I recently noticed that the vote button tooltips have changed. They now say:

Downvote this <post-type> if you find it unclear or not useful

Upvote this <post-type> if you find it clear and useful

The text changes depending on which vote you cast.

For context, the old tooltips on questions used to say:

This question shows research effort; it is useful and clear


This question does not show any research effort; it is not useful or unclear

The new text isn't that different for answers.

IMO, the former wording was way more informative (esp. for the OP).

Is "lack of research" for a question not a good reason to downvote a question anymore?
The help center still states1:

Please look around to see if your question has been asked before.

Were there some resilient statistics done about users who complained about the old wording?

1)The link to the SE internal search is a bit misleading IMO. It's well known that you'll get far better search results using extenal search engines like Google or DuckDuckGo.

  • 31
    upvotes this question for showing research effort – Glorfindel Jun 17 '20 at 19:03
  • 42
    Dear Stack Exchange: Please change these back, and please remove the "Thank others for their answers" and any similar pop-ups. They are distracting, make the site more stressful to use, push away newcomers, and are simply not needed. – Panzercrisis Jun 17 '20 at 20:39
  • 9
    Basic tooltips like before work great. But these new ones look and act more like pop-ups and spam. – Panzercrisis Jun 17 '20 at 20:41
  • 6
    @Panzercrisis You may try to post the essence of your comments here as an answer. They just need to be reformulated a bit. – πάντα ῥεῖ Jun 17 '20 at 21:21
  • 11
    @πάνταῥεῖ Oh, I'm sorry, I was being sarcastic, I'm still reeling from whenever they tried to remove the highest upvoted answer (Jun 26 '19 at 14:23) that said that research was one of the most important parts to this whole site. I'm trying to call attention to the fact that this isn't the first time that SE has thought that research might not be all that important. I agree with every single one of your points. – zero298 Jun 18 '20 at 16:10
  • 5
    Why is research important? – gnat Jun 18 '20 at 20:43
  • 3
    I want to mention something that I don't see on MSE but was brought up on MSO - after you vote, the text changes to "This answer is useful" or "This answer is not useful" for upvote/downvote respectively. – VLAZ Jun 19 '20 at 9:26

Thank you for the feedback on this and the impact changing the wording has here. We want to be sure that copy is clear, concise, and useful and this seems like a change we need to discuss more before making changes. We also realize that, as a text-based platform with complex systems, precise wording is necessary to ensure proper usage of tooling. We'll keep this in mind for wording changes in the future.

The motivation to update the voting tooltip copy was to improve its tone and provide additional clarity against the new Reactions test. The intention was not to downplay the value of research effort or change voting criteria, which was an oversight on our part. This was considered a general copy improvement and wasn’t flagged internally as something that might be problematic so we have rolled back the tooltip language to the original copy. Additionally, we have decided to revisit any copy changes around voting at a time when we’re specifically focused on that area, and can build in more time for feedback.

