Re: [WSG] divs, layers, trans and strict
Bob, I just opened both pages in Safari, and they seem to be/look the same to me, so the first question is which browser(s) are you testing in? Justin On 02/01/2005, at 1:15 AM, designer wrote: Hello all, and a good new year to you all! I am 'playing' with some layers which sit on top of each other to create a specific effect. I have converted the layers to CSS in the header (instead of inline styling) and it works fine as HTML4.01 trans. it can be seen at [1]. However, as soon as I make it strict, the positioning changes, as shown in [2]. Both files validate OK, and so does the CSS, so why are they not behaving properly? - It's probably staring at me, but I just can't see it! Would any of you alert new year revellers take a look for me please? [1] http://www.treyarnon.fsworld.co.uk [2] http://www.treyarnon.fsworld.co.uk/index_strict.html Many thanks for your time and expertise, Bob McClelland, Cornwall (U.K.) www.gwelanmor-internet.co.uk ** The discussion list for http://webstandardsgroup.org/ See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** --- Justin French, Indent.com.au [EMAIL PROTECTED] Web Application Development Graphic Design ** 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] New Zealand Web Standards??
On 07/12/2004, at 2:19 PM, Darren Wood wrote: I get _very_ depressed when i see high profile[1] new zealand sites completely drop the ball[2]... I'm guessing there's an equal percentage of Australian sites who neglect standards. I don't think it's a NZ-specific issue, and nor do I think us Aussies are particularly standards compliant. Obviously us here on the list and many more are, but for every design studio that embraces standards, there are many more who do not. David Trewern Design (dtdesign.com.au) is a high profile Australian studio that builds great looking sites for high profile companies, but even their brand new sites are based on tables and a lack of standards, yet they're getting some of the biggest and highest paying projects in Australia. I just let them do their thing, and I'll do mine. Frankly, I don't want or need them to embrace standards -- it creates a point of difference between myself and them, and the longer they build non-compliant sites, the more legacy mark-up and bloat they leave behind (possibly for people like me to clean up a few years from now). What they do doesn't directly affect me in any way, so like I said, I'll just do my thing, and they can do theirs. (I'm sure their must be more, and I'd love see them so I can feel a little better about being a web developer in NZ.) Am i missing something? Am i over sensitive? Why don't you see this as a good thing? It sounds like you one of only a handful of developers that get it... that creates a point of difference amongst your competitors, and ultimately an advantage (you can build and maintain sites faster than they can). If I were you, I'd stop complaining and get in there! --- Justin French, Indent.com.au [EMAIL PROTECTED] Web Application Development Graphic Design ** 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] Correct use of fieldset
On 23/11/2004, at 2:32 PM, [EMAIL PROTECTED] wrote: Every example I've come across so far (and I've seen *a lot* of examples) has used it within this context. Is it just that grouping input is the most logical and common used example? Or can it be used in other contexts? It's only intended to group form controls, and I think the spec is pretty clear on that. The FIELDSET element allows authors to group thematically related controls and labels. In the case of DLs, the spec gives a bit of leeway as to what they can be used for, but there's no such leeway here. Now, I can't see anything that implies the FIELDSET *must* contain FORM controls, but the only way to be sure you're valid is to validate the mark-up. Personally, I imagine that the FIELDSET tag would throw off a number of browsers, and most definitely confuse users relying on screen readers, etc, and don't see the point in bending a tag like this to do things it wasn't meant to do. For example, can it be uses to group a number of related links together? A better way to group links is in a UL or OL, then style to suit (even display:inline;) Justin French --- http://justinfrench.com http://indent.com.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] Suckerfish owning li ul tags
On 21/10/2004, at 4:19 PM, Paul Ross wrote: Hello folks, I've used the infamous Suckerfish dropdown menu on a couple of sites and have come across one glaring issue. The suckerfish CSS owns the ul and li tags so you can't style them anywhere else on the page. Anyone else had the same problem and what is the best solution? I haven't tried son of suckerfish yet - maybe that has the same issue?? Wrap the navigation in a div with an id like nav to act as a container, and change the suckerfish CSS selectors to reflect the container: ul { /* all lists */ padding: 0; margin: 0; list-style: none; } ... li:hover ul, li.over ul { /* lists nested under hovered list items */ display: block; } becomes #nav ul { /* all lists */ padding: 0; margin: 0; list-style: none; } #nav li:hover ul, #nav li.over ul { /* lists nested under hovered list items */ display: block; } make sense? Justin ** 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] 256 colours or the whole enchilada?
On 21/10/2004, at 4:34 PM, Andreas Boehmer wrote: Out of curiosity: what's your stand to the 216 web colours? Do you stick with them or do you go the full 16 bits? I personally have stopped limiting myself a long time ago (unless absolutely necessary), but keep coming across articles warning me from doing so. What's your thoughts? I ditched them a long time ago -- the reality is that 99% (or greater) of most audiences would be on thousands as a base. However, I'm careful about the contrast difference between non-safe objects, so that if a device rounds the colors to part of the 216 pallet, it would round one up, and one down, creating contrast, rather than rounding both to the same safe color. It's not an exact science at all, but I would never set a copyright notice or mission critical text in something with low contrast to it's background (like a very light grey on white), because that's just asking for trouble. An interesting side-note is the introduction of hand-helds... sure, most PCs might be running millions of colors these days, but are all those mobiles and PDA's? I bet they aren't :) Justin ** 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] accessibility question: same link phrase more than once
On 23/09/2004, at 2:28 PM, Lea de Groot wrote: On Thu, 23 Sep 2004 14:10:24 +1000, Justin French wrote: 3 read morespan on 'title of my post'/span then hide the span with display:none; from modern browsers, while still having entirely accessible source No, wait - surely the image replacement techniques have shown that display: none is not a good way to make things accessable? Arrgh! Good point. Has the same been proven with visibilty:hidden; ? Perhaps it could be some DOM scripting instead? Just bangin' out ideas :) Justin ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] IE border dotted
On 20/09/2004, at 2:02 AM, Cameron Muir wrote: Does IE support 'border-style: dotted' ? In my IE it renders as dashed. Mozilla/Firefox are fine, of course. I've certainly never seen it work -- definitely not in a standards-compliant way. However, before you start bashing IE, consider this quote from the w3 spec http://www.w3.org/TR/CSS2/box.html#propdef-border-style: Conforming HTML user agents may interpret 'dotted', 'dashed', 'double', 'groove', 'ridge', 'inset', and 'outset' to be 'solid'. So, IE isn't exactly compliant (it should render dotted as solid), but I guess they figured dashed is closer than solid. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Overflow on select or options
On 13/09/2004, at 10:09 AM, [EMAIL PROTECTED] wrote: I want a default width, i.e. all selects to be 250px, if there is overflow I want to see it... Then use min-width and realise that it'll work for everyone except IE :) --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Content Management tools for non-tech authors
I imagine new versions of Contribute are pretty good (there's a demo, why not try it out?). It depends on how much they need to do -- if it's just basic formatting of text, I'd say Textile[1] or Markdown[2] which are ASCI-to-XHTML converters with simple shorthand for links, bold, italics, etc. Textile especially is quite in depth if needed (tables, styles, and all sorts of stuff) but it generally just gets out of the way and let's text be XHTML. Textile is available as part of Textpattern[3], a GPL'd CMS written in PHP so you should be able to use it in some way in your app, but I'd review the license for yourself. It's also been ported to Perl and even Python I think. Markdown is written in Perl, but there are ports as well. 1. http://textism.com/tools/textile/ 2. http://daringfireball.net/projects/markdown/ 3. http://textpattern.com/ All the in-browser WYSIWYG editors seem to be either cross-browser OR standards-compliant, but not both. Since I have users on all sorts of operating systems (a lot on Mac OS 9 with IE5Mac) they're just not an option at all. Justin French On 09/09/2004, at 7:35 AM, Joseph Lindsay wrote: Does anybody have any experience with any content management tools that produce standards compliant code, and can be used by non-standards-savy authors? Does macromedia contribute produce good code? Editize? any other tools out there? --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] When to start?
On 06/09/2004, at 9:43 PM, Dylan Egan wrote: Hi, If you have no worries about shunting out a massive percentage of IE (and other browser) users to adopt a new technology with very little support, then go for it. Yes, that's sarcasm. It's more about what should we be limiting ourselves too. I know most people use IE out there, but should that really stop us? If we sit back and wait and wait, I dont think anything will come of it. You should be limiting yourself to tricks tools which are widely accepted by your audience, or degrade gracefully so that your audience can still access the content/functionality regardless. AFAIK, XForms does not fit in this category at all, and *I* wouldn't use it on anything less that a corporate intranet with guaranteed support by all clients. Im not so much going to be forcing people who are visiting these sites to use the ideal browser, but if they want an enjoyable experience they would need to use that browser. You need to clearly define enjoyable. If you mean it works, then I think you're headed down the path of inaccessibility. If you mean works faster/easier/smarter, then sure! I just want to get an idea of how many people are inserting hacks into their code to satisfy the older browser department and what are they going to do when a new technology comes around that one browser will never support, but is still the major browser throughout the net. I generally don't hack for backwards compatibility in the mark-up at all. I do a little CSS hacking, but it's limited to the bare minimum required, and they're clearly defined and separated into their own file for easy recognition and later removal. But you're not talking about hacks on HTML or CSS for backwards compatibility, you're talking about using a technologies which the browser HAS NO CLUE ABOUT. Not good I say. As for new technologies, it's going to be really tough -- we're *getting there* with support for CSS2, XHTML, etc, but pretty soon browser manufacturers will be far down the path on CSS3, XHTML2 and many other newer technologies, and we'll be looking at an uphill battle once again as we wait for browsers to catch up. Personally, I wouldn't use XForms anywhere that wasn't a 100% controlled environment. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] When to start?
Dylan, If you have no worries about shunting out a massive percentage of IE (and other browser) users to adopt a new technology with very little support, then go for it. Yes, that's sarcasm. You haven't told us what sort of site it is, but there ARE circumstances where this would be fine (if you could guarantee browser support via plug-ins or browser choice, like a corporate intranet), and there are some sites that could get away with forcing tight browser restrictions (like a site just for your small circle of friends, all of which have the necessary browser requirements). But the reality is that any site aimed at the general public needs to cater for the widest possible user base, without requirement of obscure plug-ins or edge browsers. Justin On 05/09/2004, at 7:37 PM, Dylan Egan wrote: Hi, I was wondering what would be the best time to start using XForms? Thankfully IBM and Novell are lending a hand to Mozilla to get XForms into the core and I hope this isn't a long development process. I am unsure as to what I should limit myself to in terms of standards compliance and semantic markup. I am working on a project with a few people and they're sort of wanting to make a partial living out of it, whereas im in it for the coding and this is stopping me from adopting XForms straight into the framework because of their potential customers being a mass IE user base. I know there are a few plugins around for IE, but should I really worry if IE doesn't support something that is, in my mind, wonderful for the internet? Im just thankful I could convince the designers to use XML/XSLT on the template side, which allows us to add more portability to the software. Anyways, just checking in to see if anyone is adopting XForms at this moment, going with Web Forms 2.0, or going to just force themselves to adopt MS technologies. Regards, Dylan. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Restricted HTML Editor?
