6
votes
Topic tags - tagging the domain?
I've noticed that some people are adding topic tags to identify the domain of an article. For example, if it's from the New York Times, they'll add a "nyt" tag, or if it's from The Guardian, they'll add a "the guardian" tag.
Why? What's the purpose of these tags? Do people really filter in or out topics based on what website they come from? "Show me all articles from the New York Times." "Show me all articles except if they're from The Guardian."
Is this really a thing that people do?
It has nothing to do with the domain, those are just major publications putting out the articles.
Its less so people filter out sources (although, there is no reason it cannot be used that way if say someone hates fox) it's more there so if you are ever looking for something you can find it.
"Hmm... what was that Guardian article that @EscReality posted last week"
That sort of thing.
At least, that's why I do it and I only really do it with major publications (CNN, NYT, BBC, The Guardian etc etc).
Domain, source, publisher, whatever you want to call it: it's the place the articles originate. I would also point out that, in the cases I'm talking about, the publisher is the domain: the Guardian publishes at www.theguardian.com; the New York Times publishes at www.nytimes.com.
Wow. People certainly are different. To me, it would be, "Hmm... what was that article about Harry Potter I saw in ~books last week?" I'm a lot less concerned with where an article comes from (The Guardian, EscReality), and a lot more focussed on what it's about (Harry Potter books). Do you really track articles by who published them and who posted them here?
Publisher matters a lot when dealing with current events/politics, not so much with Harry Potter I guess :)
I can understand why you'd want to consider the source when reading a political article. However, you can see the source in the post's URL and at the top of the article page while you're reading it.
We're talking about topic tags, which are used for filtering and searching. Are you likely to search for articles from a particular news site? How would you use a topic tag that said "nyt" or "the guardian"?
I was thinking about this earlier because I think I've done it a few times but it also strikes me as a bit weird. Still, I think it's worthwhile:
It's the easiest way to filter things out for stylistic reasons. If I find that many submissions about misogyny or whatever else tend to be written in a smug or condescending way that I just don't want to see or find to be useful, in the absence of source tags, I would probably filter out the "misogyny" and "sexism" tags. That's a bit of an unfortunate end. Because a lot of submissions on a topic were not to my taste for stylistic reasons, I've filtered out an entire category of content. But if I'm able to filter out x blog and y news site, I can stamp out the repeat offenders and then judge case-by-case.
The only other way to accomplish this with the current tag system would be requiring descriptor tags for tone and style, but you can see how that would turn into fights quickly. Putting a "smug" tag on someone's post probably isn't going to make them very happy and others will probably disagree about that anyway. If it were bad enough to be clear to everyone, it would probably be deleted anyway.
It's a little less useful on publications without a clearly defined voice, but I can definitely see myself doing just that for a specific blog that's annoying me.
But this would mean we'd have to go further than @EscReality's practice of only really doing it with major publications: we'd have to tag every single submission with the website it came from.
The site is already in the URL of the submission. The URL itself does the tagging for us.
The names don't look nice if we just stick the website name on as a tag, it's better to see 'The Guardian' than just their web address. That means we just build a little alias-dataset to give sites their 'proper' names in the tag listing - if they are popular enough. Editor-level mods could set this for us when they encounter new sites in the wild. It's kinda cool to do it this way because you'll notice when a new site appears and suddenly starts getting traction.
I understand how we could automate the topic tags. That's not the point. The point is that I don't understand why we need these domain/publisher/source tags in the first place. It seems like work for no benefit.
Hacker News allows you to click on the source’s domain, and I do find it useful sometimes, but rarely. Usually I just click it in the case of non-mainstream sites to see what has been discussed about it.
That's probably the easiest kind of tag to enforce or eventually automate, if deemed useful enough.