How "Voting" in Online Forums Upholds the Status Quo

Thought these were great insights into course UX and for our community forum…and point to why I’ve always been uncomfortable with voting and Q&A formats…

"Implementing voting in online forums, evidence suggests, is a good way to keep them white and full of dudes.

Rather, the decontextualized up or down votes on news stories, people, or lines of code…reinforce the status quo by choosing the topics that are up for discussion before nuanced debate can take place. You can vote up or down, but both choices reinforce the binary that directs and frames conversation around topics that are already established as being of interest or concern to a white male readership."

Hi @vanessa,

It was an interesting read. If you are against voting within online systems as a Ux convention, what do you feel is a more useful mechanism for users to discriminate between high and poor quality content? The article was certainly thought provoking, although to be honest, I am immediately put-off when reading critiques such as this when the author decides to over-use emotive language. When this happens, it immediately reeks of subjectivity, which then leads the article to lose credibility for me.

I think it is a bit extreme to say that implementing voting promotes straight, white, male readership. If you take a look at the gender breakdown for other social media sites which employ analogous voting systems, this hypothesized causal relationship does not consistently occur.

In the article you reference, the author is attacking Reddit. However, according to the above data, the Reddit gender breakdown is 64% male 36% female, whereas Pinterest has the reverse skew, with 21% male and 79% female. It does not, therefore, seem that the inclusion of a simplistic voting system for content causes a certain type of readership.

Personally, I make use of the Stack Exchange platform quite extensively, and I find that they upvoted questions (not only answers) contribute to ensuring quality content on the site. When there is so large a collection of data, there needs to be a discriminatory mechanism to assist users in filtering out the less valuable content.

I feel that the author is misinterpreting the data by failing to consider that social networking sites are largely emergent, automatically fulfilling a purpose greater than that of its constituent parts. And so, if it is discovered that a specific site fulfills the needs of straight white males, then it does not necessarily mean that some racist authority deemed it so. It is simply a property of the emergent system. Similarly, with other social “ecosystems” such as Pinterest and others with differing demographics, just because they do not have a perfectly balanced representation from every demographic category, does not, in and of itself, imply that there is something fundamentally wrong with the emergent system, it is simply how it is, and usually there are good reasons.

To me, the author’s argument is similar to complaining that an online forum for gay rights has too many gay users and not enough straight users, and then attributing this “problem” to some arbitrary technical mechanism. The reason Reddit is used more by young, white males, is because it fulfills the needs of young, white males. Firstly finding fault with this, and secondly attributing causality in such an arbitrary fashion does not seem useful to me.

The author goes on to make the claim that: " You can vote up or down, but both choices reinforce the binary that directs and frames conversation a[ro]und topics that are already established as being of interest or concern to a white male readership." This should be of no surprise that the topics are of interest to white males, as the usership is primarily comprised of white males.

“The voting system is left as the proving ground for otherwise amorphous social phenomena”

  • It seems bizarre to me that this should be a criticism in this age of wikipedia and the like, where the mechanisms are in place to allow for the system to be self governing.

:“Worse, the voting system itself reiterates the pervasive social mechanisms of control that perpetuate rape culture, producing new instances of psychic violence.”

  • This seems rather extreme to me. He is basically saying that self-governance promotes a culture of rape and violence. My guess would be that the author would not have the same things to say about Wikipedia or Facebook. The reason, I assume, he has these things to say specifically about Reddit is because their user demographics is not aligned with his personal agenda, whatever that may be.

“Sites like Reddit will remain structurally incapable of producing non-hegemonic content because the “crowd” is still subject to structural oppression”

  • The author here criticises Reddit for not producing non-hegemonic content. This is like criticizing an ant colony for not producing honey. It is not its nature nor purpose. The internet is a large place. If one is wanting non-hegemonic content, then do not look for it in places such as Reddit. In addition, shortly before this extract, he points out that in each sub-category of Reddit, there are only a few highly active users. It is therefore logical that the content of each of these sub-categories would more greatly reflect the mindsets of those active users than be representational of a global community. However, the author fails to make this connection, and instead places the resulting ‘non-hegemonic’ nature of the content squarely on the voting mechanism. Bizarre.

“Voting on social-media sites eventually unites and distills the user base to a small minority of like-minded content-producers who seek out story “candidates” with the terrible accuracy of a beltway pollster.”

  • This is one of the useful things that the internet offers us; that we can create communities of like-minded individuals which are not limited by geographic location. To criticise this property of the internet seems unconstructive.

The author ends off by stating “When their communities die, let’s hope they die of apathy.”. I find this quite ironic, in an unfortunate and sad way.

Basically, what I am trying to point out with my comments on his article is that none of the data he presents implies the causal relationship between the voting mechanism and the user demographic. In statistics, the principle is that “Correlation does not imply Causality”.

