RE: [WSG] Default state of radio buttons. (Maybe OT?)
I reckon this discussion is relevant here too. The one I'm struggling with at the moment is a form with Select Gender M / F. It's a government web site so I'm trying to be careful. Checkboxes are inappropriate. I could use a select with M and F but then what for unspecified or don't want to answer or I just ignored that question? Also, whatever is chosen takes 2 clicks. Unchecked radio buttons are a very usable choice, you hit one or the other or neither and move on. Mike said ...it is sometimes invalid communications/user interface to have one and only one 'checked' item at all times. I said, and still hold to the view, that sometimes a form has to be presented with none of the radio buttons 'checked'. I agree. The standard (the HTML spec) says that's OK, and that's how we want it to work for usability. All good. But RFC1866 says the user agent should make the choice - user agent must check the first radio button of the set initially. So some standards-compliant browser is going to over-ride my standards-compliant form and check one of the radio buttons? Isn't this a conflict between the standards for UI and User Agent? -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of John Horner Sent: Wednesday, 2 February 2005 5:25 p.m. To: wsg@webstandardsgroup.org Subject: Re: [WSG] Default state of radio buttons. (Maybe OT?) That's the relevance to standards - i.e. that if it's only standard if there is a default radio button and never valid if none of them are 'checked' then the standard is wrong and ought to be changed. I heartily agree, Mike. Have You Validated Your Code? John Horner(+612 / 02) 9333 3488 Senior Developer, ABC Online http://www.abc.net.au/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
I reckon this discussion is relevant here too. The one I'm struggling with at the moment is a form with Select Gender M / F. It's a government web site so I'm trying to be careful. Checkboxes are inappropriate. I could use a select with M and F but then what for unspecified or don't want to answer or I just ignored that question? Also, whatever is chosen takes 2 clicks. Unchecked radio buttons are a very usable choice, you hit one or the other or neither and move on. Mike said ...it is sometimes invalid communications/user interface to have one and only one 'checked' item at all times. I said, and still hold to the view, that sometimes a form has to be presented with none of the radio buttons 'checked'. I agree. What happens in this scenario I check a radio button Then I think, no, I don't want to check any of them How do I uncheck the radio button checked without setting one of the others in the same group? Reset the whole form? So now, by supposedly giving me the option of not making a choice, you force me to make a choice. Radio button groups exist for precisely the situation where there is one and only one option that must be chosen. That's how they have always worked. The web has enough trouble with people inventing their own buttons, styling scrollbars with Microsoft's CSS like scrollbar style abominations, inventing their own scrollbars (why does every flash site have to do that?). The UI conventions of the desktop have been around for a generation now. They represent the baseline of user expectations about how an interface should look, and work. Their appearance and behavior are burned deep into the unconscious of all computer users. Let's not keep reinventing the wheel. HTML and the platforms on which our browsers run provide perfectly good UI widgets and behaviours that users are used to. Now to try to be useful In this scenario The one I'm struggling with at the moment is a form with Select Gender M / F. It's a government web site so I'm trying to be careful. Checkboxes are inappropriate. I could use a select with M and F but then what for unspecified or don't want to answer or I just ignored that question? Also, whatever is chosen takes 2 clicks. Yes, checkboxes are not appropriate, as there should be at most one answer. Two checkboxes could allow 2 answers. What is wrong with a popup menu, the initial option being No answer and the other options Male and Female? The UI itself tells them they need not answer. If you had radio buttons, you'd need a label that said you do not have to answer this question and then you have to hope the user sees it (whereas in the popup menu case, they see it while making the choice.) And we also avoid the problem of them checking a radio button then not being able to uncheck it. They also need only click once. So it seems that in this case the popup menu is superior to misusing radio buttons. John -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of John Horner Sent: Wednesday, 2 February 2005 5:25 p.m. To: wsg@webstandardsgroup.org Subject: Re: [WSG] Default state of radio buttons. (Maybe OT?) That's the relevance to standards - i.e. that if it's