Respecting poster's style

Question

When editing a question or answer, one usually fixes also occasional slips in capitalization, for example "Latex" into "LaTeX" or "Tikz" into "TikZ"; frequently one sees "i" for "I".

If I interpret correctly the site's tradition, a message should not be edited just to fix misprints or minor glitches (except, perhaps, in titles).

A recent answer by a very respected member of the community was edited twice in order to "fix" her peculiar style and twice rolled back (once by myself, the second time by the OP).

I think we should respect a poster's style, particularly in this case.

Answer 1

I think I remember a comment of B/barbara somehwere, in which she stated she only uses standard capitalization whenever she's writing on behalf of the AMS. I'm not sure whether or not this is a reason to deviate from this site's style so decisively, but I'd recommend she add a note to her profile about her doing so. Also, it'd be good to a notice (or a link to this discussion) when rolling back an undesired capitalization edit to a previous version. To do so, you need to use the [edit] link at the revision you want to reinstate instead of the [rollback] link. You're still effectively doing a rollback, but you can add an edit summary explaining the matter. This way, future capitalizations can be avoided. All this is assuming the not-capitalization is condoned.

In general, you're in line with When is (and isn't) it acceptable to edit?: Only edit for "Fixing typos or grammar errors" if the question is on top of the list anyways (or if you're editing for another reason already). However, if a question is really terribly full of spelling mistakes, I tend to edit it anyway; I like to keep the Wikipedia aspect of tex.sx in mind, i.e. as a reference site beyond just helping one particular user. So without a reason like B/barbara's, I might well edit a post (of decent length) that doesn't have any capitalization.

Answer 2

I'd like to sound a note of dissonance.

I understand [Bb]arbara's reasoning. But I think that it is flawed for two reasons:

1. It is not clear from a post itself that there is a potential for two voices. A post with no capitalisation simply looks like a post with no capitalisation. One has to look further, and know what to look for, to understand the reason behind it. So people, trying to be helpful, are going to be continually editing her posts to capitalise them. My guess would be that there are very few occasions on this site where [Bb]arbara needs to put on her Official AMS Hat (here's a chance for Paulo to show off his graphic skills again) and Declaim With Authority - indeed, I can think of only one occasion where this was needed ("Is it legal to TeX the TeXBook?"). On all other occasions, the fact that she works for the AMS is useful information (her answers on AMS stuff are more trustworthy because of it) but at no point would I think that there was a risk that her posts would be viewed as some official AMS viewpoint.

2. It makes the post very hard to read! The people editing it do so for a reason. As web browsers aren't as sophisticated as TeX, the need for proper grammar is all the more important as all the line spacing, punctuation spacing, and so forth is hopeless. I agree that small grammatical changes should only be made if bigger edits are going on, or if the post is near the top of the front page anyway, but there were a lot in that post. I would have considered editing it.

   The conjunction of this with Yiannis' answer is a happenstance that I cannot pass by. Imagine if barbara had written Yiannis' answer. It would have begun:

   "dear polish friends, i'm doubly sorry"

That's not to say that I think that barbara should properly capitalise her posts. But I think that we have a higher duty to make the site clear and understandable than to respect any particular person's quirks so if someone edits them just to put in the capitalisation, then actually I would let that stand. To ensure that it is clear that barbara's posts are her own and not the AMS's then a short disclaimer in her profile page would suffice. Anyone who knows that she works for the AMS will check there, and anyone who doesn't know that will not be in danger of getting confused anyway. Moreover, I think that the default assumption on the SE sites is that people speak for themselves and not their employers. These aren't "Ask an Expert" sites, they're "Ask a Fellow Traveller". A fellow traveller might be able to tell you the best pubs from their home town, but you'd never mistake them for the mayor. On the other hand, knowing the best pubs is more the sort of thing we're interested in here than the statistics on how much Real Ale is sold in comparison to Fake Lager[1].

[1] A tautology for effect.

Answer 3

I would like us to keep in mind, that we would like to respect the reader.

