Who gives a “gosh-darn”?
Nov 03, 2009 In Process By Tim AidlinBroadcasting an audience's live, unfiltered feedback is compelling and useful, but it also presents some challenges, most of which revolve around profanity, search results, and spam.
Being able to see, collect, and analyze a wide spectrum of data concerning your brand, event, or website is very useful. Increasingly, social media tools such as Facebook, Twitter, & Channel 9 make gathering this information—and keeping in touch with our audience in general—easy.
These same tools make it extremely easy for our audience to keep in touch with us. At conferences, for example, we can stream live content that shows what other attendees are thinking & doing back to the audience. News stations do this all the time, displaying live Twitter feeds of certain words marked with keywords, such as "#election".
Broadcasting an audience's live, unfiltered feedback is compelling and useful, but it also presents some challenges, most of which revolve around profanity, search results, and spam:
Profanity
How do we deal with bad words & cursing? Just being a jerk for being a jerk’s sake? What about “rated X” avatars? Do we want profanity displayed to our conference audiences?Search Results
What happens when your Twitter search or “hashtag” produces unexpected results? Searching for "#election" in New York, after all, will probably produce "#election" results for all kinds of elections—not just those in New York. If we can't register hashtags (which would be a nightmare to maintain and control), how do we deal with irrelevant search results?Spam
Tweet spammers tend to piggyback spam messages onto “trending topics”, which are the most searched-for keywords at any given point (i.e., @britneyspears, #healthcare, #keyword). How do we arrange our hashtags so this doesn't happen?
Censorship
The easiest solution is to simply apply a filter that knocks out bad words and items with multiple subjects/search-terms. But this presents a whole new problem: censorship.
It's clear that heavy censorship defeats the purpose of live, community-generated content—to give an open and honest glimpse into your audience's thoughts. Still, there's definitely a need to filter for obvious, extreme instances irrelevant or pornographic content; otherwise, people will stop reading the feeds.
But what about less obvious instances of profanity? How can we be mindful of our audience's sensitivities, when everyone's sensitivities are different? How can we filter, but not censor, content? And where do we draw the line?
Who Cares?
One solution is to just let the community work this problem out itself. After all, who really cares about "inappropriate content"? The answer is unclear.
If you are a person who cares (or doesn't) about inappropriate content, I'd love to hear from you. What content is in bad taste? What is offensive? Do you hate cursing? Where should we draw the line? Let us know by leaving a comment below. Feel free to curse if you really want to be a jerk.
Follow the Conversation
6 Comments so far. You should leave one, too.
Hmmm… I believe Americans are far more sensitive to inappropriate content than anyone else. In my opinion (as a Brazilian that lived in the US), even children don’t explode or become Charles Manson when they read/hear profanity. Some good parenting and context should take care of any situation.
As for adults, most adults sensitive to profanity are not sensitive to Intelligent Design proponents, Glenn Beck or Bill O’Reilly (or their cuckoo counterparts abroad, we have them too). So, should we really care about them?
Are you asking how to really make a comment wall? If so, that’s actually a hard dynamic problem, one that changes with scale.
At the Adobe MAX conference last month, once the suggested Twitter hashtag broke into Twitter’s Trending Topics it started drawing the spammers. I didn’t see the wallboard myself, but I did see the Twitter stream. The popularity itself made it a target.
One possibility is skipping universal search, and instead registering users. This will work until the spammers understand how popular you are and how to game the process.
Fighting-words are another issue, whether standard AngoSaxonisms or terms used by one subset prejudicially against another subset (“teabaggers”, “birthers” eg). One possibility is to asterisk the f*** out of such s***ty word choices, although you’re still left viewing the spewings of an ugly mind.
Perhaps an approval queue, with its associated high labor costs, is the most practical way to go…?
jd/adobe
PS: I received an error when submitting this comment in Opera… am trying it in Firefox now.
I personally don’t care about bad language being used but I think it is expected to be controlled somehow in business situations.
I appreciate the comments. What we’ve found is most people really don’t seem to care, and actually — on occasion — find profanity funny and/or engaging. However, the people that take offense to profanity seem to really take offense.
I agree, too, that the problem grows exponentially with scale. For instance, I might not care so much broadcasting to an audience of 100, but an audience of 6 million is a different story altogether.
You also bring up an good point about “AngloSaxonisms” which are sometimes hard to filter automatically, as well as dealing with #hashtags getting noticed by spammers as they become trending topics.
We’ve found that the best route for catching everything that might be offensive is to make it a manual process using custom software. Do you think that there would be interest in using this software by the public? Would you want to be able to save a search-query and filter offensive, irrelevant, or unwanted content?
I’m not overly sensitive about language, etc. Etiquette is expected in business/professional settings, and that is ok. For searches that are screened, I wonder if good content or a site that would be an excellent resource might be missed because it is flagged for inappropriate content – although I don’t know how computers in public libraries or K-12 schools are managed for such things.
One thing we learned during the PDC was that the “inappropriate language” that we did, in fact, see, was absolutely innocuous. Stuff like “PDC knocked me on my ass.” or “Holy *, we got new touchscreens.” I think that sort of language is totally reasonable :-)