only standard if there is a default radio button and never valid if none of them are 'checked' then the standard is wrong and ought to be changed. I heartily agree, Mike. Have You Validated Your Code? John Horner(+612 / 02) 9333 3488 Senior Developer, ABC Online http://www.abc.net.au/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** John Allsopp :: westciv :: http://www.westciv.com/ software, courses, resources for a standards based web :: style master blog :: http://westciv.typepad.com/dog_or_higher/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Radio button groups exist for precisely the situation where there is one and only one option that must be chosen. That's how they have always worked. John hit the nail on the head. If you come to a situation where you're attempting to use radio buttons and the above does not apply, then you need something other than radio buttons. Radio buttons are like on/off light switches. The light switch is either on or its off. 1. If you're in a situation where you want a default value checked, but want to allow the user to deselect it without inserting another value, use a check box. 2. If you want a default value checked, but want to allow the user to alternately choose a different value (up to four choices), then use a radio button. More than four choices, use a select menu. While it appears that the standards allow you to NOT assign a default value to a radio button set, I'd say this is non-intuitive from a user experience standpoint. Plus, a value will be chosen by the browser anyhow, if I'm reading the previous posts correctly. So in this specific situation, where the user is allowed to choose male, female or no answer, number 2 above seems to be the most applicable. And seeing as they are allowed to not specify gender, I would default to no answer. Even if they might have provided an answer and simply forgot to, you've signified that the answer is not important/relevant, so it shouldn't make any difference to you if they willingly chose to specify gender or not. HTH Regards, Ron ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Ron, I agree with your summary except for More than four choices, use a select menu. Surely it is better to show as many options as possible at the same time? Stating four as a reasonable limit seems a little low. Although with select inputs I would state Selects with less than 5 options should be coded as radio buttons it does not mean quite the same thing. If anyone has a little spare time I'd appreciate comments / corrections on a proposed set of Accessible and usable form guidelines I've compiled: http://www.websemantics.co.uk/tutorials/form_guidelines/ there's 17 of them so far. This has been a WIP since September. I'm hoping to gain further clarity from this discussion over how best to define radio inputs and boolean choices. mike 2k:)2 marqueeblink e-mail: [EMAIL PROTECTED] site: http://www.webSemantics.co.uk /marquee/blink -Original Message- From: Pringle, Ron [mailto:[EMAIL PROTECTED] Sent: 02 February 2005 15:00 To: 'wsg@webstandardsgroup.org' Subject: RE: [WSG] Default state of radio buttons. (Maybe OT?) Radio button groups exist for precisely the situation where there is one and only one option that must be chosen. That's how they have always worked. John hit the nail on the head. If you come to a situation where you're attempting to use radio buttons and the above does not apply, then you need something other than radio buttons. Radio buttons are like on/off light switches. The light switch is either on or its off. 1. If you're in a situation where you want a default value checked, but want to allow the user to deselect it without inserting another value, use a check box. 2. If you want a default value checked, but want to allow the user to alternately choose a different value (up to four choices), then use a radio button. More than four choices, use a select menu. While it appears that the standards allow you to NOT assign a default value to a radio button set, I'd say this is non-intuitive from a user experience standpoint. Plus, a value will be chosen by the browser anyhow, if I'm reading the previous posts correctly. So in this specific situation, where the user is allowed to choose male, female or no answer, number 2 above seems to be the most applicable. And seeing as they are allowed to not specify gender, I would default to no answer. Even if they might have provided an answer and simply forgot to, you've signified that the answer is not important/relevant, so it shouldn't make any difference to you if they willingly chose to specify gender or not. HTH Regards, Ron ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
