I just want to emphasis the kind of customization you are speaking about (add or remove custom fields from standard views) is just very straightforward and very effective in the proven OpenERP module system on the contrary, see explanations here:
http://doc.openerp.com/developer/2_6_views_events/views/view_inheritence.html

Any view XML structure can indeed be extended at any point by a custom module, respecting dependency order and all…
Just the difference between a hyped immature system and a mature one…

Raphaël Valyi

Imagina que un cliente no quiere la mitad de los campos de la pestaña producto. Como la pestaña y la tabla son de core, lo suyo, parece, es realizar la modificación de que tal campo no se vea en la pestaña asociada a tu módulo, no? Es decir, que con tu módulo, el campo X de la tabla Y ya no se muestra –> adivina el mensaje de error… DBPrefix… es decir, que te tienes que crear la ventana entera para quitar un campo estándar de una pestaña?!?!

O editar core… con el lío de las actualizaciones que comentas… así, algo tan tonto como dejar de mostrar un campo en una pestaña, parece, se convierte en replica la ventana para poder hacerlo…

Salu2

May be one more thing about OpenERP: I think the quickest way to connect to the Spannish community is through this site:
http://www.openerpsite.com/
They did a meeting a few weeks ago:
http://www.openerpsite.com/conferencias-openerp/469.html
I just discovered that they also commented one of your post:
http://www.openerpsite.com/post-sobre-openbravo-su-vision-del-sw-libre-y-su-relacion-con-la-comunidad/503.html
And finally they are organizing a formation next week:
http://www.openerpsite.com/cursos-presenciales-openerp-a-nivel-usuario-en-catalunya-bilbao-y-madrid-juniojulio-2009/486.html

I think that you can transpose a good 80%% of your business knowledge to an other ERP like OpenERP. Now, that’s true, aside from Python which is arguably one of the simplest languages to grasp, learning the ERP framework and the implementation details of the main module take has a learning curve too, I understand it’s really hard to learn on spare time.

Ok, let us know whatever you do.

Raphael Valyi

my name is Raphael Valyi, I had the opportunity to study Openbravo a bit along with other open source ERP’s. I should say I quickly understood the architectural flaws (inefficient modules in a procedural programming paradigm, non competitive “appliance” offer…) of Openbravo you mention and for that reason decided I’ll never even try to integrate it (chances are you can survive an ERP failure only once) before an other say 2 years if Openbravo finally comes up with something strong (considering their venture capital, they could eventually survive, others did much better with way less, so let’s give them a chance, but let’s not be their collateral victims for now).

Finally, I advise you to have a look at OpenERP instead. Not perfect either, but way better and already competitive for some situations (which already represents a Huge market. More promising also. I’m in touch with Spanish integrators and they are getting better and better instead. In France there are no known Openbravo deployement, while OpenERP deployments are popping all over the place, with a few professionnal integrators too.

I came from the Java camp too but learnt Python for OpenERP. Python itself is only 1 or 2 weeks of learning for an experienced OOP programmer, meaning it’s peanut in the whole ERP learning curve (and will only making you win time instead of fighting against the tons of Openbravo legacy pl/SQL). And if you does it for OpenERP, I believe you will be rewarded much more.

I could highlight a few decisive advantages of OpenERP:
- OpenERP is object oriented all the way. It’s like if the DAL system of Openbravo were already running and really used by the code since 3 years and even better than that (view inheritance, better migration system, BPM engine, interface is component oriented…). Overall I would tell that it’s a solid 3 times quicker to developp a custom feature X in OpenERP than in Compiere clones. Given that all those ERP’s have missing feature, this is something decisive how quick you add those.
- OpenERP has a much wider feature scope than Openbravo, I would say something like 3 times wider. See CRM or projects for instance. But only MRP or chained pickings are more powerful I think.
- OpenERP is more integrated, this is probably due to its OOP structure.
- OpenERP is already modular and modularity in OpenERP is not a myth, it already works for real, overall the maintenance headache you speak about doesn’t exist in OpenERP and module are reused for real.
- OpenERP has no Venture Capital. When I see how wrong gone Compiere and Openbravo and a few other Venture OSS projects of that kind, I think that’s definitely an advantage not owing a big money to rich morons.
- OpenERP is lead by the initial tech guy, not like Openbravo which has been designed by the Compiere guy and then abandoned to non exceptional architects.
- A few months back I annotated the Openbravo 2.50 roadmap to highlight what was already working in OpenERP:
http://docs.google.com/View?docID=ajb639cjf9fb_132ct6kdxgx&revision=_latest

Now yes OpenERP does have a few pitfalls too, let’s list some:
- they are hardly professionnal enough, but they are getting better and really the product is more important, isn’t it? And given that they have no Venture, the main guys tend to overwork quite a bit, for that they should be forgiven, open source is not easy every day.
- they just started over-marketing too, but really far from the Openbravo mystification though. Getting in touch with the real community behind it will tell you what works and what is not 100% OK yet. And you can expect things announced to work finally after some maturation.
- there are some regressions and lack of testing (while more than Openbravo I believe), again they are on the good track and it should be fixed over the year.

Ok, enough said. I’m glad to see someone who really know Openbravo giving detailed explanations about what doesn’t really work yet and where are the marketing mystifications.

If you try OpenERP, you might recycle a large part of what you learnt in something that work already instead, eventually coming back to Openbravo in a few years if it matures enough if you still love it.

Also OpenERP is going to just run over the Java stack too, by the end of the year, see my blog post if you are curious about this:
http://rvalyi.blogspot.com/2009/05/openerp-jython-unladen-swallow-where-do.html

IMHO, Openbravo has an acceptable technology considering the dinosaur ERP’s usually are. Openbravo might even be a bit better than those proprietary.

However, lot’s of those proprietary ERP’s gained a wide and mature feature scope that is still lacking in Openbravo, for lot’s of them it took them decades and hundreds of millions of euros (SAP invest $ 1.5 billions in R&D per year).

Now, when targeting SME’s and SMB’s just like Openbravo is doing, this is just not competitive (yet at least) to spend so much integrator engineering time in building in one shot the missing features, that’s why I didn’t even try it.

Ok, anyway, good luck to you, that’s just an advise in case you face some dead end with OB. I hope OB is going somewhere still because it would be nice to have several open source ERP actors in the future and because a few honest guys invested a lot in its fate too.

So far they seem to keep progressing, but as I said the road is still very long for them (considering their codebase and because they didn’t anything exceptional themselves yet aside from loud marketing).

My take is that they need some 2 years at this progression rate (when the code base will be rewritten in Java over the DAL + mature module system at the very least) to compete against the proprietary offer. But considering that I see no reason why OpenERP wouldn’t lead by a fair margin (they are developping faster), I have some concerns for such an expensive business model.

Keep posting nice blog posts and sharing real open source ERP experiences.

Raphael Valyi
(I’m working for an integrator like you, in France, but will be running on my own in Brasil in less than a month).