One interesting point that he does make is his reference to the votes as being “decontextualized”. He mentions that by voting on a topic before a discussion has arisen results in “decontextualized votes”. Which basically means “worthless votes”. However, I feel that instead of the votes being without context, it is rather the case that the author fails to see the context. It is not that the votes provide insight into the as yet nonexistent conversation, but rather, serve a self-governing function of filtering topics which appeal to the specific sub-community (for which the sub-reddit exists) from those which do not.

If I think, once again, of my involvement in the StackExchange community. I would hate to use the site if the topics which appeared on the front page were globally representative. If this were the case, they would all be of a simplistic, technical-support nature. Rather, I enjoy the more deeply technical topics which appear on the front page when I visit the site. Thus, the site caters to an online community which is more technical in nature than most people. To criticize the site for this would be absurd.

Regards,
Ralfe

1 Like

I’m not a big reddit fan/user, but I did use stack exchange some time ago. The voting system is one of the things that caused me to disengage.

I don’t ask questions if I can find the answer myself - that means not asking obvious questions for which the answers exist. That also means not getting votes for asking a questions that everyone has. Secondly, I tried answering questions for a bit, but it is a fiercely contested in unconstructive ways. People write quick answers and then edit it later simply to be the 1st one who answered. And when someone asks a difficult question, chances are that only they will upvote the answer.

Now, this is not a big problem, except for the fact that stack exchange staples your score to your forehead. It becomes their (and other people’s) measure of what you know. I’d rather not engage in a community like that than to play silly games to get a high score.

Wikipedia articles don’t get voted on as far as I know. To see if an article is any good, you can look at the discussion page and the revision history. But Wikipedia is different in that they aim for all their articles to be high quality rather than trying to bring the good ones to the front.

With regards to P2PU or more generally, finding good online courses, I would like something that helps sift through all the options, but at the same time, I would be greatly discouraged if I create a course that is very specific and I don’t get any votes and some short and basic courses have high scores.

Hi @dirk,

  • Unfortunately most people are not like you, and ask many, many, many questions that a quick google can answer. However, there are a lot of questions that are asked on StackOverflow that are fringe cases, or situations where unexpected behaviour is found, or where something is trying to be achieved which is not well documented. In such cases, I think having a community such as StackOverflow is great. I think in any community, there will always be a large portion of irritating users who do not make the best use of the technology or platform.
  • This hasn’t been my experience. Although, it might just be that the questions that attract my attention have not had this kind of thing going on. I’ve been quite impressed on occasion with the effort that some users make when answering well thought-out questions.
  • I suppose this can be a bit frustrating when one has not been participating in the community for a long time. I like the fact that they do this. There is/was a very interesting organization called Whuffie Bank (https://twitter.com/whuffiebank) which attempt(ed) to create a social currency out of reputation scores such as the StackExchange score. The idea behind it was quite intriguing. If you think about it, anyone can protest to being an expert on anything on the internet. However, “the proof is in the pudding” as the saying goes. And in this case, the score from StackExchange is the proverbial pudding :smile:
  • I was not trying to suggest that Wikipedia has a voting system. I was simply referring to the fact that Wikipedia is a very well respected resource which emerged from “crowdsourcing” and which is self-governing. I made reference to this fact, as the author of the article in question criticized the fact that assessment of content is left up to the collective.
  • Are you suggesting that because your hypothetical course is so specific, it does not receive a high number of votes, and thus you are disponent? Well, voting is a very general term, and there is no specific requirement to have the course valuated on the number of votes. I depend quite heavily on Amazon’s voting system for books when I am buying online. Most books that I buy are super obscure and only a few people voted. However, I am not interested in the number of votes, but in the score each voter gave to the book.

Regards,
Ralfe

@ralfe (and @dirk)

I think the graph you shared is interesting because to me, it suggests a relationship between hierarchy/collaboration and participation by gender.

Pinterest, Tumblr, Flickr and similar communities have UX gestures that are framed in terms of “sharing” (vs competing). Pin it, reblog it, respond in longer formats. These gestures express interest and support without the insinuation of hierarchy.

There’s also something here about objectivity that I think is useful to think about. Several of these communities are in search of the “right” answer, which can make them somewhat hierarchical/competitive.

Hi @vanessa,

That is a very interesting observation. Following on from your observation, it seems like all the blogging platforms, and art sites (deviantART, Flickr, etc) are also more female dominant. In addition, if you look at the subject matter, the most male dominant social networking sites (Github, Stack Overflow, Hacker News, Slashdot) are all of a very technical nature, and related to IT, which is traditionally very male dominated.

Regards,
Ralfe

@vanessa - an interesting chicken or egg question being whether these sites targeted a female audience first and therefore designed in a way they believed would be “female friendly”, or whether they designed in a particular way first (for reasons other than gender) and just happened to attract a female heavy audience :smile:.