saz: briefly talking about the F2F meeting
<shadi> http://www.w3.org/WAI/ER/2005/03/scenarios.html
met with a few other groups
<shadi> http://www.w3.org/WAI/ER/2005/03/requirements.html
<shadi> http://www.w3.org/WAI/ER/2005/03/issues.html
saz: rough requirements
... some known issues
... need to finish requirements in this quarter
... question is the deliverables
... are we going to have one document, or split it in two documents, like the
RDF
... it would be double work (at least)
... two docs to edit, editing is tough work
... we should have more volunteers for editing
ak: we would it necessarily be double work
saz: two monsters in the requirements
... 1. locating subject of test result
2. persistence of test results
saz: describing these requirements
... e.g. 1. where in the table does the fault exist
... some sort of XPointer, XPath locating the fault in the page
... 2. persistence e.g. a pointer that says if the test is still valid
... has the page been changed?
... important when u have dynamically generated pages
... have to look how far we want to go into this issue
ma: we have to develop some technique for
locating elements in the HTML code
... are they both related
saz: yes, persistence needs to be based on some
mechanism for locating the fault in the page
... often the page for dynamically generated pages, when u cannot compare
anymore
jk: is persistence really manageable?
... earl date element in the subject path, is not clear taht a test is for a
document at a specific time
... document may change later
... i don't think there is a persistence
ma: the order of elements, some test could be
the same other change
... i agree with jk, we can do something, not much
cr: we need something to have some sort of
persistence
... e.g. if u change an image you may still need a longdesc
saz: some of this are already in the
requirements doc
... context sensitive info,
...
... i agree with jk, ma, we cannot have absolute persistence
... reevaluation will need to happen if we change the page
... some of this eval is expensive, results not easy to get, can we do more
than the date
jk: maybe we can have some option of module
... persistence of pointer should not be a requiremen
cr: ... yes i agree
... maybe just one requirement (XPoint)
ma: we can implement the location, it is important, we can impl it we persistence in mind
saz: i agree, let's still leave it open for
now
... let's dig up previous discus, resources
... like QA, SW people, resources
... check what options we have
... what do u think?
everybody agrees
saz: EOWG resource "Evaluation of web sites..."
<shadi> http://www.w3.org/WAI/ER/existingtools.html
saz: existing tools page, which we ERT
maintain
... this page has become long and unstructured
... difficult to find what u need
<shadi> http://www.w3.org/WAI/ER/tools/tools.html
saz: we can get this list sortable, according
to list of criteria
... i tried to find as much info as I could - links for sorting this page
according to database
gb: regarding the location - do u have some heuristics that we could test?
saz: there are solutions by the previous group
2-3 years ago
... somebody should research in these resources
cr: we have a strategy with pointers, can i post that?
saz: yes, thank you
... back to -- sort the table of tools in different ways
... during F2F we did some basic requirements on this
<shadi> http://www.w3.org/WAI/ER/2005/03/tools.html
saz: we need to get stuff going
cr: this can become a table
saz: so far info is unstructured, but now we
are spliting in certain fields
... it becomes difficult to fit this info in the table
... tool characteristics (last section), characteristics to sort by
... a lot of information there, difficult to lay it out
... anyone interested in this?
gb: interested
saz: now we only talk about sorting, but also
talked about filtering, search facility etc.
... sometimes we wont have all the information
... because developers may have not submitted this info
... can put it as a check whether you want to filter a sort
... maybe have a separate page, instead of having a long listing - like in
search engines for example
ma: are you considering other characteristics
saz: yes, for example platform is a characteristic that is not there yet
ma: one charact, if tool is free and if it available in the internetm, which URL
saz: currently it goes to the URL of tool, but type of licensing is not provide
<shadi> http://www.w3.org/WAI/ER/2005/03/scenarios.html
saz: existing scenarios, I have put links of
scenarios , there are more in the mailing list
... we want to compile all the different scenarios into one big catalogue
... we need unique examples, not the same use cases
... then we can start working on EARL, the primer, etc
... we need editors for EARL requirements doc
<shadi> http://www.w3.org/WAI/ER/2005/03/requirements.html
saz: if anybody has experience on writing req docs
<shadi> http://www.w3.org/WAI/ER/2005/03/issues.html
saz: researching known issues, there is more
than in the document
... need to search in previous mail archives
... submit more issues in the mailing list
<AWK> Shadi, I'm happy to help with work, but my schedule for the next few weeks is uncertain since I'm expecting a child just over a week from now.
saz: before distibution of tasks, are there any
clarifications that need to be done?
... congrotulations, for baby
jk: we have to write some introductory stuff
... can write for the primer document
saz: do u want to work on scenarios for the primer
jk: yes we can do that
saz: we will see how to coordinate this - whatever u produce send it in the mailing list
jk: ok
<shadi> ACTION: JK will start working on EARL Scenarios [recorded in http://www.w3.org/2005/03/08-er-minutes.html#action01]
saz: chris is it interested to u the ???
cr: sure
saz: any findings you do point in the mailing list so we know it's done
<scribe> ACTION: Chris RidPATH and Gabriele will be working on locating results and the persistence of the results [recorded in http://www.w3.org/2005/03/08-er-minutes.html#action02]
saz: anybody for requir docs
... I will take up that myself
<scribe> ACTION: Shadi and Myriam Arrue take up in editing the requirements document [recorded in http://www.w3.org/2005/03/08-er-minutes.html#action03]
saz: awk would you be willing for some editing?
awk: yes
saz: locating and persistence are the big
issues
... we should all pay special attention to that
... it adds a lot to the current spec
... goodbye, no call for next week
... next meeting 22 March