30

Someone just edited a link in one of my answers to point to a 'better reference'.

Why are C++ STL iostreams not "exception friendly"?

I've no problem with links being changed in general, but this smells like part of a concerted attempt to redirect traffic.

Is this kind of activity acceptable on SO?

  • 7
    Yeah, someone is putting in a bit of effort there: stackoverflow.com/users/283145/jons34yp?tab=activity Don't know if this was somehow discussed within the C++ community. – Bart Aug 27 '13 at 13:54
  • I want to say there are some people who believe that cplusplus.com is an inferior referrerence, ala w3schools. NB - I don't know if that is actually true. – MikeTheLiar Aug 27 '13 at 13:54
  • 2
    Don't think it's malicious attempt to "redirect traffic", more likely someone who honestly believe that other site is the ultimate and official resource so doing his best to link everything to it. Unless you think "your" site is better or find anything wrong in "his" site, let it be. – Shadow Wizard is Ear For You Aug 27 '13 at 13:56
  • 14
    Myeah, looking through the C++ lounge transcripts, it seems cplusplus is discouraged and cppreference is preferred. Most likely a user just trying to genuinely improve the references in posts. – Bart Aug 27 '13 at 13:58
  • 20
    @Bart While the community may prefer one, unless there are pretty significant problems with the one it would probably be best to just limit it to comments, rather than going around changing all links to a site you think is just "not as good". Is there concensus in that community that this link is so bad it ought to be removed on site (ala w3schools)? – Servy Aug 27 '13 at 14:00
  • 1
    @Servy Funnily enough w3schools is references in some of the talk about cplusplus. But I don't use either resource enough to have a strong opinion one way or another. – Bart Aug 27 '13 at 14:03
  • 9
    @Servy: There certainly are substantial problems with cplusplus.com. There was a big question/answer about how terrible it was that ended up being removed due to off-topic or something. The bottom line is that information from cplusplus.com cannot be trusted to be accurate. – DeadMG Aug 27 '13 at 14:17
  • 8
    As someone actively involved in cppreference, I am fairly certain that nobody on the team would be doing that - besides questionable ethics of such editing, we're simply too busy. I think @ShaWizDowArd is right. – Cubbi Aug 27 '13 at 15:02
  • 3
    @DeadMG "information from cplusplus.com cannot be trusted to be accurate" Welcome to the internet... – Roddy Aug 28 '13 at 8:09
37

The 'problematic' editor here.

For what is worth, 'traffic redirection' is indeed a semi-accurate description of what I'm doing. However, the motive is not profit in any way - I genuinely think that cppreference.com is a better reference. The primary reason is not that the content there is often more accurate, but that it pushes better programming practices and the overall spirit of the website is more professional. I would just love that Google returned a link to cppreference as first result for each query related to C++, because this would lead to better educated C++ programmers and more opportunities to do cool stuff in C++.

Incidentally, cppreference does not have ads, therefore any chance I could theoretically edit links for profit is very, very remote. Not that I would do this in practice ;)

  • 33
    Just a hint for any such future endeavours: Should you find that there are a lot of problematic references used, and should you wish to change them en masse, please come to Meta first. Propose it here, discuss the pros and cons, and when a consensus is reached, go ahead. Heck, some of us would be glad to help you out. (Not that there's necessarily a problem with what you've done so far though). – Bart Aug 27 '13 at 15:32
  • 7
    I personally find cppreference.com superior to cplusplus.com. It's wiki nature allows anyone to improve should something be wrong or missing. – Cole Johnson Aug 27 '13 at 16:47
  • 2
    @jons34yp Thanks for pitching in. Upvote and accepted for a sensible reasoned answer. No ads on cppreference yet, but these kind of sites can be big business (look at SO!). Website valuation tools put cpluplus.com at ~$1M ...extra-large pinch of salt needed with these, as always. – Roddy Aug 27 '13 at 20:51
  • 17
    @Jons34yp: I would seriously recommend not doing this in the future. Comments maybe, but cplusplus.com isn't bad enough to justify immediate replacement IMO. Just discussion. – Mooing Duck Aug 27 '13 at 21:19
  • @MooingDuck I'm too lazy to edit links, but when I find one that is actually bad, I won't hesitate a single second to replace it by a better one. – sehe Jan 26 '16 at 12:41
  • I'd prefer some more details and justification here. – cp.engr Sep 6 '18 at 14:15
  • FYI: cppreference has ads now, and it's lame. All good things must come to an end, I guess. – HostileFork says dont trust SE Sep 21 '18 at 12:57
26

When the original reference is to cplusplus.com, then YES. That site is notorious for being inaccurate and generally bad in a very large number of ways, at least amongst those of us who have had the misfortune of visiting it. I downvote all answers that link there without hesitation, much like I do those that recommend the Google Style Guide. So it turns out that guy only fixed the question and not the answer.

Unfortunately, most of the material on SO discussing how terrible that site is was closed because folks thought it belonged on Programmers, but there was once a big question with answers showing how bad it was.

