Why Left Search Buttons Perform Faster Than Right

16 Nov 2010 - 11:06pm
3 years ago
72 replies
3806 reads
uxmovement
2010

Research has shown that having or not having a search button makes no difference to pro users. However, for novice users, right search buttons are generally slow and inefficient. To me, the solution is simple. A left search button reduces visual fixations, puts the label first and makes doing searches faster for users. 

The pattern left search buttons follow is similar to IXDA's search where the combo box and icon come before the text field. The label and button have more visual weight and are what users need to see first before they carry out a search, which is why I believe they should come before the text field. Despite the fact that most websites and search engines have it backwards, what are your professional and experienced independent thoughts on this matter?

http://uxmovement.com/design-articles/left-search-buttons-perform-faster-than-right for further detail.

 

 

Comments

17 Nov 2010 - 6:05am
Natalia Rey
2010

Yes It's true.Here I attached an example of this, the yahoo web for mobile devices.

17 Nov 2010 - 7:05am
William Hudson
2009

It sounds to me that it would not be very accessible. For a screen reader user, a label, text field and search button are needed in that order. (It's not impossible to have a design that would hide a trailing search button for screen readers, but I really do doubt the benefit of designs that are inconsistent to save a dubious amount of time. Presumably, if it was really significant, the study would have reported it!)

Also, fixations aside, I hate having the search button before the text field as I am expecting it to be to the right. Maybe consider a shorter text box (which would reduce both fixation time and mouse travel) or not spending so much time studying eye-tracking results and just look at time on task.

Regards,

William Hudson Syntagm Ltd User Experience Strategist UK 01235-522859 World +44-1235-522859 US Toll Free 1-866-SYNTAGM Web www.syntagm.co.uk/design Email william.hudson@syntagm.co.uk Skype williamhudsonskype Twitter SyntagmUCD

Syntagm is a limited company registered in England and Wales (1985). Registered number: 1895345. Registered office: 10 Oxford Road, Abingdon OX14 2DS.

UX, UCD and Usability Courses in London and Vancouver www.syntagm.co.uk/design/schedule.shtml

17 Nov 2010 - 8:12am
holger_maassen
2010
17 Nov 2010 - 10:05am
Billbarany
2008

Your own research doe not agree with your statement that "Left Search Buttons Perform Faster Than Right".

After reading your paper the correct statement should be "The enter (return) button is faster for pro-users." The pro users in your study never use the left button "or the mouse". As for the novice user and the western left to right reading standard, I still think it is correct to have right side buttons to accommodate their needs. The speed in form completion is because the users hand doesn't have the extra step of reaching for the mouse. Not so ground breaking.

17 Nov 2010 - 10:05am
Andante
2008

Your graphic makes it look more efficient, but in reality is not communicating correctly. We read and write from left to right so our minds are already condition that when you finish typing (if you reach out for the mouse to click the button) you will look to complete the action after the text field, not before. I think that this pattern will work very well for non-latin alphabet languages where they read and write from right to left, like Arabic. Also, you have to put into perspective that non-visual browsers like JAWS, will read the button before the text field, and this will confuse the user.
Though, you may have a stronger argument if the button is left-align under the field.
From my own experience, I've found that placing the button before the text field lead the user to make mistakes. Which is in violation of one of the most important heuristic principles we have learn love.
There you have my two cents,E 

18 Nov 2010 - 7:32pm
ekim
2010

My 2 favorite replies from "anthony" in the comments section:

Any­one who is will­ing to think and use their brain can see this, regard­less of research specif­i­cally say­ing this. Some­times you have to think and use your brain to under­stand things.

and

I hon­estly guar­an­tee you that if you were to do an A/B test with a right search but­ton vs. a left search but­ton the left search but­ton would per­form faster. If you want evi­dence, then test it your­self. Don’t expect me to do every­thing for you.

I think these replies only prove how baseless his articles are.  Does anyone else feel like this is detrimental to the UX community?  It's sad to think some designers may take his suggestions to heart.

21 Nov 2010 - 11:05am
dmitryn
2004

Agreed.

It's unfortunate that anyone can put up a blog with "UX" in its title and expound views that have been proven wrong in research literature, and would never be espoused by any experienced practitioner, as "thought leadership".

But, I guess that's just a side effect of the inclusive nature of our community.

Dmitry

On Thu, Nov 18, 2010 at 8:08 PM, ekim wrote: > My 2 favorite replies from "anthony" in the comments section: > ... > I think these replies only prove how baseless his articles are.  Does anyone > else feel like this is detrimental to the UX community?  It's sad to think > some designers may take his suggestions to heart. > > ((

29 Nov 2010 - 12:47pm
uxmovement
2010

Eugene, I'm sorry you think that way. I think you are expecting UX Movement to be something it isn't. It isn't a testing and research facility. It's just a blog.

I wish I could oblige you and test this theory, but I don't have the necessary equipment, tools nor access to users to conduct a full-on study. However, if you are truly interested in testing I encourage you to test it yourself. When you do, I'd be happy to publish the results whether they fare for or against it.

