Next steps

[Jochen Wiedmann <jo...@gmail.com>]

Hi,

looking forward, the most important steps for the VXQuery project seem to me:

- Create some documentation, at least for the following use cases:

  * Use VXQuery to select data out of Java beans (JAXB classes?)
  * Use VXQuery to select data out of DOM documents

And then, ASAP, cut  a first release, no matter the projects then current state.

That way it should be possible to gain a little bit more interest by
others. I'd strongly recommend to think in that direction.

Jochen



-- 
In other words: what could be seen as a socially debilitating failure
of character can certainly work to your advantage too. (Linus
Torvalds, but the use in the signature tells something about me as
well.)

Re: Next steps

[Cezar Andrei <ce...@gmail.com>]

IMHO, some documentation on what and how it works today and a release is
more important than another percentage of passing tests. A release means
the project is serious and it passes the Apache release requirements, i.e.
licensing etc and also is a good exercise to go through for new projects
like VXQuery.

Cezar

On Wed, Aug 8, 2012 at 5:53 PM, Till Westmann <ti...@westmann.org> wrote:

> Hi Jochen,
>
> I agree with both high level points:
> 1) creating more (some) documentation and
> 2) creating a release.
>
> However, I think that these 2 use cases won't be among the things we can
> document.
> As we had decided in February, we changed the focus of VXQuery from
> "running on different data models" to "running on large amounts of data in
> a clustered environment". (Unfortunately we only updated the project page
> yesterday, so the fact that we actually did change the focus was somewhat
> hidden ...)
> For this new focus, we anticipate that VXQuery will more or less
> standalone and that data will come in from external sources (mostly files).
> So it is unlikely, that data will come in as JAXB classes or DOM instances.
>
> But we really need to document what works today (e.g. approx. 4000 of the
> 19000 tests of the XQuery test suite) and how it can be used.
> Also, I think that we are only a few weeks away from enabling another
> large chunk of the test suite and from increasing the usefulness of what we
> have. And I think that it would be a good point in time for our first
> release when we get there.
>
> Does this make sense?
>
> Cheers,
> Till
>
> On Aug 6, 2012, at 2:50 AM, Jochen Wiedmann wrote:
>
> > Hi,
> >
> > looking forward, the most important steps for the VXQuery project seem
> to me:
> >
> > - Create some documentation, at least for the following use cases:
> >
> >  * Use VXQuery to select data out of Java beans (JAXB classes?)
> >  * Use VXQuery to select data out of DOM documents
> >
> > And then, ASAP, cut  a first release, no matter the projects then
> current state.
> >
> > That way it should be possible to gain a little bit more interest by
> > others. I'd strongly recommend to think in that direction.
> >
> > Jochen
> >
> >
> >
> > --
> > In other words: what could be seen as a socially debilitating failure
> > of character can certainly work to your advantage too. (Linus
> > Torvalds, but the use in the signature tells something about me as
> > well.)
>
>

Re: Next steps

[Vinayak Borkar <vb...@yahoo.com>]

> But we really need to document what works today (e.g. approx. 4000 of the 19000 tests of the XQuery test suite) and how it can be used.

As of last night, we are at 5500 passing tests!

Re: Next steps

[Till Westmann <ti...@westmann.org>]

Hi Jochen,

I agree with both high level points: 
1) creating more (some) documentation and 
2) creating a release. 

However, I think that these 2 use cases won't be among the things we can document.
As we had decided in February, we changed the focus of VXQuery from "running on different data models" to "running on large amounts of data in a clustered environment". (Unfortunately we only updated the project page yesterday, so the fact that we actually did change the focus was somewhat hidden ...)
For this new focus, we anticipate that VXQuery will more or less standalone and that data will come in from external sources (mostly files). So it is unlikely, that data will come in as JAXB classes or DOM instances.

But we really need to document what works today (e.g. approx. 4000 of the 19000 tests of the XQuery test suite) and how it can be used. 
Also, I think that we are only a few weeks away from enabling another large chunk of the test suite and from increasing the usefulness of what we have. And I think that it would be a good point in time for our first release when we get there.

Does this make sense?

Cheers,
Till

On Aug 6, 2012, at 2:50 AM, Jochen Wiedmann wrote:

> Hi,
> 
> looking forward, the most important steps for the VXQuery project seem to me:
> 
> - Create some documentation, at least for the following use cases:
> 
>  * Use VXQuery to select data out of Java beans (JAXB classes?)
>  * Use VXQuery to select data out of DOM documents
> 
> And then, ASAP, cut  a first release, no matter the projects then current state.
> 
> That way it should be possible to gain a little bit more interest by
> others. I'd strongly recommend to think in that direction.
> 
> Jochen
> 
> 
> 
> -- 
> In other words: what could be seen as a socially debilitating failure
> of character can certainly work to your advantage too. (Linus
> Torvalds, but the use in the signature tells something about me as
> well.)