What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 129 Stack Exchange communities.

TLDR Post-Mortem I failed a reopen review audit because I neglected to extend the benefit-of-the-doubt to the OP of a relevant question about an interesting new language, but instead treated him like some noob who wants the rest of us to answer his basic homework questions. I might solve this by reviewing fewer questions, other questions, or taking a bit longer, perhaps stopping for a while when noise to signal seems too high, because maybe SNR isn't actually so bad, and I'm just fatigued.

-- Original post below --

Is there a way to pass a primitive parameter by reference in Dart? is a reopen audit review question, intended to test whether the volunteer reviewer is "paying attention", i.e. robo-reviewing.

Here is the body of the review audit question, in case some people can't see the link:

I would like to pass a primitive (int, bool, ...) by reference. I found a discussion about it (paragraph "Passing value types by reference") here: value types in Dart, but I still wonder if there is a way to do it in Dart (except using an object wrapper) ? Any development ?

This question features:

  • no source code, neither problematic code nor what he's tried to pass parameters by ref
  • a link, but no quotations from it that might help -- or at least avoid link rot

The close reason provided was "unclear what you are asking".

This looks like an invalid close reason, but so what, the question fails overall site standards. It looks like a classic lazy question. Although I'm unfamiliar with Dart, I would expect the language web site to have a FAQ and docs on something like how to pass parameters.

I voted "leave closed" and was then given the canned admonition:

STOP! Look and Listen.

This was an audit, designed to see if you were paying attention. You didn't pass. There are no major problems with this question. You should click Reopen, or make minor corrections via Edit and Reopen.

Don't worry, we've already handled this post appropriately – but please take a minute to look it over closely, keeping in mind the guidance above.

But I respectfully disagree.

This seems to be a correct vote.

The question fails the off-topic/provide your code test and the old "too localized" test (by failing to extract any goodness from the cited link that might help a future visitor).

There is a long discussion in comments. I did not read them all, there is a lot of noise to filter, and it only promotes the idea that maybe there was something unclear in this question. At best the OP has told us that he read a google groups posting and got confused about a specific task or paragraph, that you, dear reader, get to go find. I can see some people thinking 'I cant read his mind. Show us/tell us what, exactly, you found confusing.'

So was this just a bad review audit question?

Are volunteer reviewers fully correct to vote on what they think would be best for the site, or is the reopen review to be narrowed to 'did he fix the previous issue?'

share|improve this question

1 Answer 1

up vote 8 down vote accepted

The close reason provided was "unclear what you are asking".

This looks like an invalid close reason, but so what, the question fails overall site standards.

I take issue with this.

The question is clear as to what it's asking - if one can pass a primitive value in Dart by references without wrapper classes - so the close reason is wrong.

I don't see how the site fails overall site standards, though. Nor do I see it being a "lazy" question per se.

I'm also not a fan of the flippant "so what" portion - but that's more a personal matter than a procedural matter, so I'll disregard it...

The question is asking something legitimate - if Dart supports passing primitive types by reference. The discussion linked to only hints at something like this, but doesn't fully answer the question.

This question features:

  • no source code, neither problematic code nor what he's tried to pass
  • parameters by ref a link, but no quotations from it that might help -- or at least avoid link rot

First, if one is asking a conceptual question like this, why would source code be relevant? Not all questions have to have source to be a good, on-topic question at StackOverflow.

Second, your point about link rot may be valid, but there are ways to rectify that - and that alone doesn't make the question bad.

This seems to be a correct vote. The question fails the off-topic/provide your code test and the old "too localized" test (by failing to extract any goodness from the cited link that might help a future visitor).

Where did that test come from? This is the first I've heard of it. (I do refer you to my previous comment about code in questions, though.)

I don't see how it could be "too localized", either - at least, not by your reasoning. The user coming in to see this question should get the context not only from the question they're reading (which may be missing a bit due to the potential for link rot, I don't disagree), but also from the answers, to which there is plenty of useful information for someone stumbling upon the question.

So was this just a bad review audit question?

