Choosing Tryton over OpenERP

[Gergely Kis]

Hi,

We are in the process of selecting the right ERP for our company.
We initially started evaluating OpenERP, but as I was getting to know
the project a little bit better, more and more problems came to my
attention, which I tried to summarize in an email:
https://lists.launchpad.net/openerp-community/msg02588.html

A nice member of the Tryton community (I don't know if he wants to
remain anonymous) pointed me to your project as an alternative to
OpenERP.
I of course knew about Tryton as a fork of OpenERP, but for some
reason I never thought it was still maintained. I have to say that I
was very pleasantly surprised.

I am trying to summarize my current understanding of the project and
also I would like to ask your opinion about the issues I see:

1. My first impression was when I started to read about Tryton, that
as a software project it is executed much more professionally, for
example:
- clear release schedule
- migration support
- sound technical decisions (e.g. using decimals instead of floats, or
the introduction of PYSON)
- database independence (although we do like PostgreSQL and intend to use it)
Do you see any specific technical areas where either Tryton or OpenERP
is clearly ahead of the other?

2. Unfortunately Tryton has a much smaller community (at least an
order of magnitude smaller) than OpenERP. This of course shows in the
raw count of implemented modules / features. What are the trends in
the community? Do you see an increase in users / service companies?

3. On the other hand our requirements are not handled out of the box
by OpenERP either, so we have to develop our own modules anyways
(either by ourselves or contracting a partner). I think these modules
would be interesting for the other members of the community as well --
either as contributed modules or even as part of the official Tryton
modules. First I would like to share the specification / requirements
with you. Is this mailing list the right place for that or
tryton-contrib / tryton-dev?

4. The lack of a mature web ui is a problem, because a web interface
is one of our core requirements. I already tried out the new JQuery
based web ui, but it looks like as if the css styling is not yet done.
Or I made a mistake during the install, in which case I would like to
know what I missed.
I know that the development of this client received some funding last
year. Is it planned to be finished in the next release cycle?
Obvoiusly, if we decide to go with Tryton, we plan to contribute to
this client ourselves as well.

5. Of course an alternative to using the official web client is
Nereid, which is something we still have to evaluate.

6. No Hungarian localization seems to be available. Are there other
fellow Hungarian developers in the community, who would be interested
in joining forces?

7. I checked the websites of the companies listed on the tryton.org
services page, and it looks like most of you still provide services
for OpenERP as well. What is the reason for this? Are you only
supporting your existing OpenERP clients, or there are specific areas
where (even with its issues) OpenERP is still a better choice? Which
are these areas?

8. I already checked out openerp2tryton.com, which I found to be a
great resource. In your experience: is it possible / practical to
develop functionality while targeting both Tryton and OpenERP. In
particular it would be great if the Hungarian localization (chart of
accounts, translations) could be shared as much as possible --
provided that the contributors agree to dual license their code under
GPL3 / AGPL.
Or have the 2 systems diverged too much to allow such direct code sharing?

9. If the AGPL license is not a problem for us, is it practical to
port features from OpenERP to Tryton? Or again: it is more practical
to just use the ideas / design from the existing module (if it makes
sense), and develop from scratch? I know that the usual answer to such
questions is: "it depends on the module". What I am interested in is
some kind of statistics, like: we made X modules for Tryton, and Y of
that was based on an OpenERP module (a real port), and Z of that was
new implementation with only some ideas taken from the OpenERP
counterparts.

Thank you for taking the time to read my long list of questions, and I
am looking forward to your opinions.

Best Regards,
Gergely

--
Kis Gergely
ügyvezető / CEO
MattaKis Consulting
Email: gerge...@mattakis.com
Web: http://www.mattakis.com
Phone: +36 70 408 1723
Fax: +36 27 998 622

[Cédric Krier]

On 25/04/13 09:49 +0200, Gergely Kis wrote:
> We are in the process of selecting the right ERP for our company.
> We initially started evaluating OpenERP, but as I was getting to know
> the project a little bit better, more and more problems came to my
> attention, which I tried to summarize in an email:
> https://lists.launchpad.net/openerp-community/msg02588.html

It sounds like the usual concerns about OpenERP.

