@janneman and @dhariwal, I agree with most of your remarks. About code quality: my vision is that Fabien is a bit like an artist: he got the overall architecture extremely right where so many team of engineers failed before and are yet to fail. In one way in invented the Ruby on Rails productivity for the legacy ERP world and got a respectable set reusable modules running upon that.
Fine, now, may be like an artist, he lacked a lot of rigor in the day to day implementation and so did others from Tiny (obscur commit messages like "modifs", variables names such as "todo" come directly from him). Meaning inside the good architecture there are a lot of coding errors, poor respect of conventions (naming, coding, PEP8, English, Terminology...), poor enforcing of test first approach... Because the scope has gone very large upon that basis which is still far from perfect, I hope this is at least acknowledged by Tiny at least toward the community members proposing merges, work and ideas. Acknowledging it and teaming with the community to fix it is an reasonable perspective, but it as to be done. Otherwise OpenERP will reach a quality asymptote were one bug fix lead to one regression and no feature can be added with an acceptable compatibility / without regressions. I've seen so many softwares trapped at that asymptote, I don't want to invest in one that will be trapped here too before even it can turn a mass market product. ------------------------ Raphaël Valyi CEO and OpenERP consultant at http://www.akretion.com -------------------- m2f -------------------- -- http://www.openobject.com/forum/viewtopic.php?p=50481#50481 -------------------- m2f --------------------
_______________________________________________ Tinyerp-users mailing list http://tiny.be/mailman2/listinfo/tinyerp-users
