Tuesday, July 16, 2013 11:59 AM
@Matthew, "narrowcasting". What a perfect way to describe that alternate purpose! I do know people who use 'narrowcast' or unique unused hashtags for personal purposes, usually to create a small private collection for personal use. I do this when preparing for some presentations, to collect core thoughts or resource exclusive to that event. It makes sense, but then the question is, if someone wants to narrowcast, why use a standardized set of terms?
@Colleen, yes, Matthew and I have gone back and forth around the topic of community managers and moderators. Thomas Vanderwal also spoke with us a bit about this and related concepts — the importance of the tags arising FROM the community, not being targeted AT the community. What makes this project(?) so exciting is the way both patient communities and provider communities have expressed interest in collaborating to develop a concise set of shared tags for common conversations. I don't know if anything will come of it, but the conversation itself is valuable.
@Michael, you bring up several intersting points. Regarding character count, yes, daisychaining WILL take more characters than the alternatives. The point is the way the conversations work and the way the technology currently manages hashtags. The boolean AND/OR dynamic, opening and closing conversations. Each tag is essentially a channel of its own. Some tools allow us to search by tag combinations, which then creates a new "channel" without needing a unique tag just for that topic. It simplifies the process of expanding and narrowing search.
For example, in Twitter, I can search #bcsm https://twitter.com/search?q=%23bcsm&src=typd
I can search #bcsm and #brcahttps://twitter.com/search?q=%23bcsm%20%23brca&src=typd
I can search #bcsm and #supporthttps://twitter.com/search?q=%23bcsm%20%23support&src=typd
Or I can search #brca and #supporthttps://twitter.com/search?q=%23brca%20%23support&src=typd
Or I can search #cancer and #supporthttps://twitter.com/search?q=%23cancer%20%23support&src=typd
What I find in each of those streams is very different. If I'm not finding what I seek with one term, I can mix and math with others. This is also a model that is currently in standard practice throughout Twitter and social media in general, the idea of combining tags to create unique concepts, as well as being similar to MEDLINE search structures. There are benefits and drawbacks to both approaches. With Matthew's description of narrowcasting, I'm thinking we might want two different sets of 'tags' for broadcast and narrowcast functions, with the narrowcast terms being shorter. The problem with short tags when not combined is that they are rarely unique and tend to generate a confusing amalgamation of strange tweets on diverse and unrelated topics.
Going back to the prostate cancer example, frankly, I can imaging a clinician including "clinically localized" as a concept, but I can't see many patients understanding the significance or attempting to build a conversation around it. I can easily imagine a patient combining a condition tag with a new diagnosis tag. Indeed, poking around, I just found an example (and as far as I knew, #newdx was a new tag).
I suspect that as this evolves, we may find certain tags to be initially primarily of interest to clinicians, and thus less useful in open public communication spaces like Twitter. It may not make sense at this time to focus on tags for explicitly or exclusively clinical topics, at least not until the docs in the conversation feel they need them. Or, an alternative view, having such tags pre-established may ATTRACT clinicians to the social spaces for conversations. I don't know. It could go either way. I was pleasantly surprised to find that "clinically localized" is actually used quite a bit in Twitter, just not as a tag. https://twitter.com/search?q=clinically%20localized&src=typd
The nice thing about hashtags is that they ARE useful in a variety of spaces, far beyond Twitter. Hashtags can be used to aggregate conversations across many platforms. Matthew told me a vision he has of websites for the top level hashtags, with background information and live content fed from across the web. This wouldn't necessarily just be Twitter posts, but Flickr images, Google+ posts, Youtube videos, and more. So having a tag ontology actually supports your idea of pushing the conversation out, and (obviously) you and Matthew seem to be thinking along parallel lines. Perhaps existing web sites on specific conditions might agree to partner with the #hlthtop project to sponsor or host content for/from a specific high level tag.
PS. Matthew and I have started to use the #hlthtop tag to gather conversation around this. Mostly just because it makes it easier for us. #HLTHTOP = Health Tag Ontology Project