> A nice member of the Tryton community (I don't know if he wants to
> remain anonymous) pointed me to your project as an alternative to
> OpenERP.
> I of course knew about Tryton as a fork of OpenERP, but for some
> reason I never thought it was still maintained. I have to say that I
> was very pleasantly surprised.

It will be great if you can tell us why did you think it was dead and
how we could have prevent you to think about that. This could help us to
improve the communication.

> I am trying to summarize my current understanding of the project and
> also I would like to ask your opinion about the issues I see:
>
> 1. My first impression was when I started to read about Tryton, that
> as a software project it is executed much more professionally, for
> example:
> - clear release schedule

Yes timebased.

> - migration support

We don't want vendor locking system because it kills the collaboration.

> - sound technical decisions (e.g. using decimals instead of floats, or
> the introduction of PYSON)

Yes we make it first for us to be able to develop peacefully.

> - database independence (although we do like PostgreSQL and intend to use it)
> Do you see any specific technical areas where either Tryton or OpenERP
> is clearly ahead of the other?

Heee, yes I think so :-)

- really modular on_change design
- Active Record pattern
- Dict fields
- Dynamic digits
- One way to manage states of fields
- views not stored in the database
- XML record protected by default
- etc.

There are also big differences in the design of the key modules like
sale, purchase, stock especially the way they interact together.

> 2. Unfortunately Tryton has a much smaller community (at least an
> order of magnitude smaller) than OpenERP. This of course shows in the
> raw count of implemented modules / features. What are the trends in
> the community? Do you see an increase in users / service companies?

We see a slow but stable increase over time. But as we are quite strict
on the inclusion of service companies (require contributions), there is
no place for «parasite».

> 3. On the other hand our requirements are not handled out of the box
> by OpenERP either, so we have to develop our own modules anyways
> (either by ourselves or contracting a partner). I think these modules
> would be interesting for the other members of the community as well --
> either as contributed modules or even as part of the official Tryton
> modules. First I would like to share the specification / requirements
> with you. Is this mailing list the right place for that or
> tryton-contrib / tryton-dev?

It depends if your goal is to have it included in the core then it is on
tryton-dev. If you don't think it is not generic enough then it is for
tryton-contrib.

> 4. The lack of a mature web ui is a problem, because a web interface
> is one of our core requirements. I already tried out the new JQuery
> based web ui, but it looks like as if the css styling is not yet done.

Yes we are focused to cover the minimal requirement to have a minimal
working client at first. So there is no styling at all except the
default JQuery one.
And we rely on the community to help us on the styling.

> Or I made a mistake during the install, in which case I would like to
> know what I missed.
> I know that the development of this client received some funding last
> year. Is it planned to be finished in the next release cycle?

We wanted to have finished our main target for the release 2.8 (50% of
the funding), we are almost. And for sure we will have finished the
funded part for the next release.

> Obvoiusly, if we decide to go with Tryton, we plan to contribute to
> this client ourselves as well.

Great.

> 5. Of course an alternative to using the official web client is
> Nereid, which is something we still have to evaluate.

Be careful, nereid is not a web client. It is a tool to build website
linked to Tryton like an webshop.

> 7. I checked the websites of the companies listed on the tryton.org
> services page, and it looks like most of you still provide services
> for OpenERP as well. What is the reason for this? Are you only
> supporting your existing OpenERP clients, or there are specific areas
> where (even with its issues) OpenERP is still a better choice? Which
> are these areas?

Can not answer this as we (B2CK) don't provide services on OpenERP.

> 8. I already checked out openerp2tryton.com, which I found to be a
> great resource. In your experience: is it possible / practical to
> develop functionality while targeting both Tryton and OpenERP. In
> particular it would be great if the Hungarian localization (chart of
> accounts, translations) could be shared as much as possible --
> provided that the contributors agree to dual license their code under
> GPL3 / AGPL.
> Or have the 2 systems diverged too much to allow such direct code sharing?