  • 11
    @Des Tyvm. That was the official statement I was looking for. – πάντα ῥεῖ Jun 22 '20 at 21:21
  • 6
    This is a fundamental aspect of using the site. It's astounding that such wording isn't treated at the highest priority. And by the most senior of product/system designers. (Something the company seems to have no awareness of.) Questions are bad enough, do you know what it would be like with no research? No downvotes plus having to do people's research for them before being able to close? The latter is bad enough as it is. – philipxy Jun 23 '20 at 1:54
  • 1
    While you're rephrasing arrows: The old phrasing was poor--though useful as a canonical example of poor user guidance on SE sites. It was not clear whether semi-colon meant "and" or "or" or just what it was demarking a list of in either & correct interpretation relied on already knowing what it meant. ("This question shows research effort; it is useful and clear." "This question does not show any research effort; it is unclear or not useful.") And now you use 2 styles of labels, what clicking communicates vs instructions. And you say "if" but you mean "iff/when". (Too bad: no English expert.) – philipxy Jun 23 '20 at 2:28
  • 20
    Update: tooltip language has been rolled back – Yaakov Ellis Jun 23 '20 at 15:00
  • @YaakovEllis I'm hoping it's temporary, but I'd like to hear something official – Pureferret Jun 24 '20 at 8:16
  • 1
    @pureferret see this post that we are commenting on – Yaakov Ellis Jun 24 '20 at 8:48
  • 1
    @YaakovEllis I assumed you meant rolled back to the new version (so old-> new->old-> rollback->new) as I'm still seeing the new ones that don't reference research. – Pureferret Jun 24 '20 at 9:35
  • 1
    @Pureferret No. Rolled back to old. where are you seeing this? The tooltip text that changed is on the question. The answer tooltip had not referenced user research previously. – Yaakov Ellis Jun 24 '20 at 11:48
  • I guess the problem here was that it was labeled a copy. Everyone who uses the site and reads meta knows that the tooltips are the only immediately visible thing really explaining the votes. These tooltips texts are frequently used to explain the meaning of votes to users. A change in them should not have been labeled a copy. – Trilarion Jun 27 '20 at 7:33
  • 2
    @Trilarion “A change in them should not have been labeled a copy” — What do you mean? The change itself was a copy edit. It was never labeled “a copy”. – Sebastian Simon Jun 27 '20 at 14:59

As a mod on a scientific stack, this drives me nuts.

Research effort is built-in to our community's self-determined quality control, and this tooltip was the most concrete support for that approach in the actual site machinery.

I don't recall (and please point me to it if it exists) any community discussion about this change, and I certainly wouldn't want it at Biology.SE. I constantly downvote questions that do not show research effort and I think it is a critical feature that distinguishes StackExchange from "homework help" sites that exist mostly to guide students towards bypassing learning for the sake of a "grade". I specifically use this site and answer questions here because of that feature, and refuse to participate on other sites that lack it.

(in a comment on the OP, @gnat linked to Why is research important? which, although it is specific to a local community, also has a lot of cross-site application. I think it's a good read in addition to the reasons I mentioned)

  • 12
    As a mod on History:SE, I couldn't agree more. – sempaiscuba Jun 18 '20 at 19:54
  • 10
    "I don't recall (and please point me to it if it exists) any community discussion about this change" Me neither, that was one of the motivations to post this question here. – πάντα ῥεῖ Jun 18 '20 at 19:56
  • 2
    @πάνταῥεῖ I figured as much, definitely a point I wanted to emphasize while allowing that I might be mistaken. But it seems that indeed I am not, and this was not discussed at all openly. – Bryan Krause Jun 18 '20 at 20:43
  • 5
    I fully agree with this answer, as a mod on two scientific stacks – AliceD Jun 19 '20 at 8:46
  • 3
    (+1) I find that half the time the OP has researched their question, but fails to explain what's still puzzling them: i.e. they haven't asked their real question. This failure to get to the crux of the matter, if unresolved, leads to answers that neither satisfy the OP nor are of much use to anyone else. – Scortchi - Reinstate Monica Jun 20 '20 at 8:52

Changing the message has removed the reason why users edited their posts and added the research, in order to improve or make it comply with their site's guidelines.

Whenever I saw a post which I had downvoted show effort and research (which was not the "I googled" variant), I would very often retract the downvote and if the post was improved measurably, I would cast an upvote. I don't think I'm alone in behaving this way. Many veteran users hang around because they care about quality and want their site to thrive and be useful to visitors.

What's the most tangible and easiest way for improving any post?

Show the research.

It's so easy; posting the research may take over an hour if the user hadn't done any previously but if it was prior to posting and the OP didn't mention it in their post, the time lost in editing will be a matter of minutes.

The new message is opinion-based. It's one grade up from saying Downvote if you didn't like it.

Downvote this if you find it unclear or not useful

How is a commenter supposed to help a user improve their post quickly?

User: I'm sorry, the post is not useful
New contributor: What do you mean? It answers the question. It is the correct answer.
User: Yes, I know that but "why" is it correct? Can you please show some supporting evidence?
New Contributor: Evidence? I am a English native speaker, I have spoken the language all my life. Why should I waste my time on this site just to please a clique of stuck-up users?

