[daisy] regarding the DB2 issues -- @Aaron

David Goodenough david.goodenough at linkchoose.co.uk
Mon Jan 22 07:26:47 CST 2007


On Monday 22 January 2007 12:26, Steven Noels wrote:
> On 22 Jan 2007, at 12:59, David Goodenough wrote:
> > On Monday 22 January 2007 11:34, Steven Noels wrote:
> >> I'm very much open to any kind of debate about how we could resolve
> >> this situation, and whether we are erroneously doing MySQL-specific
> >> stuff. We don't have the resources however to continuously develop
> >> and test Daisy against other databases than MySQL at the moment.
> >> Committing your patch won't change that.
> >
> > Well that is a shame.  I was building a system around Daisy and
> > Postgresql, but in light of your statement I will have to reconsider.
>
> In the light that no-one forced you to take a look at Daisy, I'll
> wisely ignore the flamebait. ;)
It was not intended as flamebait, if you notice I started with "it is
a shame", and that is how I intended it to be read.  I also said
reconsider, not abandon!
>
> > Being Database agnostic is rather like being web browser agnostic.
> > You
> > can take the view that the majority of your users can be satisfied by
> > using one particular flavour, but you leave yourself open to the kind
> > of problem that is now hitting those who relied on IE6 specifics and
> > are finding that IE7 had taken rather more notice of the standards
> > and suddenly their web apps stop working (or at least stop doing what
> > was expected of them).
> >
> > It is a shame that there is no equivalent of the w3c HTML checker for
> > SQL (or at least none that I have come across) which would obviously
> > make a developers life much easier.  However there are a number of
> > people out here who can test what you produce against other DBs, and
> > that will catch most of the problems and produce a more future proof
> > product.  It does seem a shame not to take advantage of this.
>
> David, I don't see where I state that we won't support other
> databases. Quite the contrary.
Your statement above reads "We don't have the resources however to 
continuously develop and test Daisy against other databases than MySQL at the 
moment".  That is what I was reacting to.  

My comment was to the effect that it is better to start from a basis of SQL
that will run anywhere, rather than having to go back and fix things for
other DBs which follow the rules more closely later.  Design it in day one,
not backfit.
>
> I see myself stating that cross-the-board support of relevant
> databases is a worthy goal, however we should address that
> comprehensively. Sending in a big patch again 1.5 and expecting us to
> take that patch and port it to trunk isn't exactly "comprehensive" in
> my book.
I was not discussing the merits or otherwise of a particular patch.  It 
was the approach (MySQL first then fix it for others) that I was questioning
(and no more than questioning either).

David
>
> My mail was a honest and friendly plea for patience and smaller, more
> regular contributions. Otherwise, Daisy will trample on its own feet
> from code-rot and half-finished features. *That* would be a real
> disservice for Daisy's users, no?
>
> </Steven>


More information about the daisy mailing list