Generating tests [was: Re: Review SPARQL Query?]

On 1/16/2011 12:01 PM, Andy Seaborne wrote:
> Birte,
> I think it's better to wait until the doc is finished. If, however, it
> fits better with your time, then there are just one or two specific
> points that need work (+ general editorial) so reviewing the rest could
> work.
> Overall:
> What we are missing most is tests. In the test work so far, we have
> ended up sometimes identifying differing expectations of the design with
> the WG. As LC is suppose to be design-stable and changes expensive, I
> think it's important to have test coverage before LC to make sure the WG
> is in proper agreement on the details.

Hi Andy (and everyone else),

Do you have a suggestion for how we can generate tests? Perhaps we ought 
to enumerate all of the SPARQL 1.1 features (beginning with query), and 
then split them up amongst all of us on the WG on Tuesday -- if each 
person came up with a few tests per feature, it would probably get us a 
long way to where we want to get.

What do people think?


> Andy
> On 13/01/11 15:24, Steve Harris wrote:
>> That's correct from ,my point of view.
>> - Steve
>> On 2011-01-13, at 11:07, Birte Glimm wrote:
>>> Andy, Steve,
>>> just to confirm that I did understand this correctly in the last
>>> teleconf. I wait with the review until you tell me that the doc is
>>> ready, right?
>>> Cheers,
>>> Birte
>>> --
>>> Dr. Birte Glimm, Room 309
>>> Computing Laboratory
>>> Parks Road
>>> Oxford
>>> OX1 3QD
>>> United Kingdom
>>> +44 (0)1865 283520

Received on Sunday, 16 January 2011 21:33:29 UTC