I think that’s dependent on platform. It may also be dependent on what the particular forum/discourse installation is for.
For example: A software feature request function (a common usage), a developer sees the feature with 0 votes, that is effectively 0 interest if they’re using pure data metrics (not the fact that the person who posted it is interested).
Devs: “Alright ,lets sort these feature requests by votes”
In some cases, from a development standpoint a not voted for item may actually wind up being higher priority than voted for too, but seeing that there’s actual interest is usually more useful.
Now if you’re talking about opinion polls and stuff, you’re absolutely right. I’d frown upon it.
The first time I used it (in a software development forum a few years back) my immediate thought was; “Why didn’t I vote for my own feature suggestion? Am I not interested in what I just posted?”
So, ultimately I think this leans into the following:
Provide the option.
Allow the discourse administrator to turn it on or off based on their own decisions. And even allow them to decide if the vote counts against the total or not.
From an administrator standpoint, more options (when well organized) are always good.
I’m aware the same argument could be made that 0 vote is basically the same as 1 vote when everyone votes automatically anyway, it just was something that confused my brain when first used in feature suggestion. Hell perhaps even allow it configurable on a per-forum basis.