On Thu, Jun 3, 2010 at 8:17 PM, Pierre Joye <pierre....@gmail.com> wrote:
> On Thu, Jun 3, 2010 at 7:58 PM, Ferenc Kovacs <i...@tyrael.hu> wrote: > > > So? > > Now we shouldn't try to improve things because they aren't perfect? > > Because it is a very bad thing, by design and we will have to keep it > during years. I think this exact point was clear enough. > > >> We need to stop this and start fixing it today. > >> > > > > Go ahed. > > Do you have any timeline for it? > > The idea of discussing something is not about defining a dead line or > ask to provide patches/code. This discussion has a long history, and it would be better to actualy do something about the problem, because it seems that it won't going to disappear by itself. I don't want to be rude, but the PDO v2 spec was last modified in 2008.01.23. the problems with the current PDO architechture are the same that Wez pointed out in his FAQ and spec. "PDO was built around the ideal of being able to provide a common data access paradigm across a variety of data providers. The goal was to provide the lowest common denominator of features using a common API, while still being able to provide access to database/vendor specific features." With the current model, there are some db specific features, like the unbuffered query attribute for MySQL, but we are lacking and rejecting others, because some point in the future somebody will implement the perfect solution. It is about constructive, mature and > quiet discussions. There is no need to rush but to really discuss and > plan what we need (a bit). > > Agree, but currently I don't see any point in waiting, if we don't do anything else beside waiting. > I also have to say that I am very happy to see Jonathan (doctrine > lead) finally posting on internals about PDO, or about databases need > in general. I hope we will get more developers like him being active > in discussions. Their feedback will be very valuable to make the right > design decisions for what the modern applications and frameworks out > there can and will use. > > +1 ps: Maybe it's not obvious, but I __really__ would like to see the PDO moving forward. Tyrael
