-8
\$\begingroup\$

The question in question is this. I specifically relate to the edit #2. I guess the intentions of the moderator were to improve the post, but I am unhappy about the way it looks now. The headlines are kitschy, and the title is flat.

There is some soul that I put into writing. I see a post as a synergetic entity; one cannot just edit one part of it and disregard the way it plays against the rest. This edit thus harms me. It is especially unfortunate that the editor did not leave a note so I may not even address them to reconcile this predicament.

Was it indeed necessary for the moderator to apply the edit? Is it not objectionable by a mere user? To what extent is a question I pose mine, and to what extent it belongs to a highly reputable passer-by?

Overall, I must say I am frightened by the zeal of the moderation comparing to Stack Overflow. I am not willing to post questions only for them to be trimmed to an informal standard. I am a person.

\$\endgroup\$
  • 3
    \$\begingroup\$ that edit trimmed tons of unnecessary fluff from your title and made the "headings" over codefiles (you used bold markdown) into non-monospace headings. If you want to get the original look back, you should just put the three headings in backticks. \$\endgroup\$ – Vogel612 Mar 6 '18 at 20:40
  • \$\begingroup\$ I would ask a down-voter to reveal themselves and leave a comment to inform me of the underpinnings of their unfavorable judgement of my question, so that I may obtain knowledge beside grief. \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 20:51
  • 3
    \$\begingroup\$ First things first: downvotes on meta signal disagreement (at least generally speaking). They're not generally an indication of question quality... Re the remaining points: Nice stalking, I'd prefer to go by my moniker and I'm writing up an answer now... \$\endgroup\$ – Vogel612 Mar 6 '18 at 20:53
  • \$\begingroup\$ @Vogel612 I removed the offending comment. This harm from my side is unintentional, and I apologize deeply. \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 20:59
  • \$\begingroup\$ @Vogel612 May I ask you this: is every highly reputable passer-by entitled to judge what is necessary and what is fluff in my question's titles? And, did you pay any attention to the second paragraph of my question at all? To the second part of your comment: I don't have a problem reverting edits and fixing markdown; I am proficient enough in both. But I thought edit wars are discouraged, and this is why I put forward this question instead. \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 21:00
  • \$\begingroup\$ No harm done, it's just a preference. Thank you for respecting it :) See my answer for a more in-depth discussion \$\endgroup\$ – Vogel612 Mar 6 '18 at 21:09
9
\$\begingroup\$

Let's check the edit diff and see what it's about:

title diff

+ Type safe, easy to understand matrix multiplication
- Here is a type safe, easy to understand matrix multiplication. Is it correct? Is it actually as type safe as I think?

The questions in the title are too much text to be displayed correctly on the front page. They also do not make your question title more indicative of what the question is about.

These are questions that should generally be formulated in the question body, because they are specific concerns.

body diff

+ ### Data.Nat
- *`Data.Nat`*
...
+ ### Data.Vector
- *`Data.Vector`*
...
+ ### Data.Matrix
- *`Data.Matrix`*

These changes are accessibility changes. As of the original post body, these pseudoheadings were bold monospaced text. With the edit, they are level 3 headings. That has some advantages. For one, headings are often converted to fragment addresses (on SE that is not the case yet). Furthermore they are handled differently by screen readers.

Lastly they are semantically more correct and additionally they are automatically bold.

Assuming you really liked the monospaced look, you can improve upon the original edit, by explicitly marking the heading text as monospaced like this:

### `Data.Nat`

Now to your questions:

I see a post as a synergetic entity; one cannot just edit one part of it and disregard the way it plays against the rest.

True, but this edit is hardly worth mentioning, IMO. It significantly improves the question presentation on the front page and slightly improves the presentation on the question page, without even touching most of the post.

This edit thus harms me.

If that is what you truly feel, then you should roll the edit back. As you correctly note in the question comments here on meta, rollback wars are not something you want to get into, though.