John Horner mailto:[EMAIL PROTECTED] on Tuesday, February 01, 2005 5:06 PM said: 1) Three radio buttons, one of them pre-selected for the value nothing or incorrect input -- the user is encouraged to select one of the other two and an error given by the script if they don't 2) A select menu, with one option pre-selected for nothing as above. The first seems kind of illogical to me, and the users will be more used to the second. I decided to go with #2. The options being, -, Home, and Agency. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
On Wed, 2005-02-02 at 21:35, John Allsopp wrote: The UI conventions of the desktop have been around for a generation now. They represent the baseline of user expectations about how an interface should look, and work. Their appearance and behaviour are burned deep into the unconscious of all computer users. Let's not keep reinventing the wheel. Well said John. This kind of mentality is the key to creating intuitive interfaces that everyone can easily understand and use. On the other hand, you could copy some of these.. Interface Hall of Shame http://digilander.libero.it/chiediloapippo/Engineering/iarchitect/shame.htm Regards Chris Blown ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
I think it's the good old checked=checked attribute that you add in your default radio button's code. HTH Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Iain Gardiner mailto:[EMAIL PROTECTED] on Tuesday, February 01, 2005 12:04 PM said: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. By mailing address I meant: postal mail. And by both options will probably be chosen an equal amount of times, as has been the case in the past I meant: Neither option is generally chosen more than the other. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
I think this is like FAQs - my FAQ is never there. Likewise, pre-set a control to option A and I'm equally likely to want option B. IMHO pre-setting options for the user (unless they're VERY obvious) is like making assumptions about them. That said, RFC1866 says 'CHECKED' is optional but then says At all times, exactly one of the radio buttons in a set is checked. If none of the INPUT elements of a set of radio buttons specifies 'CHECKED', then the user agent must check the first radio button of the set initially. Does anybody know why one button has to be checked at all times in a circumstance like Chris referred to? Thanks Rowan Quoting Iain Gardiner [EMAIL PROTECTED]: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
With radio buttons, no value is passed to the form's action page unless one of the options is selected. This will normally cause an error in the processing page unless special consideration is given to this possibility.Normally if there are radio buttons on the form, it is best to ensure that at least one option is selected.CheersMike KearAFP Webworks Pty LtdWindsor, NSW, Australiahttp://afpwebworks.comIndustrial strength coldfusion, .asp, .asp.net, php hosting from A$15/month - Original Message From: wsg@webstandardsgroup.orgTo: "wsg@webstandardsgroup.org" wsg@webstandardsgroup.orgSubject: RE: [WSG] Default state of radio buttons. (Maybe OT?)Date: 01/02/05 20:56I think this is like FAQs - my FAQ is never there. Likewise, pre-set a controlto option A and I'm equally likely to want option B.IMHO pre-setting options for the user (unless they're VERY obvious) is likemaking assumptions about them.That said, RFC1866 says 'CHECKED' is optional but then says "At all times,exactly one of the radio buttons in a set is checked. If none of the INPUTelements of a set of radio buttons specifies 'CHECKED', then the user agentmust check the first radio button of the set initially."Does anybody know why one button has to be checked "at all times" in acircumstance like Chris referred to?ThanksRowanQuoting Iain Gardiner [EMAIL PROTECTED]: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help The discussion list for http://webstandardsgroup.org/See http://webstandardsgroup.org/mail/guidelines.cfmfor some hints on posting to the list getting help** Message sent using UebiMiau 2.7.2 ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help **
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Or pass hidden parameters onto the action page ... these then can be over ridden if the radio is selected From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mike KearSent: Wednesday, 2 February 2005 2:51 AMTo: wsg@webstandardsgroup.orgSubject: RE: [WSG] Default state of radio buttons. (Maybe OT?) With radio buttons, no value is passed to the form's action page unless one of the options is selected. This will normally cause an error in the processing page unless special consideration is given to this possibility.Normally if there are radio buttons on the form, it is best to ensure that at least one option is selected.CheersMike KearAFP Webworks Pty LtdWindsor, NSW, Australiahttp://afpwebworks.comIndustrial strength coldfusion, .asp, .asp.net, php hosting from A$15/month - Original Message From: wsg@webstandardsgroup.orgTo: "wsg@webstandardsgroup.org" wsg@webstandardsgroup.orgSubject: RE: [WSG] Default state of radio