The original goal of writing clean HTML was for static, document-centric 
websites where you could do things like run a filesystem search or a web 
spider to create an index of all your documents, use an HTML editor, or 
perhaps do a search and replace. And of course it helps with SEO. But how 
relevant is it to a web application where "view source" won't show you much 
and you'll only see the DOM in the debugger?

On Wednesday, October 10, 2012 2:22:32 PM UTC-7, Roger wrote:
>
> There is no such thing as "producing clean vs not clean" html unless you 
> rely on other peoples widgets.
>
> 100% of my widgets are a UIBTemplate.. of my creation… I use GWTQuery (or 
> jquery) to add/remove elements from my widgets.  Thus, the HTML is exactly 
> as clean as any HTML that any non-gwt application would use/produce.
>
> Roger
>
> On Oct 10, 2012, at 4:57 PM, Thomas Broyer <[email protected]<javascript:>> 
> wrote:
>
>
>
> On Wednesday, October 10, 2012 10:13:36 PM UTC+2, Shaun Tarves wrote:
>>
>> There is no doubt that what GWT does, it's really good at. However, some 
>> things that I've found GWT really isn't good at:
>>
>> 1) Producing clean HTML
>>
>> The structure of GWT "page views," especially with GWT widgets, is really 
>> poor. The DOM gets bloated with lots of extra elements that are used for 
>> focus and positioning. There are ways around this, but I feel like I'm 
>> constantly fighting with GWT to generate HTML structure on my terms.
>>
>> For example, some of the most lauded constructs in GWT are the Cell-based 
>> widgets (CellTable, and CellList, specifically). With CellLists, you are 
>> stuck with divs. There's no way around it. So that means if you want to 
>> make a good data model-backed list and render it as a UL with LIs, you're 
>> SOL.
>>
>
> It's a false problem. GWT widgets are generally good as far as 
> accessibility is concerned, and let's put it clearly the only reason on 
> having a "semantic" DOM tree is for a11y.
>
> 2) The history mechanism is really powerful, but it's important to get 
>> your URL structure correct from the start. The built-in history token 
>> parser is a little too rigid in that it forces the first part of your URLs 
>> to be of the form xxxx:yyy and then anything you want after that. When you 
>> dive deeper into GWT, you'll understand the limitations of the 
>> PlaceHistoryMapper and why you might want to avoid the tokenizers and write 
>> your own parser.
>
>
> On the plus side: it's pluggable. (it wasn't at first, you had to 
> re-implement the whole PlaceHistoryHander+PlaceHistoryMapper)
>  
>
>> 3) The GWT CSS compiler doesn't understand any CSS3 attributes. Also, 
>> browser-specific attributes (such as the * hack for IE) throw warnings on 
>> compiling. It's not really GWT's fault (it's a Java compiler issue), but be 
>> aware nonetheless.
>
>
> You don't need browser-specific hacks, simply use "@if user.agent ie6 
> ie8". The real issue is with selectors. FYI, gradients can now be used 
> without literal() in 2.5.0-rc2: 
> http://code.google.com/p/google-web-toolkit/issues/detail?id=5771
>
> -- 
> You received this message because you are subscribed to the Google Groups 
> "Google Web Toolkit" group.
> To view this discussion on the web visit 
> https://groups.google.com/d/msg/google-web-toolkit/-/A-GepWmKMf0J.
> To post to this group, send email to 
> [email protected]<javascript:>
> .
> To unsubscribe from this group, send email to 
> [email protected] <javascript:>.
> For more options, visit this group at 
> http://groups.google.com/group/google-web-toolkit?hl=en.
>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Google Web Toolkit" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/google-web-toolkit/-/WgLQsqv2dBoJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/google-web-toolkit?hl=en.

Reply via email to