On 27/08/2004, at 11:32 AM, Simon Chalmers wrote: There were posts on this mailing list a week or 2 back re HTML WISIWIG editors, but most give away too much control to the user and produce non-css-based HTML. Most of them are customisable to cut-down the features to only what you want to allow. There's also Textile as an option http://textism.com/tools/textile/. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Absolute positioning vs floats
Anura, I noticed someone made the comment that the preferred floats to absolute positioning. I have just created a new design using absolute positioning. It 'seems' to work across IE, Mozilla, Opera and latest Netscape (I'm trying to forget about NS4.7). But what is the consensus amongst my esteemed colleagues here? Am I walking into a trap? Are there flaws in absolute positioning so terrible that something will break dreadfully somewhere? The traps are mainly centred what happens to your positioned layout when the text is resized in the browser (either by text-zoom, page-zoom, or even just a minimum font size that is large). The trap here essentially is that an item with a fixed height (like 50px) will eventually be too small to contain it's text. With a floated layout, you can more easily stack blocks down the page which will clear each other regardless of how large the text is inside them. Typically I prefer floated (or a combination of absolute positioned items in relative positioned boxes), because it's a lot easier to reproduce classic layouts which have a footer below all other elements. If you layout still works when text is zoomed to 200-300% of it's original size in Mozilla/Opera/IE, then you don't really have much to worry about, IMO. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] [OT] NZ vs Aust
What part about THIS TOPIC IS CLOSED do you not understand Cameron? On 25/08/2004, at 5:11 PM, t94xr.net.nz webmaster wrote: Guys, To put a final end to this topic You Aussies will never have us Kiwis. We would rather bring our country down to get away from yours. Its embarrasing when people mix us up when they mention down under. And besides, theres a clear line between us Kiwis and you aussies. We are a much superior people. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] List Idea
Update: I've started a list/group. Anyone interested can subscribe here: http://lists.indent.com.au/mailman/listinfo/interface On 25/08/2004, at 3:04 AM, Justin French wrote: I'm sitting here ummming and ahhhing over interface designs for complex forms here at 3am (yes, workaholic), and I thought to myself it'd be great if I could just email this screen grab to my peers for some quick feedback. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] List Idea
Hey Gang, Sorry for going a little OT, but I'm sure this is of interest to heaps of you... I'm sitting here ummming and ahhhing over interface designs for complex forms here at 3am (yes, workaholic), and I thought to myself it'd be great if I could just email this screen grab to my peers for some quick feedback. I don't know if we need a full-on group/list (although I certainly can create one on my server), but at the very least, I thought a handful of you might be interested in a mini group for quickly bouncing ideas of each other. Nothing to full-on, no 20-paragraph rants, minimal code, just quick questions/screengrabs/replies to do with interface design and usability. If there's a demand for such a list or group (even just 5 people who don't mind CC'ing each other), please email me off-list with the subject interface group, and I'll see what I can do. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Online browser XTHML editor
On 24/08/2004, at 1:21 PM, Sarah Peeke (XERT) wrote: Thanks Neerav. I think xstandard delivers XHTML 1.1 or strict. I'm using XHTML 1.0 transitional for most of my work. Any other ideas? How is that a problem? To the best of my knowledge, Strict is just a subset of Transitional, so there shouldn't be any compliance problems (especially for content related mark-up). I'd be very surprised if anything output by xstandard would not validate to XHTML Transitional. --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] DOM setAttribute in IE?
Here's a function: function helpLinks() { if(!document.getElementsByTagName) return; var anchors = document.getElementsByTagName(a); for (var i=0; ianchors.length; i++) { var anchor = anchors[i]; if (anchor.getAttribute(href) anchor.getAttribute(rel) == help) { anchor.setAttribute( onclick,window.open(this.href,'popupwindow','width=400,height=400,scr ollbars,resizable'); return false;,0); } } } It works perfectly well in everything I can get my hands on except for IE, where it fails to set the onclick event to all A elements with a rel attribute of 'help'. Changing anchor.setAttribute(...) to anchor.setAttribute('target','_blank',0); DOES work (the link opens in a new window), so it would appear that IE doesn't like setting onlick attributes this way. Can anyone either: - suggest an alternate way to achieve this, or - suggest a good mailing list to seek further help on (like a DOM list) --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] melbourne/AU based flash guru?
Sorry for two job postings in one day, but I'm looking for a Melbourne (or Australia) based Flash guru for a bit of offlist advice for an upcoming project. If the project goes ahead, there'll be 10 hours (and probably a whole lot more over the following months) of work available. In particular, I'm interested in: 1. standards based embedding (see previous discussion today) 2. DOM scripting to Flash (Inman Flash Replacement [1], etc) 3. Database-driven, query-string-driven and/or XML-driven Flash applications (one movie/app can display varying content with varying) 4. Flash as a multimedia delivery platform (short video, audio, e-cards, etc) I must stress the importance of #3, and imagine that most of this would require some pretty heavy action scripting. I'd much rather give the work to someone on this list, which is why I'm posting here first! Obviously, please reply offlist for more details and a little discussion. [1] http://www.shauninman.com/mentary/past/ifr_revisited_and_revised.php --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Techniques for Styling Columns in Tables?
I asked the same question about 3 days ago -- see applying styles to the 3rd column of a table Justin On 17/08/2004, at 10:24 AM, Geoff Deering wrote: Hi, Does anyone have any reference for applying styles to columns in tables in a semantically correct way, without having to code class attr into each TD.? Can it be done? Geoff ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help ** --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] applying style to the 3rd column of a table?
On 13/08/2004, at 4:31 PM, scott parsons wrote: you can but only in IE due to IE having some weirdness occuring in the way they layout the page. BUT if you style the columns using the IE method, and style the third td (td+td+td etc) which will be understood by most modern browsers you should be able to get the column styled for everybody. I think I'm happy with right-aligned with td+td+td for decent browsers, and leaving it left-aligned for IE and all other almost-browsers :) Thanks everyone! --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] applying style to the 3rd column of a table?
Hi Folks, Is there any way (without ids or classes) to target the 3rd (for example) column of a table to apply styles? What I'm hoping for is something like... table td[3] { text-align:right; } ... but I can't see anything like that in my references. TIA --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
[WSG] HTML CSS references
Hi all, For a few years now, I've been using a combination of Mozilla and a couple of Sidebar tabs (HTML 4.01 and CSS2) for referencing the standards, which has been a great combo, but I think it's time to move on. 1. It's the only reason I have Mozilla open all day 2. It's the only reason I have Mozilla in my Dock 3. Firefox is much less bloated, and has the webdev toolbar which I use all the time In other words, I'm trying to cut my dev browsers down from 2-3 to 1. What I need to use Firefox or Safari exclusively is a replacement for the Mozilla sidebar references for (X)HTML and CSS, I guess in the form of a website, or downloadable reference to be used in a browser. What does everyone else out there use for reference? I can't stand having books open, so it has to be GUI based, preferably browser based, and preferably using frames (only because it's fast and I'm used to it). --- Justin French http://indent.com.au ** The discussion list for http://webstandardsgroup.org/ Proud presenters of Web Essentials 04 http://we04.com/ Web standards, accessibility, inspiration, knowledge To be held in Sydney, September 30 and October 1, 2004 See http://webstandardsgroup.org/mail/guidelines.cfm for some hints on posting to the list getting help **
Re: [WSG] Are wireframes necessary when using web standards?
I try and move the site away from paper and photoshop into lots of grey boxes using XHTML and CSS as soon as possible. I guess it's somewhere in between a wireframe and prototype. Considering everything is just groups of DIVs with IDs and classes, standards make it really easy to shuffle the boxes around and prototype *LIVE*... I had a client who had some input after my first prototype. Instead of going away re-jigging the whole prototype, I opened up Firefox with the CSS Editor side bar and MADE THE CHANGES ON THE SPOT (with a few tweaks to the mark-up). We achieved so much in one meeting that the client and myself were both amazed. My guess is about 2-3 meetings worth of changes and iterations in less than 2 hours. The client was able to SEE changes and assess them in realtime. If you're quick with basic CSS layout, I can 100% vouch for CSS/XHTML based prototypes/mock-ups/wireframe hybrids. On 06/08/2004, at 4:40 AM, Ian Fenn wrote: Hello, I'm in the process of redesigning a website. My client wanted something to show internal stakeholders so I started doing a few wireframes but suddenly wondered, Why am I doing this? Why don't I just build the website using web standards? A day later I finished a working prototype of the website in question. The client is happy but another producer has been quite vocal with his opinion that the prototype was built too early. From my perspective, a prototype has more value than wireframes. Web standards make development much more rapid, so we can respond quickly to any other needs thrown up before going into production. What do you think? --- Justin French http://indent.com.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] Changing the DOCTYPE messes up the page
On 04/08/2004, at 10:32 PM, Jim Barricks wrote: Hi Justin, I can't believe I put that page up. The changes were all putting a /li at the end of the lists. They were all minor and I have corrected them easily. BUT the problem is still there. :( Well, you didn't tell us that the watermark was generated content from a JS file, and I didn't look. Even though your source validates, you must understand that the (X)HTML that the javascript program writes internally doesn't appear in the source that W3C validates. Taking a quick look at the JS, there are plenty of things that aren't valid HTML at all. Lot's of items don't have quotes, there are some deprecated tags and attributes, you've got upper and lowercase tags and attributes all over the place, etc etc. It would also appear that all the styling for the watermark is done from within the JS file too. In short, your problems are here. The HTML that the JS writes must be valid for the DOCTYPE you're using (although it's very hard to check this, if not impossible), and the CSS you use to style the watermark inline must also be valid. I'd start be getting it to work without any JS, then modify your JS to use the valid HTMl you've tested and validated. --- Justin French http://indent.com.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] Changing the DOCTYPE messes up the page
On 05/08/2004, at 12:29 AM, Martin J. Lambert wrote: Justin French wrote: Well, you didn't tell us that the watermark was generated content from a JS file, and I didn't look. Even though your source validates, you must understand that the (X)HTML that the javascript program writes internally doesn't appear in the source that W3C validates. In short, your problems are here. The HTML that the JS writes must be valid for the DOCTYPE you're using (although it's very hard to check this, if not impossible), and the CSS you use to style the watermark inline must also be valid. Actually, it shouldn't be hard to check at all: view source of the rendered page, save to a file, and upload the file to the validator. Fix errors. Repeat until it passes. I'm pretty sure the source is the source of the HTML document, not the source of the HTML document AND generated source from client side scripts. It certainly isn't in any browser I've got on hand. --- Justin French http://indent.com.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] XHTML 1.0 Transitional and autocomplete
On 27/07/2004, at 11:08 AM, Peter Asquith wrote: Hi all I'm in the process of validating the markup in a suite of on-line assessment tools, which includes an ability measure. As you can imagine, in situations where those being assessed share the same computer, it's not acceptable for IE users with AutoComplete enabled to have the previous candidate's answers defaulted! The autocomplete attribute is not part of the XHTML 1.0 Transitional DTD and therefore any input tags containing autocomplete=off will not validate. The best I can think of is to sniff for IE (much as I'm loathe to revert to last century's techniques) and insert the attribute on a case by case basis. Does anybody know if there are workarounds for this or is this just one of those things? DOM scripting could dynamically add the attribute, which is 100% compliant. Something like: function autoCompleteOff() { if (!document.getElementsByTagName) return; var inputs = document.getElementsByTagName(input); for (var i=0; iinputs.length; i++) { var input = inputs[i]; if (input.getAttribute(type) == text) { input.autocomplete = off; } } } window.onload = autoCompleteOff; This of course won't help for IE with JS off, and you're not tackling any other browser with autocomplete (Safari, Mozilla, Firefox???), but it's a start. --- Justin French http://indent.com.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] Changing the DOCTYPE messes up the page
Jim, Switching the DOCTYPE may trigger slight changes to the way the browser's rendering engine will perform. But I don't think that's the underlying problem here. The problem is you've switched to XHTML transitional, but you haven't updated mark-up in each page to validate to that new DOCTYPE. There's 447 errors from the W3C validator: http://validator.w3.org/check?verbose=1uri=http%3A// www.barricksinsurance.com/jokes_NEW.html In other words, you don't have a well-formed, valid page, so it's no wonder that you've got some problems getting standards-based technologies like CSS to work properly. My guess is that once the page has been cleaned up to validate, the CSS issues will disappear. If they don't, we'll all be here to help solve the problem at that point. As far as converting your page goes, most of the errors are pretty self explanatory. And you should be able to go through them one at a time and clean them up with very little effort. Don't be put off by the high number of errors too -- you'll find that fixing one error will automatically cascade down to other errors in the page. A great start may be to batch process all your pages through a mark-up cleaner/fixer, which will solve most of your problems without any real effort. On the Mac there's Tidy as part of BBEdit, but I'm sure there's plenty of Windows and Web-based solutions as well. Justin French --- http://justinfrench.com/ http://indent.com.au/ On 04/08/2004, at 12:47 PM, Jim Barricks wrote: I'm in the process of converting over my old CSS/HTML site over to CSS/XHTML standards and trying to validate. On this page I have an watermark ad on the right side which works perfectly on: http://www.barricksinsurance.com/jokes_OLD.html But when I change the DOCTYPE from DTD HTML 4.0 Transitional over to DTD XHTML 1.0 Transitional the watermark switches over to the right side and stays immobile like this: http://www.barricksinsurance.com/jokes_NEW.html Has anybody any ideas what is happening and how to correct it? I still have 600 pages to convert. :-( Thanks best, Jim Barricks * 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] After CSS?
