW needs to simply do a pre-cull on all entries. There are over 2,000 in the picker this year… there is simply not time for anyone that truly wants to find/pick/vote for great panels to do so… at most maybe you'll review a tenth of them.

In the end, seems SXSW has abdicated a conference organizer's primary job (curating really good content) in favor of cheap marketing. Shame really.
@TomMartin

Reply
September 2, 2009

Tom Martin

Agree that the voting unfairly favors those with large, loyal followings and could quite likely simply add to the echo chamber effect. But more importantly than that — think SXSW needs to simply do a pre-cull on all entries. There are over 2,000 in the picker this year… there is simply not time for anyone that truly wants to find/pick/vote for great panels to do so… at most maybe you'll review a tenth of them.

In the end, seems SXSW has abdicated a conference organizer's primary job (curating really good content) in favor of cheap marketing. Shame really.
@TomMartin

Reply
September 2, 2009

BryanPerson

A huge problem this year comes down to filtering. Why on earth couldn't the SXSW organizers have done a more thorough job of filtering the initial proposals? 2000-plus proposals to choose? Realistically, how many people are ever going to weed through them all to find the best ones (I know I planned to, but …)

Outright pimping for votes ain't my style. But given the alternative of not doing so and seeing the time I spent brainstorming ideas with colleagues, preparing the submission, and reaching out to potential panelists having–let's be honest–ZERO chance of success, I've resorted to pimping.

@Mack: At LiveWorld we submitted 6 proposals (spread across 4 or 5 team members), hoping that 1 or 2 would make the cut for this first round of public voting. Were all 6 brilliant and featuring a so-called “tight focus” that the SXSW praised each of the submissions for? Realistically, probably not. I would expect the SXSW organizers to sort out the stronger proposals in the initial phase.

Reply
September 2, 2009

BryanPerson

A huge problem this year comes down to filtering. Why on earth couldn't the SXSW organizers have done a more thorough job of filtering the initial proposals? 2000-plus proposals to choose? Realistically, how many people are ever going to weed through them all to find the best ones (I know I planned to, but …)

Outright pimping for votes ain't my style. But given the alternative of not doing so and seeing the time I spent brainstorming ideas with colleagues, preparing the submission, and reaching out to potential panelists having–let's be honest–ZERO chance of success, I've resorted to pimping.

@Mack: At LiveWorld we submitted 6 proposals (spread across 4 or 5 team members), hoping that 1 or 2 would make the cut for this first round of public voting. Were all 6 brilliant and featuring a so-called “tight focus” that the SXSW praised each of the submissions for? Realistically, probably not. I would expect the SXSW organizers to sort out the stronger proposals in the initial phase.

Reply
September 2, 2009

jaybaer

Good post. I agree it's a mess, and I like the solution of hiding speaker names. Only shortcoming I see with that approach is that it then becomes a headline writing contest. But they may be the lesser of two evils.

Reply
September 2, 2009

jaybaer

Good post. I agree it's a mess, and I like the solution of hiding speaker names. Only shortcoming I see with that approach is that it then becomes a headline writing contest. But they may be the lesser of two evils.

Reply