Perhaps the only legitimate criticism to left search buttons I've heard is that when users search they use the text field first and the button second. So, the question is does the order of the task affect the user's search performance. My emotional judgment says yes it does, however, my logical judgment says no it doesn't. This is because the button, in this context, is not location dependent. In other words, as long as the button and field are aligned it communicates that the field and button are related and that the relationship of the two elements are for search. Users will still recognize what a button is and what a text field is. The order does not change any of that.

The order of the button and field also does not change how users think about the search task. Their mental model for searching will always be to input a search word and then press the button. The order does not change the way they think. It only changes the way designers think about it because we have never thought about it doing search any other way. It's important to remember that we are not the user, which is why as designers we have to consciously remove all emotional bias aside from our thinking.

To the claim that left search buttons are faster. I think that is pretty evident in the diagram and the article, as well as the past research on right search buttons, so I don't think I need to get into that. Hope this clears up most of your concerns.

Anthony

29 Nov 2010 - 8:03pm
ekim
2010

Anthony, you're right--I do have an expectation but it's based on how you've sold yourself and the site.  Please read my original thoughts from a few weeks ago at http://www.ixda.org/node/28262.

29 Nov 2010 - 8:05pm
Dan Saffer
2003

On Nov 29, 2010, at 2:03 PM, uxmovement wrote:

> I think you are expecting UX Movement to be something it isn't. It isn't a testing and research facility. It's just a blog.

So we are supposed to take the word of a blogger with no last name and thus no known credentials, with no data to back up his claim, to overturn 15+ years of precedent? Please.

Aside from the credibility issue, Alan Cooper's dictum still stands: Use conventions unless the proposed solution is radically better. Is this solution radically better? Would it save more than a few microseconds of time, which would probably be eaten up by the time taken to mentally adjust from the newness of the order? I think not.

Dan

Dan Saffer Principal, Kicker Studio http://www.kickerstudio.com http://www.odannyboy.com

5 Dec 2010 - 1:52pm
uxmovement
2010

Dan,

My approach is a little different. I actually don't want you to take my word for it because of my credentials. I want you to take your own word for it because you've logically and intuitively thought about it. I've never believed in accepting something because of the credentials of an authority figure because they can't always be trusted, aren't always right and think with emotional biases. Doesn't that mean that could apply to me? Yes it does. It's possible I can be wrong, but I don't think I am in this case. I go out of my way to make sure that I think and judge without biases and with pure logic and intuition. I believe other people are capable of doing this and should do the same.

Your basic question: How can we be sure that left search buttons benefit the user and is better than the way search is done now without any user testing? To answer that, I ask how can we be sure that sunlight feels warm if we have never felt sunlight before? We can either use our knowledge, logic and intuition and conclude that it is, or we can experience it for ourselves. The best and surest way would be to experience it for yourself. That experience is something I can't give you. You're going to have to experience it for yourself. All I can do is share with you my experience. Whether you believe me or not, that is entirely up to you. If you're skeptical because you don't trust your knowledge, logic and intuition, then it means you need to experience it for yourself. But please don't make the false assumption that just because you haven't experienced it, it means it's not true, or the other false assumption that just because someone has credentials, and is a person of authority then it means it's true. Instead, please read and think about it independently or experience it for yourself to decide whether it's true.

Anthony

5 Dec 2010 - 6:05pm
Dan Saffer
2003

I think you are misguided on several points.

  1. You've made a claim that the way interaction designers have been designing products isn't the best way. Great, I'm all for rethinking assumptions. However, the burden of proof lies on you, not on me nor on other members of this organization, to prove your claim. That's how it works in science and law and presumably should work here as well. You have offered no prototype, testing results, or live product to prove your claim. "Experience it yourself" isn't an argument; it's an abdication of responsibility. Until you can offer more than your guess at how it might work, it is just that: opinion.

  2. The validity of opinions rests on authority and reputation. By posting anonymously, you effectively have neither, weakening your case.

  3. I do trust my knowledge, logic, and intuition. It's yours I don't trust. See #1 and #2.

  4. No "authority" is telling us search buttons go to the right of the text entry field. It's a practice that was developed over 15+ years' time, by individual practitioners because it seemed to work. It's now become a de facto standard, adopted by thousands. So there's no "authority figure" other than our own practice and ourselves. If there was a proven, better way suggested by a practitioner with either proof or a reputation to back it up, we'd listen. If Jared Spool or Luke Wroblewski came out with an article telling this community we should rethink search boxes, you bet your ass we'd listen. Likewise, if someone posted an article saying, "I've been doing A/B Testing with the search box, and look at how much better putting the search button in front performs," we'd also listen.

  5. As I said in my initial response, any new solution that overturns a convention must be demonstrably better. We can't take your word for it because by posting anonymously, you have no standing. We can't decide for ourselves because you've offered us nothing but a hypothesis without any data or demonstration (prototype or product) for us to decide.

I look forward to your rectifying any one of the points made above, if you can.

