Often there is a quoted question (e.g. a block quote with >) from the textbook with certain typographic flaws which I would edit without any doubts, among which are the following edits where no info is added or removed, but the literacy is improved to meet current standards:

  • Use of similarly-looking symbols: $G^Θ$$G^⦵$;
  • Math mode for chemistry: $CH_4$$\ce{CH4}$;
  • No space between units and numerical values: $4^\circ\mathrm{C}$$\pu{4 °C}$;
  • En-dash or hyphen in place of the minus: $\text{-}x$$-x$
  • etc.

However, my recent edit to the question Which “exotic salt” can lower water's freezing point by –70 °C? is considered by OP as inappropriate. Editorial parts that were received negatively are:

  • Replaced en-dash - with minus for the negative values.
  • Added space between number and the unit (°C).

A screenshot of the comments sections:


So, how ethical is it to alter the quoted text in order to meet the standardized typography and assure literal writing? Should the original text be left untouched? And what to do if the source couldn't be preserved exactly, for example if there were a Unicode symbol or a sequence that cannot be identically replicated due to the set of tools available for the formatting of Chemistry.SE?

  • 4
    $\begingroup$ It's a good question, thanks! ;-) $\endgroup$
    – uhoh
    Commented May 5, 2019 at 5:07
  • 2
    $\begingroup$ data.grammarbook.com/blog/definitions/sic Using this after each incorrectly formatted equation is one way to indicate that the editor who typed the information in was aware that something was amiss in the original text. My personal feeling regarding the actual issue is that, almost always, verbatim quotes should remain verbatim. $\endgroup$
    – jonsca
    Commented May 5, 2019 at 7:17
  • $\begingroup$ @jonsca Gut [sic] p01nt [sic] :) $\endgroup$
    – andselisk Mod
    Commented May 5, 2019 at 7:25
  • $\begingroup$ @andselisk Haha, nice! Great that you brought this up, thanks! $\endgroup$
    – jonsca
    Commented May 5, 2019 at 16:42
  • 4
    $\begingroup$ I view such corrections as superfluous but an interesting tutorial on how to do the markup correctly. Frankly I think you'll just wear yourself out if you want to be the markup cop. $\endgroup$
    – MaxW
    Commented May 5, 2019 at 21:16

4 Answers 4


Generally, we cannot avoid changing the typography of any quoted text because we do not use the same fonts, font sizes, distances, line breaks, colours, etc., and because we might have different technical limitations. Therefore, any quote is always a new typographic representation of the original text. A rule like “no typographic changes to quotes” is practically impossible. Nevertheless, we really might need to define a limit of the permissible typographic changes.

When looking at all the available sources in the literature or on the internet, we can always find many typographic mistakes. However, there is no value in preserving such mistakes. Unfortunately, we cannot fix all the mistakes that we see; nevertheless, we can contribute to the efforts that such mistakes are not spread further.

Usually, when looking at a quote in a post on this site, we cannot be sure whether a typographic mistake was already present in the original source or was just introduced by the author of the post.

We are responsible for the correct typography on this site. We should make sure that our typographic representation of the quoted text is correct.

You should change any clear typographic mistakes (i.e. how the quote is written down), but you must not change what the original says (i.e. when you read it out) and means.

For example:

  • Change “tri-choloro-methane” to “trichloromethane”, but do not change it to “chloroform”.

  • Change “300°C” to “300 °C”, but do not change “reaction at 573.15 K” to “reaction at 300 °C” even if it would be a reasonable change of the implied precision.

  • Change “a mixture of ethanol & water” to “a mixture of ethanol and water” but not to “a mixture of water and ethanol”.

  • Correct any missing or surplus space.

  • Correct wrong hyphens, dashes, and minus signs.

  • Correct wrong use of italic or upright fonts for symbols or nomenclature.

Nevertheless, do not change the typography of a question if this typography is the actual topic of the question (e.g. “Is F the correct unit symbol for degree Fahrenheit?”).

