Hi, A long-term improvement of the rendering infrastructure certainly appeals to me. If the effort to obtain better performance in mass document production does not degrade the performance in other use cases, there is no problem to try and implement it in the code. If the effort may destabilize the code for a longer time, a branch would be appropriate.
The considerations about our goals are more in the area of resource planning. And I have only my own resources to plan with. Regards, Simon On Tue, Nov 27, 2007 at 11:10:55AM +0100, Jeremias Maerki wrote: > Dear Foppies, > > it has come to my attention that not everyone seems to be happy that > some of us are looking into a new design for the intermediate format > which on first glance only helps those who are doing mass document > production. OTOH, these considerations help in a long-term improvement > of our rendering infrastructure. Several features are currently not > adressed: tagged PDF, z-index, accessibility (role, natural language) > etc. But yes, these are mainly side-effects. The main driver is want for > speed. No use pretending otherwise. > > And that doesn't really help towards FOP 1.0. In June 2006, some of you > have listed the features they want done before a FOP 1.0: > http://wiki.apache.org/xmlgraphics-fop/ReleasePlanning > What remains is: > - integer keep values > - changing available IPD between pages (Simon is working on that) > - page-number-citation improvements > > Only the collapsing border model is now there and I think that > fo:wrapper is mostly ok now (haven't checked lately). I still consider > none of these points a requirement for a 1.0 release, not for a software > so widely adopted. A 1.0 is just a signal which is listened for by some > but not necessarily for the reasons that caused the above list. But even > so, once we're publishing 1.0 I can already hear people saying: "what? > It took them 9 years to come up with version 1.0? And it still only > implements a subset of FO?" Whether we implement all of the above > features or not, the reaction will be more or less the same. Ok, > maybe I'm just venting my frustration again. So, back to business: > > Is anyone else against Chris, Adrian and I going after a new > intermediate format approach in FOP Trunk? If necessary and if it helps, > we can do that in a branch. We do have a certain dilemma here: the ideal > world looks different than the business world. Of course, we're all > trying to pay attention to the long-term goals of the project but at > times shorter term issues can become more important to those who are > sponsoring my efforts, for example. With the amount of work piling up on > my desk from my clients it's sometimes difficult to retain some > motivation and energy to do much else for the project. Bigger things > anyway. But on the other hand, much of what the sometimes silent > sponsors contributed in the background brought FOP where it is today. > > I think it would make sense if we revised our plan towards FOP 1.0. We > should make sure we are all on the same track again. We neglected that a > bit in the last 18 months. After all, the project charter says we should > have an up-to-date project plan. > > WDYT? > > Jeremias Maerki > -- Simon Pepping home page: http://www.leverkruid.eu