Dan

Dan Saffer Principal, Kicker Studio http://www.kickerstudio.com http://www.odannyboy.com

26 Dec 2010 - 12:48pm
uxmovement
2010

Dan,

First of all, I don't think you speak for everyone here. Nobody here has delegated you that power to speak for everyone, so I think it would be more appropriate if you only spoke for yourself. The way an intellectual community works and has always worked in sciences and the like is to build new theories and make new discoveries off of other peoples ideas. This means that in order for a field or a community to progress, people in that field need to help each other and stand on each others shoulders to add to or improve theories to make new discoveries. I have suggested a new idea. Sure, it might be incomplete, it might lack in some areas, but it doesn't stop here. It's up to others in the field who actually care about making new discoveries to take this idea further, or to collaborate with others to build and develop a theory. Instead of working together, you seem to want to jostle, which is not on my agenda. Those who truly care about progressing the field of interaction design would want to experience it themselves and see for themselves if this theory holds up, or simply understand and analyze the reasoning. 

So far, it seems like you've been quick to tear everything down. You haven't done it on the basis of inconsistent or biased reasoning, or anything specific relating to the ideas of the article in particular, you've done it for external reasons. Reasons such as me not revealing my identity to you and you not seeing any statistical data. Somehow in your mind you have equated the following: Person who suggests a new idea identify revealed = Idea is true. If not, then idea is false. Person shows statistical data = Idea is true. If not, idea is false. Proceed to raise a stink to the following person. Surely, I'm not the only person that sees the faults with this logic. The problem is you don't want to deeply talk or think about the idea itself. You want to justify it's validity based on external reasons, so that you don't have to think. Judging it externally is easier for you than it is to truly understand it. It's very difficult to have an intellectual discussion with someone who thinks this way, which is why we aren't having an intellectual discussion right now.

It seems to me that you don't value reason at all. Which is ironic because reason is the most important thing man can have. It conducts our actions and behavior throughout our life. And even after research and testing is done, we have to interpret the data carefully with our reasoning abilities to draw accurate conclusions. So, not only in life, but even in the scientific process reasoning is not lost. You have not looked at the article in light of its reasoning. I'm not even sure if you understand the reasoning. Instead, you have judged the article based on your external biases. For me, that really distorts your credibility.

You continue to say right search buttons are a convention of 15+ years time used on thousand of websites on the web. Yes, that's a fact. But just because everyone is doing it and has been doing it for a long time doesn't mean it's the best way. It just means it's a very common and popular way. To find better ways of doing things, one has to be open-minded and flexible. Without this attitude, it's hard to grow and improve things. I think as the field of design starts to become more scientific, philosophical and intellectual, people should be more open-minded to new ideas, and work with each other to build up, not destroy, new ideas that improve our current way of doing things. And Reasoning, the sole authority for why we do what we do, should be valued among the community.

 

 

26 Dec 2010 - 8:05pm
Kunal Kapoor
2007

@ UXMovement (the hidden person), 
Thanks for finding the time + patience to write this email. 
Sincerely,
Kunal.
On Sun, Dec 26, 2010 at 11:29 PM, uxmovement <anthony@uxmovement.com> wrote:

Dan,

First of all, I don't think you speak for everyone here. Nobody here has delegated you that power to speak for everyone, so I think it would be more appropriate if you only spoke for yourself. The way an intellectual community works and has always worked in sciences and the like is to build new theories and make new discoveries off of other peoples ideas. This means that in order for a field or a community to progress, people in that field need to help each other and stand on each others shoulders to add to or improve theories to make new discoveries. I have suggested a new idea. Sure, it might be incomplete, it might lack in some areas, but it doesn't stop here. It's up to others in the field who actually care about making new discoveries to take this idea further, or to collaborate with others to build and develop a theory. Instead of working together, you seem to want to jostle, which is not on my agenda. Those who truly care about progressing the field of interaction design would want to experience it themselves and see for themselves if this theory holds up, or simply understand and analyze the reasoning. 

So far, it seems like you've been quick to tear everything down. You haven't done it on the basis of inconsistent or biased reasoning, or anything specific relating to the ideas of the article in particular, you've done it for external reasons. Reasons such as me not revealing my identity to you and you not seeing any statistical data. Somehow in your mind you have equated the following: Person who suggests a new idea identify revealed = Idea is true. If not, then idea is false. Person shows statistical data = Idea is true. If not, idea is false. Proceed to raise a stink to the following person. Surely, I'm not the only person that sees the faults with this logic. The problem is you don't want to deeply talk or think about the idea itself. You want to justify it's validity based on external reasons, so that you don't have to think. Judging it externally is easier for you than it is to truly understand it. It's very difficult to have an intellectual discussion with someone who thinks this way, which is why we aren't having an intellectual discussion right now.

It seems to me that you don't value reason at all. Which is ironic because reason is the most important thing man can have. It conducts our actions and behavior throughout our life. And even after research and testing is done, we have to interpret the data carefully with our reasoning abilities to draw accurate conclusions. So, not only in life, but even in the scientific process reasoning is not lost. You have not looked at the article in light of its reasoning. I'm not even sure if you understand the reasoning. Instead, you have judged the article based on your external biases. For me, that really distorts your credibility.

You continue to say right search buttons are a convention of 15+ years time used on thousand of websites on the web. Yes, that's a fact. But just because everyone is doing it and has been doing it for a long time doesn't mean it's the best way. It just means it's a very common and popular way. To find better ways of doing things, one has to be open-minded and flexible. Without this attitude, it's hard to grow and improve things. I think as the field of design starts to become more scientific, philosophical and intellectual, people should be more open-minded to new ideas, and work with each other to build up, not destroy, new ideas that improve our current way of doing things. And Reasoning, the sole authority for why we do what we do, should be valued among the community.

 

 

(((Pl
6 Dec 2010 - 4:05am
Richard Carson
2010

Hey Anthony,

I thought your illustrations presented were a little exaggerated, but did consider left search buttons. Here are my thoughts with illustrations below.

LEFT AND RIGHT BUTTONS EQUAL IN SPEED http://imgur.com/k90Ol.jpg

  1. The length in process of defining the form and button should to be the same, whether going left to right, or right to left. 2. Depending where the user places their cursor may make a difference, but here I assume that the user will place somewhere in the middle of the form for both cases. 3. Then the user will hit search when finished typing. Both layouts contain the same amount of steps to make a search. However, assumes that users blindly type words in fields without spell checking/ reviewing them as they type.

LEFT BUTTON IS FASTER (comparison) http://imgur.com/QQzdS.jpg

I brought over the same principles from above, incorporating the idea that users will spell check the letters they type in fields, within the process of 1. defining the form, 2. place cursor/ typing, and 3. hitting the search button when finished typing, as an addition fourth step. Having a left button shows that spell checking, is streamlined within the process. While the right button forces the user to zig zag back and forth from spell checking their typed words and hitting search.

MATCHING BUTTON AND TEXT JUSTIFICATION BRINGS PERFORMANCE http://imgur.com/ynxCf.jpg

However, if right search buttons had their keywords right justified, when typed in its form field, then it would be as equally streamlined as having the search button on the left. It's not the fact of having a left or right search button will perform faster, but rather when you match left buttons with left justification causes better performance when compared to a right button with left justification.

Richard Carson

8 Dec 2010 - 2:05am
cfmdesigns
2004

This looks to only be true for zero-length or very short search phrases, where the distance from the end of the typed phrase to right end of field (and thus the button) is significantly longer than the distance to the start of the field. It also discounts users having their "eye on the prize" -- the direction of typing leading to the right button vs. away from the left button -- and whether that has an effect on typing speed and accuracy and just on momentum of moving the cursor. For short fields or moderate length search phrases (a few words), the 3-to-4-to-search distances should change a lot.

Of course, it also leaves out that most semi-power users won't give a flip, because we'll press Return/Enter instead of acquiring the mouse, moving it, targeting it, and clicking.

26 Dec 2010 - 1:10pm
uxmovement
2010

Richard,

I really enjoyed and appreciate your in-depth analysis of the idea. It really shows that you understand the idea and care about it. You are building up the idea using your own experience, reasoning and analysis, which is exactly what we need in this community. 

Your discovery that the left justification is the primary determinant of a left button's success is important, and offers additional detail to the idea that I did not have before. This is how we're able to learn and improve our current way of doing things. 

I'd like to note that someone had suggested on the blog to use a right search button with right justification. However, I knew this was bad for users because right justification breaks the cursor, which would make spelling correction impossible to do.

 

26 Dec 2010 - 9:05pm
Kunal Kapoor
2007

Richard,
Let me add 2 cents of my discovery. We always place radio buttons on the left of a listing page. Correct?
Well, I had some 45 year olds, non tech savvy users come into the UT lab. 4 of them. All 4 of them, convinced us, that the radio button on a listing page (example - hotel listing, flight listing), should be on the right hand side. Saves time to scan. Well, yes, only 4 of them.
I was so damn excited, I felt like a 6 year old kid days, after months(ssss). Only to be butchered by the non-discovery mode personnel.
But yes, if and when I find time, I am going to conduct some of my research and see if this holds true for that demographic. Is that not personalized user interface?
Sincerely,
Kunal.
On Mon, Dec 27, 2010 at 12:22 AM, uxmovement <anthony@uxmovement.com> wrote:

Richard,

I really enjoyed and appreciate your in-depth analysis of the idea. It really shows that you understand the idea and care about it. You are building up the idea using your own experience, reasoning and analysis, which is exactly what we need in this community. 

Your discovery that the left justification is the primary determinant of a left button's success is important, and offers additional detail to the idea that I did not have before. This is how we're able to learn and improve our current way of doing things. 

I'd like to note that someone had suggested on the blog to use a right search button with right justification. However, I knew this was bad for users because right justification breaks the cursor, which would make spelling correction impossible to do.

 

(((
27 Dec 2010 - 2:06am
Richard Carson
2010

Hey Kunal.

That's awesome. We should hope to always feel like 6 year old kids. They learn better.... I wanted to conduct some prototype research to test ideas. I'm finding that people are coming out with some fascinating tools, I've yet to test.

It's funny how personalization is coming to mean something new and different these days.

Richard

On Dec 26, 2010, at 9:32 PM, Kunal Kapoor wrote:

> Richard, > Let me add 2 cents of my discovery. We always place radio buttons on the left of a listing page. Correct? > Well, I had some 45 year olds, non tech savvy users come into the UT lab. 4 of them. All 4 of them, convinced us, that the radio button on a listing page (example - hotel listing, flight listing), should be on the right hand side. Saves time to scan. Well, yes, only 4 of them. > I was so damn excited, I felt like a 6 year old kid days, after months(ssss). Only to be butchered by the non-discovery mode personnel. > But yes, if and when I find time, I am going to conduct some of my research and see if this holds true for that demographic. Is that not personalized user interface? > Sincerely, > Kunal. > On Mon, Dec 27, 2010 at 12:22 AM, uxmovement wrote: > >> >

27 Dec 2010 - 1:05am
Richard Carson
2010

Thanks! I get to be inspired for another day with your response to the subject. I think the concept deserves exploration, and hope to consider it for a project one day.

This topic recalls an article written a long time ago (c.2002-2004). It was about how Google has been through a lot of ideas getting to where they are now. One of the ideas, talks about a left search button recommended by one of its top programmers or engineers. The idea was said to improve performance, by a great percentage. however, the search button was kept to the right side, because Google did not want a button to be streamlined with the left-sided search results on the page, thus improving readability. As you can see, the right button nearly clears all the content and positions itself outside and right of all the contents.

The left button concept was still considered brilliant by the interviewee (some head of Google). I think the findings resulted in having the search button placed below, but closer to the center of the text field, and placed left to the "I'm Feeling Lucky" button on its homepage.

Richard

On Dec 26, 2010, at 2:03 PM, uxmovement wrote:

> Richard, > > I really enjoyed and appreciate your in-depth analysis of the idea. It really shows that you understand the idea and care about it. You are building up the idea using your own experience, reasoning and analysis, which is exactly what we need in this community. > > Your discovery that the left justification is the primary determinant of a left button's success is important, and offers additional detail to the idea that I did not have before. This is how we're able to learn and improve our current way of doing things. > > I'd like to note that someone had suggested on the blog to use a right search button with right justification. However, I knew this was bad for users because right justification breaks the cursor, which would make spelling correction impossible to do. > >
> >

1 Dec 2010 - 8:05am
Sascha Brossmann
2008

On Mon, Nov 29, 2010 at 23:38, uxmovement wrote: > I wish I could oblige you and test this theory, but I don't have the > necessary equipment, tools nor access to users to conduct a full-on study.

Bullshit. Of course you do, everybody able to set up a website has all that is necessary. You are probably just too lazy to get to the nitty-gritty. Trolling around is just so much easier and more fun, isn't it?

><((((º>

'nuf said.

Sascha

18 Nov 2010 - 7:50pm
ekim
2010

No, wait, this is my favorite (from http://uxmovement.com/design-articles/avoid-vertical-bars-for-a-minimalist-navigation):

Test­ing has it’s time and place. What you get from test­ing is quan­ti­ta­tive data. What you don’t get is qual­i­ta­tive data.

18 Nov 2010 - 8:13pm
penguinstorm
2005

We read and write from left to right 

Yeah well, some of us do. That's a pretty Euro-centric viewpoint you've got there Enrique.

18 Nov 2010 - 8:42pm
llschertler
2008

It is truly unfortunate that anyone in the UX and/or IA field would recommend or suggest something that has actually been tested and proven to be more of a liability than an asset.

The idea that some believe that "Test­ing has it’s time and place. What you get from test­ing is quan­ti­ta­tive data. What you don’t get is qual­i­ta­tive data" is LUDICROUS! 

I truly believe (based on research and my own experience) than user testing is the primary source for acquiring the most informative and qualitative data if conducted properly.

If there is the belief in the UX/IA community that "Test­ing has it’s time and place. What you get from test­ing is quan­ti­ta­tive data. What you don’t get is qual­i­ta­tive data" then I believe those individuals need to reevaluate their roles as user experience professionals.

Just my $2

uxGirlie

19 Nov 2010 - 5:05am
Alan James Salmoni
2008

Or perhaps a short course in statistics. Qualitative data is simply data that does not describe quantities. For example, sex/gender, occupation, nationality. Does anyone really want to argue that male / female can be weighed? Is female worth 2 males? Or a lawyer worth 3 accountants? While qualitative data can be summarised in a quantitative way (e.g., the study measured 3 females and 3 males), this is a summary of data and the underlying construct is qualitative.
And as for the more vague qualitative data (i.e., intangible notions, subjective feelings), these are also qualitative but testing can give a good researcher a 'feel' of the subjective base among participants. Of course, this is easily influenced by the researcher's own notions and may be of less value because it's hard to tease out the error. It is possible, it just takes a lot of effort and probably some form of measurement anyway to get a handle on it (see: inter-rater reliability).
Alan

20 Nov 2010 - 8:05pm
msweeny
2006

Hello All,

I love this conversation because it has interaction designers and IAs and user experience architects talking about search design and interaction. For all too long, Google has stupefied us with the simple box and the plain vanilla search results that everyone else copied.

I have read through the threads and will read the research with great interest. My philosophy is: 1) have a search capability and 2) have it at the top of the page, left or right.

Does anyone else know what is meant by "novice search user?" Can there be such a thing? Is it someone using the Web for the very first time?

Just wondering...

26 Nov 2010 - 3:34am
Ali Naqvi
2008

Quote: However, for novice users, right search buttons are generally slow and inefficient. To me, the solution is simple. A left search button reduces visual fixations, puts the label first and makes doing searches faster for users. 

I haven't read the article but are you basing your findings from tests done on users with a vertical left to right writing/reading system? Or can you claim the same for those who use a horizontal or a vertical right to left writing/reading system?

The reason why I ask, is that once during an Eye Tracking course, we studied immigrants from the Middle East and their navigation paths were different than the local Scandinavian users....

Ali

 

 

29 Nov 2010 - 5:56pm
bminihan
2007

I don't know if it's been said yet, but the crux of your argument seems to be based entirely on an exaggerated perception of the distance between search buttons and boxes.

That is, in the visual diagram of your argument, you have a blown-up search box and button, with very long lines running between the word "design", the button, then back to the box again.  In reality (in all of the sites you quoted, google, etc), the physical distance on the page is an order of magnitude smaller than that, when measured as the number of degrees the user has to move his or her eye or head.

Even simpler, I can glance at a search box and button in any combination on all of your popular sites, and glean the meaning of all without having to "look allll the way back from one to the other" (paraphrasing).

I would agree with your premise, if I physically had to move my head or eyeballs to find a search button on the right, but claiming that "popularity does not make a practice right" based on a very tenuous effort estimate seems a little...crazy.

16 Dec 2010 - 10:34pm
Floris Kruger
2010

Oh dear, I think I'm lost. Would someone kindly direct my toward the UX community for mature professionals? Good grief...

Guy, it's admirable that you're at least pondering the question, but I have to agree that your left-sided call to action is a no go. Other guy, no need to discredit someone for lack of a last name or impeccable credentials - what makes a good UX Designer is precisely one's own ability to discern what works and what doesn't. I'm sure you don't have to rely on someone else's research and credentials to make intelligent UX decisions, do you?

17 Dec 2010 - 7:17am
dennis_breen
2010

I think Floris 0117 just called Dan Saffer "other guy". Ha ha. I wonder how many design books you have to write to get some respect in this forum. 

18 Dec 2010 - 2:43am
Floris Kruger
2010

What commands respect is to treat others with respect in spite of one's credentials.

18 Dec 2010 - 9:01pm
ekim
2010

Floris, I feel that you've stepped into the middle of this conversation and you're missing out on the context of it all (and please understand that I'm not saying you should've known everything about this issue to begin with).  uxmovement has started several threads the past few months advertising his articles so it's not like he's a brand new user that we all decided to jump on simply because of his lack of credentials.  In reality, he's invited the UX community to think critically of what he's published which is exactly what we're doing.

I'd encourage you to go to uxmovement's website and judge the breadth of his work for yourself if you haven't already.  Take some time to peruse the comments other folks have left there (although it should be noted that he does moderate them which is why some of the argument has spilled over here).  There are some pretty obvious patterns in the type of comments you'll get, chief of which is probably that he doesn't back up what he says.

So it's really not about his credentials, although that's one common measurement we use to gauge a person's expertise, right?

Eugene

3 Jan 2011 - 3:46pm
Floris Kruger
2010

Thanks, Eugene. You're quite right... I interjected from a keyhole perspective and feel I should have shown more faith in the community's discretion anyway. I did look at his site and was taken aback by the amount of ideas being passed off as facts. Even his About writeup is a contradiction in terms. So my apologies to all - especially Dan.

Eugene, I'm personally proud of my own experience and humble accomplishments, but the measure of one's expertise is useless information that doesn't in and of itself solve any actual design problems. The manner in which credentials are accumulated, can be subject to factors that have nothing to do with exceptional mental ability. Just as there are those with serious credentials who couldn't design their way out of a cardboard box, there are those who can without. An emphasis on credentials can deter progress, because now we're waiting for a few people at the top to tell us what to do next, when we should be thinking for ourselves. In fact, history has shown that certain movers and shakers are those who step outside of our hierarchical comfort zones to perform acts of wonder no one ever dared dream of. I don't care what someone's resume says - just solve the design problem already. An Adobe employee said it beautifully when he described what we do as "Commonsense enhancements". My clients pay good money for creativity and commonsense, not credentials.

As for uxmovement, however... no proof, no play.