  • $\begingroup$ Hmm, it seems like that's pretty much what I did with my edit. Would you suggest to leave a comment regarding the editorial changes that have been made (or [sic], as @jonsca noted), or the SE edits tracking system would suffice? $\endgroup$
    – andselisk Mod
    Commented May 5, 2019 at 17:05
  • $\begingroup$ The SE edits tracking system will always show what changes you made, although the system doesn't allow swapping back and forth from the markup and the rendered versions. // If the OP reverts back then fine. // So far as I know the site doesn't have a style guide. So I view it as nit picking to make such small changes. $\endgroup$
    – MaxW
    Commented May 5, 2019 at 21:48
  • $\begingroup$ I never had a problem with editing per se, it's the editing without some kind of indication visible to the reader (irregardless of SE's edit history feature) that changes have been made concerns me as explained further here. A careful read of my comments on the original question shows this, though it wasn't captured in the question. Would you say that all of the above ("defined limit(s) of the permissible typographic changes") can be done without a visible indication to readers that edits have been made? $\endgroup$
    – uhoh
    Commented May 6, 2019 at 0:33
  • $\begingroup$ As a partial mea culpa I did a quick check and noticed that while I started off in Stack Exchange indicating when I'd bold-faced a short segment of a long block quote with something like "(emphasis added)" (other users sometimes say "my bold") I have lapsed several times recently, leaving no indication that the boldfacing was added. That isn't covered under typography correction, but I wanted to put that out there before I get zinged for it. (example) $\endgroup$
    – uhoh
    Commented May 6, 2019 at 1:21
  • 1
    $\begingroup$ @uhoh - Again assuming the changes were some innocuous markup changes, why not? If you want the exact quote then use an image of the quote. That is a different problem since text in images is not searchable. // I do think that the system itself leaves a visible indication that the post has been edited in that the last editor's name is shown above the OP's name. You can always check the edit history. // I hate to sound crass but it isn't like we're editing the audio tapes from Nixon's Watergate recordings. I think improving readability and meaning is fine. $\endgroup$
    – MaxW
    Commented May 6, 2019 at 1:31
  • 1
    $\begingroup$ @MaxW there's nothing wrong with having a good clear SOP now, to better address the Why not?'s of the thousands of other users, some of which may be quite generous. I like this answer, just exploring it a bit further. $\endgroup$
    – uhoh
    Commented May 6, 2019 at 1:40
  • $\begingroup$ @Loong "Correct wrong use of italic or upright fonts for symbols or nomenclature." What if the author uses a wrong but consistent typography and we would make a mess by doing it right ? $\endgroup$
    – Poutnik
    Commented May 10, 2019 at 4:00

I suggest let's look to journalism

(no, not the tabloids - but the concept)

First, editing another users post for typos and typesetting is... working as intended I would say. The issue is when we quote 3rd party sources - that are not likely to respond for clarity.

It is common practice in the field of good journalism to try to preserve when citing a source. Preserve the intended meaning of what was spoken, or written. Preserve the identity of the source, preserve his language, etc. A French newspaper "fixing" the Canadian French of a Quebecois would be unethical. So, while preserving you can certainly shorten sentences, but their literal meaning must be in line with their intended meaning. That is a responsibility you accept when you edit. Often, editors will add for clarity in square brackets pieces of sentence that needs to be reshuffled when editing. The use of the literal [sic] is slightly arrogant to me - it merely points out that the source made an obvious error and no effort could be undertaken to fix it. But it is hard sometimes, when a persons way of speaking is part of the message, part of his language. So - [sic] is sometimes useful - but not as a general rule.

For this particular instance - I would have prefaced the blockquote where it is cited with [edited for typesetting]. As long as the original script is available and linked, I would feel rather liberal about doing some cleaning - as long as it is communicated to be so.

I quote you, your last paragraph as example.

[What] to do if the source couldn't be preserved exactly [...] due to the set of tools available for the formatting of Chemistry.SE?

My suggestion is: We preserve as much as we are able to, but we correct the typesetting - as long as we notify this with edit brackets where applicable. We should really take care if we are quoting a source that is not linked or open - because the whole idea is that the reader should be able to make up his mind based on the source information - our edits are purely for reader comprehension (...and to be a good example of proper typesetting, if I dare say so).


Looking at the Chicago Manual of Style, there are a couple of sections on permissible changes to quotations. Section 10.7.4 says that in a passage quoted from a modern book, journal, or newspaper, obvious typographical errors may be silently corrected.

Section 10.8 says that typographical style may be changed to agree with the style of the work in which the quotation occurs. Elements of typography which are the publisher's doing and not the author's choice do not need to be reproduced exactly.

My conclusion is that andselisk's changes are okay according to a widely-accepted style guide.

  • 3
    $\begingroup$ +1 for citing authoritative source $\endgroup$
    – uhoh
    Commented May 9, 2019 at 11:49

My feeling is that editing a block quote and not leaving an editorial mark of some kind to indicate that you've altered it is not proper editorial behavior, and falls into the category of good intentions on a slippery slope. Here's why.

If you look at my comments, I've simply suggested that altering a passage that is quoted without leaving some kind of editorial mark does not make the quote more accurate, and that you should not do that. There are editorial indicators to indicate that you've altered a quote.

Editing a users post constructively is fine, but modifying text by a third party that the OP has quoted without leaving some visible indicator is bad practice.

As I also mentioned in comments, it's a "slippery slope":

[@uhoh 1. The quote is more accurate than ever.]...(Which "exotic salt" can lower water's freezing point by 70 °C?)

The quote itself is now less accurate of course, even though one feels one is making the content more accurate. It's like thinking that throwing out an "obvious flier" in your data is making your data more accurate.

As other things in Stack Exchange, editing another user's post should involve a light touch, and editing their quote of a third party to make the way you want it to look and doing so without leaving any visible indication that the quote has been modified is a bad practice.

If it doesn't really need to be edited, don't. If it does, make it visibly clear that an alteration has been made. Stick to best practices.

visible meaning something that the reader can see while reading the altered quote, as opposed to it existing somewhere in an edit history.


You must log in to answer this question.

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