buttons. (Maybe OT?)Date: 01/02/05 20:56I think this is like FAQs - my FAQ is never there. Likewise, pre-set a controlto option A and I'm equally likely to want option B.IMHO pre-setting options for the user (unless they're VERY obvious) is likemaking assumptions about them.That said, RFC1866 says 'CHECKED' is optional but then says "At all times,exactly one of the radio buttons in a set is checked. If none of the INPUTelements of a set of radio buttons specifies 'CHECKED', then the user agentmust check the first radio button of the set initially."Does anybody know why one button has to be checked "at all times" in acircumstance like Chris referred to?ThanksRowanQuoting Iain Gardiner [EMAIL PROTECTED]: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help The discussion list for http://webstandardsgroup.org/See http://webstandardsgroup.org/mail/guidelines.cfmfor some hints on posting to the list getting help**Message sent using UebiMiau 2.7.2** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
IIRC though, while RFC 1866 says exactly one the checked attribute is optional in W3C guidelines. HTML 4.01 (and thus also the XHTML 1 series) state: [http://www.w3.org/TR/html4/interact/forms.html#radio] If no radio button in a set sharing the same control name is initially on, user agent behavior for choosing which control is initially on is undefined. So if you leave them both undefined, some UAs will select the first, some will select neither. I can't see many doing anything else. It's not standards aversion and I can't see why that's not ok. HTML4.01 spec goes on to say that authors should ensure one is selected by default but I'm not sure why you need consistant behavior in this case. If worst comes to worst, just have whichever one comes first in the page checked=checked. You'll be inconveniancing roughly half the people no matter which is checked by default. If neither are then it's the same inconvenience for everyone. I guess I'm saying that I'd consider this a moot point - check whichever you'd like and the user will deal with it. Regards, mjec On Wed, 2 Feb 2005 08:54:31 +1100, Wybrow, Mark [EMAIL PROTECTED] wrote: Or pass hidden parameters onto the action page ... these then can be over ridden if the radio is selected From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mike Kear Sent: Wednesday, 2 February 2005 2:51 AM To: wsg@webstandardsgroup.org Subject: RE: [WSG] Default state of radio buttons. (Maybe OT?) With radio buttons, no value is passed to the form's action page unless one of the options is selected. This will normally cause an error in the processing page unless special consideration is given to this possibility. Normally if there are radio buttons on the form, it is best to ensure that at least one option is selected. Cheers Mike Kear AFP Webworks Pty Ltd Windsor, NSW, Australia http://afpwebworks.com Industrial strength coldfusion, .asp, .asp.net, php hosting from A$15/month - Original Message From: wsg@webstandardsgroup.org To: wsg@webstandardsgroup.org wsg@webstandardsgroup.org Subject: RE: [WSG] Default state of radio buttons. (Maybe OT?) Date: 01/02/05 20:56 I think this is like FAQs - my FAQ is never there. Likewise, pre-set a control to option A and I'm equally likely to want option B. IMHO pre-setting options for the user (unless they're VERY obvious) is like making assumptions about them. That said, RFC1866 says 'CHECKED' is optional but then says At all times, exactly one of the radio buttons in a set is checked. If none of the INPUT elements of a set of radio buttons specifies 'CHECKED', then the user agent must check the first radio button of the set initially. Does anybody know why one button has to be checked at all times in a circumstance like Chris referred to? Thanks Rowan Quoting Iain Gardiner [EMAIL PROTECTED]: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not sure if this is off topic or not, but let me know if it is. I'm wondering what the suggested default state of a group of radio buttons is? Let me use a current, specific example. In a form I'm writing I have one set of radio buttons. The current options are 'Home', or 'Agency'. The radio button is meant to designate what type of mailing address the customer has provided. Right now I've got neither option being defaulted to. I know that radio buttons should have exactly one option chosen at all times, but in this case it doesn't make sense to add a third option of 'None', or have the group default to one option or the other. How should I handle this? Should I bite the bullet and have the options default to one of the options (both options will probably be chosen an equal amount of times, as has been the case in the past)? Or maybe I should go to a drop down list with three options? 1. '-', 2. 'Home', 3. 'Agency' Your feedback is appreciated. Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail
RE: [WSG] Default state of radio buttons. (Maybe OT?)