Once written, hundreds or thousands of times read. It's good for the site and for the users and visitors, that the posts are as readable as possible. This includes writing in English as it's the language of this international site, capitalization as it's common practice, and proper punctuation.

I respect Barbara and also her writing style, and I also respect our readers. And I respect if an editor makes a post more readable for the public. The site is like an online publication, rather than a mailing list or discussion board. Is writing everything in lowercase allowed in the Wikipedia? In the PracTeX online journal? In TUGboat?

If we start forbidding edits to lower-case style, we consequently might have to approve all capital letter style as it's just the same other way around. It could go even further — to approve omitting punctuation.

Barbara has developed her style to let people distinguish the role she's using when writing. However, only if the reader has read her profile, or somebody told him, can the reader know that. A safer way would be a small information in official posts. As Andrew said, that we are speaking for ourselves is the default assumption.

My compromise suggestion would be both to respect lowercase writing (though it might be hard to decide if it's writing style, like Barbaras, or if it's just careless writing by anybody else) but also to respect if an editor matches the post to the site's style and the public style. Editing and improving posts is an essential feature of the site.

Answer 4

Given that this seems to come up every now and again, generally brought up by users from Stack Overflow who wish to enforce the Right Way™ of doing things at TeX.SX,¹ I thought it might make sense to try to provide a more comprehensive answer, taking into account some of the discussion that has ensued, in particular the discussion in:

1. Why were my edits rejected? and
2. Consensus vs accepted answers on meta

There are three overarching points in this post. I've demarcated these three points with section headings.²

A positive argument based on copyright and attribution

First, I think Paulo Cereda has already made a great case for respecting a poster's style in his answer to Why were my edits rejected? given reasons of copyright and attribution:

I see the SE network as an awesome Q&A knowledge base, a new tendency. But I fear that people misunderstand the real concept and see threads as collaborative wiki. If it's a wiki, everything is fine, because you are collaboratively editing stuff. The Q&A model is based on people asking and answering questions. The site footer gives us a hint:

user contributions licensed under cc-wiki with attribution required

If we take a closer look on these links, we find the following statement:

Attribution - You must attribute the work in the manner specified by the author or licensor (but not in any way that suggests that they endorse you or your use of the work).

When we edit things from other users, we are signing a risky contract: we are somehow influencing the author's contribution - for better or for worse. But note that it's up to the author to decide if the modifications to his/her answer should really take place.

A negative argument based on readability and comprehension

Now, Paulo does go on to mention that we should nonetheless keep in mind that edits improving spelling and grammar are probably Good Things™.

So we can ask whether editing barbara's answers to fix capitalization would fall into the category of improving spelling and grammar. The answer is no, at least not if we take improve to mean something like 'make more readable, make more comprehensible, etc.'.

The fact that writing systems are technological inventions is an underappreciated fact. They postdate human use of language, and they are, ultimately, nothing more than technological inventions that encode a set of conventions that we use to represent the language in our heads.

Now, certain conventions, such as consistent spelling, are arguably useful. Conventions that demarcate sentence boundaries are also arguably useful. In fact, they reflect prosodic cues that one gets in spoken language.³

The conventions that are used by this particular technological invention for encoding English happen to employ two methods of demarcating sentence boundaries: capitalization and punctuation. So, in a sense, we have some redundancy.

There are only a few edge cases where the use of these two methods is not redundant. Specifically, in cases where the conventions used for transcribing English employ a period to mark an abbreviation, such as in "et al." or "etc." we might see non-redundancy. An example of this might be:

in this case, you don't want to use the commands like \bf, \it, etc. you should use \textbf instead.

Here, it's not immediately clear whether the second sentence is a separate sentence or not, just from inspecting capitalization and punctuation. Nonetheless, we can still figure out what was said, and it's pretty easy to determine, based on context, what the sentence boundaries are. One user has claimed that:

Barbara's posts as they currently stand are well-nigh unreadable gibberish.