18 Dec 2010 - 10:00am
dennis_breen
2010

Since humour seems to be absent from this thread - Floris, I don't disagree that respecting each other is important. But this thread started with someone making a usability assertion with no evidence to back it up. They were rightly challenged to present some evidence to supports this new found "fact". Dan Saffer is a respected thought leader in this industry and he took time to point out that our community needs more than a passionate opinion to change practice. He also offered a detailed critique with some suggestions on how to gain credibility. I respect that. 

It's also worth noting that our anonymous uxmovement friend started this with the statement "Why Left Search Buttons Perform Faster Than Right". It turns out that the "why" is "because I think so". I don't respect that. As designers we need to find traction for ideas in sometimes hostile environments; places where the loudest opinion has a tradition of winning. We need to be able to explain the why in clear, compelling, evidence-based terms. If we can't do that we should go back to picking colours and fonts or whatever we used to do. 

Now, if our friend had stated a theory and asked for thoughts I could respect that. This is not a community that regularly craps on new ideas. But if you parade your opinions around as facts you'll get a cold shoulder. So thanks to the "other guys (and gals)" on this forum who keep us all honest.

18 Dec 2010 - 1:05pm
Kunal Kapoor
2007

Point to note is, that where do you draw the line? A different between doing Usability Tests in 1 country, with about 300 people, or may be 3000 people, is almost as good as 'personal opinion'. 
Beyond, that, no matter what, respect needs to be shown, and education needs to be imparted. Politely.

19 Dec 2010 - 8:29am
Dave Malouf
2005

Kunal, this doesn't make any sense. Are you suggesting that doing 3000 usability tests is akin to "personal opinion"? 
As far as country specific, well, ya gotta start somewhere to get data and decide if it merits moving on or not.

-- dave

19 Dec 2010 - 10:05am
Kunal Kapoor
2007

Dave,

I am not sure what the number is. I infer that you need to start in multiple countries, before suggesting if it is worth moving ahead or not.

sincerely,

Kunal.

On 12/19/10, Dave Malouf wrote: > Kunal, this doesn't make any sense. Are you suggesting that doing 3000 > usability tests is akin to "personal opinion"? > As far as country specific, well, ya gotta start somewhere to get data and > decide if it merits moving on or not. >

20 Dec 2010 - 9:55am
Dave Malouf
2005

Studies in anthropology and psychology have proven time and time again that there are more similarities between "defined" groups than differences. Dost think you are being over sensitive to the "international" equation. Tons of international brands, apps have had broad appeal, success w/o differentiation. I would suggest what is more important is a single heterogenous population than going out of your way to multiple countries. Just purely from an ROI perspective.

20 Dec 2010 - 11:05am
Kunal Kapoor
2007

Chief,

I am sure you are convinced.

Amen.

Sincerely,

Kunal.

On 12/20/10, Dave Malouf wrote: > Studies in anthropology and psychology have proven time and time again that > there are more similarities between "defined" groups than differences. Dost > think you are being over sensitive to the "international" equation. Tons of > international brands, apps have had broad appeal, success w/o > differentiation. I would suggest what is more important is a single > heterogenous population than going out of your way to multiple countries. > Just purely from an ROI perspective. > >

19 Dec 2010 - 5:41am
Floris Kruger
2010

Dennis, thanks for your thoughtful response. I wouldn't have brought this up if I didn't believe Mr Saffer is perfectly capable of delivering his point of view minus the condescending tone.

Dan, your experience, accomplishments and perceived seniority in this community are precisely what hold your conduct to a high standard of accountability.

Kunal, I think you said it best. At a minimum, politeness cultivates a receptive mindset. Anything less creates communication barriers. The negative tone in their delivery most likely impaired uxmovement's ability to be receptive to their otherwise valid and perhaps well intended arguments, depriving him of the opportunity to learn from this experience in a constructive manner.

19 Dec 2010 - 9:05am
Ian Fenn
2007

On 19 Dec 2010, at 11:02, Floris Kruger wrote: > Dennis, thanks for your thoughtful response. I wouldn't have brought this up if I didn't believe Mr Saffer is perfectly capable of delivering his point of view minus the condescending tone.

For the record, I didn't find Dan's tone condescending. I am, however, slightly irritated that a relatively sensible discussion has been derailed by allegations otherwise.

>

19 Dec 2010 - 12:05pm
Dan Saffer
2003

On Dec 19, 2010, at 3:32 AM, Floris Kruger wrote:

> Dan, your experience, accomplishments and perceived seniority in this community are precisely what hold your conduct to a high standard of accountability.

Was my tone condescending? Probably, and I apologize for that. (You should have seen the first draft.)

However, the IxDA exists to educate its members, and dissemination of untested ideas presented as fact by an anonymous poster is just bad practice. As part of a professional organization, this list shouldn't have any anonymous members, in my opinion. The value of the list is in getting information and answers to questions by peers—knowledgable, trusted information. If you don't have that, what good is the list at all?

As it stands, there is no way to modify, endorse, or rebut anyone's answers on the list, other than to send another email. Compared to sites like Quora, it's a primitive system. I'd urge you all to look at Quora, by the way, and see how they handle this. It's certainly become my go-to source for interaction design questions, simply because I trust the answers more because they're signed and responders vote on them. It's combination of experts and wisdom of the crowd. Going forward, it would be great to see IxDA adopt a similar model.