I personally don't think it was. You may have gotten used to the reviews where the question is obviously bad and structured like that, but that only implies you didn't try to understand the context of the question as well as you think you should have. I kind of want to see more questions like this in the review queue - it's an example of a question that's valid, has an applicable answer to it, and is terse.

share|improve this answer
    
Does foo(bar) vs. foo(&bar)? become more interesting when it is in an esoteric language? Maybe that's what I missed. I agree that you are probably right on 'too localized', that others would benefit from the answer if for some reason it is difficult to pass by ref in Dart. –  Paul Aug 26 '13 at 4:43
    
Well, Dart isn't esoteric, going by the traditional definition of "not intended for general use", this is very much intended for general use. Second, I do feel that the question posed (foo(bar) vs. foo(&bar)) is relevant given that there was a discussion about it within Dart's community. –  Makoto Aug 26 '13 at 4:48
    
OK, how about 'newly popular' language? Wikipedia Dart says it was introduced in Oct, 2011. So it is still pretty new. Wikipedia also says it compiles to Javascript, which explains a lot about why you might only be able to pass a primitive type by reference by wrapping it in an object. That is, after all, what one would generally end up doing in JS. –  Paul Aug 26 '13 at 4:55
    
"As of May 2013, Microsoft Internet Explorer, Mozilla Firefox, Opera Software's Opera browser, and Apple Safari have no plan to embed a separate Dart VM." -- Wikipedia. Although it would still run when compiled to JS. But still somewhat specialized, I guess. At least Google is behind it. –  Paul Aug 26 '13 at 5:00
    
Thanks for providing your answer. At the least I'm more aware of Dart. –  Paul Aug 26 '13 at 5:03
    
There is another language that is compiled to javascript: Coffeescript. Also, there is emscripten. so, compilation to javascript doesn't make a language esoteric. –  Jan Dvorak Aug 26 '13 at 5:20
    
I'll hazard that if I have to use A to get B to work, then there is less demand for B than if B were self-sufficient. Here's some search engine numbers: Google has 1.48 billion hits for Javascript, 588 million for Java, 141 million for C++, 114 million for Python (including the snake), 99 million for Perl, 63 million for Dart (13 million for the car, 23 million for the lang), 1.5 million for Coffeescript, 110k for enscripten. Even so, I'm sure some brilliant work goes on in these niches, and Coffeescript has that clean psuedocode look that python often achieves. –  Paul Aug 26 '13 at 5:32
1  
@user137487 if A is 100% widespread, I'll take a guess and say that the dependency of B on A does not make B any less popular. In any case, nothing like this affects the fact whether questions about it should be closed for clarity. As long as there is more than one person or company using some language, chances are it belongs to Stack Overflow even under the old "too localised" banner. Questions about Dart are not unclear by default. –  Jan Dvorak Aug 26 '13 at 5:53
    
@JanDvorak Does a question about Dart (or C++/Fortran/etc) require the OP to demonstrate code he has tried? That was my concern for which I indicated "keep closed" on the reopen test. It would be great if there were a "keep closed, this post has other issues" button, but there isn't such a button. Now Makoto has said he disagreed with my action, that a succinct question need not include code. That's fine. What do you think? –  Paul Aug 26 '13 at 6:01
    
Personally, I think this discussion shouldn't be in my global inbox, but that's just my two cents... –  Makoto Aug 26 '13 at 6:14
1  
@user137487 if this is your concern, then the audit was a great success. Proof of effort are merely words to explain to newbies why their poor and heavily duplicated questions (or even homework) were closed. Questions don't always need source code as a proof of effort. Here the asker did show effort by linking the article he has found. Great (widely applicable, clear and not-heavy-duplicate) questions don't even need a proof of effort (though it still helps to show your research). Only close for the lack of research when it is obvious the asker is lazy. This asker isn't lazy. –  Jan Dvorak Aug 26 '13 at 6:47
    
@user137487 the question is: widely applicable (check), clear (check), not heavy-duplicate (check), not lazy (asker did research), useful (if someone is interested, they will find this question) –  Jan Dvorak Aug 26 '13 at 6:52
    
@user137487 here's an example of a question I found via google when researching my problem. This is the purpose of stack overflow, to create such questions (but not if they are already present). Naturally, I upvoted the question. stackoverflow.com/questions/621299/… –  Jan Dvorak Aug 26 '13 at 7:03
    
@user137487 "proof of research" is merely a way of saying "if you're too lazy to look for an answer, we are too lazy to find the canonical duplicate of this question." –  Jan Dvorak Aug 26 '13 at 7:05
    
Thanks, Jan, for taking the time to express this. I think you are correct. –  Paul Aug 26 '13 at 7:11

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .