Community mailing list archives
Re: Should Odoo keep the old API for v10?by
Thanks for your comments Graeme Nhomar and Ermin. You are right that customers pay and modules get created or migrated. The fact that this is not happening on OCA also shows that most of those non migrated modules on OCA seem to be of no use for customers OR and this is for sure the case - they have no need to convert them as they know that your agencies are successfully blocking the move to Python 3 since longer time!
If Odoo stays 2.7 this will continue for decades like that. Best example was the move from PHP4 etc to PHP 5,2. Unbelievable but still we get sites rinning on it but most sites have been already been migrated to 5,2+ and the migration jobs increased tremendously after 5.2 and no more backward compatibility of PHP products have been announced.
The same will happen after Odoo will release their first Odoo Core Dev Vetsion for contributions to the public.
We wont invest in further 2.7 stuff and many others do exactly the same. Instead we recommend customers also other solutions so they can make their own decisions.
If you want to have funding for migrations simply support actively the move to Python 3 and stop blocking it.
Support the request of many here on the lust which ask since years for a roadmap milestones and sprint release dates of an Odoo version running on Python3.
We might talk only for us and some others who agree but we recommend customers to test out also other products which run already on Python3 or which run mith Maria DB / MySQL as this is anither culprit often wanted from customers as they don't like to switch to a hoster which provides Postgres.
nor they are interested to move in the clouds.
and most of all they are customers which are happy with an easy to handle and fast and nit with a blown up Odoo. Customizations are usually done since nine in LGPL and therefore only available for several customers - that is the bad thing of constant license mives of Odoo.
and as you already verified most OCA modules seem to be complete useless for companies as they don't use them anymore in 9 like they perhaps have used them in their previous versions. customers checkout other products evaluate prices evaluate the future of the product etc and in those parts Odoo with 2.7 is simply on death row woth very clear end as most customers which contacted us recently pointed out due to the clear statements of Ubuntu concerning their focus - and that focus is Python3 and 2.7 is simply out even it us still part of the package but without further improvements.
When will you start doing your homework. Update all your site to Odoo 9 before you even dare to discuss a move to Version 10.
It is you who block progress as it seems to be only your customers which need those OCA modules entirely, but it also seems that you are not capable to make them interested for a migration to the new API and exactly this will also continue after Version 10 got released with new API. you will still keep them running their already deprecated 5,6,7 systems and for sure you will start again exactly the same blocking discussion when than Version 11 gets discussed to be 3 or 2.7. Again you will try to keep 2.7 as otherwise you won't be capable to satisfy your customers as they simply would either move to other products or even worse for you to other agencies who kept up with the Odoo progress since years.
They will find those agencies easily on OCA as those are the developers who already upgraded their stuff to the new API.
Odoo needs NOT a discussion concerning the new API as it already exists. But we need a timeframe, roadmap with milestones and release date for a Odoo 10 version running ob Python 3 as soon as possible
Nobody who promotes the migration to 3 needs this Python 3 in final release tomorrow as it has been pointed out already many times. but we need a dv version of the Odoo Core at least which is running on Python 3 to start developping Python3 modules and convert existing new API modules even from OCA which are really needed and usefull to Python 3.
Don't blow up Odoo with your deprecated stuff which your customers don't want to get migrated anymore.
Do what most want since Odoo came out with its website builder. They want a fast reliable and easy to handle Odoo. nothing more
they want psckages for certain verticals and no more those ultra mega odoo stuff which you run still without the need of a midern website builder interface.
I understand that you only need the intranet stuff- thats fine but no more the vision only of people who like Odoo. Most new customers are interested because of the easy way of its website builder together with the standard backend features and of course its shop and this will be the future of Odoo!
if you don't like it stay woth what you have or follow your own advice and move or create a fork for your 2.7 or old API.
We need Python 3!
What is the roadmap?
What is the timeframe?
odoo customers like to hear about exactly that and I hope Fabienne will respond soon to this!
LisAndi Co. Ltd.
15/21 M.2 Viset Road
Rawai, Muang, Phuket
+66 (0)81 606 3378
+66 (0)76 38 33 08
+49 (0)711 50 88788 50 -2
+49 (0)711 50 88788 59 Fax
On May 12, 2016, at 4:57, Graeme Gellatly <firstname.lastname@example.org> wrote:
Lol very frustrating for customers to see that slow process. Here is a clue, OCA modules are migrated when a customer(s) pay for it. If no one is willing to fund the migration, ergo, no one wants it migrated badly enough at that time so the speed or otherwise of migration is irrelevant. Free is not free.
Generally it is larger integrators and developers with larger customers funding those who don't want to pay for anything. Their migration timelines are naturally slower than the little free bunch who want everything for nothing yesterday. If there are modules you need migrated, either fund it or do it. Complaining is pointless.On Thu, May 12, 2016 at 3:27 AM, Andi <email@example.com> wrote:Thanks for your last statement nhomar, it shows exactly the ignorance and blocking power for the future of Odoo which seems to be indeed an OCA problem. Simply have a look how many modules have been converted to run in Odoo 9 until now. It is very frustrating for customers to see that very very slow process and you indeed start discussing Version 10.Sorry this is actually the suicide you are talking about. Yes we are developers but some which producer consumer oriented and easy to handle Products.Have a look to the survey about what is wished for the future of Odoo - sorry don't have the tweet link here please check the odoo tweets. Result afaik was that it should be FAST and EASY. In other words it is perhaps just the contrary of what you like to have.You at Vauxoo have perhaps Mega Instalations might be but how old are they. Perhaps not even updated to 9 i guess.So what the heck the discussion about a 100% new API discussion if most agencies seem not even casble to handle quite simple updates to Version 9 which holds the new API.With Version 9 customers have a stable and meanwhile well tested product. Version 10 on 2.7 but new API would be not much different beside the fact that the Enterprise MRP Version and perhaps other Enterprise only features would be added.On our customers side and as I know from others too it is no big interest to feat Enterprise only stuff.of course an Odoo 10 with Python 3 would have the same 100% only new API. As Odoo already stated THEIR modules are already focussed and ready for version3.So what do you want? get even further behind and still feeding those who were incapable to adapt even smaller changes in the past and still most of their so called mega sites are running on OLD API stuff ????We and I know lots of others have not even more one site or mega installation running on the OLD API? We worked hard and customer friendly to get things done constantly. We call it consumer friendly.In other words the discussion new or old API is complete nonsense for us, especially as long as 90% of OCA or other modules from companies which post in this Thread to keep 2.7 are not even running in Version 9 - properly.Instead of focussing and pushing a 2,7 Odoo 10 version with new API it would be adviced to KEEP Version 9 as long as possible and integrate here 2.7 improvements and most important get as FIRST the OCA modules updated and also in your case all your Vauxoo Modules working properly in Version 9!!!It is complete suicide right now to discuss a new 2.7 Version as this would delay the OCA conversion process which just has been started many months ago for many more years.Years Odoo customers will get even more behind. than as they are currently already.With your perspective your customers can calculate with a stable version 10 Odoo on 2.7 and most OCA and Vauxoo modules running about 2020!!! not earlier!If they would use Version 9 they could achieve exactly the same hopefully already in 2017 latest in 2018 ( depending of course also on how Vauxoo would convert their modules to the new API - which is your own problem which you need to solve internally )Now take the case of an Odoo 10 running on Python 3. It would be well tested and ready also in 2020 perhaps even much earlier as many more developers would be interested to an 3.x Odoo 10 and even better many more investors could be aggregated to sponser 3.0 developments.For a 2.7 Odoo Version they would not even invest a brief view to a concept. Pleae stop blocking initiatives and ideas concerning Python 3. And stop calling things you don't like Feeding the troll!Already many customers follow exactly this Thread here as it helps them to evaluate if Odoo would be the right choice for the future.Like mister Rotterman already asked again: What is the roadmap and what is the timeframe!!! for Odoo running on Python Version 3.x - this is actually the question most of them are intetested sice years already - actually since Tryton runs on Version3 very successful.And you can be assured that