Dan

Dan Saffer Principal, Kicker Studio http://www.kickerstudio.com http://www.odannyboy.com

19 Dec 2010 - 8:33am
Dave Malouf
2005

To get back to the topic. I don't understand something. why are we assuming that the button & label are always together?

Also, even if the button is the label, I have to squarely stand w/ convention att his point.  the text box to button is such a ubiquitous pattern that its gestalt [text box here] [btn] is symbolic in nature alone.

as for the credential bit. We all started sometime and need a forum, but what bothers me is not having a name. Anonymity to me makes it hard to want to push forward the conversation, but I realize in replying that I'm speaking more to the others in the community and less directly to whoever this person is.

-- dave

19 Dec 2010 - 10:05am
Kunal Kapoor
2007

People evolve, and may be the conventions, too need to involve people, and evolve, to be effective, and efficient.

That is how designers contribute to potray what people really want, conventional, or un-conventional.

Sincerely,

Kunal.

On 12/19/10, Dave Malouf wrote: > To get back to the topic. I don't understand something. why are we assuming > that the button & label are always together? > > Also, even if the button is the label, I have to squarely stand w/ > convention > att his point.  the text box to button is such a ubiquitous pattern that its > > gestalt [text box here] [btn] is symbolic in nature alone. > > as for the credential bit. We all started sometime and need a forum, but > what > bothers me is not having a name. Anonymity to me makes it hard to want to > push forward the conversation, but I realize in replying that I'm speaking > more to the others in the community and less directly to whoever this person > > is. >

20 Dec 2010 - 10:00am
Dave Malouf
2005

Kunal, no one is suggesting that breaking away from convention is bad, but you need to do so w/ evidence.

you and this other guy who appears to want to remain nameless have only stood on 2 topics: change happens (and can be good) which no one is arguing against and that unless you are going to test everyone data is meaningless which is an absurd statement which just goes back to "why bother at all" which is in direct contradiction to UCD theory and practice that IxD is built on. Let's just do art!

-- dave

20 Dec 2010 - 11:05am
Kunal Kapoor
2007

Chief,

I prefer Partipatory design, which, in my humble opinion covers the short comings of UCD.

In today's world, more than ever, you need science + art.

Of course, do resist, saying stuff like, me and this other person....

Sincerely,

Kunal

On 12/20/10, Dave Malouf wrote: > Kunal, no one is suggesting that breaking away from convention is bad, but > you need to do so w/ evidence. > > you and this other guy who appears to want to remain nameless have only > stood > on 2 topics: change happens (and can be good) which no one is arguing > against > and that unless you are going to test everyone data is meaningless which is > an absurd statement which just goes back to "why bother at all" which is in > direct contradiction to UCD theory and practice that IxD is built on. Let's > just do art! >

21 Dec 2010 - 11:25am
Dave Malouf
2005

Kunal, that is not an answer. Some would call Participatory design just a specific method within UCD.

What's worse is that it has a strict limitation on #'s which totally counters your initial point. 

-- dave

21 Dec 2010 - 12:05pm
Kunal Kapoor
2007

It is an answer. 
It does not counter in my mind, only in yours.
Sincerely,
Kunal.

On Tue, Dec 21, 2010 at 10:31 PM, Dave Malouf <dave.ixd@gmail.com> wrote:

Kunal, that is not an answer. Some would call Participatory design just a specific method within UCD.

What's worse is that it has a strict limitation on #'s which totally counters your initial point. 

19 Dec 2010 - 11:48am
Jared M. Spool
2003

I've disliked the switch from people's names to handles with the Conan upgrade.

Personally, I think the quality of posts has degraded since people are now hiding behind handles.

That's my opinion. I have no data to back it up.

Jared

19 Dec 2010 - 2:05pm
Cwodtke
2004

which research shows having a not having a submit button makes no difference? How many users were in the study? Was it significant?
In your theory, i do not see any testing or data, just a hypothesis. It's an intriguing one, but I wish you wouldn't say it "performs" better without having run a statistically significant number through a multivariate test. 
Still, cool theory!

On Tue, Nov 16, 2010 at 8:08 PM, uxmovement <anthony@uxmovement.com> wrote:

Research has shown that having or not having a search button makes no difference to pro users. However, for novice users, right search buttons are generally slow and inefficient. To me, the solution is simple. A left search button reduces visual fixations, puts the label first and makes doing searches faster for users. 

The pattern left search buttons follow is similar to IXDA's search where the combo box and icon come before the text field. The label and button have more visual weight and are what users need to see first before they carry out a search, which is why I believe they should come before the text field. Despite the fact that most websites and search engines have it backwards, what are your professional and experienced independent thoughts on this matter?

http://uxmovement.com/design-articles/left-search-buttons-perform-faster-than-right [1] for further detail.

 

 

(((Ple
Syndicate content Get the feed