Sorry if I'm a little late to the part on this one, but I think you can rest a little on CSS now -- CSS3 is s far away from being usable, because we're still trying to get IE6 to play well with CSS2, and we're STILL seeing really buggy browsers like IE5 all over our logs. There's much more to web design than CSS/XHTML though -- here's some food for thought: 1. Accessibility (508, WAIG, etc) will eventually become law, if it hasn't already in your area. Most of the requirements are really easy, so there's no excuse for not getting at least 95% accessible. 2. Usability (making something *truly* intuitive to understand and interactive with) is the difference between a good interface, and a bad one. You'll spend your whole life perfecting this skill, so you might as well get started now! 3. DOM scripting can be used to *enhance* the experience for UA's with JavaScript available, while continuing to remain 100% accessible to those without. 4. Flash replacement techniques are going to be BIG. They can allow you, the designer, to meet some of the more challenging needs of your clients (like graphical headings in an obscure font) with minimal impact to your time and budget. 5. Defensive Design User Experience Design is related to interface design, but it's pretty much it's own thing. What happens if a form is filled out incorrectly, or there's an error, or a user enters something unexpected or malicious into a form? See http://37signals.com/book/ to understand what I mean. 6. Server-side scripting. Websites need to fulfil much larger goals now than ever before, and usually this involves some sort of server-side data manipulation (everything from a contact form to a product database). Most clients will also want a CMS. You need a basic understanding of PHP or a similar server-side language to: a) meet basic client needs b) collaborate efficiently with programmers on larger projects and teams 7. Content is King. Writing or editing for the web is a fine art. You may think this isn't your job, but if you manage websites, then I think it is part of your huge job description. You should at least be able to edit and re-word content to suit the web, and recognise bloat. That should keep us all occupied for about the next 400 years :) And by the way, thanks, I think you've just inspired a new article for me to write on my site tonight :) Justin French http://justinfrench.com/ http://indent.com.au/ On 23/07/2004, at 8:45 PM, 7 sinz wrote: Hi all, Im an 19 yer old desinger, with a particular interest in web design. For the last 8 motnhs i've been huddled up in my workspace practising my art learning the ins and out of CSS and pretty much learning the language to a T. Anything i used to do with nested tables I now write with CSS, layouts dependant on the viewport are atill time consuming to make, and depending on browser support/target audience scale of the project they can be time consuming, still i managae to pull through. Fixed width layouts are no brainers to use, i feel once you've comfortably mastered positioning in CSS you pretty much can design any layout you cut outta of Photoshop/other image editor. Not to say that thats all you need to know when developing with CSS, but it is a main part of structual design, once you can write it fluently theres no stopping or holing back with what kind of structure your static design may be. Now, it took me a while to get here, but we all got to start somewhere. But what now? How can you prepare for the next specification, new attributes and selectors, is there a test suite for CSS3?. Im a graphic designer, who's passion for web development introduced me to one of the strongest client side languages available to any developer, now that im at a point of speaking fluent CSS what do i tackle next, what new CSS3 flavours can i focus on for the next wave of innovation? --- Justin French http://indent.com.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 *
[WSG] tooltips
Is there any way to disable tool tips for a certain element or class of elements? I've got my own hover content on a nav bar, but it gets obscured by the browser's tool-yip for the title attribute. And no, I can't leave the title element empty, because I'm using it's content for the hover :) --- Justin French http://indent.com.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] Titles Acronyms Abbr etc
On 21/07/2004, at 4:10 PM, Bert Doorn wrote: G'day Looks distinctly like a case of totally unnecessary to me but we have a difference of opinion in the office... Even if it was written as VDS, it would be an abbr(eviation): abbr title=Vent Door SystemsVDS/abbr Actually, VDS is an acronym of Vent Door Systems, not an abbreviation. --- Justin French http://indent.com.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 *
[WSG] semantic way to mark up form help?
Hi all, I'm trying to decide on a nice semantic way to mark-up a short (usually only a few words) block of help text in the context of a web form. I currently use a label to label the input, and a paragraph or div to mark-up the help text: form... div class='formitem' label for='f-title'.../label input id='f-title' type='text'... / p class='help'This is the title of your news post, which does not accept HTML input/p /div /form But logic tells me that in the above example, the p help text is not associated with the form widget or the label at all. The only way I can see this being done is by including the help text in the label, but this will restrict me in terms of layouts. Honestly, the most logical way I can see to do this is to have them in three cells of a table row, since at least they'll be associated in a row. fieldset's would also be nice, but they're intended for groupings of form elements, and using them for each text input seems like a load of bloat. I've been looking at many examples of correct, semantic forms, but can't see anything like this out there. TIA --- Justin French http://indent.com.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] semantic way to mark up form help?
Thanks to everyone for their input, especially Steven insight into the world of screen readers. My summation of all this is that the elements need to appear in this order: Label Text Extended help Input or widgets I really like the idea of wrapping the label tag around the whole lot, so that everything is associated with that single form element. It would also remove the need for div class='formitem', a wrapper I used around each form element (label input), since I can style the label with display:block; label for='f-title' News Postbr / This is the title of your news post, which does not accept HTML inputbr / input id='f-title' type='text' ... / /label Although I'd like more control over the elements involved, so maybe something like (similar to Steven's example): label for='f-title' h3News Post Title/h3 p class='help'This is the title of .../p input id='f-title' type='text' ... / /label ... but maybe that's an abuse of H3 and P, and I should stick with generic DIVs or SPANS? Justin French On 23/07/2004, at 12:24 AM, Justin French wrote: Hi all, I'm trying to decide on a nice semantic way to mark-up a short (usually only a few words) block of help text in the context of a web form. I currently use a label to label the input, and a paragraph or div to mark-up the help text: form... div class='formitem' label for='f-title'.../label input id='f-title' type='text'... / p class='help'This is the title of your news post, which does not accept HTML input/p /div /form But logic tells me that in the above example, the p help text is not associated with the form widget or the label at all. The only way I can see this being done is by including the help text in the label, but this will restrict me in terms of layouts. Honestly, the most logical way I can see to do this is to have them in three cells of a table row, since at least they'll be associated in a row. fieldset's would also be nice, but they're intended for groupings of form elements, and using them for each text input seems like a load of bloat. I've been looking at many examples of correct, semantic forms, but can't see anything like this out there. TIA --- Justin French http://indent.com.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 * --- Justin French http://indent.com.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] Re: Does a CSS file load all graphics or only page relevant ones?
On 20/07/2004, at 2:03 AM, Chris Stratford wrote: I cannot provide any proof of this. but I believe it would only load it when its displayed. the CSS wont parse and execture BANNERPAGE2, unless its caled from Page2.HTML I too believe this is the way it's *intended* to work, although there are some browsers out there which misbehave. Safari 1.0 was one such browser (Mac OS X Jaguar), downloading all background images at once, but it was fixed in 1.1 or 1.2 (Panther), so it now only downloads the images as needed. --- Justin French http://indent.com.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] text field size tag
On 14/07/2004, at 5:13 AM, Ted Drake wrote: In the 5th edition of O'Reilly's HTML and XHTML the definitive guide, they suggest listing a size and max length for the text input field. I would like to define the width with css and leave out the size attribute on my input fields. I'm trying to remove as much styling from our forms as possible and this works in my testing so far. I'm wondering if anyone has any more concrete opinions on the practice of defining width of input and select fields with css instead of the size attribute. In my opinion, you still need to set a default width for the element using the size attribute, for those without CSS. Yes, it will be overridden with CSS for 99% of your browsing audience, but it safer to put *something* in there as default, since you have no idea how a browser will behave without it. Most browsers seem to put in a sensible default width if you exclude the size attribute, but I think better the devil you know in this case -- since you can't possibly test on every useragent available, let's give them a sensible default, then style over the top of it. --- Justin French http://indent.com.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] Frameworks of Design for Easy Reuse and Change in CSS
On 10/07/2004, at 1:29 PM, Geoff Deering wrote: Do others think about CSS, design and layouts from the point of view of flexible, extensible frameworks? I'm sure a lot of us try and design our own sites like that so that we can reuse our own components. I think it's possible for such frameworks to exist, but the reality is that each user of the framework will fall into one of two categories: 1. They're using the framework, but the framework is overkill with many containers and elements which simply aren't needed in the scope of the project, or 2. They're trying to use the framework, but technical limitations won't let them achieve the visual or business goals they've set. Plone, Textpattern, etc could all have nicer default layouts and frameworks, but the reality is that nearly every project is different, and needs it's own solution. ZenGarden-like interchaning stylesheets on many similar sites would be great, but it's all pie-in-the-shy to me -- it will be overkill for some, and too limiting for others. --- Justin French http://indent.com.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] Frameworks of Design for Easy Reuse and Change in CSS
On 12/07/2004, at 9:22 AM, James Ellis wrote: Hi I do just this for a reseller based product. One set of markup styled with CSS based on the client's 'look n feel' requirements. All done with floats and works off the bat with all modern browsers and some older ones like IE5.x for both platforms. It took me a while to do it but once done I haven't looked back as previously the markup was copied per reseller. Takes a day or so for me to work up the style for a new site based on the client's spec. It can be done, you just have to make the markup as free of any style as possible. i.e just boxes in a page. Exactly. A framework will not work, but having CMS's offer default templates which have very clean, simple mark-up and a simple default style sheet will make is easy for developers to come in over the top with style sheets which can re-skin that content, and if needed, add to the templates. However, if the CMS vendor doesn't have a visually pleasing default skin, they probably won't get enough sales. It's a fine line, which requires a lot o though on behalf of the template designers.' In other words, it probably should not be done by one of the lead programmers, unless they're also well versed in clean mark-up, CSS positioning, accessibility and standards. You could consider designing them some new default templates :) FWIW, that's why the blogger re-launch included skins designed by some of the industry leaders, like Zeldman and Bowman. Cheers James Justin French wrote: On 10/07/2004, at 1:29 PM, Geoff Deering wrote: Do others think about CSS, design and layouts from the point of view of flexible, extensible frameworks? I'm sure a lot of us try and design our own sites like that so that we can reuse our own components. I think it's possible for such frameworks to exist, but the reality is that each user of the framework will fall into one of two categories: 1. They're using the framework, but the framework is overkill with many containers and elements which simply aren't needed in the scope of the project, or 2. They're trying to use the framework, but technical limitations won't let them achieve the visual or business goals they've set. Plone, Textpattern, etc could all have nicer default layouts and frameworks, but the reality is that nearly every project is different, and needs it's own solution. ZenGarden-like interchaning stylesheets on many similar sites would be great, but it's all pie-in-the-shy to me -- it will be overkill for some, and too limiting for others. --- Justin French http://indent.com.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 * --- Justin French http://indent.com.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] id or class on html or body
On 09/07/2004, at 5:55 PM, Mordechai Peller wrote: Putting an id or class on the html or body tags is a useful way of targeting slight variations in style rules with resorting to a second style sheet. I remember seeing a discussion of the pros and cons, but I can't remember where. Does anybody have a link? Or perhaps the discussion can begin anew. I use id and class attributes on the body all the time. I don't really like using them for something like body id='3-col-red' because that's a presentational thing (IMHO), but I do use it for something like body id='home', in which case I can target elements on #home separately to those on #about or #forum, which allows drastically different styling for each page. For me, it's a way of keeping my ids and classes as generic and minimal as possible, but still only using one stylesheet. The alternative (for my uses) is to have many stylesheets for each section, but this is an absolute pain, unless there are heaps of styles, in which case smaller stylesheets on deman seems much smarter. I know others use them to: - identify the domain (so others can easily override the stylesheets with a single user style sheet), but I *really* question how many people would bother to do this. eg body id='www-meyerweb-com' see http://www.meyerweb.com/ - trigger different layouts eg body id='col-SMX' see http://www.adaptivepath.com/ - trigger active and inactive tabs on a menu navigation (can't think of an example) Hope this gives you an idea or two :) --- Justin French http://indent.com.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] IE linked image border grief
On 24/06/2004, at 5:02 PM, Craig Stump wrote: Unless I'm missing the point entirely, what's wrong with: a img {border:none;} Been there, tried that. The border is being applied to the a, not the img. --- Justin French http://indent.com.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] Site Deconstruction, those crafty Germans
On 25/06/2004, at 10:06 AM, Shane Helm wrote: Are you serious? Is this possible? On Jun 24, 2004, at 5:48 PM, ckimedia wrote: Hi, I found this wonderful site (http://www.mbusa.com/brand/index.jsp) listed at the WSG section for full CSS sites(http://webstandardsgroup.org/resources/#cat9). As today is my Review and Research day, I've been peeking under the hood. If my interpretation of the rather elegant code is correct, this site has a second layout that is rendered if FLASH is not present. Can some one please confirm or correct my observation. I've sent an e-mail and poked around for other examples, but have come to rely on this rather savvy bunch for my final analysis. I have no idea if it's *possible*, but I've just disabled Flash in Safari IE5Mac, and all I get is a white page with the footer HTML -- no Flash, and no content in replacement of Flash. So, at the very least, it's not working well -- if at all. Can't find a way to disable the plug-ins in Firefox and Mozilla, so who knows if it works for them, or for IEWin (the big target I guess). --- Justin French http://indent.com.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 *
[WSG] IE linked image border grief