Michael Cordover mailto:[EMAIL PROTECTED] on Tuesday, February 01, 2005 2:51 PM said: IIRC though, while RFC 1866 says exactly one the checked attribute is optional in W3C guidelines. [snip] I guess I'm saying that I'd consider this a moot point - check whichever you'd like and the user will deal with it. Sounds good to me. Thanks! Chris. ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
There are reasons why you might not want to select a 'default' on radio buttons. It can distort your data.For example, if you have option 1 checked as the default, and a user forgets to choose one of the options, they're selecting option 1 anyway. This may be erroneous data. (Or it may not matter in which case, do whatever you like). For example in a survey, if youindicate any default answers, you are automatically slanting the results, and if someone doesnt make a choice to a question, they wont get a warning popup, instead they will have a selection of your default entered. How will you know which of the results are where users have selected option 1, and which are where they have forgotten to give an answer at all?That's not to say you shouldn't use default answers. I'm saying you shouldn't always use defaults just because the 'normal' practice is to do so. If the radio button gathers important data, another practice could well be to offer no defaults, but instead show a warning or some kind if the user doesnt select one or the other and force them to choose before you process the rest of the form.CheersMike KearAFP Webworks Pty LtdWindsor, NSW, Australia Http://afpwebworks.comIndustrial Strength hosting - coldfusion, .asp, .asp.net, php from AUD$15/month - Original Message From: wsg@webstandardsgroup.orgTo: "wsg@webstandardsgroup.org" wsg@webstandardsgroup.orgSubject: Re: [WSG] Default state of radio buttons. (Maybe OT?)Date: 01/02/05 22:53IIRC though, while RFC 1866 says "exactly one" the checked attributeis optional in W3C guidelines.HTML 4.01 (and thus also the XHTML 1 series) state:[http://www.w3.org/TR/html4/interact/forms.html#radio]If no radio button in a set sharing the same control name isinitially "on", user agent behavior for choosing which control isinitially "on" is undefined.So if you leave them both undefined, some UAs will select the first,some will select neither. I can't see many doing anything else. It'snot standards aversion and I can't see why that's not ok. HTML4.01spec goes on to say that authors should ensure one is selected bydefault but I'm not sure why you need consistant behavior in thiscase.If worst comes to worst, just have whichever one comes first in thepage checked="checked". You'll be "inconveniancing" roughly half thepeople no matter which is checked by default. If neither are thenit's the same "inconvenience" for everyone.I guess I'm saying that I'd consider this a moot point - checkwhichever you'd like and the user will deal with it.Regards,mjecOn Wed, 2 Feb 2005 08:54:31 +1100, Wybrow, Mark[EMAIL PROTECTED] wrote: Or pass hidden parameters onto the action page ... these then can be over ridden if the radio is selected From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Mike Kear Sent: Wednesday, 2 February 2005 2:51 AM To: wsg@webstandardsgroup.org Subject: RE: [WSG] Default state of radio buttons. (Maybe OT?) With radio buttons, no value is passed to the form's action page unless one of the options is selected. This will normally cause an error in the processing page unless special consideration is given to this possibility. Normally if there are radio buttons on the form, it is best to ensure that at least one option is selected. Cheers Mike Kear AFP Webworks Pty Ltd Windsor, NSW, Australia http://afpwebworks.com Industrial strength coldfusion, .asp, .asp.net, php hosting from A$15/month - Original Message From: wsg@webstandardsgroup.org To: "wsg@webstandardsgroup.org" wsg@webstandardsgroup.org Subject: RE: [WSG] Default state of radio buttons. (Maybe OT?) Date: 01/02/05 20:56 I think this is like FAQs - my FAQ is never there. Likewise, pre-set a control to option A and I'm equally likely to want option B. IMHO pre-setting options for the user (unless they're VERY obvious) is like making assumptions about them. That said, RFC1866 says 'CHECKED' is optional but then says "At all times, exactly one of the radio buttons in a set is checked. If none of the INPUT elements of a set of radio buttons specifies 'CHECKED', then the user agent must check the first radio button of the set initially." Does anybody know why one button has to be checked "at all times" in a circumstance like Chris referred to? Thanks Rowan Quoting Iain Gardiner [EMAIL PROTECTED]: Oops, sorry I didn't really read your question thoroughly. Surely an e-mail address will be either a personal or a business address. Personally I'd set the default to personal as this seems to me the most likely option. Iain -- Iain Gardiner http://www.firelightning.