UPDATE: Amber's original question has been undeleted and the new question closed as a duplicate. It looks like that new question has since been deleted (honestly, it didn't need to be this complicated...). The old question has been updated with the new question's text, so to experience the fullness of this post, you need to click the question's edit link and look at the history. 😜 Anyway...
Recently @LiveInAmbeR posted a question about how robots could detect medical issues. After a series of comments, Amber chose to delete the question. I was disappointed by that as I don't believe Amber should have done so.
Because the linked question has been deleted, you may need a minimum amount of reputation to see it. In other words, if you click the above link and the page appears blank, it's because you don't yet have enough experience on the Stack. I believe I've provided enough information to be helpful without the need to read the original post.
I do not have Amber's permission to use the post as a case study and if Amber asks I'll have this Meta post deleted, but I think it's a good case study in what weak comments can do to a question. Thanks, Amber, for granting permission.
First Comment
This seems to be about a third-party property, off-topic here.
This comment was most likely made because the question has at least the following:
Many robots in fiction have this superpower where can stare at a cake and tell you its exact chemical composition (likely using some kind of spectrometer). However, since the robots themselves don't need to eat, how can they tell that a human being is malnourished?
...
Detective Del Spooner : Human beings get hungry. Even dogs get hungry, but not you, you are just a machine. An imitation of life. Can a robot tell if a human is starving? Can a robot tell if a... human has a case of low blood sugar?
Sonny : Can you?
–That's probably how the quote went, right?
Amber did not ask a question about a 3rd-party or commercial world. Amber did refer to a common trope in Science Fiction and did use a clever bit of writing from the movie I, Robot to demonstrate the reason why the question had been asked. Amber also referred to Asimov's laws of robotics (even though they're not really germane to the question). So, the first commenter jumped to the conclusion that the question was about a commercial world and failed to realize that their comment was inappropriate. Amber's question did not violate the 3rd-party/Commercial World restriction.
Second Comment
How can a robot tell whether a person is alive? How can a robot tell whether a car needs oil? How can a non-car robot tell whether something is a car? How can a grey metal robot tell whether something is blue or yellow?
Curiously, the commenter is frankly restating the very question Amber is asking. I admit that I'm about to judge the commenter without sufficient information, but it appears the commenter was trying to point out the belief that the question, perhaps, couldn't be answered. In reality, "robots" do all those things all the time (e.g., your car knows when it needs oil...). The argument that a robot isn't a car (it's not checking it's own oil!) is the whole point of the question. Therefore, this comment suggests the commenter didn't understand the potential of the question or its value if someone could answer it, but it might have left Amber with the idea that the question was somehow unworthy.
Third comment
how can a human tell whether something is a car?
This was in response to Amber wondering why the second comment had been posted. How does a human tell whether something is a car? Well... we have a "database" of associative images that identify objects as "cars" on a sliding scale of probability depending on the quality of the pattern match. We call this "thinking." Computers do this all the time. They identify cars, faces, stock value patterns, and lots of other things. In other words, for some reason the commenter didn't realize that this ability already exists. But it might have left Amber wondering even more if the question was worthy.
Fourth Comment
The fourth comment was deleted before Amber deleted the question. I posted an observation that, as asked, there were two questions that were distinct enough that the should be separated before I cast a vote to close for asking too many questions. Amber immediately edited the question to focus on the most important issue and I deleted my comment, up-voted the response comment (and the question), and started writing an answer.
If you're thinking one of the reasons I'm writing this post is that I'm peeved I couldn't post my answer. OK, you're right. I am peeved... but I also think this issue is a very good case study that could lead to better Stack behavior.
Fifth Comment
You're making assumptions about robots based on existing fiction. There is no such thing as the first law of robotics. Something to that extent is likely to be implemented, but not at all inevitable. Similarly with robophobia.
Amber's response to this was "Not worth asking. Got it." after which Amber deleted the question. I believe that was (IMO) the wrong conclusion and the wrong action to take. This is a case of judging the backstory. Ultimately, the entire discussion in Amber's post involving the laws of robotics had nothing to do with the question. If an argument could be made in this regard, it would be that Amber didn't specify the technological level the respondents should deal with. I started writing a response because I thought today's technology was a perfectly reasonable springboard to answer the question.
My point and the reason for the case study
What you'll find in common here is that, with the exception of my own comment (which gave specific direction as to how to overcome my concern, direction that was adopted), the comments were vague complaints based on a failure to read the question from the perspective of "what's really being asked here?" The consequence of them was to lead Amber to believe that the question wasn't worth asking.
When, in fact, it was very much worth asking.
Conclusion
We all make mistakes. I've made plenty of my own. But if you, dear reader, feel inclined to leave a comment that even remotely expresses the idea that a question is inadequate, please have the moral fortitude to clearly explain what your issue is, why it's an issue, and how the OP can fix it. No users should be members of a peanut gallery.