Hi all, I have the following rules (amongst many): a { text-decoration: none; } a:link { border-bottom: 1px dotted black; } a:visited { ... } a:hover { ... } No surprises so far -- it's a classic dotted border on links. The problem I have is that IE displays the dotted bottom border on images that contain no text, I guess that's a reasonable thing, and maybe not a bug, but it's a pain in the ***. If we had *ascendant* selectors, my problem would be solved, but we don't: imga{ border:0; } Bowman stopdesign.com solves the problem with a 'noline' class on his linked image tags, but in this case, I can't edit the source of the image tag, as it's auto-generated by the CMS. Bugger. Searching Google for this is pretty much hopeless (the keywords are too generic), so here I am, begging the WSG for some help :) --- Justin French http://indent.com.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] Inheritance question
On 17/06/2004, at 1:11 PM, Tim Yang wrote: I remember reading about a syntax that when applied to an element would force all the elements contained within that element to inherit the properties of the container element. Is there really such a syntax? if so, how do I write it? I tried googling for it, but all I got were pages describing what inheritance was. You can set a property for all children of a selector with an asterix: #something * { ... }; but this could be overridden by another selector further down the stylesheet, like #something div { ... }. I don't know of any way to *force* inheritance, or for something to inherit all parent properties... usually inheritance is done one rule at a time. --- Justin French http://indent.com.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] XHTML Transitional - Strict
On 17/06/2004, at 7:00 PM, Jamie Mason wrote: Hello all, I was hoping one of you could tell me, or know any url's that would be helpful on moving from XHTML Transitional to Strict. What are the main things to look out for when moving over? What is allowed in Transitional that has to be removed for strict? Make a copy of your site, change the doctype, and start validating, fixing each error as you go, until it's valid strict. It's really very easy to achieve. --- Justin French http://indent.com.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] scrolling area
On 15/06/2004, at 4:21 PM, Jake Badger wrote: From having a quick play with those scrollers they seem to do exactly what you want when both CSS and JavaScript are off but not when just one is off. I guess the question is how many people actualy browse like that (one on/ one off). Lots is my guess. I spend quite a lot of time with JS off, but have never turned off CSS for more than 10 seconds while testing a web page design. Lots of companies will turn off JS for security reasons, but I'm yet to hear of an IT department turning of CSS for security reasons! --- Justin French http://indent.com.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 *
[WSG] scrolling area
Hi all, I've got a client who wants a scrollable area of text within a web page layout. My instant reaction was to use overflow: auto; or an iframe to solve the problem, but he doesn't like the visual appearance of the GUI-native scroll bars, and I'm having a few problems with browser inconsistencies (mainly NN / Mozilla) showing the horizontal scroll bar when it isn't needed, etc. So, I'd like to experiment with a javascript/css based solution which (preferably) is 100% accessible, based on a scroll box with simple up and down arrows, etc. I've got a few ideas on how it could be done (whilst remaining accessible to those without JS), but I'd love to see if there's something already out there. As a side-note, is there anyway to force NN/Mozilla to only show the vertical scroll bar on an overflow: auto; box (I'm thinking something like overflow-vertical: auto; overflow-horizontal: clip;)?? --- Justin French http://indent.com.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] scrolling area
On 15/06/2004, at 1:23 AM, Tim Lucas wrote: Justin French spoke the following wise words on 14/06/2004 11:29 PM EST: So, I'd like to experiment with a javascript/css based solution which (preferably) is 100% accessible, based on a scroll box with simple up and down arrows, etc. Travis Beckham's divscroller works a treat: http://www.squidfingers.com/code/dhtml/?id=divscroller2 http://www.squidfingers.com/code/dhtml/?id=divscroller Thanks, but they're not very accessible at all. At the very least I would hope that (when JS isn't available) the text would still be readable (eg: break the layout to ensure accessibility). I think what needs to be done is start with an accessible layout, then have javascript come in over the top and add any new mark-up and styles that are needed, but I'm a little over my head there. --- Justin French http://indent.com.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] Linking External CSS
On 04/06/2004, at 10:02 PM, Jamie Mason wrote: What's the difference of linking by; @import url(styles.css); And link rel=stylesheet type=text/css href=styles.css / If any? @import(...) is not recognised by some older browsers (like NN4), it's helpful for keeping old browsers away from the new-school CSS for starters. --- Justin French http://indent.com.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] Budget Design
On 01/06/2004, at 8:55 AM, Luke Moulton wrote: At this present time, for our team a small budget website consists of a 5 to 10 page (for want of a better word) XHTML/CSS site with CMS and can cost the client anywhere from AU$3.5K to AU$5.5K depending on customisation. I'm also in AU, and this appears to be a fair *market* price for such work, although in my case it's more like 10-15 pages. Of course, I can't tell you whether this is a fair price for the work you perform, or if you're covering your costs and making a profit, but $3-5k seems to be what the market will handle here right now for something with basic CMS and a general quality (XHTML, CSS, WAG, etc). --- Justin French http://indent.com.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] accessibility feedback
On 01/06/2004, at 11:26 AM, Andreas Boehmer wrote: I was wondering whether you could give me some feedback on a website we have created: http://www.jet.org.au. We have tried to make it as accessible as possible, but better than any Bobby or W3C validation is probably going through your critique. We are still working on improving it, but please feel free to be tough critics! From a quick glance, without looking at any code or styles... - the small type in the logo is virtually impossible to read (1280x1024 17 flat panel)... either you want people to read the text (so make it bigger), or you don't (so get rid of it). I know it's not a TRUE accessibility issue, but since there are validators for those (which you should use), I don't need to go into all that here :) - Link colours in the breadcrumbs, and on the homepage (for starters, eg Login Now!) don't have a visited link state -- I bet there are a lot more examples all over the site. - some of the font sizes (especially in the left navigation bar) are pretty small (around 9px I guess?) in Safari -- whilst I haven't bothered to see how you've styled them, it's certainly something to watch out for across all browsers with a default font set small-ish. These comments shouldn't detract from what is looking like quite a nice site. In other words, you're getting there for sure! --- Justin French http://indent.com.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] Should web standards cost more?
On 01/06/2004, at 11:46 AM, Neerav wrote: So has any member of the WSG done that, justifying charging $X more than eg: the design agency down the road which makes webpages by putting sliced photoshop images into tables by saying something along the lines of I design to Web Standards and can show my product validates against these standards, this is like an industrial business getting ISO Quality certification for their processes and means my product/service satisfies web industry standards for quality of code, can other companies with proposals for your project do the same ? Yes. Every business needs to create a point of difference between itself and it's competitors to stand out. This would be one way to do it. Specials prices, stunning sample designs, pre-built code and templates to save the client money, quick turn-around times, money back guarantees, proven track records and proven experience would all also be valid ways of differentiating yourself from your competitors. But be careful, if every web firm suddenly promotes itself as being standards compliant, then there is no longer a differentiation. How many web form websites have you seen with this line? We create visually stunning, easy to use websites with intuitive user interfaces which load quickly Yep. Uh-huh. We all all say that. It might buy you a few months of differentiation, but I'm still not sure that the average client cares about standards -- they care about price design far above usability, accessibility and standards in my experience, and so they should. But what will the home page look like? :) --- Justin French http://indent.com.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] Standards compliant content management system.
On 01/06/2004, at 2:22 PM, Ned Lukies wrote: Has anyone come up with some solutions for helping the client maintain their own content while still retaining standards compliance? a) Macromedia Contribute could work out fine for small stuff, and is really the only GUI option I'd consider -- everything else will be browser based, and result in similar problems to what you're already experiencing. b) Textpattern [1] is a new open source CMS with Textile [2] in it, a standards-compliant web text generator that runs off a sort-off web/email shorthand thing using pure ASCI, so it's cross platform, and the XHTML it generates is nothing short of beautiful. You may consider a complete Textpattern site, or just stripping on Textile for use inside your own system. It's written in PHP by Dean Allen. c) Markdown [3] is a simple Textile-like web text generator written in Perl by John Gruber. It has a smaller range of features than Textile, which is probably a good thing. d) Simon Wilson has also written a few articles and snippets on creating simple, standards compliant mark-up from user input, some of which may rub off on you. [4] 1 http://www.textpattern.com/ 2 http://www.textism.com/tools/textile/ 3 http://daringfireball.net/projects/markdown/ 4 http://simon.incutio.com/archive/2004/04/13/myriadOfMarkupSystems --- Justin French http://indent.com.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] Correct way to swap style sheets based on Browser?
On 31/05/2004, at 12:18 PM, [EMAIL PROTECTED] wrote: What I would like to do is, limit the browsers that see the fully styled version of my site to the ones I know work correctly others will get the un-styled version. I am aware I can do this in JavaScript, is there a way to do so in CSS? Logically I want to say: If (ie4,5,5.5,6+, NN6, 7+ , Opera7+, Mozilla 0+ etc ){ use this style sheet } Elseif (NN4.7) {use this style sheet} Else { dont use style sheet} Thanks CSS and Web Standards is a challenge by I am improving! Hi, I'd really recommend you put IE4 in the same category as NN4, because it's CSS support is flakey, and CSS's @import function is not supported by either browser, so it will solve 99% of your problems. This will be a basic style sheet seen by all browsers: link rel='stylesheet' media='all' href='/css/basic.css' / Then we @import an advanced style sheet over the top for modern browsers. IE4 and NN4 won't see this style sheet, because they don't support the @import function. style type='text/css' @import url(css/modern.css); /style No scripting, no server-side stuff, nothing but pure CSS. Now, pre-empting your next question, yes, you can even target IE versions 5+ (individually or collectively) with IE's conditional comments, which allows you to overlay stylesheets for different IE versions to plug up holes in their box model (or anything else). Quick example: !--[if IE 5]link rel='stylesheet' media='all' href='/css/ie5.css' /![endif]-- See http://msdn.microsoft.com/workshop/author/dhtml/overview/ ccomment_ovw.asp for more info. No CSS hacks, no javascript, no PHP/ASP/CF/Java/Perl, and the conditional comments validate perfectly well -- it's the best separation of old, modern and buggy browser stylesheets I've seen so far, and it forms the basis of every new site I start. --- Justin French http://indent.com.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] Fwd: remove p widows/orphans?
Zulema, First and foremost, I am a designer, so please take this advice from someone who understands really beautiful typography, but also understands when the web won't comply. This is definitely a case of don't bother. For starters, you're typesetting those orphans and widows on YOUR screen, with YOUR fonts, with YOUR resolution, with YOUR browser. The problem is, what looks perfect on your screen will be an absolute mess on someone else's, because your hard-coded br /'s will more than likely appear mid-line on someone else's system. This is an unavoidable reality which you need to accept and move on: br /'s are doing WAY more harm than good - they're not a typesetting tool, they're designed to be used in things like poetry, where a break in the flow of the text is intended. Most typographic resources discuss orphans and widows in reference to lines which are separated from the rest of their paragraph over a page break or column break. However in this case, I assume you mean single words that are pushed down to a line all by themselves, like the last word in a paragraph, right? Assuming this is what you want to prevent, then the quickest hack I can think of is to use a no break space (nbsp;) inbetween the last two words in each paragraph, forcing them to appear on the same line. However this isn't much of a solution. As you no doubt know, the problem with this is that sometimes forcing the 2nd-last word down as well will cause the preceding line to appear quite short, in which case you'd bump other words around until the paragraph was well-balanced. This is simply not possible with HTML, due to the lack of control you have over the user's environment and settings, and the fact that you can't *visually* look at the paragraph on everyone's screen and decide what's best -- it's out of your control. Modern DTP programs like InDesign currently make visual corrections to the line breaks in a paragraph automatically, so my suggestion here is to simply leave it alone, and let the browser's own line-wrapping mechanisms decide where to wrap. Over time, as browsers improve, it's my guess that they will handle line-wrapping, orphans, widows, etc, much like InDesign does, but until then, your br /'s are definitely doing more harm than good. On 28/05/2004, at 7:39 AM, !!blue wrote: Hi, Is there a way to eliminate widows/orphans in paragraphs? Not when the html page is printed out, but when viewed in a browser. I wanted to know if anyone's run into this and knows how to fix/kludge/hack? (preferably something that would validate of course) If not, I'll keep on adding br /'s where necessary as I've done in the past. =:-O the horror!! thanks, Zulema --- Justin French http://indent.com.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] Browsers Emulator
On 26/05/2004, at 10:15 AM, Peter Firminger wrote: It's very difficult (impossible) to emulate all the bugs in a browser without running the browser. Emulators can emulate the required behaviour but generally not the bugs. So unless you actually do what people like browsercam have done and set up a bank of machines running the browsers and screenshot them, it's a bit pointless. It would be conceivable (but a LOT of work) for someone to build an emulator that took advantage of all opensource browsers' rendering engines in one neat package, but the reality is that a lot of browsers (IE in particular, but plenty of others) are closed-source, so there's very little we can do about it. And as you say, the only true way to test a browser is with the browser running natively in it's intended environment, so an emulator will never achieve accurate results, so it's pretty much useless. What I'd like to see is a program for Mac OS X that will load a URL into all currently running browsers (including those in X11, VirtualPC and MacClassic, so it's truly cross-platform), take a screen shot, and present them back in one tabbed interface as a series of screen shots which the designer can view to get a decent overview of what's going on browser-to-browser in one location. Hit refresh, and all N browsers refresh, take a new screen shot, and load back into the tabbed interface. A small go to this browser button would bring that particular browser/window to the front, so that the user could study it in the flesh. Even better would be if this previewing program took notice of mouse position, and did long screen shots that continue down below the fold. Screw opensource -- I would pay serious cash for such a tool! This is MUCH better than browsercam, because - it doesn't require a subscription, network connection, internet access, etc - it doesn't require any bandwidth, and would be virtually instant (my guess would be a 5-20 second delay) If any crazy Mac developers out there are listening, GET TO WORK :) --- Justin French http://indent.com.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] Accessible alts and titles...