Sharing code is really difficult because the API was completly
redesigned in Tryton (and almost unchanged in OpenERP).
Sharing data, it is probably possible if the schema is still similar.

> 9. If the AGPL license is not a problem for us, is it practical to
> port features from OpenERP to Tryton? Or again: it is more practical
> to just use the ideas / design from the existing module (if it makes
> sense), and develop from scratch? I know that the usual answer to such
> questions is: "it depends on the module". What I am interested in is
> some kind of statistics, like: we made X modules for Tryton, and Y of
> that was based on an OpenERP module (a real port), and Z of that was
> new implementation with only some ideas taken from the OpenERP
> counterparts.

We (B2CK) ported all the core modules and we just kept the very general
ideas and made our own design (based on our experiences).
This makes Tryton looks like OpenERP design at first sight but really
different once you dive in it.

So I will strongly encourage to just have a look at what is in OpenERP
but not too much. And also look at other system like Openbravo,
Compiere, SAP etc. to get a good picture of all possibilities.

--
Cédric Krier

B2CK SPRL
Rue de Rotterdam, 4
4000 Liège
Belgium
Tel: +32 472 54 46 59
Email/Jabber: cedric...@b2ck.com
Website: http://www.b2ck.com/

[Gergely Kis]

Hi Cédric,

Thank you for the informative answers.

2013/4/25 Cédric Krier <cedric...@b2ck.com>:
[...]

>> I of course knew about Tryton as a fork of OpenERP, but for some
>> reason I never thought it was still maintained. I have to say that I
>> was very pleasantly surprised.
>
> It will be great if you can tell us why did you think it was dead and
> how we could have prevent you to think about that. This could help us to
> improve the communication.

Well, this is a difficult question. When you look for an open-source
ERP, you instantly get hits for Compiere and its derivatives
(ADempiere, OpenBravo), then of course OpenERP. More comprehensive
articles will mention Tryton as an OpenERP fork, and don't really go
further than that.

Take a look at this page for example:
http://en.wikipedia.org/wiki/List_of_ERP_software_packages

Tryton is at the end (alphabetic order :( ), and the only information
available is that it started as an OpenERP fork, which is probably not
the most important thing about it.

One improvement idea would be to write case studies and/or
introductory articles in high-profile open-source magazines. For
example Jonathan Corbet of LWN.net fame is looking for a new
accounting/ERP package for their operations. He already wrote a number
of articles as part of the "accounting quest" series:
http://lwn.net/Articles/496158/
http://lwn.net/Articles/501681/
http://lwn.net/Articles/516659/

If you could get him to cover Tryton (and even better: get a positive
review from him :) ), I think you would see an instant increase in the
interest for the project.

[...]

>
>> 5. Of course an alternative to using the official web client is
>> Nereid, which is something we still have to evaluate.
>
> Be careful, nereid is not a web client. It is a tool to build website
> linked to Tryton like an webshop.

Yes, I understand.

I also saw, that there are already apps in development, building on
Tryton and Nereid, like nereid_project which could be used instead of,
or in parallel to the generic ERP web client for project management.

[Sharoon Thomas]

On Thu 25 Apr 20 09:49, Gergely Kis wrote:
> Hi,
>
> We are in the process of selecting the right ERP for our company.
> We initially started evaluating OpenERP, but as I was getting to know
> the project a little bit better, more and more problems came to my
> attention, which I tried to summarize in an email:
> https://lists.launchpad.net/openerp-community/msg02588.html
>

> Do you see any specific technical areas where either Tryton or OpenERP
> is clearly ahead of the other?

Software quality is the single most important difference for us at
Openlabs. The simplicity of the core modules and the KISS approach to
design makes Tryton a far better choice than OpenERP if you plan to
customize. We have seen that the development effort is lower with Tryton
and we don't get surprises as we extend the standard modules.

Another important difference for us is the ability to test the software.
Tryton has the capability to be both unittested and proteus allows you
to write higher level integration tests. This has made software
mainteinance a breeze for our support customers.

