RE: wap development

Hi Tim

Good to hear that DELI is useful to you, unfortuantely I am working on
something quite different now so DELI hasn't progressed in a while.

Some other possibilities, which have been mentioned here on this list
before, are

- Use a tool that makes RDF processable in XSLT such as TreeHugger or
RDFTwig
http://rdfweb.org/people/damian/treehugger/
http://rdftwig.sourceforge.net/

- Use a tool for querying RDF such as RDQL, RxPath or Versa
http://www.hpl.hp.com/semweb/doc/tutorial/RDQL/
http://rx4rdf.liminalzone.org/RxPath
http://www-106.ibm.com/developerworks/xml/library/x-think10/index.html

(BTW, for others not familiar with DELI, more information on the DELI work
can be found on my webpage below)

Dr Mark H. Butler
Research Scientist, HP Labs Bristol
http://www-uk.hpl.hp.com/people/marbut 

-----Original Message-----
From: www-di-request@w3.org [mailto:www-di-request@w3.org] On Behalf Of
tim.glover@bt.com
Sent: 08 June 2004 12:46
To: deepix76@yahoo.com; Rotan.Hanrahan@MobileAware.com;
hendry@cs.helsinki.fi; www-di@w3.org
Subject: RE: wap development



Hi,

I am also new to WAP, and CC/PP, and I am looking at the same problem. 

In the Java/XML world, the java DELI application can be used to extract
information from profiles, and this has been integrated into COCOON, an XSLT
architecture for transforming XML from one form to another. 

	 XML->PROFILE->XSLT->WML/HTML/...

Personally, I am prejudiced against cocoon. It seems peverse to me to use
XML as a processing language or storage language, when other much better
tools (like java and RDBMS) exist. And...I could go on. But this may be just
prejudice!

DELI is also used in DICE, which is a java architecture for pipelining
processors. 

	DATA+PROFILE->Process1->Process2...->WML/HTML/...

I like this, but it relies on passing data as a simple string between
processors, which seems a bit crude...

Another soution might be the W3C's CSS Media Queries specification, which
allows parts of an HTML (WML?) file to be made conditional on the Profile
attributes. This looks good - but seems to have little software support at
present, even for HTML. Also, it means that all the data for all devices is
transmitted to the client, which then selects the bits it needs - costly in
bandwidth and difficult to read and debug. 

I am experimenting with a "template" system I have lashed up using DELI,
based on a sort of cross between DICE, cocoon and media queries. For each
html page (wml deck)  there are a number of templates giving the gross
features of language, data selection and layout. The choice of template is
based on the Profile. Within each template, the content and layout can be
"fine tuned" with a syntax similar to media queries. But I am still in the
early stages. 

I would welcome any other ideas.

Tim Glover 

-----Original Message-----
From: www-di-request@w3.org [mailto:www-di-request@w3.org]On Behalf Of Deepa
Srikanth
Sent: 08 June 2004 12:10
To: Rotan Hanrahan; Kai Hendry; www-di@w3.org
Subject: wap development



Hi,
 I am new to the WAP development.I was looking into
the CC/PP,RDF formats for Useragent profiles.
My question is how is the origin server going to
format the WML pages using these RDF information -
screen size,pixels etc.
As a developer how will we incorporate the display of
a text box(to show it small in size) in a mobile
through our wml code?How does these information in RDF
files help the developer and how is it sent out to the
wap gateway?
could any of you explain?
Thanks
Deepa Srikanth

--- Rotan Hanrahan <Rotan.Hanrahan@MobileAware.com>
wrote:
> 
> Overloaded terms are a problem.
> 
> "Proportional", in the context of text, is commonly
> used with respect to the size
> of individual characters, generally with reference
> to their width.
> 
> "Scale" refers to gross adjustments to size with
> respect to an agreed base. Thus
> "100%" would refer to the base size. There is no
> necessary implication that "100%" 
> means legible, though this would be a natural
> assumption.
> 
> The use of "scale" to refer to the ability of a
> systems architecture to grow/shrink
> in size and still perform well, is sufficiently
> different in context from the use of 
> "scale" to refer to text size. I do not think there
> would be any confusion.
> 
> So, I would recommend that you define a base size,
> or at least some concept of a
> base size, that refers to the optimal size for
> visual reading on a device. Declare 
> this base to be 100%, and then use the term "scale"
> to refer to relative sizes 
> expressed in percentage w.r.t. the base.
> 
> Resizing of text is, in my opinion, a shared
> responsibility. The selection of size
> may be an adaptation operation to ensure that the
> relative importance of text is 
> reflected in the visual space allocated to it.
> Changes in size may occur to provide 
> more room for other aspects of the presentation. I
> suggest that the device should 
> also be able to scale the text, but the motivation
> for this capability is the 
> requirement of the user to be able to see the text
> under varying ambient conditions 
> (including visual impairment of the user). It is the
> combined influences of these 
> variables that determines the ultimate text size,
> and I would further suggest that 
> preferences of the end user should take precedence
> in situations where such 
> preferences have been expressed. In the absence of
> such an expression, the origin 
> server, intermediate adaptation and client device
> contribute to the final 
> presentation.
> 
> Personally, I find the inability of a client device
> to scale text (or images)
> locally to be a serious deficiency.
> 
> Finally, if it were possible, I would devise a
> solution whereby my preferred visual
> range could be expressed, and then the active
> components (server, adaptation, 
> client) would cooperate to scale text within this
> range as part of the presentation 
> of content. But who would want to undergo an eye
> exam each time you start using a 
> browser?
> 
> ---Rotan
> 
> Thought: of course, if the device could
> communicate/record my regular local changes
> in text scaling, one could gather statistical
> information from which one could infer 
> my visual range and thus adapt the content
> accordingly. There's nothing new about 
> adaptive feedback systems...
> 
> -----Original Message-----
> From: Kai Hendry [mailto:hendry@cs.helsinki.fi]
> Sent: 08 June 2004 11:38
> To: www-di@w3.org
> Subject: For want of a better word
> 
> 
> 
> I have chosen the word "scale" to describe how
> content is proportionally
> defined to be able to be device independent.
> 
> So if text is sized at 100% it should be very
> readable / legible default setting
> on a desktop browser, as well as a XHTML micro
> browser on a mobile. Hence
> it scales across devices...
> 
> Am I correct for using the word "scale"? It seems to
> be an overloaded
> term. http://dictionary.reference.com/search?q=scale
> 
> It's often referred on the web to whether hardware
> or software can meet
> large amounts of users.
> 
> Here I am asserting that it's the devices browsers responsibility to
> resize the text accordingly, not the content
> authors:
>
http://natalian.org/archives/2004/06/07/can-text-scale/
> 
> W3 resources such as:
> http://www.w3.org/TR/acdi/#diac-3.29
> Certainly doesn't make it any clearer. :/
> 
> 
> 
> Kind regards,
> 
> 
> 
> 
> p.s. Btw http://www.consensus-online.org/ is still
> not working ...
> 



	
		
__________________________________
Do you Yahoo!?
Friends.  Fun.  Try the all-new Yahoo! Messenger.
http://messenger.yahoo.com/

Received on Tuesday, 8 June 2004 08:40:38 UTC