On 26/05/2004, at 2:34 PM, James Cowperthwaite wrote: ... for, ahem, images that may only be design elements of a page. Would it be better to have a blank title and alt attributes on images that don't add any /real/ value to the site? Or better to say: title=graphical site element? Actually, I'd take them completely out of the mark-up, and insert them as background images on existing content items in the page. If that can't happen (there's lots of cases where it can though), I'd consider placing an empty div in the page mark-up, and using it as a handle for the pure presentational graphic to be done with a background image once again. Why? Because they're not CONTENT, they're presentation :) I know you're still putting some additional mark-up handles in there that don't specifically relate to content, BUT that's what a DIV is for -- structure :) img src='images/useless.gif' alt='useless' height='20' width='20' / could be replaced with div id='something-useless'nbsp;/div and styled with #something-useless { width:20; height:20; background: url(images/useless.gif) no-repeat top left; } It depends HOW accessible you want your pages to appear to screen readers, non-visual devices, text-only devices, etc. Justin French --- Justin French http://indent.com.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] Re: digest for wsg@webstandardsgroup.org
On 23/05/2004, at 10:23 PM, Michael J. Hußmann wrote: Personally, I have no particular interest in discussing theoretical issues regarding web standards; if someone can steer me to a list better suited to dealing with the more practical issues of using CSS, I would be grateful. see http://www.CSS-Discuss.org --- Justin French http://indent.com.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 *
[WSG] alternate style sheets expert here?
Do we have an expert on alternate stylesheets in the house? I'm a little confused. What I'd like to do is link or @import a few base style sheets for the general (default) site design first. Then on top of that I'd like to link to a series of alternate style sheets that ADD TO those base styles. Seems easy enough, and the code on http://www.simplebits.com/ seems like a good starting place, but when I look at his alternate style sheets, they're not *alternate*, they're *alternate additional*, in other words, the alternate style sheets act as a group of alternatives (of which 0-1 can be selected) which will be applied *in addition to* the base style sheet. Is this a reasonable summary? Can someone provide a clearer definition? To add to the confusion, I'm hoping to tie this all into a PHP session and membership preferences, but that's another story all together :) --- Justin French http://indent.com.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] Re: digest for wsg@webstandardsgroup.org
On 21/05/2004, at 12:36 PM, jneen wrote: my five cents worth I would prefer to see a more typical thread based web forum where I just sign in and can read at my leisure. Any posts made by me should list my login name and I think that only member's should be able to post. I also find the number of emails I am getting harder to digest and would prefer a thread based approach so that I can see all replies on a given topic in one hit. Get a better email client -- threading of emails based on the subject and other headers has been possible in email programs as far back as my very first email in about '96 or '97 I guess. Seriously, threading in email is pretty standard stuff. 1. Set up a new folder / mail box for all the WSG email. 2. Set-up a filter/rule to make sure all email with the subject [WSG] get pushed into your new folder 3. When viewing the WSG folder, make sure you're viewing the emails in a threaded view. If your email program doesn't support threading, get one that does! Judging by your headers, you're using Thunderbird, so all should okay there. Want to be even more productive? 4. Flag the threads you're interested in so it's easy for you to scan for topics of interest. 5. Set up your mail client to only show threads with new messages, so that the WSG folder is a lot less cluttered. Now you have a system which is FAR more productive than a clunky forum could ever be. You can download all messages for viewing offline (on the train in the morning?), you can do a full text search via your mail client (again, offline), you can have instant and reliable private messages, and you can keep track of all your post all via a very fast email client, rather than a slow web based forum. If you want a forum, then why did you join a mailing list? There are plenty of such forums already in existence on the web -- we certainly don't need another one here. Email is fast, clean, free from bloat, and allows me to quickly keep on eye on the mailing list whilst working -- a lot harder to do with a web-based forum. --- Justin French http://indent.com.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] 'It Works in Gecko Browsers ...'
On 19/05/2004, at 8:24 PM, Mordechai Peller wrote: Some of signs that is might slip are increasing computer literacy in the general public, increased awareness of Mozilla and Opera (media reports, Opera on mobile phones, etc.), and increased acceptance of Linux. We can aid this further by educating our friends, family, and clients. Opera will never do it. The UI is butt ugly, the usability is woeful, and the whole thing feels a whole lot cheaper.\ The only way I can see a browser beating IE is if it looks, feels and behaves like IE in every way possible. They don't need to reinvent the wheel in terms of UI design and interaction -- they need to mirror it, which in turn lowers the learning curve required to switch. This is the big flaw in Mozilla and Opera right now -- they look and feel different, and people are afraid of change. sidenote My biggest gripe with Mozilla et al is that they don't use the Win/Mac standard GUI form elements and widgets, which not only cheapens the look (IMHO), but instantly causes the browser (and all the user's favourite web pages) to feel unfamiliar or foreign). /sidenote What they DO need to do is beat IE in regards to security, performance, preferences (cookies, scripting, security, etc), and yes, standards compliance, and sell the browser on these points. I've got some nice ideas on how Opera or Mozilla could be marketed to the masses, but I see reason to give those away for free :) --- Justin French http://indent.com.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] 'It Works in Gecko Browsers ...'
On 19/05/2004, at 11:59 PM, Rimantas Liubertas wrote: Opera will never do it. The UI is butt ugly, the usability is woeful, and the whole thing feels a whole lot cheaper.\ Have you seen opera 7.50? And opera on mobile phones is reality, not something will never do it. The only way I can see a browser beating IE is if it looks, feels and behaves like IE in every way possible. They don't need to reinvent the wheel in terms of UI design and interaction -- they need to mirror it, Oh, please. I've swithced to FireFox (Firebird then) a year ago just because it looks, feels and behaves way better than IE. And even before the switch I've been using NetCaptor (commercial software), which added some features to IE. I got those for free with Mozilla. Which was the last version of Mozilla/Opera you have tried to use? You're missing the point. If Opera and or Mozilla want to TAKE MARKET SHARE FROM HAPPY IE USERS, then they won't be able to do it by alienating the potential user with a brand new interface to learn. Essentially, this would of course be a backward step for Opera and Mozilla in some ways (giving up on some of their innovations and UI concepts) but the reality here is that for Opera and Mozilla to take a share of the IE market, they need to make the transition easy. Just like Explorer, but safer. Just like Explorer, but faster. Just like Explorer, but better. Just like Explorer, but secure. ... would all be a perfect concepts for a browser trying to steal people out of the IE market. Much more effective than: Opera. A whole new way to surf the web. Of course it'd still have to remain free, and they'll never be able to overcome the fact that IE will be bundled / integrated into Windows forever :) --- Justin French http://indent.com.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] should I track down this problem?
On 20/05/2004, at 2:56 PM, Universal Head wrote: Curious how others would approach this? I've just finished a simple site that is perfect in WIN - NN 7, IE 5.01, IE 5.5, IE 6 MAC - Safari 1.2.1, Mozilla 1.4, IE 5.2 And a friend looks it in on Mac IE 5 and finds a big problem. IE 5 and 5.1 are pretty buggy compared to 5.2+. My question is, do you draw the line or not? Do I spend more time time trying to track down this one dumb version number browser's problem, or do I just say it's not worth it? If the site is currently unusable with older version of IE 5 Mac, then you have to do SOMETHING. My advice would be to hide the stylesheet completely from IE 5.2 via server-side scripting if at all possible, or perhaps hide it from all IE browsers via a MacIE hiding hack if server-side isn't feasible. The ROI (return on investment) is virtually nil for you to track down and attempt to solve these bugs in a browser that is very outdated, discontinued, and a very small % of users (my guess is 0.10.5% -- unless you're targeting a Mac-oriented audience with older OS 9 machines!). Do what you can, but don't loose sleep over it -- just make sure the content is accessible primarily. --- Justin French http://indent.com.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] 'It Works in Gecko Browsers ...'
On 19/05/2004, at 10:49 AM, Mike Pepper wrote: I may suggest you tip that on it's head. Dead serious. I build in IE then ensure I adjust accordingly. I know ahat will happen in the Geckos. If you start with IE then patch to Mozilla et al, then your thinking is too near-sighted. The standards are there, let's use them, then apply patches to make non-compliant browsers behave nicely. What happens in X years time when IE6 is irrelevant, and you've got to re-visit a whole bunch of stylesheets and bastardised mark-up getting rid of all the IE-centric bloat to ensure it works on the popular browsers of period. I know IE is a *huge* market leader, and I *do* make sure my sites work in IE, but I most definitely tackle 99%-compliant browsers as a whole (Mozilla family, Safari, Opera, Omni, etc) first, because it's a FORWARDS compatible business practice. I use zero hacks, and try and keep the style sheets as simple as possible. THEN I create a separate style sheet for IE 6 (linked after the main sheet, so cascading applies to it), which is hidden inside an IE-only conditional comment. THEN (if needed) I create an IE 5/5.5 style sheet (which cascades over the top of the other two) which deals with older versions of IE. Again, this is done with a conditional comment, so that only older IE browsers download it and read it. What I'm achieving is a definite separation of long term, forwards compatible, future-ready style sheets from those which patch up older or less compliant browsers and will have a shorter life cycle. In X years time when IE5/5.5/6 has disappeared off the radar, I can quite easily drop the stylesheet(s) all together, or make amendments without hacks and complex rules. If you start with IE browsers, you're investing your time (and your clients money) in non-standard (or at least bloated) stylesheets which may create a burden in the future. How will your hacks and IE-centric rules be interpreted by future compliant browsers and useragents (the ones which haven't even been invested yet)? That's the whole point of standards -- you don't have to worry about that. IMO, develop to the standard, then apply simple patches for difficult browsers for a pleasant future -- less bloat, simpler stylesheets, zero hacks, less dependance on *today*'s market leader. --- Justin French http://indent.com.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 *
[WSG] back to basics
Hi all, I feel incredibly inept for even asking this question, but I can't find a solution. I have an application that widely uses single quotes for tag attributes, which is perfectly valid XHTML, eg: input type='text' name='surname' value='French' / However, I can't find a way of escaping a single quote inside the attribute, which is causing problems with surnames which have an apostrophe, eg: input type='text' name='surname' value='O'Riley' / My first reaction (after years of PHP scripting) was to escape it with a slash: input type='text' name='surname' value='O\'Riley' / -- doesn't work. My next step was to replace it with an entity, but there isn't one (apart from #145;, which seemed to cause problems later). The simple answer seems to be use double quotes for all attributes, so that at least the nested double quotes can be replaced with a quot; entity, but I can't believe this is my only option -- am I missing something?? --- Justin French http://indent.com.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] Web safe colours - still relevent ?
On 17/05/2004, at 1:21 AM, theGrafixGuy wrote: I would argue that it depends on your target audience. Suppose you are developing for an audience that mostly uses cellular phones and PDAs. Few of these devices support more than 256 colors, so the web-safe colors are relevant for those devices. But if you are building for those devices, why wouldn't you just build a WAP site??? Nope. Many of these devices are running full-on or cut-down versions of desktop browsers like Opera. The hand-helds have gone beyond WAP. Yes, there are plenty of WAP devices still, and it's still a market, but it's growing beyond WAP very quickly. --- Justin French http://indent.com.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] reference entity year end with ; ???