>
> 2. Unfortunately Tryton has a much smaller community (at least an
> order of magnitude smaller) than OpenERP. This of course shows in the
> raw count of implemented modules / features. What are the trends in
> the community? Do you see an increase in users / service companies?

There is a steady increase in the number. Numerous requests come often
from companies requesting to be listed on the page, but meritocracy
being the factor to be listed there, the criteria falls back to making a
significant contribution.

However little the community is, it is an active and vibrant one which
constantly

> 5. Of course an alternative to using the official web client is
> Nereid, which is something we still have to evaluate.

Nereid is not an "alternative" to the web client. It extends tryton as a
web framework and allows you to build web applications over Tryton.
Commonly built application are:

* eCommerce systems
* Customer and Supplier portals
* Employee self service portals
* Project management systems etc

Feel free to ask if you need any support in developing with nereid and
we would be happy to help you

> 7. I checked the websites of the companies listed on the tryton.org
> services page, and it looks like most of you still provide services
> for OpenERP as well. What is the reason for this? Are you only
> supporting your existing OpenERP clients, or there are specific areas
> where (even with its issues) OpenERP is still a better choice? Which
> are these areas?

Openlabs has customers on both OpenERP and Tryton. We are in the process
of moving our customers to Tryton and we have happy customers who moved
to Tryton.

> Or have the 2 systems diverged too much to allow such direct code sharing?

In my opinion, the systems have diverged too much to have direct code
sharing, especially with the never versions. The only common link seems
to be the inheritance pattern and the usage of python and xml.

Welcome to the Tryton community.

Thanks,
--
Sharoon Thomas
CEO & Chief Software Architect
Openlabs Technologies & Consulting (P) Limited
Ph: +1 813 793 6736 Extn. 200
t: @sharoonthomas

[Axel Braun]

Hi

Am Donnerstag, 25. April 2013, 09:14:30 schrieb Ilya Melnikov:

[...]

> First and the main - poor documentation.
> I read documentation on site and completely understood nothing.
> Tried to understand "by mouse clicking" and finally confused.
> The problem was complicated by lack of translation (docs and soft)
> for my language.

My personal impression is that everything is focussed very much on the
developer, not on the user. The functional benefits Tryton offers are not
immediately visible to a more business user. Compared to other ERP projects,
we have a lack in 'sales', demonstrating the capabilities and the ease of use.

> I would like to see more detailed documentation. Еven though in English.
> But with more details and examples.
> OpenERP has books with screenshots and step-by-step instructions,
> Quick start guides, some ready-made recipes for common cases, video
> tutorials.
> Tryton has list of "Field: description".
>
> Second reason which stops me - lack of builds for Ubuntu.
> And don't say me "go http://www.tryton.org/download.html". :)
> There is antique version of Tryton in Ubuntu repo.
> Currently 2.2, released one and a half year ago.
> This suggests me that "Tryton abandoned".

A proposal to all contributors: Next to the sources available from the Tryton
website, I made an attempt to start building packages on the Open Build
Service (OBS). http://build.opensuse.org/
OBS supports all major distributions (including Ubuntu), and by this gives us
the opportunity to build 'all' distributions from one set of source data.
Feel free to share your effort on
https://build.opensuse.org/project/show?project=home%3ADocB%3Apython_packages
At the moment I started with openSUSE, but more distributions can be added.

Cheers
Axel

[Raimon Esteve]

> First and the main - poor documentation.
> I read documentation on site and completely understood nothing.
> Tried to understand "by mouse clicking" and finally confused.
> The problem was complicated by lack of translation (docs and soft)
> for my language.

> I would like to see more detailed documentation. Еven though in English.
> But with more details and examples.
> OpenERP has books with screenshots and step-by-step instructions,
> Quick start guides, some ready-made recipes for common cases, video tutorials.

From trytonspain people are working dynamic user documentation (also available technical user)

This doc is build depend modules are installed because there aren't two companies equals.

http://doc.tryton-erp.es/

Of course, you could translate spanish to another language ;-)

Every module is a doc dir where are available rst files about documentstion (using sphinx inherit)

[Raimon Esteve]

> Openlabs has customers on both OpenERP and Tryton. We are in the process
> of moving our customers to Tryton and we have happy customers who moved
> to Tryton.