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Chris W. Parker Sent: 01 February 2005 19:12 To: wsg@webstandardsgroup.org Subject: [WSG] Default state of radio buttons. (Maybe OT?) Hello, Not su
Re: [WSG] Default state of radio buttons. (Maybe OT?)
On Tue, 1 Feb 2005 17:24:35 -, Mike Kear [EMAIL PROTECTED] wrote: For example in a survey, if you indicate any default answers, you are automatically slanting the results, and if someone doesnt make a choice to a question, they wont get a warning popup, instead they will have a selection of your default entered. How will you know which of the results are where users have selected option 1, and which are where they have forgotten to give an answer at all? Good surveys need don't know/prefer not to answer for every question and that might be a good default. -- regards, Kornel Lesiski http://browsehappy.pl ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
I'm a bit baffled by this question -- in a way it's not about CSS or HTML or standards at all. If you want to get good data from your form, and you have two radio buttons, then neither should be checked by default. Your CGI script or whatever should do the checking and return the form with you must select a button to proceed. But should the browser really auto-select radio buttons for the user if none is selected? That seems like a very strange decision to me, and wouldn't allow your script to check the form as above. So that argues that you should use a different kind of control, for instance a SELECT menu. The value of none can be the default, and your script can chuck an error if it sees none submitted. Have You Validated Your Code? John Horner(+612 / 02) 9333 3488 Senior Developer, ABC Online http://www.abc.net.au/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
Perhaps, Kornel, but in that case how to you tell the difference between responses where people preferred not to say/didn't know (i.e. an answer to the question), and where people didn't answer the question or didnt notice it?There are cases where a default is a bad thing, and you need to be able to handle forms with no answer, rather than providing a default. In another example, providing a default may slant your responses in favour of the default, because respondents think that's the answer you're wanting. Ever-obliging they tend to give the answer they think the questioner wants.--CheersMike KearWindsor, NSW, AustraliaCertified Advanced ColdFusion DeveloperAFP Webworkshttp://afpwebworks.comColdFusion, PHP, ASP, ASP.NET hosting from AUD$15/month - Original Message Good surveys need "don't know/prefer not to answer" for every question andthat might be a good default.--regards, Kornel LesiƱskihttp://browsehappy.pl Message sent using UebiMiau 2.7.2 ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
John, If you want to get good data from your form, and you have two radio buttons, then neither should be checked by default. Your CGI script or whatever should do the checking and return the form with you must select a button to proceed. the difference between a radio button set and checkboxes is (going way back to Mac UI guidelines in the 80s) is that a radio button group always has a value, and the values are mutually exclusive (you can't check two radio buttons in the same group), whereas checkboxes in a group can all be on, all be off, or any combination of on and off. It's a long time since I have done some intensive forms work, but IIRC, browsers generally managed this for you if you set up your radio button sets properly. I know, its strictly off topic, but its about conventions, which are a kind of standard john John Allsopp :: westciv :: http://www.westciv.com/ software, courses, resources for a standards based web :: style master blog :: http://westciv.typepad.com/dog_or_higher/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