There's nothing ultimate or official about cppreference, although the only other reference sites I am aware of (MSDN and cplusplus) both have some issues (cplusplus.com is a whole lot worse). For example, MSDN shows docs for Visual Studio which isn't always the same thing as C++, and it's not as nice to use or navigate. And cplusplus.com is riddled with inaccuracies and bad examples and such.

That guy did you and everyone reading a big favour.

Edit: But this was one guy- not some co-ordinated effort or discussed with us.

  • 5
    So you would compare this to, say, replacing links to w3schools? – Andrew Barber Aug 27 '13 at 14:16
  • 8
    @AndrewBarber: I have never used w3schools so I can't really comment. But purely judging by its reputation as a terrible resource, then yes. – DeadMG Aug 27 '13 at 14:18
  • 4
    Cool. That's kinda where I am with cplusplus.com :) – Andrew Barber Aug 27 '13 at 14:19
  • 1
    Used both. Neither of them are advisable for those who do not know better, and if you know better why look their in the first place? – thecoshman Aug 27 '13 at 14:48
  • 7
    You downvote purely because of the URL without actually looking to see if the content is incorrect? – Justin Meiners Aug 27 '13 at 17:21
  • 4
    @JustinMeiners: A link to that URL is incorrect content. There's no point posting one right answer if you're gonna lead the questioner to a bunch of wrong information. – DeadMG Aug 27 '13 at 17:24
  • 7
    @DeadMG "A link to that URL is incorrect content" nothing on the site is correct? Wow bold claim. – Justin Meiners Aug 27 '13 at 17:25
  • @JustinMeiners: Nice strawman. It is not correct enough to be usable, and has more than enough incorrect content to bite anyone trying to use it. – DeadMG Aug 27 '13 at 17:28
  • 3
    @Justin: If you link someone to a reference site, you are linking to the entire site. – DeadMG Aug 27 '13 at 17:31
  • 4
    @DeadMG true, but if a link answers a single topic correctly then why not? If they click it, learn what they need to and go then where is the harm? Just because the rest of the site may contain some inaccuracies doesn't devalue that page, and certainly doesn't warrant a downvote if correct. – Justin Meiners Aug 27 '13 at 17:35
  • 3
    @syam I disagree partially. If I have a question about std::string and the std::string page answers it correctly, im not going to be misled by a mistake in std::vector. I get the arugment though, send them to a site with fewer errors so other lookups will find correct info etc, but if the answerer linked there I think its fine (if its right). I also think we may be overestimating the number and level of errors on the site – Justin Meiners Aug 27 '13 at 18:51
  • 6
    Downvote without hesitation no matter which page on that site was linked? That's really not helpful... Stick to comments to educate people unless the specific linked article/page is completely wrong. – Georg Fritzsche Aug 28 '13 at 9:39
  • 18
    Apparently I can't link to Stack Overflow answers anymore, because then I'm linking to all of SO and there might be a wrong answer on it somewhere – Michael Mrozek Aug 28 '13 at 15:36
  • 7
    "but there was once a big question with answers showing how bad it was" -- If it's the same one I'm thinking of, every single point on it was fixed long ago(when cppreference.com was in its infancy and nearly unusable), and people continued to link to some archived version of the SO post as evidence of how bad cplusplus.com was without actually checking the changes that the site had made. – Benjamin Lindley Aug 29 '13 at 20:13
  • 5
    @AndrewBarber: Yes, it's the C++ equivalent of w3schools. Though I should point out that the accuracy of the reference part of cplusplus.com has improved greatly in the last couple of years. Alas the user-contributed examples, tutorials and forums are still a problem, no more so than because the site looks (via its name) as if it is somehow canonical, official, recommended-practice, etc. – Lightness Races BY-SA 3.0 Feb 12 '14 at 14:39
16

There have been cases when cplusplus.com provided better information than cppreference (recent example, which was, of course, promptly fixed, since cppreference is publicly-editable). In general, as far as I've seen, cplusplus.com improved a lot in the last three years since cppreference.com became an open wiki.

If I were to notice such edit, I would actually visit both pages and revert if cppreference isn't offering anything better than cplusplus in that case.

4

It's just a suggestion. If you don't like his edit, roll it back.

  • 23
    I don't fancy an edit war, and the editor has done this across lots of questions. – Roddy Aug 27 '13 at 13:56
  • 1
    @Roddy At the end of the day, in an edit war, the author of the post generally "wins", unless there is compelling reason not to, i.e. they are trying to vandalize their own post. If, after rolling it back, the user suggests the edit again, just flag the post and a moderator will deal with the issue. If you disagree with the edit, you have every right to roll it back, as suggested. – Servy Aug 27 '13 at 13:57
  • 4
    Just a minor nitpick; the word "suggestion" in your answer here... umm... suggests that the user does not have enough reputation to do full, unreviewed edits. – Andrew Barber Aug 27 '13 at 14:18
  • 1
    @AndrewBarber I know. I mean his "edit" as a whole is merely a suggestion to improve the quality of the post. – bobobobo Aug 27 '13 at 23:07

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