10
\$\begingroup\$

I've noticed a couple of experienced longtime users answering questions which they surely know are duplicates of other older questions.
In at least one case the same user answered essentially the same question in much the same way - although I think their earlier answer was better and more detailed - in fact this user has answered this "same" question at least 3 times now ...

Why?

Examples:

Is it easier to re-answer the question than it is to find the duplicates - even if you've written one or more of those duplicates yourself?
What can be done to encourage the appropriate use of SE's "duplicate system"?
Is SE's question & answer search feature so bad that it can't find duplicates?

I've asked essentially the same question over on DIY's meta , as I see the same sort of behavior over there too.

\$\endgroup\$
11
  • 1
    \$\begingroup\$ Not to mention the fact that some of them also get highly upvoted despite being duplicate but for being so interesting. \$\endgroup\$
    – Mitu Raj
    Jul 28 at 18:56
  • 8
    \$\begingroup\$ @MituRaj That's pretty much why these users have such a high reputation too. \$\endgroup\$
    – pipe
    Jul 29 at 0:45
  • 2
    \$\begingroup\$ And another ton of duplicates on ohm's law, potential difference and electrons movement could be found too. \$\endgroup\$
    – Mitu Raj
    Jul 29 at 4:26
  • 1
    \$\begingroup\$ At times when we are posting a question , similar question suggestion does not have this duplicate question. We might also search with keywords , but then any question can be asked in so many multiple ways. I also get confused before asking a question since its too obvious to be asked before , but unable to be find it in the site. Idk any solution for this issue ! \$\endgroup\$ Jul 29 at 8:08
  • \$\begingroup\$ In this process, SO looks more intelligent. Several times I had to post the link after finding the duplicate after a bit digging in EESE. \$\endgroup\$
    – Mitu Raj
    Jul 29 at 12:01
  • 12
    \$\begingroup\$ The fact of it is that the search function on all stacks just straight up sucks. Even if I know I posted an answer to a similar question, I can never find it using the stack search. I have to go to google and use "site:electronics.stackexchange.com" and search terms including my user name to find my own posts. \$\endgroup\$
    – JRE
    Jul 29 at 14:22
  • 2
    \$\begingroup\$ You will find that I answered several of those questions but the replies are different in each case due to differences in the questions. Sometimes I forget that I have posted an answer on the topic, sometimes I can't find it and sometimes I can and reuse the schematics (perhaps with alterations) but rewrite the text. I agree that SE's search makes searching my own posts is difficult and finding them on Google search is even more difficult due to my username! \$\endgroup\$
    – Transistor
    Jul 30 at 9:21
  • 1
    \$\begingroup\$ I know I answered this a few years ago ... now where is the answer again? Sometimes it really IS quicker to answer again than go searching... \$\endgroup\$ Jul 30 at 15:29
  • \$\begingroup\$ Any number of times I have found, questions locked as Duplicates, which when you check are not because the question has a different emphasis, or the original does not have a good answer (no or low votes) to the original question. \$\endgroup\$ Aug 7 at 23:07
  • 1
    \$\begingroup\$ Well, I am not saying it applies to any particular user, but one of the reasons could be getting more rep without much extra effort. \$\endgroup\$
    – Eugene Sh.
    Aug 9 at 18:53
  • \$\begingroup\$ @Transistor “even more difficult due to my username” — haha, yeah I bet it is! \$\endgroup\$ Aug 18 at 4:48
11
\$\begingroup\$

I agree. I suspect its because answering takes less time than actually finding the appropriate duplicate. Also, it's not crystal clear how to actually provide the link to the dup in the close menu (yes, I know, it's not THAT difficult, but the lack of instructions might be an obstacle to some).

\$\endgroup\$
2
\$\begingroup\$

I agree, SE's duplicate feature mostly just compares 2 posts which use the same words, not necessarily having the same meaning. I have noticed this when flagging duplicates, the right question isn't suggested as a duplicate by SE.

Here's another duplicate, which I flagged:

Why does lowering VDD increases the delay for digital circuits? is a duplicate of Why does decreasing the CMOS supply voltage also decrease the maximum circuit frequency?

It wasn't closed as a duplicate. And SE did not suggest the right question as a duplicate, because the words are different.

\$\endgroup\$
2
\$\begingroup\$

Personally, I think that almost-duplicates are a good thing to have a round. Here is why:

I would consider as duplicate question such questions that can be fully answered by copy-pasting a previous answer.

But I see several reasons, why there are questions that fall just shy of this criterium:

  1. Users are encouraged to ask new questions (as opposed to necro-commenting) when an aspect of a previous answer is unclear to them. New users are not even able to necro-comment.

  2. Most situations are indeed different, so that a previous answer is slightly invalidated by the given circumstances. Just consider the boatload of questions on PCB stackups and planes. I genuinely think that these are only rarely duplicates.

  3. Things change. EE is a very alive field and standards that are 3 or 5 or 10 years old, might be outdated or even discouraged. Therefore, even an actually duplicate question might deserve a different answer than some years ago. And authors of old answers are not always around to update them.

  4. As with most things, "the truth is somewhere in the middle". When doing proper research about any topic, it is hugely important to consult many sources and carefully analyze where they agree and where they don't and why. Having many almost-duplicates around is a very rich source for such studies. They do show up on google when researching prior to asking questions, but of course it would also help if the live in-site search while typing the question would also give more relevant results, as others have commented. If the in-site search was a bit smarter, it could even add an automatic 'possible-duplicate' tag to questions that have a lot of overlap with existing answered questions.

\$\endgroup\$
0
\$\begingroup\$

It would be preferable if users would check for duplicates before they post an answer, this isn't easy to do because of the amount of questions. If you see a question that seems like it could be a duplicate, take some time and do some searching before answering.

Here is what a duplicate is:

According to Stack Exchange co-founder Joel Spolsky, we should only close real duplicates, and according to co-founder Jeff Atwood, there are three kinds of duplicates: cut-and-pastes, accidental duplicates, and borderline duplicates (requiring judgement as applied by the community).

Questions may be duplicates if they have the same (potential) answers. This includes not only word-for-word duplicates, but also the same idea expressed in different words.

Source: How should duplicate questions be handled?

\$\endgroup\$
-2
\$\begingroup\$

I'll add to this list: Does the voltage difference have an effect on the electrons' speed?

The original Q had weak answers, this 'dupe' one much stronger ones.

\$\endgroup\$

You must log in to answer this question.

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