At 11:36 AM +1100 2/2/05, John Allsopp wrote: John, the difference between a radio button set and checkboxes is (going way back to Mac UI guidelines in the 80s) is that a radio button group always has a value [snip -- I did know the difference between checkboxes and radio buttons!] If you're saying that a set of radio two or more buttons must always be shown with at least one pre-selected, i.e. as soon as the page loads, one is already selected, then there are only two options for the original poster to get good information from their form: 1) Three radio buttons, one of them pre-selected for the value nothing or incorrect input -- the user is encouraged to select one of the other two and an error given by the script if they don't 2) A select menu, with one option pre-selected for nothing as above. The first seems kind of illogical to me, and the users will be more used to the second. Have You Validated Your Code? John Horner(+612 / 02) 9333 3488 Senior Developer, ABC Online http://www.abc.net.au/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
John, [snip -- I did know the difference between checkboxes and radio buttons!] I was almost certain you did :-) If you're saying that a set of radio two or more buttons must always be shown with at least one pre-selected, i.e. as soon as the page loads, one is already selected, then there are only two options for the original poster to get good information from their form: Well, by convention, that is what radio buttons should do. 1) Three radio buttons, one of them pre-selected for the value nothing or incorrect input -- the user is encouraged to select one of the other two and an error given by the script if they don't 2) A select menu, with one option pre-selected for nothing as above. If there are 4 or fewer responses, that is where radio buttons have traditionally been recommended. For more, popup menus. I'd suggest a radio button with no response or some such, which is selected, and then the other two The first seems kind of illogical to me, and the users will be more used to the second. Given this limitation of radio buttons, I wonder why it has not turned up in UI discussion much. I've never seen it as an issue before (althoough I do see it as one in this instance) j John Allsopp :: westciv :: http://www.westciv.com/ software, courses, resources for a standards based web :: style master blog :: http://westciv.typepad.com/dog_or_higher/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
I hesitantly suggest a good place for this discussion would be on Justin French's Interface list. http://lists.indent.com.au/mailman/listinfo/interface Cheers Chris Blown ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
Well I got involved in it because (i thought) someone said at the beginning of this thread that it was only valid markup if a set of radio buttons had one and only one 'checked' item.My point was that regardless of the validity of the code, it is sometimes invalid communications/user interface to have one and only one 'checked' item at all times. I said, and still hold to the view, that sometimes a form has to be presented with none of the radio buttons 'checked'.That's the relevance to standards - i.e. that if it's only standard if there is a default radio button and never valid if none of them are 'checked' then the standard is wrong and ought to be changed.CheersMike KearWindsor, NSW, AustraliaCertified Advanced ColdFusion DeveloperAFP Webworkshttp://afpwebworks.comColdFusion, PHP, ASP, ASP.NET hosting from AUD$15/month - Original Message From: wsg@webstandardsgroup.orgTo: "WSG" wsg@webstandardsgroup.orgSubject: Re: [WSG] Default state of radio buttons. (Maybe OT?)Date: 02/02/05 03:56I hesitantly suggest a good place for this discussion would be on JustinFrench's Interface list.http://lists.indent.com.au/mailman/listinfo/interfaceCheersChris Blown Message sent using UebiMiau 2.7.2 ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list & getting help **
Re: [WSG] Default state of radio buttons. (Maybe OT?)
That's the relevance to standards - i.e. that if it's only standard if there is a default radio button and never valid if none of them are 'checked' then the standard is wrong and ought to be changed. I heartily agree, Mike. Have You Validated Your Code? John Horner(+612 / 02) 9333 3488 Senior Developer, ABC Online http://www.abc.net.au/ ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **