Daniel Egger wrote: > > I think you need to > > consider the experience that menthos has with this situation. If we > > consider what we might end up with in the future (many more tips, more > > complicated tips, more languages), it makes sense to plan po right now. > > I'm never planning for now but always for the future. I think we have the problem right here. Although future concept "if we did this from scratch" ideas are always interesting, they are hardly ever a solution to the situation today, simply because they are by definition not implemented, and often hardly have any compability with the existing software. So I think you should devote your resources to implement this software, rather than trying to enforce the use of the not-written-yet software today. It sure sounds backwards to me. > gettext and po > files are a dead end for modular applications because they only behave > well for monolithic and small applications; both of which GIMP > definitely isn't and for sure even less will be in the future. Evolution certainly isn't monolithic nor small, and yet it has scaled well to almost 3000 messages as of today. > I like the idea of the XML tip-files with the xml-i18n-tools which will > transform between XML and .po for now because it's a good idea for the > future and as soon as the translators see the deficiencies of .po > and/or new good tools for XML files we can easily ditch them and > go for the right thing(TM). I sure agree with you here, but I'm fairly confident in that there will never any "ditching" until said alternative tools are a reality and tested in practice, and have extensive support. Christian
