Although it certainly makes sense to repair what appear to be "obvious typos", this is a slippery slope... so it is probably better to give an exact quote, and then the "sic" (and possibly an explanation...) Less efficient, perhaps, but more "traceable/replicable".

There's also the alternative to not quote, but paraphrase in your own words and give a reference.

A third possibility is to quote but instead of [sic] to bracket-replace the incorrect segment.

Not sure if this is answer-worthy yet, but it's worth noting that using [sic] has also been used as a shorthand way to question the intelligence of the person making the quote. As in, "This person is so stupid they can't even spell their own argument!" I can pull up a number of Washington Post articles where [sic] is used to that effect with a casual google search.

It's way too easy to screw this up. Think of all the people who have "corrected" the "obvious typo" of "iff". Please don't misquote people, even if you're trying to be helpful.

@GGMG, it's definitely answer-worthy. That was my exact thought. Please go ahead and write the answer. :) virmaior's suggestion above may integrate well with your observation as it avoids that problem.

Not sure if it's answerworthy, but "email the original author and ask them"? That way they can check you have got the sense correct, or alternately can explain why you haven't.

A fourth possibility is to forget that [sic] exists and simply don't modify the quote at all.

@GGMG Since when has the "Washington Post" been an citable academic standard? Give me the science section and I will find for you so much errors, misunderstandings and omissions that I can use the notes as crane counterweight. It does not matter what the WP or other newspapers are doing or if personal pride is challenged, accuracy is paramount.

switching a few words around is not a "typo". Could you cite the typo? Anonymity is guaranteed when citing such a short piece, no one's pride will be harmed but at least users will actually get to judge if it is effectively a typo, (which I wouldn't be 100% certain) or merely a printing error... IOW fixing an entire sentence is not fixing a typo.

@Mari-LouA: well, unless you're still using physical movable type, in which case switching a few words around can indeed be an actual typographical error. I'm not sure whether copy-and-paste errors can be considered a legitimate analogy to that class of ancient typo, but I certainly produce a fair few mangled sentences that way, and I think of them as typographical in nature rather than as (say) linguistic errors.

If you change a quote, you're no longer, well, quoting. This is the best and most obvious reason to never change a quote.

In this instance, I would probably paraphrase the author.

it's like the Buffet Warren billionaire said: "the more earn you, the more you drive up Hills in the Hollywood here."

As it is from a book it may be worth checking if there exist a newer edition where the typo is corrected.

For all the programmers here: what do you quote when the original quote says [sic]?

@Mehrdad, remember the [sic] goes outside the quotes, so one could say "Bob said, 'Alice siad, "hlelo" [sic]' [sic]" without much ambiguity, albeit without much beauty.

I think that @sgroves's point is the most important one. Out of respect for the author, and academic integrity, one uses quotes to indicate only quotations, with emendations as necessary. I think that re-phrasing it makes it clear: "I want to cite a paper that has a wrong result. Would it be OK if I silently fixed the result, and cited that?"

@LSpice: Oh... I wasn't assuming there are actual "" quotes present. Like what if you're quoting the passage The way to indicate typos is via writng [sic] so that the reader knows it is intentional? Like, it has a typo, so the middle part becomes writng [sic] [sic]? (And imagine it's long so there would be no surrounding quotes either, but rather a paragraph indentation.)

@Mehrdad, I'm not sure I understand. It seems to me that any quotation should be set off somehow, with quotation marks as usual, or with backticks and italics as you have done, or anyway by some typographic device, in which case the modifying [sic] is simply not subject to that typographic device. If the quote by any such means, then I would argue that what is happening is not quotation, but paraphrase (or, if the source is not attributed, plagiarism).

@LSpice: Long (block) quotes don't involve quotation marks; they get indented or something. Imagine this is in the middle of a block quote. How would you put the extra [sic]? You can't un-intent the middle of a line and then re-indent the next word.

@Mehrdad, I think I see the issue. My understanding is that [sic] modifies and follows the entire quote; it simply indicates "I have quoted exactly what was written"—with, presumably, the implication that apparent errors were already in the original. It need not be placed directly after the apparent error; so, "'2 + 2 = 3 is well known' [sic]" rather than "'2 + 2 = 3' [sic] 'is well known'".

@Mehrdad your "for programmers" question is a great one. It is discussed at length here.

@DanRomik: Thanks for the link! :)