Same us, Zikzakmedia from Spain.

We release/develop and Nan-tic, Spanish localization modules to spanish companies.

[Albert Cervera i Areny]

A Dijous, 25 d'abril de 2013 09:49:28, Gergely Kis va escriure:

> 7. I checked the websites of the companies listed on the tryton.org

> services page, and it looks like most of you still provide services

> for OpenERP as well. What is the reason for this? Are you only

> supporting your existing OpenERP clients, or there are specific areas

> where (even with its issues) OpenERP is still a better choice? Which

> are these areas?

 

In our case, we're now moving to tryton and we're encouraging new customers to use Tryton instead of OpenERP (many contact us because of OpenERP). Also, OpenERP v6.0 is the last release we support, so although we're in a transition phase, we're really almost only offering Tryton to new customers.

 

Let me say that I was talking to a lead of a large company in Spain last week and they had more or less your feelings. They preferred Tryton over OpenERP for code quality and architecture, and I think there are lots of chances Tryton will keep growing because arguments are solid enough when properly explained.

--

Albert Cervera i Areny

http://www.NaN-tic.com

Tel: +34 93 553 18 03

 

http://twitter.com/albertnan

http://www.nan-tic.com/blog

 

[Giedrius Slavinskas]

2013/4/25 Axel Braun <axel....@gmx.de>

> First and the main - poor documentation.
> I read documentation on site and completely understood nothing.
> Tried to understand "by mouse clicking" and finally confused.
> The problem was complicated by lack of translation (docs and soft)
> for my language.

My personal impression is that everything is focussed very much on the
developer, not on the user. The functional benefits Tryton offers are not
immediately visible to a more business user. Compared to other ERP projects,
we have a lack in 'sales', demonstrating the capabilities and the ease of use.

It is really bad attitude. You must see Tryton as a tool and not as a product.

When the open source project becomes a product, it starts to die. You can see this

process with the release of OpenERP v7. It became a product which is pretty easy to

setup for non technical user. That means many companies were founded which are not

doing any development, but only selling it. It is hard for real dev company

to compete with such re-sellers.

So if there are no dev companies who will develop this "product"? For OpenERP,

the partnership program works, but as i understand, Tryton is in the difference way.

I hope, Tryton will be even more developer focused project. It must be a tool, framework

which makes the business application building process fast and cheap. Also taking the

maintanance process in account.

At first glance, it may look that OpenERP is easy to install and configure. But try to implement more

complex process or do maintainance and you will see all the "benefits" of being product :)

p.s. this is my personal view. Tryton's one can differ.

--
Giedrius

[Jean C]

It is really bad attitude. You must see Tryton as a tool and not as a product.

I personally consider that tryton as a framework (which is a tool) is different than

tryton as an ERP (which is more of a product).

The framework part (that would be server + client + core modules) must provide a complete

documentation fully developer-oriented. The modules which makes it an ERP (sales, stock,

party etc...) should come with a user documentation, as they are supposed to be used "as is"

with little to no extra developments.

At Coopengo, we use the tryton framework so the technical doc suits us pretty well. We plan

to use some of the accounting modules soon but here we would like both dev and user doc.

We got people who do not have the time / knowledge to read the code but we need their advice

on how tryton fare as an accounting software, and right now they just cannot use it until we dev

can write down the how-tos.

That being said, we do not want (and I do not think anyone here wants) to go down OpenERP

path and sacrifice the stability of the framework for better sales.

My 2 cents

Jean CAVALLO

Coopengo

 

[Giedrius Slavinskas]

2013/4/26 Jean C <jean.c...@coopengo.com>

It is really bad attitude. You must see Tryton as a tool and not as a product.

I personally consider that tryton as a framework (which is a tool) is different than

tryton as an ERP (which is more of a product).

The framework part (that would be server + client + core modules) must provide a complete

documentation fully developer-oriented. The modules which makes it an ERP (sales, stock,

party etc...) should come with a user documentation, as they are supposed to be used "as is"

with little to no extra developments.

