RE: Explicit vs Implicit URIs (aka "blanket statements or not")

 
> Carlos Iglesias wrote:
> > Why *not this one*?
> > 
> > <blockquote> F04: EARL 1.0 will support aggregation of test results 
> > according to different criteria (for example with respect to the
> > subject) </blockquote>
> > 
> > Aggregation according to the subject --> Aggregated subjects e.g.
> > Domain, Subdomain, etc...
> 
> When you give a definition of what is your understanding on 
> the topic, but the DNS definitions are not clear enough on this, i.e.:
> - A domain includes all its subdomains?
> - We need to worry about top-level-domains, country code 
> top-level domains, etc.?
> 
> In any case, to aggregate, is not the same as to compact.


In any case, this seem to be more a linguistic discussion :o)

 
> > ...
> > No, they're not included because they are subdomains. This 
> is why the 
> > proposal [1] includes a "Domain" Scope and a "Subdomain" 
> Scope. If you 
> > define a "Domain" Scope just example.org is included, if 
> you define a 
> > "Subdomain" Scope then images.example.org and www-inter.example.org 
> > are also included.
> 
> That is your interpretation of domain. However, [1] is not 
> that categoric on whether a subdomain is not part of a 
> domain. For others, it might be so. 


Who are the others? Amenabar's film? ;o) --> Spanish joke
Seriously, any reference?

Any way, we could name them Host and Subhost if you prefer.


> By the way,
> - example.org is a domain, and a subdomain of the TLD .org
> - www-inter.example.org is also a domain, and a subdomain of 
> example.org
> - example.co.uk is a domain and a subdomain of the ccTLD co.uk
> 
> Thus your definition does not hold.


Thus my definition holds

Domain --> example.org

images.example.org, www-inter.example.org and example.co.uk are not included

SubDomain --> example.org

images.example.org and www-inter.example.org are included. Example.co.uk is not

Subdomain --> .org

example.org, images.example.org and www-inter.example.org are included. Example.co.uk is not

Anyway, I'm not trying to do an oficial definition of Domains and Subdomains, I'm just noting that they're usual concepts in web validation and that we should incorporate them to EARL (I want this semantics). We could give these concepts whatever name and definition you want.


> > ...
> > You could just record the variables you were using during 
> the crawler.
> 
> And who does guarantee that your crawler gets all resources? 


No, it just give you the necessary conditions to reproduce it if you need to do so.


> We have tested several crawlers in big portals (10,000+ 
> resources) and you get different results with all them (not 
> to mention they run out of memory).


Mainly due to different configuration parameters, timeouts, etc.


> That is why we developed our own anyways ;-)


That obviously get different results too.

 
> >>> EXAMPLES: * "example.org/page.html passes WCAG 1.0 CP 1.1" - the 
> >>> "usual" way of using EARL to record test results will
> >> still be
> >>> available * "example.org/page.html passes WCAG 1.0 Level A" - a 
> >>> contraction of individual checkpoints is possible due
> >> to the WCAG
> >>> hierarchy
> >> It depends. There is still an open discussion on TestCase and 
> >> TestRequirements. Until that discussion is closed, some of 
> the above 
> >> examples might not be valid.
> > 
> > 
> > I can't see how the discussion on TestCases and TestRequirements 
> > affects to the examples above. Could you elaborate?
> 
> Well, simply is EARL reporting *test results*, which lead to 
> some compliance with some given requirement (e.g., WCAG 2.0), 
> or directly compliance with the requirement? This issue is 
> still open in my understanding.


The requirement is Web Accessibility
The Test Cases could be WCAG 1.0, WCAG 2.0, Ley Stanca, WCAG 1.0 Level A, Checkpoint 1.1, Checkpoint 3.2...
EARL reports test result of whatever of the previous Test Cases.


 
> > I think we shouldn't focus on covering conformance claims 
> needs, but I 
> > don't see any problem if, as a result of our work in other 
> areas, EARL 
> > is also useful to express conformance claims.
> 
> I do see a problem, though ;-) namely, the misuse of EARL.


If you see a conformance claim as a high level report there is no misuse at all.

Regards

Carlos too.

 
--------------------------------------

Carlos Iglesias

CTIC Foundation
Science and Technology Park of Gijón
33203 - Gijón, Asturias, Spain 

phone: +34 984291212
fax: +34 984390612
email: carlos.iglesias@fundacionctic.org
URL: http://www.fundacionctic.org

Received on Wednesday, 10 May 2006 10:52:03 UTC