nobody arguing in favor of a 100% new API and Python 3.x Odoo with out compatibility breaks will leave the Thread or Odoo itself. They love Odoo and will stay with Odoo but they really don't like those throwing towards those who bring up critical stuff and which did their jobs propperly with updating all their customer sites regularly and have no need in a 100% Odoo on 2.7 as they run since years on the new API already.This is the right Thread for all those who are interested in the future of Odoo and this Future can ONLY be an Odoo 9 with new API and some backward stuff but all on 2.7 and more and more OCA modules getting converted to 9 - even it is a very very slow process.and in parallel a 100% Only new API and on Python 3 running Odoo 10.Odoo itself seems to be ready for the switch many others too but right now missing is the Odoo Core on Python3as soon as even a develoment version of that core would be released to public for further contribution and development you would experience the real power of a switch to Python 3.You will have lits of developers developing only for Python 3 and of course with the new API.You will have much more tests running than ever in 2.7 as a Switch to Python 3 is a huge motivation for most developers I was talking the last mobths. Ok they are from India, Pakistan, Afghanistan, Iran, .. in short not from the western Europe or South America. but here the interest seems to be immense to get things done in Python3, To build customer friendly speciaized Odoo Packages and more and none of them is interested to invest any more time in old API conversion stuff to 2.7. As said their sites run already on new APIs.this fact is for sure also the reason for the huge result in having a 100% new API odoo. and only 15% don't like the elimination of the Old APi perhaps because they did not do their jobs properly in the past and are now standing for a huge problem while they still run Odoo 5, 6 and 7 sites. Simply forget about them and make a CUT - allow Odoo to be future proove and I hope Odoo will post asap a timeframe and exact roadmap with milestones, license changes which might occure again and of course for an Odoo 10 running on Python 3 and any later Version running on Python 3.For your own sites I would recommend starting to do your homework properly ( customer service ) and I recommend to you Odoo 9 as it is stable and will have soon more and more modules converted and it is running the new aPi and stil allows some old aPI stuff.Start preparing your customers to be ready and have fun with a Python 3 Odoo10.Perhaps Fabien should have formulated the survey different.Keeping Old API and Python 2.7Enhance Odoo stuff in Version 9 making it the Last version running on Python 2.7 but increae the support until 2020.Drop Old API and keep 2.7 in Odoo 10Drop Old API and use Python 3 in Odoo 10additional it would be interesting to know when people would like to see Odoo10 running on Python 3 should be released.Odoo 10 with Python 3 shoud be available in:6 months12 months18 months...and a Odoo 10 on Python 3 core version should be avaiabe for development and contributiion in3 months6 months9 months...This would bring up a much better overview and results about what is wished on developer site.Perhaps differentiate the surveyI am a developerI am a customerI am not a developer but member of an agency i.e. sales or outsoourcing project manager etc with no dev know how but know how about customer needs and skills to improve sales.Have a nice evening!On May 11, 2016, at 10:41, Nhomar Hernandez <firstname.lastname@example.org> wrote:On Tue, May 10, 2016 at 9:42 PM, Andi <email@example.com> wrote:Please don't throw to many stones in the roadmap of Odoo for its future. Customers want actually see a modern product and are asking more and mire for a Python 3 Odoo Version!
Your perspective is the perspective of "Product Consumer with Critic" not the "Software Editor" when you manage our volume of modules + OCA the perpective is totally diffferent.It is not a matter of that: We will install python3 we will run a magical script that nobody knows how the hel it was done and PUM! all is migrated!!!!And then, everybody will deploy that in seconds.!The Statement you use here is not true 100% of times, even the fast odoo's evolution is a great problems in the perpective for long term, I need to discuss with our customers sometimes (almost all our implementations are middle size) and say that the deployment is in the way you mention for the reasons you mentioned is absolutly wrong, (even If in some cases it is not).There are several types of customers all of them has its own kind of "Early adopter methods".You can be partially right, but from an Editor Perspective you are absollutly wrong, and push Odoo to be even more dast of what actually is (which is bad enough rfom some point of views and good enought from other PoVs until now) IMHO is suicide EOF..Regards.