I consider the changes to be minimal and very noninvasive.

Was it indeed necessary for the moderator to apply the edit?

I'm pretty sure that the moderator only acted in their capacity as a user here. As such the question should be "was it necessary for a user to apply the edit".

I'd say: Yes. It significantly improves the question title and on the way also fixes some minor problems in the post body.

Is it not objectionable by a mere user?

Why would it not be? I mean it's a pretty clear-cut good edit, insofar as that it accomplishes the goals defined in the editing guidelines.

To what extent is a question I pose mine, and to what extent it belongs to a highly reputable passer-by?

hmm ... to some and to some. The question itself is yours, but SE explicitly encourages other users to improve your question. Whether that entails spelling, formulation updates or other improvements to the question structure.

I am not willing to post questions only for them to be trimmed to an informal standard. I am a person.

There is a pretty formal standard. I'll link to it again, just in case: Editing Privilege Explanations.

Yes. You're a person. You're not your post. An edit to your post is not an edit to your person.

\$\endgroup\$
  • \$\begingroup\$ I have to make some footnotes. (1) It was not monospace bold, but monospace cursive, and this is because these parts are auxiliary and should only draw a small bit of attention. In no way they are headlines; they are captions. They do not represent major partitions of the thought. This edit thus creates false structure. Consider yourself: are these important thoughts that received so much highlighting? \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 21:19
  • \$\begingroup\$ (2) The questions in the title − I consider them very useful. The reader may proceed from the title straight to the code, skimming the actual text paragraphs. They will know what I want them to tell me from the start. If the editor moved the extended title to the body of the post, with appropriate highlighting, I could have understood. But they simply trimmed the title, thus reducing the meaning of my post! \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 21:27
  • \$\begingroup\$ (3) Concerning length, I don't believe my question looked inappropriate with its longish title. If there were a limitation, it should have been enforced by technical means; but the input field is spacious enough and the titles wrap nicely. I don't believe a questioner is actually restricted so severely in the length of their titles. \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 21:28
  • \$\begingroup\$ (4) There is a formal standard that this edit does not match. The changes are at the same time lossy, uncareful, and trivial. Indeed, if this is all that had to be changed, it must mean the question is good enough. In particular, there were no spelling errors, the meaning was clear, there were no minor mistakes to be corrected, and no related resources to be added. As I showed previously, the supposed improvements are rather debatable. \$\endgroup\$ – Ignat Insarov Mar 6 '18 at 21:34
  • 2
    \$\begingroup\$ @IgnatInsarov (2) I agree that at least the second question in the title should be in the post body and have therefore edited your question to add that. Regarding the question "Is this correct?" I chose to not include that because it may sound like you haven't tested the code. You could edit and ask if this is "good practice" however, or another similar formulation - but I recommend against asking "Is this correct?" If it works, then it is one way of doing it and can therefore in a way be considered "correct". \$\endgroup\$ – Simon Forsberg Mar 7 '18 at 9:36
  • \$\begingroup\$ @SimonForsberg Considering the code is in Haskell and uses some advanced type level manipulations − in essense, automated proving, − it may respond to tests well, but still not be correct in the sense that the supposed proofs are false. The reuse of the same functions in less friendly circumstances may then break the behaviour. In other words, the code may permit what I want it to, but also permit something I did not think about. I believe this definition of correctness is intuitively clear to proficient Haskell programmers, familiar with the type system tricks I attempt to peruse. \$\endgroup\$ – Ignat Insarov Mar 7 '18 at 9:45
  • \$\begingroup\$ @SimonForsberg That being said, I am happy with your change. \$\endgroup\$ – Ignat Insarov Mar 7 '18 at 9:46
1
\$\begingroup\$

Is it correct? Is it actually as type safe as I think?