Vaska, The answer is simple. Your URLs contain ampersands (), which are a character which cannot be used directly in HTML. Why? Because it's used for entities, like amp; and copy; and #8212;. Without boring you with the details, you need to use ?month=4amp;year=2004amp;a=Home, not ?month=4year=2004a=Home to pass validation (hence write well-formed HTML). No, the validator is not broken or wrong. Yes, I know every book tells you to use a plain ampersand. Yes I know it works in most browsers and situations today, but plain ampersands are not correct :) If it's a huge deal to re-write your application at this point, you might consider writing a quick function (in PHP, I'd use ob_start() with a call back function with a regex to replace all the problematic ampersands in URLs on the way to the browser), but your mileage may vary. You're better off getting it right now, rather than relying on such a beast. Justin On 13/05/2004, at 1:03 AM, Vaska.WSG wrote: It's frustrating as it can be very difficult to find information about these things via Google. Anyways, I'm getting alot of error messages when I validate - in particular I'm getting messages like this: 7. Line 50, column 40: cannot generate system identifier for general entity year td class=calndrHdra href=?month=4year=2004a=Homelaquo;/a/td [snip] Oi vey, wondering what I'm doing with this stuff anymore... Can somebody shed some light on these messages? --- Justin French http://indent.com.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] Design Philosophy
On 12/05/2004, at 11:03 PM, Alan Milnes wrote: I have seen some articles on the web that say we shouldn't care about how our web sites look as long as they use valid mark up language and separate content from presentation. Personally I want to design web sites that:- 1) Look good in standards compliant browsers. Look good is a subjective term, but yes, I'm designing websites which look good. 2) Degrade gracefully in other browsers. I design for standards-compliant browsers first, optimising for the future, rather than the past. Once I'm happy, I'll use a combination of the following to attack older, less compliant browsers: 1. @import (to hide the CSS from NN4, IE4, etc), so they just get plain text. I'm guessing WebTV falls into this category too. 2. IE-only conditional comments [1] to provide style-sheets targeted at IE5/5.5/6 if they're proving to be problematic with the main style sheet. As I'm sure you're aware, Graceful Degradation means you (and your clients) do need to let go of pixel-perfect designs on older browsers. Make sure the content is accessible first, and then see what you can do about style on top of that. 3) Are accessible to other devices (one of my readers uses Internet Television so this is a real practical issue for me). The beauty of standards is that most of the work is done for you here. If you mark-up your pages with structural, semantically rich XHTML without any presentational code, you've made a good start. Now, make sure that your pages function and are legible with JavaScript turned off, and with CSS turned off (or just comment out your style sheets). REALLY good start. Then have a glance at the 508 and WAG accessibility checklists, and cover as much of it as you can within reason (another subjective term). The biggest hurdle right now in terms of multiple devices is that a lot of hand-helds and PDAs are reading the screen media stylesheets, instead of the hand-held media stylesheets. Who knows what WebTV reads (if any). Is this a reasonable philosophy or is there something I have missed in this debate? Totally reasonable, well within reach, and it's all around you. There are thousands of beautiful, valid, standards-compliant, reasonably accessible, usable, cost-effective websites out there. It can be done. Those who argue that design doesn't matter are probably not taking into account the real-world business, branding and marketing needs of my clients, which is why I think they're wrong :) I do agree that the content and code should be the primary objectives, but we can have our cake and eat it too. --- Justin French http://indent.com.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] Textarea problem
Can you post a demo link? On 10/05/2004, at 12:14 AM, Vaska.WSG wrote: I'm having a problem with a textarea. It will display properly on the page (which has three div columns) but when a person begins to input data it automatically expands (to the right) so it occupies two columns instead of just the middle column. I've never seen anything like this before... textarea name='msg' class='t1'/textarea .t1 { display:inline; - this is a tip I got from the mail list width:98%; height:125px; } I've validated the code and this problem seems consistent in IE6 in windows 2000. Has anybody experience this same problem? I could post more code for this but I'll wait and see if there might be an easy solution to this first. --- Justin French http://indent.com.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] Re: Ten questions for Anne van Kesteren
On 06/05/2004, at 12:03 AM, Chris Bentley wrote: On 05/05/2004, at 10:09 PM, Patrick Griffiths wrote: I thought XHTML transitional _is_ XML. In what way is XHTML transitional is a less strict data format? It's a transition. It's a half-way house between HTML 4 and XHTML as it is intended (XHTML Strict). Are you saying that XHTML transitional is a less strict data format than XML too or are you off on some tangent? If the the former then please explain in it more detail, I really am under the impression that XHTML transitional is XML - that being so, in what way can it (XHTML transitional) be a less strict data format (than XML)? I *think* that the transitional aspect is related to the set of available tags, rather than it's XML suitability. A lot of behavioural/presentational tags and tag attributes were removed from strict, but left in for transitional. Whether XHTML is valid XML is beyond my knowledge, but I believe it is. --- Justin French http://indent.com.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] resources for media=print?
On 04/05/2004, at 10:50 AM, Paul Ingraham wrote: I've been studying CSS steadily for about a three weeks now, and I have yet to come across any information about writing stylesheets for media=print. I know that they exist, but I don't know anything about them. Is this because: (a) it's so simple that there's no need for education resources about on the subject? :-) or (b) I just haven't looked in the right place yet? If there are some good articles or tutorials about this, I'd be grateful if someone could point me in the right direction. Before thinking about ANY topic, go check out A List Apart... usually, someone has already thought about it, and has come up with something pretty detailed to walk your through it without any glitches. In this case, Going to Print by Eric Meyer will help: http://www.alistapart.com/articles/goingtoprint/ And as always, Google can solve just about anything: http://www.google.com/search?num=50hl=enlr=lang_enie=UTF-8oe=UTF -8q=media+print+stylesheet+tutorialbtnG=Search --- Justin French http://indent.com.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 *
[WSG] iCal and other Calendar Formats?
Hi all, I know this is a little OT, but hopefully I can justify it :) I'm looking at publishing an event calendar on a few band websites I maintain in formats other than HTML... in particular, I know that Mac iCal users and Mozilla Calendar users can subscribe to an iCal file over the web, but I was hoping there might be some options for other standard formats (XML?) that I'm unaware of, particularly targeting users on Lotus Notes, Outlook, Entourage, Palm, etc. Can anyone point me to some standards or guides for targeting these users with subscription-based calendars? And yes, since I'm serving these over the web, and am looking for standards, I think this is on topic enough :) --- Justin French http://indent.com.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] SMH launch
Peter, Here we go again :) Alternative #1 My guess is that you need to attach the background image to something other than the body (like #wrap)... my theory being that if #wrap can't disappear off the left of the window, then neither can it's background :) Alternative #2 I can't see any major use for the BG image at all... perhaps the line effect you're trying to achieve can be done some other way, negating the need for the image at all. It's a bit hard to tell what's what without spending a bit of time studying the CSS. Alternative #3 Ignore what's happening, and put a solid white background behind that left nav bar, so that when the BG image goes under, it doesn't obstruct the navigation. I'd consider this graceful degradation considering your % of Opera visitors. On 29/04/2004, at 1:27 PM, Peter Ottery wrote: when we launched theage.com.au last week Justin pointed out a way (adding 1px padding to the left of the main #wrap div) to make Firefox keep the background image aligned hard left with the content when your browser window was narrower than the content - and stopped the background image becoming mis-aligned with the content. Even tho that fixed it in Firefox the problem still exists in Opera and mac ie. Heres a screenshot of the new smh site with a browser window set narrower than the content (note the body bg mis-aligned with the left nav): http://www.c41.com.au/test/opera7_2_squished.jpg any ideas of a way to make these browsers keep the background image aligned hard left and not adversly affect anything else? pete --- Justin French http://indent.com.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] A discussion leads to an idea - Dynamic CSS!
On 23/04/2004, at 7:20 PM, theGrafixGuy wrote: I met with a programmer today to actually discuss the idea of a Portland, Oregon based WSG - he likes the idea and will talk to others about the idea. However, during our ongoing 3 hour conversation over a few pints, the discussion eventually moved from WSG to web design and then to experimental design. Then an interesting idea hit the table, what is to prevent a person from creating a dynamic stylesheet. Has anyone ever toyed with this idea before and if so what were the results??? It depends on your definition of dynamic... I've built PHP-driven stylesheets to serve a stylesheet customised to the browser or user, but I was concerned about the caching which: - if it cached, wouldn't be dynamic - if it didn't cache, would cost bandwidth and performance It enabled me to do really simple stuff, like store my colour pallet in PHP variables, allowing for hassle free redesigns and tweaking (body { background-color: ?=$col1?; ? for example). And more complicated stuff like testing for certain user-agent strings for tweaking, or even user-preference based style-sheets and style-sheets to reflect the time of day, or the weather, or whatever else... the guts of the problem (which I haven't properly investigated) was caching... I think a more realistic option would be using PHP to dynamically *link* to an appropriate STATIC style-sheet (eg ie6.css, or nighttime.css or winter.css), which would still allow full caching by the browser. Now, since I just had my wisdom teeth out last night, I should go back to the couch :) --- Justin French http://indent.com.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] marquee text
I don't have a solution in the form of code, but I do have the theory behind a solution. I also assume that you mean a horizontal block with a line of text that moves from left to right, yes? From a mark-up point of view, I'd have the following: div class='marquee' pThis is the long piece of text to be scrolled in the marquee. This is the long piece of text to be scrolled in the marquee. This is the long piece of text to be scrolled in the marquee. This is the long piece of text to be scrolled in the marquee./p /div My base CSS would just style the paragraph in a way that's accessible to all: .marquee { border: 1px solid red; } .marquee p { font-family: verdana; } In other words, nothing special at all. THEN, with javascript (this is the bit I can't give you), you can: 1. dynamically change the styling so that the paragraph is a fixed height (eg 25px), with the overflow set to hidden, which effectively chops the paragraph off on the right edge, and whatever else you need. 2. dynamically move the .marquee's paragraph to the left by 1px every .5 seconds (or something), which will simulate the marquee effect. 3. I think you can also use JS to detect the width (in px) of the para, so you should be able to reset the marquee at the end of the content, and go again. What this means is that the content is accessible to all by using simple tags, and simple CSS. Then *if* JS is available, the marquee is styled differently, animated, etc. Good luck, --- Justin French http://indent.com.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] CSS Tables
On 22/04/2004, at 11:37 AM, Noa Groveman wrote: I recently converted a directory lister script from using table tags to using CSS styled tables (display:table), and I've noticed that there is no provision for a colspan attribute. This makes sense, because tables are for displaying tabular data and not for fancy headers, but I want to do it anyway. Basically I want to make the first cell (that displays the path) maintain the entire table's width without effecting the other columns' width as it does in the table version. Note, I have a user agent switch, since CSS tables don't work with IE, for displaying the old tables or the new display:tables accordingly. Here's an example: http://eastsdomain.com/test/ . I don't get it. I don't think you need to use display:table to achieve what you want, but then again, it's not all that clear what you want. Why not post a table-version with a layout you like, then we can have a look at what to do from there. Remember, it may still be appropriate to use a table (if it's tabular data), and it may also be worth thinking outside the box, taking advantage of CSS's strengths, rather than wasting life emulating table behaviour in CSS. --- Justin French http://indent.com.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] CSS Tables
On 22/04/2004, at 12:19 PM, Noa Groveman wrote: The only advantage I can tell is that it just doesn't use any table or td or tr tags, which I have an irrational hatred towards. I've done what you suggested, but the problem is that I can't make the header have the same width as the table. I've tried using a containing box, but then the header stretches unnaturally off to the right past the table, I'm not sure why. I really agree with Noa here -- it's tabular data, so figure it out with tables, not divs and spans. --- Justin French http://indent.com.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] Attribute wrap not allowed in text areas
On 21/04/2004, at 12:33 AM, theGrafixGuy wrote: What is the recommended replacement when using a textarea? I haven't replaced it with anything... just got rid of it, which has been fine for me, since I used wrap='virtual', which seems to be the default behaviour for every browser I'm testing... further more, it's always a good idea not to mess with the default behaviour of widgets and other browser/OS level GUI elements. If you were specifically after wrap=none or wrap=physical (?), then I'm not really sure... my guess is it can be done with CSS's overflow --- Justin French http://indent.com.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] theage.com.au: new design
Peter, It's great to see another big player strutting a bit of CSS/Standards muscle! I really admire what you're trying to do with JavaScript to use a wider layout where possible (with a sensible default), but resizing the browser window (quote a common occurrence I would have thought) can produce some scary results... Resizing under IE six seems fine (haven't checked 5.x), but Firefox (for example) loses it's navigation bar (left) when resized down from 1024 to 800 wide, and worse still, it can't be scrolled to. In the process, the background image sits in a weird spot (obscuring text), and things like the search box disappear off the right of the screen. I *think* you might be able to solve the problem in one of two ways: a) on resize, reload the page (annoying, but it will fix the problem) b) place a 1px border or padding on the left edge of the layout, which will force your layout to disappear off the right edge of the window only (not the left), which will mean the behaviour is normal for a 1024 layout viewed at 800. I stumbled on option B one day when researching something related to what you're doing. Here's a quick sample: !DOCTYPE html PUBLIC -//W3C//DTD XHTML 1.0 Strict//EN http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd; html xmlns=http://www.w3.org/1999/xhtml; xml:lang=en-au lang=en-au head meta http-equiv=content-type content=text/html; charset=iso-8859-15 / titleUntitled/title style #wrapper { position: relative; width: 760px; margin: 0 auto; padding: 1px; } /style /head body div id='wrapper' a b c d e f g h i j k l m n o p q r s t u v w x y z /div /body /html Setting the padding of #wrapper to 0px and resizing to a small window will cause a b c... to disapear off the left. But leaving it at 1px will ensure that the left edge of #wrapper doesn't disappear off the left edge of the window. Your combined Mozilla/Firebird audience looks like at least 34372 unique visitors a day, so it's something you probably want to take a look at :) --- Justin French http://indent.com.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] CSS version of onfocus onblur for form fields
On 20/04/2004, at 11:49 AM, Paul Ross wrote: Hello folks, Does anyone know if there is a CSS equivalent of onfocus and onblur? There is a CSS pseudo selector :focus http://www.w3.org/TR/CSS2/selector.html#x33 I have a homepage with 2 input boxes - one with the initial value called 'Username' and the other called 'Password' (which shows up as on screen). What I would like is the user to click on the input box labelled 'Username' and the initial value then disappears and is replaced by whatever the customer is typing. Same for the password field. I have read how I can do this using the JavaScript onfocus and onblur commands but was wondering if there was a better, more accessible (?) way using just CSS. Sorry, that's a JS job, not a CSS job. If you're going to do it at all, I'd use JS to SET the default on body load AND remove the default on focus, so that non-JS users are not inconvenienced by having to select and delete the text. Better still, don't do it, because it changes the behaviour of the browser. People expect that they'll have to select+delete existing text, because that's what USUALLY happens. Changing this behaviour seems like bad usability 101 to me. --- Justin French http://indent.com.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 *
[WSG] RSS or Atom for an events calendar?
