last day (16 days later) » 

9:13 PM

to study: slippery dupe votes

yesterday, 2 hours 56 minutes total – 31 messages, 5 users, 6 stars

Bookmarked yesterday by gnat

hi @GlenH7 - thanks for examples, it has been pretty educative to get back to these
> For me, the web services aspect is what differentiates this from the duplicate. Yes, services are just classes or methods off of classes, but the manner in which they're interacted with is different. So I see them as related, but different questions.
my reading of OP clarification is, it's not about web services:
@ThorbjørnRavnAndersen a class comprised of multiple related methods and usually sitting between the controller and the dao — Bob Roberts May 25 '12 at 0:20
side note your guess isn't bad by the way, I also sensed a scent of WS there. This didn't stop me a bit though, what I learned from own experience with web services is advice to try POJO is good idea always unless the question is 200% clearly and unambiguosly specified as WS.
> This one is really unit vs. integration testing whereas the dup is "what sorts of testing are there?" And that doesn't go into the level of detail the unit vs. int question could have generated.
That one, we probably just have to agree to disagree. Your reasoning makes sense to me, but the thing is my priorities / preferences here are seem to be different from yours.
To me, this is a closure of a broad NC as a dupe of an even broader NC. Another thing that played an important role in my decision is that dupe target has pretty good introductory-overview kind answer,
A: Software Testing Techniques

JB KingBroad categories to my mind would be: Black box testing. You don't get to see the code and are just testing blindly to some extent as what is in the application or system is hidden from you. Thus in this case people don't know all the error cases and have to guess with various boundary conditi...

your note on possibility of going into the level of detail the unit vs. int question could have generated looks spot on, I only feel differently about it.
Prevention answers explosion was yet another reason I voted to close. As they say in faq,
> If you can imagine an entire book that answers your question, you’re asking too much.
> Another related, but different set here. The question I linked is really focusing on when to use. The duplicate question (to design...) is kind of "design patterns, good or bad?"
> This is looking at the broader question of "best" practices in DB app design whereas the dup is focusing on business logic in the DB layer. The dup is a subset of the broader question being asked.
To be honest, 3) and 4) closures are slippery to me as well, and 4th one exactly for the reason you mention. I would prefer these to be closed the other way round but couldn't vote the way I wanted, because of expired older votes.
My plan was like, suggest dupe close in the opposite direction and if reviewers agree then later flag to moderator to revert closure direction.
I am still considering flagging like mentioned above, but figuring flag message turned out trickier than I expected. Need to find convincing wording, that would also fit in 500 chars limit of the flag message :)
2 hours later…
11:08 PM
Regarding your other note...
> I think it's better to close as NC or NARQ when that's the case instead of finding a "duplicate" that's a bit of a stretch.
well my take on it is rather opposite. We spoke about this with Jimmy Hoffa a while ago
in The Whiteboard, Apr 17 at 16:57, by Jimmy Hoffa
@gnat I 100% disagree with Karl's claims of dupe closure being the most helpful also, I'm of the opinion the only time you should close as dupe is when the Q is one you want people to be aware are on-topic good questions. If it's a bad question closed as dupe it just tells people the question has been asked twice, so it's definitely in scope and good quality
My position hasn't changed since, I prefer dupes to other close reasons...

preference for duplicate over other close reasons

Apr 17 at 18:28, 8 minutes total – 19 messages, 5 users, 4 stars

Bookmarked 25 mins ago by gnat

It is probably worth noting that besides "user friendly / less whining" benefits of dupe closures, I also prefer dupe-closures as these seem to be of greater help in reopening. It has happened to me that I voted reopen after clarifications made by OP, it feels really great when this happens
Duplicate target question provides something concrete for OP to start off in making their question easier to understand. When they clarify stuff thinking they are just trying to explain how it differs from dupe, it usually turns into more general clarification, into adding details that make question better.
You see, other closures are like telling OP "your question is crap". While probably true, this is quite general. As opposed to that, reasonable dupe (even into other NARQ/NC) tells them like, "your question is green crap". If they respond with objection like "no it's not because it's red" they're getting one detail added to their question...
...then someone responds like "no luck buddy, one of the answers in target dupe talks about "red hard crap". If OP responds like, "no it's not it because mine is about liquid" they're making next step in explaining details that were originally missing...
...To complete my imaginary story, someone could say, "still dupe, there's an answer saying red liquid crap, no fun" and OP objecting like "No, wine certainly can be fun" we've got the puzzle solved. Oh so you're asking about wine! vote reopen follows
For the sake of completeness, I've seen the opposite, the clarification that made it obvioous that question is crap ("EDIT: What books do you recommend to learn blah blah"). Fortunately this happens rarely, much more frequently things go like turning crap into wine - that is, when OP puts some effort into studying the dupe
11:28 PM
That said, I indeed welcome negative feedback on my dupe suggestions. Thing is, my preferences on stretching dupe closures have an aspect that worries me, namely these weaken my sense of duplication. You see, the more I stretch, the weaker my judgement is getting...
...without proper control, this may eventually led to completely insane dupe votes, which kind of trashes the whole purpose of what I am doing...
I mean, it's OK to me to suggest a stretched but reasonably related dupe, so that they can look through it to maybe better understand specifics of their situation. But when dupe isn't even close, it will be useless for OP.
That's why I try to learn when my dupes are too arbitrary. I may still stick with my close vote for the reasons explained above, but I want to at least to know that this vote is a stretch

  last day (16 days later) »