"Is it correct" should never be used in a title here. Titles here should state what the code does. If someone feels that it is not correct, then we'll tell you. That question is implied everywhere, so it needs to be stated nowhere. This is also a sensitive subject since we don't review non-working/buggy code. So if it's not correct, we shouldn't review it.

If you want a specific question answered (type safety), you're on the wrong site. What we would do is a general review, possibly paying extra attention to type safety. Or possibly not.

I think that a lot of the confusion is that you're coming from Stack Overflow. They are entirely different. On Code Review:

  1. A question cannot be marked as a duplicate unless the code is identical. On Stack Overflow, it's the question that needs to be essentially the same. Even quite different code can be marked as a duplicate.

  2. We encourage more context. Stack Overflow wants a MCVE. We close MCVE questions.

  3. On-topic answers here are reviews of the code. The requirement is that there should be at least one insight about the code in the answer. We don't allow askers to limit that. Stack Overflow is all about the question; the code is only to support the question or answer.

These differences lead to us having relatively specific standards about titles. A title should describe what the code does briefly and to the point. Occasionally we allow funny titles, but we never allow titles about correctness or specifying a type of review.

If that edit had been proposed by a regular user, I would have accepted the change to the title. That's in line with site standards. If you think that particular piece of advice should appear somewhere specifically, that would make a good meta post (Please add title requirements to ...). You may want to search to see if someone else has made the same suggestion previously. There may of course be an actual reason for not doing it.

I disagree with the change to the headings, as they are now not semantic HTML (there's no second level heading; it jumps straight to third level). But there is a standard of putting file names in bold, not italics. I'm not familiar enough with Haskell to say how that applies to your code. But if it were Java, each class would be in its own file and the filename would be listed in bold. If those are not separate files, then it's not clear to me why they are in separate code blocks.

\$\endgroup\$
  • \$\begingroup\$ If you consider that "correct" may not equate with "working", you may see why "correct" sometimes bears extra meaning. A piece of code may work in most cases yet still be incorrect in a sense that it fails in some corner cases, or that it may be used in a way that the author meant to disallow because it is unsafe, or just that it is not proven to be correct. I could argue that most of the code you see in Linux kernel is not correct (try passing a few null pointers here and there), and even that Intel chips happen to be incorrect (due to recently discovered branch prediction exploits). \$\endgroup\$ – Ignat Insarov Mar 16 '18 at 4:47
  • \$\begingroup\$ In particular, my remark on correctness was supposed to point out that I know my code appears to work, but I doubt my understanding of its correctness in the sense that the type system should prove it, but I may be using it clumsily, and the proof, while valid, may happen to be unrelated to what I actually meant to prove. \$\endgroup\$ – Ignat Insarov Mar 16 '18 at 4:51
  • \$\begingroup\$ Can you point me to the standard of putting file names in bold, and specifically in heading bold? Those are separate files, but I do not believe there should be a standard that tells me how to format file names in my posts. \$\endgroup\$ – Ignat Insarov Mar 16 '18 at 4:52
  • \$\begingroup\$ I could also make a case that, insofar as I can put a sentence X (for example, "is it actually type safe?") somewhere in the post, I can put it in the title, unless there are arbitrary, non-constructive restrictions. (Which I believe the restrictions I observe here to be.) After all, a title is not merely an extended tag, and at the same time not a separate entity, but rather a specially emphasized part of the post. And it is the author who gets to decide what to emphasize in their post in order to express their thoughts best, is it not? \$\endgroup\$ – Ignat Insarov Mar 16 '18 at 5:03
  • \$\begingroup\$ In other words, a post B that contains the same plain text as post A, but with emphasis on different sentenes, may actually express an altogether different thought. So, if a post A is changed to a post B by moving emphasis, it may do harm. In particular, some subtle but important idea may be lost. It is true that, perhaps, such a post should be edited to make a subtle but important idea appear more prominent and clear. But this is not what was done. \$\endgroup\$ – Ignat Insarov Mar 16 '18 at 5:07

You must log in to answer this question.

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