This is patently false. Any user that can read and comprehend English can read and comprehend barbara's answers. Though I tried searching the internet for studies on whether there is any empirical evidence documenting a decrease in comprehension/readability based on lack of sentence-initial capitalization, I could not find anything.⁴ Nonetheless, I have an intuition—an intuition that I suspect others share, if they are honest with themselves—that there would be no significant difference in comprehension, should someone actually bother to conduct an empirical study.

Now, tohecz worries that this will lead to peOPlE adOPTing TeEN caSINg. First point to note: it is now 2014 (no longer 2012, the time of tohecz's comment), and, as far as I am aware, there has been no instance of this. Second point to note: there is actually an argument to be made that fixing someone's post that uses "teen casing" improves readability/comprehension. Again, I haven't found any empirical studies, but I have an intuition—as well as the phenomenological experience—that reading "teen casing" is harder and takes longer than reading normal casing. I suspect this would be confirmed with an empirical study, should someone really want to waste their time carrying out such a study.

Thus, this warrants editing the ridiculous behavior of the user who asked Consensus vs accepted answers on meta because they eventually edited the question to include this:

Specifically, editing such a post would improve the comprehension and readability of the post for at least two reasons:

1. the user does not conform to conventional English spelling (which makes it particularly hard for non-native speakers of English to understand); and
2. the use of code/keyboard markup might mislead one into thinking that STYLEZ, for example, is a tag or a LaTeX package or something, a hypothesis which is probably easily dismissible, but might nonetheless confuse new users on the site

So, the "consensus"⁵ on TeX.SX at the time of writing seems to be to respect barbara's style—which is readable and comprehensible—not any style. That is to say, there is not a rational argument in favor of not respecting her particular style.

A positive argument from changing the meaning of an answer

Nonetheless, if one is able to delude oneself into thinking that lack of sentence-initial capitalization somehow really turns some writing into "well-nigh unreadable gibberish",⁶ there is a further positive argument for respecting barbara's style, in addition to the one already sketched by Paulo. This one comes from David Carlisle in a comment:

@DavidMulder You are ignoring the fact that your edit was simply a bad edit, irrespective of any general conventions about fixing grammar. It changes the meaning of the answer (given the statement on the user profile). Perhaps you think that's a bad convention for a user to have but making all following readers believe that a statement is an official policy statement of the American Mathematical Society when it was explicitly marked as not being so in its original form is a bad edit and should be reverted.

The moral of the story

So, even if you don't buy the negative argument I've sketched (for whatever reason), we still have at least two positive reasons (actually, more than two, cf., fn. 2) for why we ought to respect barbara's style, which is not to say that we ought to respect any style, if that style really makes a post harder to comprehend or read (especially for non-native speakers of English). In such cases as those, editing is then probably the Right Way™ and a Good Thing™.

---

1. Indeed, percusse counts seven such users.
2. The three points in this answer are tailored to the particular point under discussion: the poster's style (in particular, barbara's style), so I will not mention any of the points that were already made in When is (and isn't) it acceptable to edit? one of which is a further reason not to edit for spelling/grammar 'errors' because "[e]diting a question/answer for such small issues has the undesired effect of bringing the question back to the homepage, which can become a nuisance for frequent users of the site".
3. Prosody is just a fancy way of saying "intonation". Intonation is often used to demarcate sentence boundaries, among other things, in spoken language. The specifics depend on the language.
4. Probably because such a study sounds like a pointless waste of time.
5. I do not like using this word to describe the state of affairs on Stack Exchange. See my answer here.
6. Or, alternatively, if one is so attached to prescriptivist ideals that one needs to change the definition of improve to be something like 'conform to redundant conventions' in order to rationalize their vehement defense of editing all of barbara's answers.

Answer 5

The overall quality of the site shouldn't suffer just for the sake of preserving one individual poster's style. If I decide to write all my posts in ALL CAPS would that be okay? How about 133t 5p34k or Spanish? What about if I decide that using a signature or tagline is a part of my personal style? No, none of those things are acceptable.

By rejecting and rolling back perfectly good edits you're discouraging users from actively improving the quality of your site.