Now, with the old tooltip message I used to say something like this:

This question does not show any research effort; it is not useful or unclear

User: I'm sorry, your post was downvoted because it lacked research. If you hover your mouse on the arrows you'll see the message. Can you please add some research to improve your post? Thank you.

In my opinion, the new tooltip messages sound just as brusque as the old one. And, paradoxically, they are less useful and more unclear to new contributors.

  • 11
    Ty Marie-Lou! "It's one grade up from saying Downvote if you didn't like it." That's exactly my point. Yesterday it was also brought up in the chat, that the new tooltip text makes the reasoning simply too broad to be useful for anyone. – πάντα ῥεῖ Jun 18 '20 at 9:23
  • 10
    It would be great if the question downvote referred to a lack of research, and the answer downvote referred to a lack of evidence. Research and evidence, that's how you make a great Q&A pair! – curiousdannii Jun 18 '20 at 11:39
  • 4
    @curiousdannii perhaps 'explanation' instead of 'evidence'. Few answers would be improved by adding "evidence", per se, but many answers could be improved by adding an explanation. – TylerH Jun 18 '20 at 14:14
  • 1
    @TylerH I don't think I can agree. If an answer needs explanation then it should probably just be deleted! Supporting evidence is what distinguishes a poor answer from a good one. – curiousdannii Jun 18 '20 at 14:40
  • 1
    @curiousdannii: Don't forget though that downvoting is the path to deleting an answer for any non-mod users since it has to have a negative score. That being said, I'd also disagree that an answer needing explanation would automatically mean it should be deleted. We occasionally (alas, not often enough) get answers that require more explanation that is added later upon request from commentors. – Rubiksmoose Jun 18 '20 at 15:23
  • @Rubiksmoose Sure, I don't think they should automatically be deleted. – curiousdannii Jun 19 '20 at 0:07

I keep checking in from time to time, and you know, it is incredibly demoralizing to see unilateral changes like this pass without prior discussion with the community. I thought we talked about this and agreed not to breach community's trust in SE company's actions in this way, many times before.

This, and the recent "clap" buttons make me really motivated to tell everyone about common incidents like these (which sadly they are) and it seems like a better idea to keep contributing to the development of competing Q&A and other types of developer hangout platforms that would directly compete with StackExchange, with the goal of taking control and never letting this kind of untrustworthy unstable behavior happen over there.

StackExchange management, in your shortsightedness to "expand community by any means possible, as quickly as possible", long term you're going to lose it completely and be out of the investments, and jobs. But for now you're fine, right?


It has been suggested to potentially post my comments on the question as an answer:

Regardless of the messages in these notifications, let's change the look and feel of the UI back to before. The older style of tooltips that this site has used work great. In particular, they are unobtrusive. For example, there is a small, roughly 1-second delay before they appear. And even when they do appear, they are small and do not grab all attention away.

However these new ones look and feel more like, unfortunately, pop-ups. They show instantly every single time the cursor hovers over one of the upvote/downvote buttons, they are much larger than before, and they immediately steal all attention from the rest of the page, over and over again.

On another question, I mentioned a few common reasons that these sorts of notifications are really a bad thing. I will use a similar, yet slightly different list here:

  1. They add clutter to the site's UI.

  2. They are typically not necessary or relevant for the user.

  3. They distract the user, and they are very stressful as a result.

Regarding reason #2, who on this site does not know what an upvote or a downvote is? This is extremely common and standard all throughout the Internet, and it is only a slightly different expression of the same mechanism as likes (such as on a Facebook post). Yet when blaring notifications instantly show every single time an upvote or downvote button is hovered over, it constantly distracts the user, and it creates stress.

If you really feel this information must be present, then please at least revert back to the standard, much less obtrusive tooltips, which have timed delays and do not use excessive space and glare.

Personally I feel like I'm using these sites less now, specifically as a result of this - yes, small, but nonetheless constantly recurring - stress factor.

You must log in to answer this question.

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