In my opinion it should be opposite. Standard modules are also the framework. They must

provide only the generic functionality for the field they are used for (the same like it is now).

Of course, for some companies this basic functionality will be enough and thats fine. It does

not make the difference for developer.

 

At Coopengo, we use the tryton framework so the technical doc suits us pretty well. We plan

to use some of the accounting modules soon but here we would like both dev and user doc.

We got people who do not have the time / knowledge to read the code but we need their advice

on how tryton fare as an accounting software, and right now they just cannot use it until we dev

can write down the how-tos.

But it is your problem, not community's. If you need docs, you must write them. And you are free

to share them with community.

My main point was that we need to stop begging such things from community by telling that other

ERPs have it and we need to compete against them.

If you need docs, write them, if you need specific module, code it.

--
Giedrius

[Nicolas Évrard]

* Max Jeanpierre [2013-04-26 13:23 +0200]:
>Hi Cédric,

Hello Max,

>>It will be great if you can tell us why did you think it was dead and
>> how we could have prevent you to think about that. This could help us to
>> improve the communication.
>>
>

>Simply have a look at the home page tryton.org:
>
>You just won't find the word " ERP".

But in fact Tryton is a bit more than an ERP, some companies are using
it to manage hospital, others are using it as an e-commerce backend or
as an insurance software.

>Instead:
>"
>The core of Tryton (also called Tryton kernel) provides all the necessary
>functionalities for a complete application framework: data persistence,
>extensive modularity, users management (authentication, fine grained
>control for data access, handling of concurrent access of resources),
>workflow and report engines, web services and internationalisation. Thus
>constituting a complete application platform which can be used for any
>relevant purpose.
>"
>
>That just made me run off the first I arrived on this page. No print
>screen, nothing "showing" the product, nothing appealing.
>
>Same on the news:
>
>"
>All shipment Many2One on Move have been merged into one single
>shipment Reference The method find_best_supplier doesn't optimize
>anymore on the delivery delay and so it fully respects the priority
>order to select a supplier. Etc..."
>
>It definitively looks like a developper toy, made by developper for
>developper, and not made for real users...

In fact Giedrius is telling exactly the opposite than what you are
saying. That is what is so difficult about the organization of the
Tryton website: make it a comprehensive source of information for both
the Tryton/Python developer and at the same time for the CIO (or
whatever you want to name the person that is in charge of choosing
the next ERP solution of his company).

Nevertheless, I tend to agree that the current website is not enough
business-friendly, but as Giedrius said: anybody can submit patches
(or maybe even just text, because we know that making a patch is not
something trivial) that would make our website look more professional
and welcoming for those kind of users.

--
Nicolas Évrard

B2CK SPRL
rue de Rotterdam, 4

4000 Liège
Belgium
Tel: +32 472 54 46 59

E-mail/Jabber: nicolas...@b2ck.com
Website: http://www.b2ck.com/

[Rajesh Kumar]

Hi Cedric, Gergely,

I'm new to Tryton and have setup both server and GTK client. I'm looking for a web client and have no idea about the jQuery UI you are talking in this thread. Could you please share the details? I'm looking to not only using it, but also contributing as much as I can. Thanks for help.

Regards,
Rajesh

[Nicolas Évrard]

* Rajesh Kumar [2013-08-19 18:17 +0200]:
>Hi Cedric, Gergely,

Hello Rajesh,

>I'm new to Tryton and have setup both server and GTK client. I'm looking
>for a web client and have no idea about the jQuery UI you are talking in
>this thread. Could you please share the details? I'm looking to not only
>using it, but also contributing as much as I can. Thanks for help.

We are currently developing a javascript webclient, named sao. This
work has been financed by the community during the latest Tryton
Unconference.

You'll be able to find the relevant code there:

http://hg.tryton.org/sandbox/sao

This client is built using jQuery and jQuery UI:

http://www.jquery.com/
http://jqueryui.com/

PS: Please do not top-post on this mailing list.

--
Nicolas Évrard

B2CK SPRL

4, rue de Rotterdam