Re: Info on Webinar & Update of Milesones! (Implementor? Please read!)

Dear Yves, All… 

I agree with you, especially with your experience of going through the process. My aim for M2 is inline with your comment "(so not getting to M2, but probably in better shape to get to M3)." We need to make sure, before the Christmas break, that the ball is rolling on implementing the data categories and checking their conformance. So I'm in agreement that M2 is somewhat flexible. Although I'll just put my cards on the table that M3 & M4 are really important. M2 is a marker along the way to M3, one which I'd like most to achieve by the break for the holidays. If you find your implementation for M2 varies in regards of your work-flow then thats great, it shows your making process :) 

I appreciate your feedback and hope this mail is inline with your view. 

Dom. 


 



M1:     15th November 2012 - (Test files reviewed) - Ongoing process. 
M1.5:  4th December 2012 - Webinar on Test Suite conformance / usage. 
M2:     17th December 2012 - Implementors have run at least one global and one local test file, for each of their data categories. One XML and one HTML as well.
M3:     31st January 2013 - Test files completely run.
M4:     8th March 2013 - Data category used in application, spreadsheet updated with conformant / non-conformant implementations.
M5:    15th March 2013 - MLW-LT review - Presentation on test process and 80% of data categories shown to be conformant. 


--
Dominic Jones | Research Assistant 
KDEG, Trinity College Dublin, Ireland.




On 23 Nov 2012, at 21:29, Yves Savourel <ysavourel@enlaso.com> wrote:

> Hi Dom, all,
> 
> Just one minor comment:
> 
>> M2:     17th December 2012 - Implementors have run at least 
>> one global and one local test file, for each of their data 
>> categories. One XML and one HTML as well.
> 
> The milestone above implies that developers implement at least partially all their data categories for M2.
> 
> While I understand this goal tries to help the WG to assess the progress in all areas, this may not be the best or preferred way developers want to implement their support. It's usually far more efficient to implement one data category completely then move to the next one, or all XML support, then HTML, etc. 
> 
> I'm not saying M2 should be changed, but don't be surprised if some don't make that deadline.
> It doesn't actually mean much anyway: one can implement the easy parts first and leave the hard parts for after; that may get you to M2 but is probably less meaningful than another implementer doing 50% of all its data categories completely (so not getting to M2, but probably in better shape to get to M3).
> 
> cheers,
> -yves
> 
> 
> 
> 
> 
> 
>

Received on Monday, 26 November 2012 14:00:56 UTC