Hi all, Hopefully this is an applicable place to discuss this. I've just started looking into RSS/Atom/etc (news feeds in XML), and everything is going pretty well, but I'm working on a website for a band, and the news feeds seem easy enough, but I'm interested in the possibility of using RSS or Atom for calendar events (shows and tours). But I'm not sure RSS/Atom can be used in this way. The theory behind shows for a band is that they need to be advertised/shown UPTO the date of the event, then they're irrelevant. The theory behind news readers is that once you've read something, it no longer exists, which is not what I'm aiming for of course. RSS2.0 only seems to have a published date, not an expiry date, so that doesn't seem to help. Atom seems to have issued and modified, but again no expiry. So, how can I take advantage of RSS/Atom/etc to keep fans up-to-date with shows/tours? Some ideas I have: a) when a show is added/edited/deleted from the show database, a news item is created advising of the new/revised/deleted show, which would of course appear in the news feed. The advantage here is that there's multiple news items which can be tracked. The disadvantage is that shows won't appear in date order (or reverse date order) -- they'll simply be added to the feed as things change. b) a news feed of *just one item* containing a summary of all upcoming gigs is offered over RSS/Atom... as the gig database changes, this news item is either *modified*, or *over-written* with a new summary... the advantage here is that the summary will be ordered correctly, and the latest (and only) news post will contain all the information in one hit. Any other ideas? --- Justin French http://indent.com.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] RSS or Atom for an events calendar?
On 14/04/2004, at 12:38 PM, Geoff Bowers wrote: You might also consider generating an iCal feed which is simple enough -- although Outlook has no idea (as per usual) the rest of the calendaring world regards iCal as a common protocol. That's a great idea, and it looks like there's some PHP classes to do it all too! Is there a similar method of calendar publishing with Outlook at all? Justin --- Justin French http://indent.com.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] Re: IE with Gecko Was: Relative Fonts
On 07/04/2004, at 5:50 PM, Leo J. O'Campo wrote: Maybe we as a standard based community should knock on $MS's door and shout I'm mad as hell and I'm not going take it anymore. Pffft. A bunch of web standards geeks like us (no matter how large, it's still just a tiny fraction of the entire web community) will have no impact whatsoever on MS's plans in regards to IE. For starters, MS will only pay attention to something which will increase or protect it's market share... with their market share up over 90% somewhere, they don't really need to do either. It would take a substantial shift of users away from IE (we're talking *millions*) before MS could even *begin* to care. Secondly, it's widely known and reported that IE is a dead weight. The new browser in the next generation MS OS will be fully integrated. I can't imagine MS spending any time on improving IE when they (and us) know it's headed for the scrap heap. The best to expect (IMHO) is security fixes. --- Justin French http://indent.com.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] Flow from bottom?
On 04/04/2004, at 10:23 AM, Tim Shortt wrote: russ weakley wrote: One method - not tested but in theory: Place the content inside a container, and apply absolute positioning to the container: #contaner {position: absolute; left: 20px; bottom: 20px;} The container will set at the bottom of the viewport or its containing box and the content will flow up rather than down. Russ I was lurking and took the liberty of trying this, and the content indeed stuck to the bottom of it's container. However, once the window was resized smaller (vertically), a scollbar would not appear. So the top-most content moved up and cut off at the top of the browser. Any ideas on how to prevent the top content from cutting off? There is a similar problem with some browsers not providing scroll bars on centred+fixed width content, which I found could be avoided if the fixed-width container had a 1px border on the sides... I don't have the source code you guys are both working of, but it's worth a try. --- Justin French http://indent.com.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] hiring a standards-savvy designer
The only way to find a designer that is sensitive to what's required with standards, accessibility, CSS-based layouts, valid XHTML, etc is to find a web designer with ALL these skills, even if you don't take advantage of them all. They must be able to hand-code valid XHTML in order to have an understanding of what's required. They must have some experience using CSS2 for layout to know it's strengths and limitations. They must have attempted to get a few sites upto 508 or WAG in order to fundamentally understand Accessibility. The bottom line is these skills *can* be *TAUGHT* to any new employee, so maybe all you really need to know is if they can hand-code HTML and CSS, and have an understanding and appreciation of standards accessibility? I think I'd just hire someone with ALL the skills, then take advantage of as much as you can -- over time your business will grow to take up all his/her skills. On 31/03/2004, at 10:52 PM, Kay Smoljak wrote: At work, we're about to hire a new graphic designer, as our guys are flat out. We're looking for someone with some markup skills as well as visual design, and as I'm the nominated standards nazi I'm charged with making sure their html and css is up to scratch. As this is primarily a graphic design position, I'm not expecting any of the applicants to be too savvy about web standards already. But I'm hoping to find someone who doesn't consider Fireworks Web export a good way to create sites, and who won't be too hard to bring around to our way of doing things. So what am I actually asking? I'm interested in what you guys consider reasonable to expect from a graphic designer who also does some overflow html. What would you be looking for? What would you ask in the interview? Thanks for any ideas, K. --- Justin French http://indent.com.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] Show/hide layers without javascript (was: [WSG] How to do some things)
Leo, On 01/04/2004, at 1:49 PM, Leo J. O'Campo wrote: To say that anyone concerned with accessibility should have JavaScript enabled **utterly misses the point**. Accessibility is about providing access to the content for the widest possible number of users, regardless of how they're accessing it. Hmm.. well Justin your missing my point. I did not use those words. My point was if a page was designed for JavaScript to be used, then disabling it, is to your own disadvantage. And you're missing mine :D Usually disabling (or not having) JavaScript is not a conscious decision -- I know there would be a minority of technically knowledgeable web users who would specifically go into their Internet Options/Preferences and disable JS by choice, but this is definitely a minority. In most cases, my educated guess is that if JS is disabled or not available, it will be a choice made by someone else: - the useragent vendor (defaults to off, or not available at all) - an IT manager, department head, or some other paranoid management type - a 'power user' friend, colleague, who modified the settings for whatever reason - the owner of the computer (in the case of shared or public computers) As such, I don't blame the lack of JavaScript on the user. Therefore, any page which relies on javascript being enabled for it's functionality is inaccessible, and I blame THAT on the web developer and site owner, not the user. It's to the *website's* disadvantage, not the *user's*, because they've just turned away a reader/customer/client/friend. --- Justin French http://indent.com.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 *
[WSG] *sigh* drop-downs
Hi all, I'm quoting for a client at the moment who seems to have her heart set on dropdown menus for the sites navigation. I've implemented such menus before, and am currently using a version of the Suckerfish menus [1], which is fine for a one-deep hierarchy, but it looks like she wants 3-4 levels of depth in the menus, which is going to: - significantly increase development time and testing time - decrease browser compatibility accessibility I would also speculate that menus with such deep hierarchy would confuse and distract users. But we all know how difficult it is to convince a client that they're wrong, especially when they see huge hierarchical menus on other corporate sites. So instead, I'd like to read up (and point her to) any studies conducted in terms of their usability in context of website navigation, perhaps even compared to other forms of navigation. I have no doubt that such menus ARE usable (we use them every day in Windows, Mac OS, etc), but as pointed out recently on this list (or another?), the menus in our OS are not *navigational* -- they're *functional*. I'm confident I can provide simple, smart navigation without them, but first I need to find some solid proof that they're a bad idea :) --- Justin French http://indent.com.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] How to do some things
On Wednesday, March 31, 2004, at 06:21 AM, Vaska.WSG wrote: I'm trying to find a method for showing/hiding layers using CSS which doesn't use javascript if it's possible (?). Ideally, a layer (let's call it A) would automatically show when a page loads, then when a link is clicked it would disappear and B would replace it. Then, another link would show A again and hide B, etc. I don't think it can be done sorry. Remember, CSS is about presentation, not interaction. The only property you really have in CSS to play with is :hover, which allows some interaction with an element in it's hovered state. [1] This has allowed many CSS-only drop-down menus to crop up [2], but since IE only supports :hover on a's, DOM javascripting [3] is required to make IE work as expected. In otherwords, you certainly can't escape JS on IE -- an exception may be all that IE7 huff that's going on at the moment. I'd suggest checking out the scripting and work down with Suckerfish drop-downs [3], but since you were looking for clicking interaction, not hovering, my guess is this won't be sufficient. 1. http://www.meyerweb.com/eric/css/edge/popups/demo.html 2. http://www.meyerweb.com/eric/css/edge/menus/demo.html 3. http://alistapart.com/articles/dropdowns/ --- Justin French http://indent.com.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] Hiding styles message to certain browsers
On Friday, March 26, 2004, at 02:50 AM, David McDonald wrote: From my perspective, the whole point of coding to standards means that it doesn't matter what browser the user is viewing your site in - they should be able to read your content regardless. Whilst a little idealistic, yes that's the whole point of standards. We can only hope that one day we don't even have to TEST the site in multiple browsers -- it will just work. But that's not today. They can still read your content is also pushing it a little, because in reality, the Mac IE5 user will see the broken styles, and *possibly* be able to read / access the page. If they can't, they'll either turn off their style sheets to access the unstyled content, or walk away. Sadly, my guess is the latter for 99.9% of that community. I certainly do hide my CSS from Netscape 4.x, but being a Mac user, I've always attempted to pull IE5 into line -- which means no hiding. There are ways to target JUST IE5Mac in your style sheets, and it can also be done with JavaScript or server-side scripting like PHP and ASP. For IE on Windows, I use IE conditional comments to directly target IE versions. I guess my point is I'd try EVERYTHING I can (JS, PHP, CSS hacks, etc) before giving up / hiding the style sheets / accepting a broken style-sheet on IE5Mac (or any other popular browser). I would however not embed all this mess in my main style-sheet -- I'd either separate it into targetted style sheets (eg ie5mac.css), or at the very least comment the hell out of it so that in 5 years time, when IE Mac and most IE Win versions are irrelevant, it's easy to trim their bloat out of your CSS. As already stated, there's minimal choices for Mac OS9 users, and anyone still on IE5Win probably has a good reason too :) --- Justin French http://indent.com.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 *
[WSG] Firefox 0.8 bug?
Hi all, I'm playing around with some web pages which have a wrapper div centering the entire web page contents via margin: auto;. My experience and testing with this method has led me to believe that's it's virtually flawless -- until I looked at my current page (and others) in Firefox 0.8 on Win XP. The bug: While viewing http://www.simplebits.com/ , http://www.9rules.com/whitespace/ , or any other site which uses this method on Firefox 0.8 under Windows XP, there's something definitely wrong. - At 1280 x 1024, the layouts are perfectly centred with no problems. - At 1024 x 768, the layouts are a *little* off-centre to the right. - At 800 x 600, the layouts are a *lot* off-centre to the right, with content disappearing off the right side -- and worse still, no scroll bars are enabled to allow me to view the missing content. The bug is not present in Mozilla 1.6 or Mac Firebird, but given the amount of fixed-width, centred sites using this method, I'm thinking this is a big problem :) Is there a work-around, or do we just wait for the next release, and hope that people don't stick with 0.8 for ever? PS: had a quick look in Bugzilla, but couldn't find anything related. --- Justin French http://indent.com.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] Firefox 0.8 bug?
Well, I'm stumped! I did a restart, and everything is fine again -- I can't repeat the problem AT ALL. Since I had recently installed Firefox 0.8 Mozilla 1.6, as well as Opera 6 7, all without a restart, something must have gone temporarily wrong. Who knows -- I spend 99% of my day on OS X :) The important thing is that it's not a widespread problem. Thanks all for their feedback! --- Justin French http://indent.com.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] Suggestions about what to do here ...
On Thursday, March 25, 2004, at 12:46 AM, Michael Kear wrote: [A] say hes got IE5, tell him to take a jump because he needs to upgrade. (Ive done that, but if theres another answer thats easy Id like to do that too hes not alone) That seems a little arrogant. [B] just let him lump it and use the CTRL-Wheel to increase the font size That's an option, but why not attempt to fix the problem? [C] change the style sheets to accommodate IE5 users. Bingo. I'd just include a separate style sheet specifically for IE5 users using an IE conditional comment (search google), which fixes the problem -- however you first need to determine what the problem is... So whats your suggestions about how I can handle this? Im not against telling him tough upgrade your browser! but if theres a fix thats fairly easy Id like to consider that. I just logged in with IE 5, and whilst the font is *small*, it's not *tiny* FOR ME. So, I've got no real idea why the fonts are small for him: 1. he may have his font size set to small 2. he may have poor eye sight, or a bizarre resolution, or a bad screen, or left his glasses on the top of his head 3. he may have some other exclusive problem that is making the text hard to read (browser setting, OS setting, beta version, etc) I'd start by checking out his browser's font size (eg View Text Size Medium), pointing out that he can increase/decrease it there to customise the browser to suit his needs. Try and find out why text is so small in his situation, and perhaps even ask for a screen grab. If that doesn't work out, offer some alternate style sheet tests with different font settings, seeing what he prefers. If it's something that could work for all users, or all IE5 users (using a conditional comment again), then implement it and you're done. If that doesn't work out, *then* consider telling him where to go -- the key is to make sure it's just him, not a whole bunch of IE5 users, before dismissing it as an anomaly. Obviously getting the HTML and CSS valid might fix the problem all by itself. --- Justin French http://indent.com.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] Standards-compliant browsers - in order!
On Monday, March 22, 2004, at 07:00 AM, Martin E wrote: 1.So, IYHOs, Is it better to code, then check, code some more, then check again, using a much more standards- compliant browser like Mozilla, or go with ie, then tweak for the rest? Yes. Start with something compliant (because it's the future), then go back to IE. Personally, I patch-up IE's rendering with a IE-only conditional comment [1], after making sure all the good browsers are happy. 2.Is Mozilla more standards-complaint than the rest, or should I rely on Opera first? The latest version of all of them are pretty good, but keep in mind that users might still have the older versions. Personally, Firefox 0.8 is my default perfect browser to check. A.Which browser (which version too), in order of compliance, rate first in standards. Is my list accurate: a.Mozilla builds (1.5, 1.7b, etc) b.Mozilla Firebird 0.7 c.Mozilla Firefox 0.8 d.Opera e.Netscape f.IE For starters, I'd guess that Firefox 0.8 comes above Firebird 0.7, and NN6 and NN7 were built off early Mozilla builds. You're also missing Safari 1, 1.1, 1.2, and OmniWeb for the Mac, a few different things for Linux, etc etc. Secondly, I don't think it's worth establishing a list / order. Pick a browser with pretty good support (for me, it's Firefox 0.8 and Safari 1.0) to develop on, then test and adjust on everything else... last but least, I look at IE 5/5.5/6, and try to patch it's rendering engine with an additional stylesheet, linked inside an IE conditional comment. --- Justin French http://indent.com.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] drop down menus
On Monday, March 22, 2004, at 09:29 AM, Universal Head wrote: I recently had a client who insisted I implement drop-down menus for the navigation on their site, even though I gave them all the reasons why I thought they were unnecessary in their case - and I was wondering what the list's thoughts were on this method of navigation. Personally think that in most cases they are unnecessary. I think a well designed site should present information in a hierarchical fashion, allowing the user to access more detailed info as they progress into the site (while still keeping all parts of the site quickly accessible in two or three clicks). I think the opposite approach, of making every section and subsection available from the homepage via drop-down menus, has the opposite effect to what is intended by confusing the user with masses of choices upfront. What do y'all think? They're generally unnecessary, unless the client wants them, in which case they're necessary :) On a more serious note, they do allow for navigation though a hierarchy of content MUCH faster than viewing 2-4 pages of sub menus before finding what you want. MUCH FASTER. It also allows for an overview of an entire section (like the contents page of a book) without making a single click. And whether the UI model of hierarchical menus is sound or not, people are FAMILIAR WITH THEM, because of Apple and Microsoft's long standing use of them in MacOS and Windows. As long as you make them behave like Windows Mac menus and keep the terms short and clear, you should be fine. I don't think they overwhelm the user at all, again, because the information is organised in small bite-size chunks. To sound like a broken record, they work for MS and Apple and their billions of users everyday, so they'll probably work for your client too :) --- Justin French http://indent.com.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] iCapture + Drop down menu
Maureen, I found the site was extremely slow to load in both Safari and Firefox on Mac, and still pretty slow on IE Win. However, the problem only existed on the first page, which suggests that whatever the problem is, it's being cached so that future pages don't suffer from the same problem. The javascript is 15k, which is pretty substantial, but I'm on DSL with heaps of bandwidth, so it's NOT a download issue, so my only guess is perhaps that my browsers are caching something *else* that's slowing down the first page -- perhaps it's the images (16k for one of them), or perhaps it's related to the JavaScript... the only way to test would be to turn off your cache and try different versions (with images, without, with JS, without, etc). If the JS is the problem, I'd seriously consider trying different scripts that are more minimalist -- the speed of your first page is VERY IMPORTANT, and perhaps bloated JS is not the solution... anyway, onto your problems: I recently put up my first two commercial sites. I just ran them through iCapture and on one of them the menu is not showing. The clients wanted a dropdown, flyout menu so it is a JavaScript menu and takes a tad longer to download - would this be the reason it is not showing or do you think there is a problem on the Macs/Safari? I would appreciate it if someone on a Mac could do a quick site check for me please, the url is www.cwfs.org.au The menu (and sub-menus) are working fine in Safari 1.0 on my Mac -- I can only assume Safari 1.1 and 1.2 are fine as well. Not sure about your other issues. --- Justin French http://indent.com.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] Just got an eMac - now I'm more confused than ever!
On Thursday, March 18, 2004, at 11:41 AM, John Penlington wrote: I normally use IE6 on Win XP Pro to code pages in valid XHTML 1.0 Transitional. Bravely, I bought an EMac to view sites I'm building on IE 5.2 for Mac and Safari 1.1.1 I'm determined to follow Web standards, but the eMac browsers have got me very confused. I read that Safari is about the most standards compliant browser you can get. My ultimate aim is to code for that and fix problems on other browsers. If you're aiming for standards compliance, **don't start with a browser that poorly supports standards**. Do your initial development on a single standards-compliant browser like Firefox, and validate your mark-up and CSS periodically. Learn about IE's major problems BEFORE you start coding -- avoiding padding will solve 99% of layout problems. Avoiding and side-stepping these problems during development will mean less headaches later. Happy? Now have a quick check in the rest of the Mozilla family on both platforms (Mozilla, Firebird, Firefox, Camino), on as many versions as you can get your hands on. Adjust as needed, validate, re-test. Happy? Now have a look on other browsers that closely support standards, like Opera, Safari, OmniWeb, etc. Adjust as needed, validate, re-test, making sure the Mozilla family doesn't break. Happy? Now you have a standards compliant site that works in standards-compliant browsers. But what about IE? Since IE is not standards compliant, the best you can really hope for is graceful degradation -- perfection across all browsers *may* happen, but let's be realistic! Check your site in IE 5.5 and 6.0 (even 5.0 if you feel it's important). Now apply any hacks [1][2] you need to get IE to look right. Most will be in regards to the box model for layout. Personally, I keep all hacks and bloat out of my main style sheet, and use an IE-only conditional comment [3] to target IE exclusively with it's own style sheet with overrides my base style sheet with patches for IE. When I've got more time, I'll document all this properly :) 1. http://css-discuss.incutio.com/?page=BoxModelHack 2. http://www.mezzoblue.com/archives/2004/02/25/sidestepping/ 3. http://msdn.microsoft.com/workshop/author/dhtml/overview/ ccomment_ovw.asp --- Justin French http://indent.com.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] SuckerFish Dropdowns
On Tuesday, March 16, 2004, at 06:53 PM, 7 sinz wrote: Hey i've been experimenting with suckerfish drop downs; first off the code works brilliantly, but i wonder if it can be achieved without the extra mark up, if there is a way??? Basicaly what im trying to do is create a CSS based drop down menu with the mark-up length of Mezzoblue's m0se menus, but the funcuality of Suckerfish menus. But im getting no where at the moment, if anyone out there is willing to throw in some suggestions it would be great. What extra mark-up? The suckerfish code is lean and mean... I just went back and had another look, and there's nothing there in the STRUCTURAL mark-up. If you're talking about the CSS, that's also pretty lean, with the exception of a few extra selectors in there for IE. That only leaves the added JavaScript, which again is there only for IE to work. If you choose to support IE, then there HAS to be JavaScript in there, and there has to be some extra CSS selectors. I'm not that familiar with the Mezzoblue version, but the same basic rules apply. --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *
Re: [WSG] SuckerFish Dropdowns
On Tuesday, March 16, 2004, at 11:05 PM, 7 sinz wrote: Dave Shea makes a drop down with 2 lines of CSS. So i wondering if you could apply the same tech, using the DOM scripting from the suckerfish menus; any thoughts? You'd have to show me a sample. --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *
Re: [WSG] loading links when loading a page.
On Monday, March 15, 2004, at 10:10 AM, Kim Buttery wrote: A number of my web pages insist that links I visited, when checking them remain in the 'visited' state in the future. I would like the links to revert to a none-visited state when a page loads. I believe there should be an 'on load' instruction but I have not been able to find the syntax in either my html references or the CSS references. Can someone assist with the correct syntax (? on load a:link ?) and tell me if it should be placed in the head content /head area. Thank you.. I'd really recommend against this, as it's changing the behaviour of the user's mail program -- generally a big no-no. For example, in Firefox 0.8's preferences, I can 'remember visited pages for the last N days'. It's a user preference, NOT a developer/site-owner preference. Of course there **MAY** be occasions where this is okay, I'd avoid it all costs if possible. Take the following example: 1. I visit a friends blog, and read the newest three posts... my browser remembers those last three posts as being visited, and offers a visual cue that this has happened. 2. I close the window, and leave the site. 3. I come back a few days later, and your script screws with the visited links, turning them all into fresh links. 4. Rather than glancing at each title/link and seeing that it is visited, I have to read the title/link, remember if I've read it, then decide what to do. If you personally want a different scenario for your browser, mess with your preferences, and set the history/visited pages preference to 0 or 1 days, and see how that goes, but please don't mess with my preferences :) --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *
Re: [WSG] Purpose of this mailing list
On Thursday, March 11, 2004, at 07:28 AM, Taco Fleur wrote: The only downside is that mailing list is that it is hard to organise it to specific folders. I have a folder for WSG but 90% of the mail ended up in my inbox instead because the from field is actually using the sender's name.\] You can also filter on the [WSG] in the subject, which is fool-proof. --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *
Re: [WSG] A kinda Simplequiz
On Thursday, March 11, 2004, at 10:39 AM, Manuel González Noriega wrote: How would you mark up an interview? a) dl dtSo, how are you doing?/dt ddFine, thanks for asking/dd /dl This seems most appropriate, given that a conversation/dialogue is given as an example usage of a DL in the specs. b) p class=qSo, how are you doing?/p p class=aDidn't you just ask me that on a)/p Yes, this seem appropriate too. c) hxSo, how are you doing?/hx pPlease, stop it./p The questions aren't really headings, so I'd say no on this one. --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *
Re: [WSG] Purpose of this mailing list
On Wednesday, March 10, 2004, at 02:21 PM, Paul Ross wrote: Having said that - when we get to 1,000 members I guess something will have to be done or the success of the list could implode in on itself and the noise traffic become too much to handle. I would suggest that you think of switching to a forum based website much like the excellent http://forums.australianinfront.com.au/Default.aspx As soon as lists move to forums, I stop posting, stop reading, and stop helping, as do many others. Web browser-based discussion lists are difficult, slow and tedious at the best of times, which is the complete opposite of mail and news groups, which were *designed especially* for threading, replies and message based discussion. Everything related to discussion happens faster and easier with a mail client than it does with a browser. Browser-based discussion has one positive; that being the fact that new subscribers can read old posts and search for topics before posting. This of course can easily be overcome with web-based archives of email lists (which is common anyway). --- Justin French http://indent.com.au * The discussion list for http://webstandardsgroup.org/ *