Fabien Pinckaers

This blog is not updated

2011-12-08T08:37:00.001-08:00

I do not blog anymore. I prefer focusing on building the OpenERP company and product.

If you want to be kept informed about OpenERP or my activities, I suggest you to follow me on twitter:
http://twitter.com/fpopenerp

OpenERP V6 CRM in Action

2010-11-14T14:16:00.000-08:00

Here is a preview of the new CRM of OpenERP v6. Enjoy !

OpenERP v6, new logging system for end-users

2010-10-16T07:18:00.000-07:00

In order to improve the useability of OpenERP v6 and make it easier to use and learn for new users, we added a logging system on main actions of OpenERP. When a user confirms a document, OpenERP will show him all the next actions that have been prepared by the system.

As an example, when a salesman convert a quotation to a Sale Order, OpenERP will tell him what have been triggered: new delivery orders planned, new tasks created, etc. The screenshot bellow, shows you what happens when you confirm a sale order with two products (PC1 & PC2), one of the two has to be manufactured in "make to order".

You can see that OpenERP tells you that you will have two products required for the manufacturing which are not available in the stock, two manufacturing orders have been scheduled (I used multi-level BoM) and the delivery order is scheduled for the 10/13/2010.

The great thing with this system is that you can click on any line of the log to zoom to the related document. If you need more information about the procurement in exception, just click on it and fix it.

As another example, here is the log that appears when you register a customer payment, that pays two invoices.

You can see that two invoices have been marked as "paid" and a delivery order has been scheduled by OpenERP. That's because the sale order SO/0017 was created with the order policy "Payment before delivery". So, the payment of the invoice triggered the delivery order.

OpenERP Automated Tests & Reports

2010-09-24T01:34:00.001-07:00

The new automated test framework for OpenERP is finished in trunk. It consists of 1604 tests spread in 146 complete scenari. Every commit run a series of tests, according to the modules changed in the commit. These tests are validating the business logic, but also the reports of OpenERP.

The automated test server is built on buildbot. Here is an example of automated test that failed: http://test.openobject.com/builders/buildbot-openobject-addons/builds/38

If you want to see all the available reports in OpenERP, just launch the following queries on a trunk branch:
> createdb trunk --encoding=unicode
> openerp-server.py -i sale,purchase,mrp,project -d trunk --test-report-directory=/tmp --log-level=test

OpenERP will launch automatically all tests scenari and save the generated reports in the /tmp directory.

We are also working on an improved "bzr stats" to get information ala Google Analytics but for the commits on a bazaar branch. The very first prototype has been released here:
https://code.launchpad.net/~openerp-community/bzr-stats/bzr-stats-analytics

This will allow us to get statistics on contributions and developers, related to the automated tests quality report in order to better do continuous improvement in our development process.

Improved Automated Tests for OpenERP: reporting

2010-09-12T05:13:00.000-07:00

Here are some news about our work in progress to improve automated tests an code review in OpenERP trunk. Let me first introduce how is our R&D organized.

At OpenERP SA, we use the SCRUM project management methodology to organize our developments on OpenERP. We use the OpenERP scrum module to organize the product backlogs. We have developers spread into 8 teams of about 6 developers per team:

Framework: scrum master = Olivier
Addons 1 (sale, stock, mrp, purchase): scrum master = Quentin
Addons 2 (crm, dms, survey, misc): scrum master = Quentin
Addons 3 (account, project, hr, events): scrum master = Quentin
Usability: scrum master = Aline
Support / Bugfixes: scrum master = Stéphane
Web Client: scrum master = Xavier
GTK Client: scrum master = Naresh

Each team work in sprints of 2 weeks. Each team works in his own launchpad branch during the sprint (example: lp:~openerp-dev/openobject-addons/trunk-dev-addons1). At the end of the sprint, the team makes a propose for mergal of his work in the OpenERP trunk official branch. In order to to control the quality of the code, every development is reviewed by the team'scrum master during the merge process. (code review)

In order to improve the quality of the tests, we setup a test server that launches a series of tests after each commit made on a team branch or the trunk official branch. This server is still under development, but you can already track the firsts results here: http://test.openobject.com/

The tests performed on each module by the test server are:

Test scenarii to check all the features and workflows, example: http://test.openobject.com/buildersresource/addons-trunk/builds/147/steps/OpenERP-Test/logs/lunch.test/text
Quality of the code and respect to OpenERP guidelines, example: http://test.openobject.com/buildersresource/addons-trunk/builds/151
Modules installation and demo data: http://test.openobject.com/buildersresource/addons-trunk/builds/151

In the past three months, our teams developed 1495 YAML tests spread into 112 yaml files and modules. We can now ensure that most common enterprise flows are tested automatically by OpenERP at every commit and installation of the software.

Today we just added automated tests on OpenERP reports. Each time you install OpenERP, it will launch the generation of sample PDF reports. In the trunk version, you have the possibility to store the generated report into a directory using the new argument "--test-report-directory=/tmp/".

If you want to test this, on the latest OpenERP trunk, you can execute the following commands under Linux:

createdb mytestdb --encoding=unicode
./openerp-server.py --init=sale -d mytestdb --test-report-directory=/tmp
xpdf /tmp/sale-sale_order.pdf

You will get a sample PDF file of two sales orders generated by the OpenERP YAML test engine.

If you want to know how it has been implemented in the test scenario, check this file in your local installation: addons/sale/test/sale_order_print.yml

The automated test server is still in beta. We plan to release the final version and start using it massively by the end of september. Our current work in progress for the automated test engine are:

get statistics on commits, test failures and contributions made by developers
allows to monitor and test community or partners branches (everyone will be able to subscribe his own branches on the integration server)
send a clear warning email to the developer that introduced a regression
add more tests on the OpenERP reports (only the sale order has been made for the prototype)
add sample of PDF reports in the documentation of the module on doc.openerp.com

We are also working on a stronger code review process to prepare ourself for the trunk Release Candidate 1. By the end of september, we will force every development to be done in the team branch and reviewed by the scrum master. (including the Bugfix team, which was commiting directly in the trunk currently.)

OpenERP's reaction to Compiere's acquisition: mixed feelings

2010-06-17T04:09:00.000-07:00

The announcement of the acquisition of Compiere by Consona Corporation leaves us with mixed feelings.

First, for us who are deeply committed to Open Source we can only feel sorry that one of the historical pioneers in this area is going to disappear. Reading the press release we understand Consona's interest in the technology but we have a hard time believing their commitment to support the community and indirect model. We know how much dedication it takes to support this business model. Even if Compiere was a competitor, from our perspective it is bad news. We believe that the more successful Open Source vendors the better for our industry as a whole.

Secondly, Compiere's buyout (and in our eyes withdrawal) happens at a very strange time. We have never experienced so much demand than during the past recent months. We see interest from major companies which strongly feel the limitations and the frustrations of dealing with SAP and Oracle, companies we never thought would consider Open Source ERP solutions a few months ago. We also experience increasing global demand: Poland, Mexico, South Africa and Philippines just to mention a few. Our opinion is that Compiere's failure is not resulting from a lack of demand for Open Source solutions, but the consequence of a poor execution and a lack of commitment toward their community.

Finally, Compiere's acquisition reinforces our determination. The Open Source ERP community needs a strong leader and we believe we can play this role. We know that we need to work hand in hand with the community to build not only a good product but the very best.

We are determined to do it by sticking to our vision, to our strategy and with the support of all of you!

CalDav and meeting improvements in trunk

2010-03-05T10:41:00.000-08:00

The calendars in OpenERP trunk have been completly reviewed. OpenERP supports:

caldav synchronisations (with sunbird, outlook, mobile phones, ...)
recurrent event (with custom rules)
alerts
invitations (internal user, partner or emails), connected with the mail gateway for invitation reply
tasks synchronized through caldav

Here is the documentation of the caldav module.

We are also working on a new look & feel for the web calendar. Here is the screenshot:

OpenERP Trunk, New Design

2010-03-02T09:59:00.000-08:00

Hello,

Here is one screen more for the new version of the web client of OpenERP. This is the dashboard view. We do our best to deliver you a working version of the trunk that integrates these new design within 2 months.

Call for contributions: automated tests

2010-02-24T01:18:00.000-08:00

Hello,

We plan to write a bunch of automated tests. These tests will be used as acceptation tests for the release of the next version of OpenERP. As every OpenERP user has different scenario according to their own business needs, we decided to make a call for contributions: Please help us to write scenario of tests.

You can describe your own use of OpenERP here:

http://piratepad.net/openerp-tests

We will manage the implementation of all these automated tests in OpenERP. This will ensure that your own usage of OpenERP is maintained and tested in the long term. The 5th of march, a developper team will be deicated to this job.

OpenERP v5.2, new UI proposition

2010-02-19T05:39:00.000-08:00

Hello,

Here is a preview of the new UI for the next version of OpenERP. We still have some details to do, but it looks promising.

OpenERP raises 3 million euros.

2010-02-16T13:38:00.000-08:00

OpenERP announces today that it raised 3 million euros. The investors are Sofinnova Partners, represented by Olivier Sichel, and the Iliad's managers, Xavier Niel and Olivier Rosenfeld. The funds raised will allow OpenERP to achieve its ambition to be one of the leading application business suite vendors worldwide.

OpenERP'strategy is based on a unique ecosystem which combines the resources of the community, of its partner’s network and of the editor. The community contributes to the ongoing enrichment of the OpenERP solution. The partners tailor the solution to customers needs and provide the customer support. The editor guarantees the quality and the long term sustainability of the software.

In the past few years, OpenERP has experienced extremely fast growth, but only self funded and with limited financial resources. This fund raising represents a crucial milestone for OpenERP's development since it will allows OpenERP to focus its resources less on project management and more on its core business as a software editor.

In particular, OpenERP will be in a position to invest in the following areas:
* Investing in the development of the OpenERP's software
* Recruiting an training a worldwide network of partners
* Developing new services such as maintenance and SaaS

We are particularly proud to have found such quality investors, with a really good understanding of the stakes, constraints and objectives relating to open source software. Indeed, as investors are highly selective, it was really equally important for OpenERP to find investors that share our Open Source vision.

As of this fund raising, the company Tiny SPRL, previously the editor of OpenERP, has been renamed "OpenERP sa". In this coming year, the company will also open new offices in the United States.

About OpenERP

The OpenERP project, founded in 2005 by Fabien Pinckaers has already been extremely successful thanks a community comprised of more than 300 members and a network of integrators established in 26 countries. OpenERP software is comprised of more than 500 modules and counts more than 1000 installations a day, what makes it one of the most frequently installed management software in the world.

About Sofinnova Partners

Sofinnova Partners is an independant venture capital firm based in Paris. For over 35 years, the firm has backed nearly 500 companies at different stages of development – pure creations, spin-offs, as well as turnaround situations – and worked alongside Europe's key entrepreneurs in the technology, life sciences and cleantech sectors. With more than €1 billion of funds under management, Sofinnova Partners' experienced team and hands-on approach in nurturing and supporting porfolio companies through to exit have created market leaders and multiplied revenues, from ladmark historical investments including Genentech and Biogen to more recent successes such as Actelion, Vistaprint, CoreValve and Fovea. With a global mindset, the firm has offices in Paris, Shanghai, China and a sister organization in San Francisco, California. Please visit www.sofinnova.fr for more information on Sofinnova's team and portfolio.

Media contact :
Aline De Bock
+32 81 81 37 00
adb@openerp.com

Open Job: "Sales Director Europe"

2010-02-10T07:38:00.001-08:00

OpenERP is looking for a new Sales Director Europe.

The OpenERP Sales Director Europe will have full responsibility for the development and performance of Open ERP sales team in Europe. Staffs and directs a sales team and provides leadership towards the achievement of maximum growth in line with company’s plan.

Job is based in the Namur Area (Belgium)

Exceed sales targets

Define the sales strategy for each line of business (Partnership, Training, Maintenance)
Define the partners strategy for each European market (profile, number of partners..)
Develop and coordinate sales selling cycle and methodology.
Produce an accurate forecast and pipeline through selected tools and processes;
Oversees all hiring and training of personnel involved in Sales
Establishes performance goals for all sales department employees, and monitors performance
Implement sales strategy from leads to orders.

Skills

Minimum of 10 years professional experience in sales, 5+ years experience in staff management and executing enterprise software sales;
Strong understanding of indirect channel partners and system integrators in Europe.
Mastery of the written English language, excellent written and verbal communication skills
Proven leadership ability to influence, develop, and empower employees to achieve objectives with a team approach.
Able to work in “start up” fast paced environment.
approachable, honest and fun team player

Contact jobs AT openerp.com if you are interested.

Useability Improvements, search views

2010-02-10T01:34:00.000-08:00

All screens and menus have been reviewed in OpenERP to improve the general useability. Here is a screencast that shows improvements made on the lists view (trunk version). You can get a wider version of this video on youtube.

A small explanation of the changes:

Every object have only one menu (instead of Draft Tasks, My Tasks, ...). But you can switch easily from one context to another one.
When you enter the list, by default, it is filtered according to the user: (In this example: my current tasks = my tasks in draft or open state)
You can change filters by fill in forms or clicking on buttons. Instead of field=value, buttons are more business oriented (example: Sales To Invoice = sale order, with state='in progress' and invoiced=False). Buttons applies the filter directly
many2one fields can be rendered as selection box to ease the search
Buttons are now available in lists. You can clickly close tasks, delegate them, ...
Advanced search criteria can be folded in groups of elements (example: group by...)
Every list view can be multi-level grouped. Example, we group tasks by projects. All float and integer fields are computed accordingly (example: remaining hours by project)
If you drag&drop a task from one group to another one, fields are updated automatically.
Some button filters may change the columns displayed. Example, click on pending tasks and you will see the column 'Delegated To' column. Or, click on Getting Things Done buttons and columns are reduced to the minimum you want to see in this environment.
The lists can easily switch, according to the context between editable lists/non editable lists. When you click on Inbox, the list becomes editable.
You can add custom filters by clicking on [+]
Sort orders are now computed on the server side (for long lists)
You can save your search filters as a filter or as a shortcut.

We are implementing the same features in the web client too.

Trunk HR Improvements

2010-01-27T09:52:00.000-08:00

Hello,

We did lots of improvements on HR (as well as others modules) in the trunk, to be released in new versions. The biggest ones:

resources : employees are resources, as well as work centers. We made a generic scheduler in the trunk that computes capacity based plannings using resources.
time scale: these resources have working time (ex: from monday to friday, 9am-6pm), capacities, ... This allows to efficiently compte project's planning (long term and short term) and manufacturing orders based on work centers capacities.
the time scale is integrated with public and employee's holidays.

For the scheduler, we took the algorythm implemented in the open source project: faces.

We did lot's of efforts in internal evaluation of employees, hr surveys and salary grid. The most significant changes in these area is the survey module that allows you to build forms/survey to:

recruit new people
do internal survey (satisfaction, efficiency, ...)
manage the internal interviews plan per employee.

For internal evaluations, we did a plan system, ex: first meeting after 6 months, next every year if everything is ok. You can define goals, review goals, ...

Others small improvements in hr:

Better integration with document management system for jobs hiring processes (database of CV's, ...)
Much more better views, menus
Review of crm.case into better redruitement forms (the object is separated)
Leaves and holidays have been improved a lot. (leaves requests, recuperation days, ...)

SUN & OpenERP Partnership

2010-01-22T07:25:00.000-08:00

Ever-shifting conditions, shorter lead times, and heightened global competition are facts of life in today's business climate. To effectively compete, you need to operate more efficiently and reduce TCO. ERP systems can provide significant improvements in operational and administrative costs, on-time delivery, and order accuracy. And running open-source Open ERP on the Sun platform can help you implement and ERP solution cost effectively.

Here is a screenshot of the announce on SUN's homepage:

You can get more information on the OpenERP + Sun Partnership. You can download MySQL branches of OpenERP on launchpad.

Community & Partner Meeting, 23-25 March

2010-01-13T01:03:00.000-08:00

Hello,

The 23th and 24th of march, we will organise the annual OpenERP Community Meeting in our offices in Belgium. The proposed planning is:

23th of March

Presentation of the new version and new features: 4h
Lunch
Each contributor will have to opportunity to present his contributions and modules, 20 minuts per module to present: 3h
Q&A, Discussion about the future of OpenERP: 1h
Diner at a typical belgian restaurant

24th of March

The whole day is dedicated to workshops by small groups: useability, accouting expert, services mgt, community organisation, ...
The goal of these workshops is to discuss priorities and write specifications for the next version of OpenERP.

This community meeting will be followed by the partner meeting, the 25th of march, to discuss business issues.

25th of March: restricted to partners

Development Strategy 2010
Services and products (Odoo, support, maintenance, certification)
Organisation of the partners network
Management of leads
Marketing activities

If you need more information and want to subscribe to one of these events (or both), you can subscribe by sending an email to nma AT openerp.com.

Useability improvements in v5.2

2009-12-29T07:31:00.000-08:00

Hello,

We are working to improve useability for the future v5.2 version of OpenERP. You can follow some of our thoughts on this launchpad item:

https://bugs.launchpad.net/openobject-client-web/+bug/501297

We did not started the design yet, but worked on some wireframes, like the one bellow:

You can also follow the discussion about this topic in the forum:

http://www.openobject.com/forum/post48716.html

v5.2 is approaching

2009-11-01T13:47:00.000-08:00

Hello,

The new version (5.2) is still under development under the trunk branch. Lots of new features have already been developed. It's time to propose your improvements for merging into the trunk branch. We plan to freeze the v5.2 in december / january.

This trunk version has lots of small improvements. The biggest changes are summarized here: https://launchpad.net/openobject-addons/+milestone/v5.2

Thanks,

Bloomberg talks about OpenERP

2009-10-30T01:04:00.000-07:00

Bloomberg, the source of news and information that drives decisions that move billions of dollars every day in the U.S., talks about OpenERP. Click on the following link to get the article.

OpenERP and Electronic Invoicing

2009-10-15T23:20:00.001-07:00

Thymbra just released an Electronic Invoice module for the argentina market. Today it has been tailored to work for the argentinian laws, but it should be very similar for other countries. At Thymbra they have been invoicing electronically for a year now. It's solid ! You can find it in the argentinian localization project in launchpad.

OpenERP and AGPL

2009-10-15T14:09:00.000-07:00

Hello,

As requested by several partners and the community, we plan to change the licence of the next version of OpenERP to the AGPL (GNU Affero General Public Licence). Our goal is to better promote and maintain the free and open source nature of OpenERP. We think this licence will better protect the community to 'evil' SaaS offers that do not want to release their source code to the community.

The GNU General Public License (currently used by OpenERP v5) permits making a modified version and letting the public access it on a server without ever releasing its source code to the public.

The GNU Affero General Public License is designed specifically to ensure that, in such cases, the modified source code becomes available to the community. It requires the operator of a network server to provide the source code of the modified version running there to the users of that server. Therefore, public use of a modified version, on a publicly accessible server, gives the public access to the source code of the modified version.

Both the GPLv3 and AGPL are OSI certified and compatible together. We advice OpenERP contributors to develop their future modules in the AGPL licence. You can get more information on this licence on the FSF website:

http://www.fsf.org/licensing/licenses/agpl-3.0.html

You can contact fp AT openerp.com, for any question about this licence change.

2009-09-08T01:18:00.000-07:00

We just made an analysis of the Open ERP visibility. The Open ERP expansion is exploding ! Here are the main figures:

20% growth in number of leads every 2 months (currently: 110 customers request per day)
1000 downloads per day of the auto-installer
One new module produced every open day (20 per month)

The most interesting statistic is the comparison of the Open ERP visibility with current competitors. Let's take Netsuite (N), an ERP offer in SaaS which is probably the nearest competitor of the Odoo offer. In 2008, their annual expenses in Sales & Marketing operations was 80.000k€ ! The funny thing is that Open ERP just became most famous than Netsuite, according to google trends:

The annual sales & marketing budget of Open ERP is 200k€. It means Open ERP has the same visibility than a company having a marketing budget 400x bigger ! This is the power of a free and killer application !

More over, the sales and marketing are our weaknesses. Unfortunatly, we don't have such statistics yet on the developments, I let you imagine the figures :)

Open ERP's new continuous testing and integration server

2009-09-02T05:36:00.001-07:00

We are proud to announce the release of our new testing and release control dashboard. You can check it in action here:

http://test.openobject.com

The goal of this dashboard is:

to be sure that everyone understands the way to contribute and share works,
define the process for releases, maintenance and contributions,
control the quality of the contributions automatically,
propose a single point of entry to all Open Object's resources.

Bugs Detection

At each commit, our automated test framework automates:

The installation of the modified modules on different environments,
The launch of assert tests and demo data provided in each modules,
The launch of unit tests provided in modules,
The installation of all languages,
The automated migration since v4.2, 5.0 to detect incompatibilities.

Our systems are testing the trunk and stable branches. We have builders for the quality certified modules, the addons-extra and addons-community ones. If one of the test fails, the quality team and the author of the bug will receive a report by email.

Quality Improvement

In order to continuously improve the quality of the code and to train contributors to do quality code, we also setup a quality testing framework. Each module are evaluated through a series of quality tests at each commit.

It performs a series of tests:

Scalability: will the module rocks if you work with millions of record,
Availability of tests and demo data,
Guidelines: do you follow the Python or Open ERP guidelines,
Security: availabilities of well defined rules,
Quality of the code: non redundant lines, etc.

This is system is based on the base_module_quality module. We are looking for more tests, feel free to contribute and add your own tests in our system.

Documents

A series of slideshow have been made to explain how we propose to organise the community.

On Demand Open ERP just launched !

2009-08-13T06:58:00.000-07:00

We are proud to announce the release of our cloud Open ERP offer: Odoo ! With Odoo, you can get a ready-to-use and complete enterprise management software in a few clicks. Odoo is based on the latest stable version of Open ERP.

Odoo is a complete offer at a unique and low-cost price:
* Open ERP Hosting with high bandwidth and servers,
* Incremental backups servers,
* Software + Infrastructure as a Service,
* Maintenance with bugfixes and automated migrations,
* Open ERP control center, etc.

The subscription to Odoo is free, you pay at the end of the month only if you are satisfied. With Odoo, you pay only what you really use, 0,60€ per hour. Odoo is free for very small companies.

For more information, please check http://odoo.com

2009-07-16T01:27:00.000-07:00

Hello,

Today, Open ERP and his CEO (Fabien Pinckaers) is in front page of the magazine "Trends Tendances", with a 4 page article. (number of the 2009-07-16).

Meet us at Euro Python - Birmingham, UK

2009-06-04T15:21:00.001-07:00

This year, the world's biggest Python event, Euro Python, will be organised in Birmingham (UK), the 28th of June, up to the 4th of July. Fabien Pinckaers, Open ERP's founder and CEO will do two conferences: one on Open Object, one on Open ERP. these conferences are scheduled for the 30th of June and the 1st of July.

The complete time table is available on the Euro Python website:

http://www.europython.eu/talks/timetable/

We will attend Euro Python for these two days. We hope to meet you in UK to discuss Open ERP issues. If you want to organize a meeting with Fabien Pinckaers during this period, contact cde AT openerp.com

API Changes Tracking

2009-06-03T13:11:00.000-07:00

Hello

We just opened a new blog to track all changes in Open ERP that may break brackward compatibility. This blog is used to track all API changes made on Open ERP. The subject must be "module, object, method", like: 'sale, sale.order, _product_id_change'. The content of the blog entry should summarize the modification. Each blog entry should be tagged by the version of Open ERP you made the change on "5.0.1". Don't tag 'trunk', but the latest stable version. to record a change, send an email to: openerp DOT team DOT code AT blogger DOT com.

You can check it here: http://openerp-code.blogspot.com/

Manufacturing Capacity Planning

2009-05-31T15:04:00.000-07:00

Here is a one minute screencast that shows the Gantt chart features of Open ERP and how we use it to manage the manufacturing planning. Enjoy the power of the Open ERP user interface. That's a good proof we can do smart applications with web interfaces today.

You must install the mrp_operations module to do this.

The screencast has two parts.

Part 1: Work Orders Planning

We compute a manufacturing order (a PC2) that has 4 work orders related. Two of these 4 work orders can be launched in parallel and the others must be serialized. We use the gantt chart to planify the operations. Open ERP is smart enough to:

Serialize some operations, recomputing dependencies
Adapt length of work orders according to day/time (because workcenters are configured to work from monday to friday, 8 hours a day)
Synchronize work orders and manufacturing orders.

Part 2: Manufacturing Orders Planning

The phase 2 show the production of a PC0, that has two others manufacturing operations in make to order (PC2 and CPU_GEN). It shows the different dependencies between the orders and how Open ERP computes automatically length, durations and partsdependencies. Of course, work orders on the different work centers synchronize automatically with manufacturing orders.

If you are interrested by logistic or manufacturing processes, I suggest you to have a look at the new Open ERP book.

Open ERP teams - Organisation of the community.

2009-05-10T02:12:00.000-07:00

I am getting more and more emails from contributors or communities that organized themselves to do some contributions, discussion groups or translations on Open ERP.

I will summarize here how we propose to work to easily collaborate and consolidate efforts. If you have a group of people, I suggest to create a team on launchpad, or join an existing one. Don't forget to subscribe your team to the openerp-community team.

Then you should do all your contributions as modules, in the community or addons-extra branch, instead of doing a dedicated branch. We have now 213 active branches on launchpad, with some containing more than 200 modules. It's quite difficult to manage, have a look at contributions and consolidate efforts.

Please also avoid to create projects. Projects are dedicated to very big projects (kde clients, web clients, ...). It's better to create groups and join your efforts in the community branches. Your modules will have more visibility. Our daily generated documentation on modules, since a few days, also reports on community branches. You can check and download our 500 modules there.

Do branch only if:

you want to modify core or modules where you don't have access rights. Then do a request for mergal.
you want to experiment some devs before commiting them to "public" branches.

Our policy is to give commiters rights based on meritocracy. Every one can commit on the community branch. Once you have contributed 2 good modules on community branch, you can request to join the commiter team that have write access on addons-extra modules.

For details about the community organisation, read our community slidecast on OpenERP.tv.

Open Object Introduction

2009-04-28T06:54:00.000-07:00

One of the biggest advantage of Open ERP is that it is built on a very powerful architecture, the Open Object platform. It allows you to customize your applications in a few hours without development or to develop new features very quickly. We just released a small explanation of the

You can get more information and see OpenObject in action on OpenERP.tv.

Designing Applications in a Few Minutes with Open Object

2009-04-17T11:10:00.000-07:00

We just released a new screencast that shows how to quickly design complete applications with Open Object, the Open ERP's framework. In 17 minutes, we develop a complete school management applications, including:

Management of teachers and students,
Management of courses, classes and sessions,
The classes availabilities and courses planning,
The document management system for training material,
Selling and invoicing courses through sales orders,
Statistics about courses.

Check this video to discover the power of Open Object to customize existing applications, or build new ones.

You can get more screencasts on OpenERP.tv.

Open ERP - A New Generation of Open Source Business Model

2009-02-26T06:36:00.001-08:00

Here is a presentation in a few slides of the Open ERP business model for 2009. The goal is to sustain the growth of the product and the quality of the services to satisfy the community, the customers and the partners.

Open Source Business Model of Open ERP

Improved Documentation

2009-02-26T06:21:00.000-08:00

Hello,

We continue our efforts to release a very good documentation on Open ERP. Today, we just uploaded an improved version of our different documentations:

* The user book on enterprise management,
* The developer book,
* The community guide,
* The modules documentation.

You can access these documentations on: http://doc.openerp.com

If you want to contribute in the documentation process, please join the launchpad project: http://launchpad.net/openobject-doc

Community IRC Meeting

2009-02-19T07:28:00.001-08:00

Hello,

We will organise a community IRC meeting every month to discuss all aspects related to Open ERP & Open Objects with our partners, the community and the management at Tiny. The first one will be organised the 10th of March at 2p.m., belgian hour (GMT+1).

We will use the official community IRC: http://openobject.com/irc/
Join us on IRC to start proposing subjects for the next meeting. We will review and post a proposition of subjects to cover. The expected duration is about 2 to 4 hours.

Thanks,

Open ERP maintenance and migrations

2009-02-10T01:46:00.000-08:00

With the version 5 of Open ERP, we setup a second level maintenance contract, that can complete partner's first level maintenance/support contract provided to customer. It's a service contract (not required to use Open ERP) to get some guarantees and to work on a professional way. In these editions, we provide:

bugfixes guarantees : according to the contract level, we guarantee a delay to manage a bug. Within this delay, we assign a full time developer on the bug that will fix it in an average of 4 hours. On previous versions of Open ERP the bugfix was at the charge of the partner (from the support hours in the partner contract). With this new system, partners get an unlimitted and free bugfixes, paid by the final customer. We setup an automated migration system within the server that can automatically download new versions and apply fixes.

Life cycle guarantee : with these new editions, we are able to provide up to 5 years of maintenance on a specific version of Open ERP. It's a
guarantee for the customer they will not have to migrate at each new version. (nobody likes to be forced to migrate often a working ERP system)

Security and bugs alerts: we reviewed all our internal procedures to be able to report to customers (or partners) new security alerts and bugs. We send the report about the security issue or bug with the fix of the problem. The application of the patch or the training of the customer should be provided by partners through their support contract.

Automatic Migrations: the maintenance contract include a guarantee on the automatic migrations from one stable version to another stable version. It already works with current 4.2 versions. We provide you the automatic migration system and the service to help you to migrate if needed.

Please note that these editions are only services. The Open ERP code used is exactly the same between all editions or for the community. We will always continue to use bazaar and launchpad to push what we do to the community.

But I think we can not achieve these 4 services and guarantees without a
service contract. The big change with version 5, is that this effort
(migration, bugfixing, life cycle guarantees) is not anymore at the charge of the partner but the end-customer.

You can compare the different contracts here:
http://openerp.com/index.php?option=com_content&id=8

More explanation

In reaction of a post we had on the forum:
> So I suggest to find a bettter term for "No guarantee" > "Supported"

A 'guarantee' is a service ! The editor can not offer 'support' for free, it's not reliable nor scalable !

It does not mean we do not provide the code or bugfix, it's just mean if you want a guarantee you need a maintenance contract. Because a guarantee is a service.

We, as an open source editor, want to provide everything for free. As a CEO, that's my business but also my passion. But I consider that a service must be paid. No matter of the form of the service. But, everything we achieved through this service is pushed to the open source community for free.

We used to work like this for the code, it will be the same for the bugfixes, the maintenance and the migrations. Of course, we continue to have R&D budgets to develop new features and improve the quality. I am just talking here about urgent bugfixes that are part of our SLA.

I understand it's a big change and people do not like changes. But note that last year, partners were complaining of the invert. Before this, bugfixes requested in an emergency by partners had to be charged in the partner contract. So the reaction was, why should the partner pay to bugfix something.

This maintenance contract sets the basis of a strong model where:

the partner do not have anything to pay (it's the customer that purchase the maintenance)
the editor can organize himself and offer a strong and guaranteed quality of services for the different maintenance aspects, because he is paid for this.

On v4.2, the more we had bugs, the more profit was generated for the editor, because the partners had to pay per hours. (of course, it was not our goal) It was good for the editor but not for the product, nor the partner. That's why we wanted to change this for version 5.

With these maintenance contracts, the more bugs we found, the more it costs to the editor ! Because we sell fixed price contracts, no matter of the number of bugs. Trust me, we will organise ourself to provide VERY good code and quality of service. And that's very good for maturity of the product.

Now, bugs-migrations-security issues will not cost anything to the partner, so they can control their costs more easily on customers projects.

So, the goals of these changes and the version 5 are:

IT'S TIME TO BE MORE PROFESSIONAL AND RELY ON A STRONG BUSINESS MODEL THAT WILL INCREASE THE QUALITY OF THE PRODUCT AND THE SERVICE, WHILE SUSTAINING THE GROWTH TO DELIVER A BETTER PRODUCT TO THE COMMUNITY.

> Automated migrations must be available to the community

I fully agree with you ! Migrations must be available to the community.

But I also consider migrations guarantee as a service and a big cost to GUARANTEE this in the future. So I thought about a mix to satisfy customers (that need guarantee, quality, quick and direct services) and the community that want to migrate for free.

That's why I think migrations should be seen like the 'Shared Funding Projects':

Those that needs a professional environment, a guarantee on the quality of services and code, have to purchase the maintenance contract.
When the maintenances contracts reimbursed the R&D costs of these migrations, we will publish them as open source for the community.

So, I estimated that we will be able to release the migration scripts with 3 to 8 months, according to the success of the editions. If it takes more than 8 months, I will take the costs at my charge and distribute everything online for free. So that the community have a 'guarantee' on waiting maximum 8 months.

This remark is only for migrations scripts and shared funding projects. For all the rest (new developments, bugfixes, ...) we push what we develop directly on launcphad.

New version 5.0

2009-02-10T01:28:00.000-08:00

This post explains all the changes that are related to this new version: partnerships, migrations, new features, etc.

About the Software

I suppose you already know most of the changes that have been applied in the Open ERP software. This new version focused on the following elements:

ACCESSIBILITY & EASE OF USE : the version comes with tooltips on most fields, a complete documentation (1500 pages) linked with contextual help, configuration wizards, pre-configured access-rights, reviewed interfaces, simple/extended mode. This should allow people to start working with Open ERP much more easily.
BETTER INTERFACE: especially with the web client with the drag&drop calendar, the gantt charts, the process view, reviewed look, better dashboards, etc.
TOOLS FOR INTEGRATORS: to allow you to integrate Open ERP by doing parametrization, not develoment: screen/object designer, server actions, workflow editor, etc.
MATURITY: about 1200 bugs have been fixed within the past 5 months.

And, of course lots of new modules and features.

About the marketing

We splitted the (allinone) Open ERP website into two websites, to target different kind of users:

Open ERP : for customers, to promote a product or a solution. You will find on this website: features, online demonstration, success stories, services offers.
Open Object : for the community, to organise collaborative efforts. New things on this website: irc channel, extensive launchpad use, technical keypoints.

These websites are all related to the same code/software. We can say that:

Open ERP is a product
Open Object is a project
On Demand is a service

We think it's important to have a different communication according to the target market. Customers will want to see advantages, ROI, services offers, success stories whereas community will want to see forum, IRC, current bugs, branches of development, ...

We also launched new services:

OpenERP.tv : a place for the community where you can upload your screencasts You can post your blog on blip.tv with the tag openerp, and we will validate it to integrate in the screencasts section.
Evaluation-Matrix.com : A pragmatic and collaborative comparison amongst most advanced ERP on the market.
Doc.OpenERP.com : a 1500 pages of documentation with the following books available: enterprise management (The Open ERP book), the community book, the installation book, the BI book, the modules list, the features list.

We splitted the planet into 2 planet, the goal is to allow every partner to contribute on the openobject planet: technical issues, new modules, customers, announce of events, ... The communication for the Open ERP planet will be done by our marketing team by reviewing, improving and filtering others contents.

About the partnership

The partnership contract has been changed to be more reliable. We kept the same ideas than the preceeding contract but improved on some points, mainly the point (units) system.

We simplifed the unit system to be much more easy and understandable. We got inspired by the ezPublish model. Now, units are computed based on all services you resell to your customer (or buy yourself), making a certified module, maintenance contracts, trainings, etc. This points system is the reflect on your activity on Open ERP:

reselled maintenance contract => number of customers
quality certified modules => number of good modules published
trainings & support => your knowledge on Open ERP

This new contract will take effect at the end of your current contract. Please note that we will be more strict with the partner level: sliver/gold. We will review the partnership level every 3 months, based on your real activity. So, people may increse their level/decrease their level according to their real activity on Open ERP in 6 months. Our goal is to be very fair and apply exactly the same rule to everyone.

The new rule to be starter/silver/gold/platinium are:

Platinium : 5% of the top partners having the most points
Gold : 15% of the top partners having the most points
Silver : partners having more than 1500 points
Starter : new partners

It's better because it scale efficiently with the growth of the partner network. We also re-evaluate partners level once every months.

In this situation, the partner level is less related to the amount the partner purchase but more to the activity and sales the partner generates.

The commission on all our products (maintenance, trainings, ...) has changed too:

before it was 20% on all our products
we increased to 25% splitted like this: 15% direct discount, 10% to be reused to buy others products (maintenance contracts, support, trainings, ...)

The commission is applied on direct purchases, or on purchases made by your customer. If one of your customer buys directly the service on our online shop, contact us to say it's your customer and we will pay you the commission at the end of the month.

About the module certification

In version 4.2, we had lots of troubles due to the quality of the modules published. Some where not fully finnished (proof of concepts), some where made by contributors and not maintained efficiently, ... It was a bad image for the product. We also get requests from customers and partners to get evaluation and feedback on their own modules.

The certification is based on a procedure of 150 technical and functionnal tests and reviewed by 2 people from our quality team. After the certification:

We promote your work through a news on Open ERP
We tag your module as quality certified in the module list
We send you technical and functionnal feedback
We include it in maintenance contracts (for corporate contracts) or send you a yearly proposition for the maintenance of the module (migration to new versions, bugfixing, ...)

It allows you to have a professional review on the quality of a module to avoid:

Complexities to migrate the module for futur version of Open ERP,
Having a module that do not fits Open ERP standards and is not used by the community,
Discovering bugs and problems once in production,
Queries that slows down the user interface when going on large production environment, etc.

Once the quality of a module is certified, we offer second level maintenance contracts on this module to ensure module lifecycle during years, migration and bugfix guarantees.

The advantage of using certified modules are:

They are bugfree and the editor provides bugfix guarantee contracts,
You will not have migrations problems for next versions of Open ERP,
They are scallable and have been tested in production environments,
We made stress tests to check they run with big databases,
They are used by the community, so they will evolve and others will contribute.

The price to certify a module is 800 EUR, but we provide discounts for big projects, having lots of modules. Note that you also have 25% discount on modules certification.

Of course, that's not something you should buy. It's better to resell it
to the customer as a quality label.

Open ERP v5 is out !

2009-02-07T08:55:00.000-08:00

That major enhancement of Open ERP can now answer to all the needs of a business. Open ERP V5 not only provides management functions, but also all functionalities necessary to a SMB, like : a process management by modules, a wiki, a webmail, a Business Intelligence (Cube OLAP), a Document Management System, an eCommerce, an idea box, etc.

Emphasis was placed on an extreme simplification of the software for the new users, a carefully designed ergonomy of the web client with, amongst others, drag&drop, Gantt graph, editable processes, etc., and more then 350 modules for specific sectors and big companies.

This new version comes with a full review of the web site giving access to more then 1500 pages of documentations on business management and a reorganisation of the community sources build upon the Open Object framework. Free cycles of conferences are planned with the version 5.0 release of Open ERP.

Unmatched functions

Thanks to its huge community, Open Object produce more then 20 modules a month. The Open Object community it is more then 1000 contributors, 126 development branches in parallel, an average of 400 new functinalities or bugfix per month, one commit every 5 minutes and functional and technical experts specialized by activity and working in teams.

The rise of Open Object and the diversity of the projects makes it an unmatched framework composed of more then 400 modules installable in a few clicks to cover all kinds of need or to simply start, with a simple module, to answer a simple need. Then, you can install other functionalities to come to a fully integreted and automized system.

Open ERP V5 is characterized by the appearance of many functionalities far beyond the perimeter of traditional management.

One can underline the following innovations:

A integrated wiki.
An integrated document management system.
A Business Intelligence (BI) using a OLAP database.
An integrated BPM (management of process).
A web portal for clients and suppliers.
Improvement of translations (1 translation file by language and module).
A touchscreen point of sale.
A full Ajax webmail .
A shared calendar.
Plugins for Outlook, OpenOffice, ms. Office, Thunderbird.
An integrated eCommerce, etc

This new release offers 3 user interfaces :

the rich application client for a day to day advanced use,
the web interface allowing a remote access and an easy deployment,
the QT client that perfectly fits in a KDE environment.

Numerous improvements have been added to the client interfaces, like :

dynamic graphs and dashboards,
really ergonomic calendar views,
dynamic Gantt graphs for planning,
workflows editors,
a fully integreted documentation,
a dynamic process view used for the end-user training,
etc.

The web version of Open ERP includes numerous functions to modify or create your own application :

an visual view editor,
an object editor,
a workflow (process) editor,
an Open Office integrated report editor
a statistics engine (BI cube),
etc.

Restructuring websites

Open ERP have made a full review of its web sites for a better structure and organisation of the ressources and community projects :

- http://OpenERP.com : there you will find the different functionalities, the documentation, the Open ERP partners and the services proposed by the editor.

- http://OpenObject.com : Open Object, the Open ERP's framework, where is found all the community sources :

documentation,
Modules list,
Forum, IRC, Bazaar.

A new IRC canal has also been opened for the community : irc://freenode.net/openobject

- http://ondemand.openerp.com : The « On Demand » offer, developped by Open ERP and Axelor. This is a « ready to use Open ERP » including hosting, maintenance and support.

Furthermore, Open ERP provides some new services to the community :

- http://doc.openerp.com : 1500 pages of documentations covering all domains of business management or about the developement under Open Object.

- http://OpenERP.tv : Screencasts dedicated to Open ERP consisting of presentation made by the contributors

- https://launchpad.net/openobject : The development plateform for Ubuntu, gathering all the developments of Open ERP and Open Objects. There is found the bugtracker, the FAQ, the ressources management, translations, etc.

Free Softwares overtakes proprietary softwares on the management softwares market

Open ERP V5 takes an unquestionable lead on its larger competitors like Microsoft Dynamix et SAP. Two very thorough studies have compared Open ERP with its proprietary and free competitors, showing an advance of several years for Open ERP:

The ERP Open Source white paper by Smile : A study carried out over 6 months by experts of Smile.
Evaluation-Matrix.Com: an community approach with a pragmatic evaluation of the management software company.

Need to know more ?

Free presentations and trainings are organized everywhere in Europe, in order to present Open ERP V5. The next dates in France or Belgium are:

On February 16: Technical training and Open conference ERP in Grand-Rosière, Belgium.

On February 20: Free Open ERP conference in Paris, France
On February 25: Free Open ERP conference, Switzerland.
On March 02: User Training (session 1) in Grand-Rosière, Belgium.
On March 04: Free Open ERP conference in Grand-Rosière, Belgium.
On March 23: User Training (session 2) in Grand-Rosière, Belgium.
On March 25: User Training (session 3) in Grand-Rosière, Belgium.
On April 01: User Training (session 4) in Grand-Rosière, Belgium.
On April 05: Technical Training in Grand-Rosière, Belgium.

For any subscription, please contact Donia : sales@tinyerp.com

The Open ERP effect

2009-02-04T12:05:00.001-08:00

I just made a small presentation that explain what's Open ERP.

The Open ERP effect

Importing data in Open ERP

2009-01-14T04:20:00.000-08:00

There are different methods to import your data into Open ERP:

Through the web-service interface
Using CSV files through the client interface
Building a module with .XML or .CSV files with the content
Directly into the SQL database, using an ETL

Open Object's ETL

The next version of Open Object will include a complete ETL system with components to allow you to easily Extract, Transform and Load data from any application to Open ERP.

This project will be released in about 3 months. If you are interested to contribute to this project, you can check the following branch on launchpad:

bzr branch lp:~openerp-commiter/openobject-addons/trunk-extra-addons/
directory: etl

We already made a big part of the specifications and a working prototype to convert and transform data from .CSV files. The ETL will be based on a generalisation of the workflow editor of Open ERP to manage any kind of jobs.

Waiting for the ETL...

The best way to import data in Open ERP is to build a module that integrates all the data you want to import. So, when you want to import all the data, you just have to install the module and Open ERP manages the different creation operations. When you have lots of different data to import, we sometimes create different modules.

So, let's create a new module where we will store all our datas. To do this, from the addons directory, create a new module called data_yourcompany.

mkdir data_yourcompany
cd data_yourcompany
touch __init__.py

You must also create a file called __terp__.py in this new module. Write the following content in this module file description.

{
'name': 'Module for Data Importation',
'version': '1.0',
'category': 'Generic Modules/Others',
'description': "Sample module for data importation.",
'author': 'Tiny',
'website': 'http://www.openerp.com',
'depends': ['base'],
'init_xml': [
'res.partner.csv',
'res.partner.address.csv'
],
'update_xml': [],
'installable': True,
'active': False,
}

The following module will import two different files:

res.partner.csv : a CSV file containing records of the res.partner object
res.partner.address.csv : a CSV file containing records of the res.partner.address object

Once this module is created, you must load data from your old application to .CSV file that will be loaded in Open ERP. Open ERP has a builtin system to manage identifications columns of the original software.

For this exercice, we will load data from another Open ERP database called old. As this database is in SQL, it's quite easy to export the data using the command line postgresql client: psql. As to get a result that looks like a .CSV fiel, we will use the following arguments of psql:

-A : display records without space for the row separators
-F , : set the separator character as ','
--pset footer : don't write the latest line that looks like "(21 rows)"

When you import a .CSV file in Open ERP, you can provide a 'id' column that contains a uniq identification number or string for the record. We will use this 'id' column to refer to the ID of the record in the original application. As to refer to this record from a many2one field, you can use 'FIELD_NAME:id'. Open ERP will re-create the relationship between the record using this uniq ID.

So let's start to export the partners from our database using psql:

psql trunk -c "select 'partner_'||id as id,name from res_partner" -A -F , --pset footer > res.partner.csv

This creates a res.partner.csv file containing a structure that looks like this:

id,name
partner_2,ASUStek
partner_3,Agrolait
partner_4,Camptocamp
partner_5,Syleam

By doing this, we generated data from the res.partner object, by creating a uniq identification string for each record, which is related to the old application's ID.

Now, we will export the table with addresses (or contacts) that are linked to partners through the relation field: partner_id. We will proceed in the same way to export the data and put them into our module:

psql trunk -c "select 'partner_address'||id as id,name,'partner_'||partner_id as \"partner_id:id\" from res_partner_address

This should create a file called res.partner.address with the following data:

id,name,partner_id:id
partner_address2,Benoit Mortier,partner_2
partner_address3,Laurent Jacot,partner_3
partner_address4,Laith Jubair,partner_4
partner_address5,Fabien Pinckaers,partner_4

When you will install this module, Open ERP will automatically import the partners and then the address and recreate efficiently the link between the two records. When installing a module, Open ERP will test and apply the constraints for consistency of the data. So, when you install this module, it may crash, for example, because you may have different partners with the same name in the system. (due to the uniq constraint on the name of a partner). So, you have to clean your data before importing them.

If you plan to upload thousands of records through this technique, you should consider using the argument '-P' when running the server.

openerp_server.py -P status.pickle --init=data_yourcompany

This method provides a faster importation of the data and, if it crashes in the middle of the import, it will continue at the same line after rerunning the server. This may preserves hours of testing when importing big files.

Wizards are dead, long life to memory objects !

2008-12-09T13:07:00.000-08:00

Here is a small note that describe briefly the osv_memory system that will replace wizards in Open ERP v5. It allows lots of things that were difficult to implement using wizards (inheritance, workflows, complex relation fields, computed fields, all kind of views (tree, graphs, calendar, gantt, etc.) But the best of all, is that it is extremely easy to develop as it's exactly the same than normal objects and views in Open ERP.

To develop osv_memory wizard, just create a normal object, but instead of inheriting from osv.osv, inherit from osv.osv_memory. The methods of the "wizard" are methods on the object and, if the wizard is complex, you can define a workflow on this object. It means the object is managed in memory instead of stored in postgresql.

That's all, nothing more than just changing the inherit.

So what makes them looks like 'old' wizards ?
* In the action that opens the object, you can put <field name="target">new</field>. It means the object will open in a new window instead of the current one.
* On a button, you can use <button special="cancel"> that makes the window close.

Some remarks:

If the wizard has different screens, you have different approaches:

putting all fields in a view but displaying groups/page according to a state field (invisible or attrs arguments). Then you just have to write the state field of your object to change the view.
using several objects for each step

As a reminder, you can trigger actions that opens object from:

buttons in forms/tree
menus
action in the right toolbar of an obejct (and contextual menu on relation field)
workflow steps (new in v5)
wizards

Of course, you can use this on normal objects or osv.memory objects. You can also use the object builder, the view designer or the workflow designer. It also works well with the module recorder and the new module builder. (I really suggest you to test this latest one, which is really powerfull, it allows you to save all customizations in a module, after having done the customization.)

And, if you add server actions in the system, you can customize and build wizards without any line of development, just by customization through the user interface ! It still has to be documented, but v5.0 will provide really really RAD tools.

All new wizards in v5 are implemented using this way. Check, for example, the configuration "wizard" in addons/profile_service.

Open ERP Books

2008-11-04T16:03:00.000-08:00

Hello,

The second book about Open ERP is on the way. We can already announce the table of content, in french as the first version will be in french:

Part 0 - Introduction
Introduction : 5 Pages

Part I - Premiers Pas
Installation : 20 Pages
Présentation des Fonctionnalités : 15 Pages

Part II: Le Jeu de Rôle (Business Game)
Cas d'utilisation: 15 pages
Gestion et Stratégie: 15 pages

Part III - Logistique et Production
Stock Management : 50 Pages
Manufacturing Mgt : 40 Pages

Part IV - Gestion des Ventes et Achats
Gestion des Ventes: 30 Pages
Gestion des Achats: 15 Pages

Part V - Processus & Gestion Documentaire
Gestion des Processus : 15 Pages
Gestion électronique de Documents : 15 Pages

Part VI - Comptabilité et Finance
De la facture au paiement : 20 Pages
Les écritures comptables : 20 Pages
Configuration : 20 Pages

Part VII - Administration du système
Administration : 30 Pages

Part VIII - Conclusion
Conclusion : 5 Pages

We expect to release it in about 2 to 3 months. A new book on the technical aspects of Open ERP is also on the way, in french and english. More information will follow soon.

Open ERP 5.0 - CRM Focus

2008-11-03T14:56:00.000-08:00

Hello,

Here is a small screencast that shows the improved CRM features of Open ERP 5.0. The CRM module comes with a configuration wizards that allows you to pre-configure the system automatically, asking you some questions so that it fits perfectly your needs.

As an example, the above screencast shows the use of the leads and opportunities. It start by showing a lead definition and creating a partner from this lead. From this partner, we schedule a meeting in the calendar.

After this, we come back to leads and generate an opportunity based on a lead. The wizard helps you to convert the lead and assign sales informations like the expected revenues. Then, we use the process management system to describe the flow of the selected opportunity.

After this, I connect through FTP to the document management system of Open ERP. The DMS has been connected to partners, so that I can browse partners by categories from the FTP access. You can browse any document of your system using this FTP connection. That's very useful to provide limited accesses to your customers to some documents of the system. It's automatically configured if you install the portal modules.

I finished by showing our new calendar synchronisation system that can be integrated with your mobile phone, outlook or any calendar system. In this example, I connected my meetings with Sunbird, the calendar of the Mozilla project.

You can also check the full size version of the screencast.

Please note that the 5.0.0-alpha version has been published today. You can download it from here: http://openerp.com/downloads.html.

When SAP meets Open ERP !

2008-10-08T10:11:00.000-07:00

Last week, we went to the meeting Solutions ERP in Paris (CNIT, La Defense). It's the biggest meeting related to enterprise management software in France. It was a very big success, where we got about 200 customers requests in three days.

It was quite funny because the Open ERP booth was just near the SAP and the Cegid one. The following photo explains well the situation: during these three days, SAP salesman saw the back of the visitors, while they were looking at Open ERP :)

Open ERP on the left - SAP on the right.

I thanks our partners that were present with us: Network Assistance, Axelor, Smile and Syleam. You can get more photos here.

PS: We are working on a SorrySAP.com campaign, any suggestion is welcome. Send them to me by email.

The OpenERP English Book is Out !

2008-09-26T09:14:00.000-07:00

The english book on OpenERP has just been released as a PDF:

Open ERP : a modern approach to integrated business management based on a free Open Source software system.

This book describes how to manage a services-intensive company effectively when assisted by a modern integrated Open Source software system.

Open ERP version 4.2 is used to illustrate a range of processes from a financial accounting base, through management accounting and comprehensive Customer Relationship Management, to HR and Project Management. All processes are completely integrated in their relationships and impact on each other.

Open ERP is free Open Source software, giving its users complete freedom to meet their own needs in their own ways. It has been designed with a robust and flexible architecture, and is built using modern resilient high-productivity technologies. This book provides guidance on the implementation of the Open ERP system - its installation and configuration to meet business needs.

You can purchase it on OpenERP's website.

Here is two samples of chapters:

Corporate Intelligence

2008-09-24T08:04:00.000-07:00

La version 5.0 de OpenERP approche à grands pas. Afin de célébrer cette nouvelle version et, pour aller de pair avec l'intérrêt croissant que remporte OpenERP dans l'éducation, nous y avons intégré les meilleures pratiques de gestion sous forme de processus utilisateurs dynamiques et configurables.

Ceux-ci vont permettre aux utilisateurs d'obtenir une vue macro de l'entreprise et de pouvoir facilement se situer par rapport à un flux de gestion transversal. (contrairement aux workflows actuels de OpenERP qui sont verticaux et par documents) Sur chaque document, vous pouvez appeller les processus de gestion qui vont décrire le comportement du système en fonction de ce que vous y encodez.

Un screencast est disponible ici:

http://tiny.be/download/flux/process1.ogg

Outre cette nouvelle fontionnalité technique, la version 5.0 de OpenERP sera packagée par défaut avec des processus prédéfinis pour l'ensemble des fonctions de gestion. En plus des ces flux proposés, par métier, nous allons y intégrer des manuels qualités par secteur d'activité et un lien direct avec les nouvelles documentations OpenERP basée sur nos livres.

Pour découvrir en avant première le fonctionnement de ce système, vous pouvez consulter sa documentation ici: http://tiny.be/download/flux/process.pdf

Vous pouvez également télécharger la version trunk de OpenERP sur notre système de développement collaboratif Launchpad. Le nouveau moteur de gestion des processus utilisateurs y est déjà intégré.

Reviewed Project Management

2008-09-09T09:56:00.000-07:00

In our process to simplify and improve OpenERP for the next version, we reviewed most screens and features. As an example, here is an explanation of the changes applied in the project management module.

First, we changed the way you estimate time on tasks. You don't anymore estimate the "% of completion" of the task but the "remaining time", which is much more easy to do. You start a task in draft by filling in the Planned Time. Once the task is open, you can not anymore change this time. The "Remaining Time" is automatically set to the planned time at the beginning and decrease when you encode your work on the task. Different factors are automatically computed based on these values: total time, delay, progressbar, time spent.

The above screenshot is in "Simplified View", if you use the expert mode, you will have more features and tabs on the task.

In the list view, we changed the colors assignation:

red: the task deadline is over-passed
blue: you have to re-evaluate your remaining time as you worked more than the expected value
grey: the task is closed/canceled

We also added a calendar view to the GTK client like in eTiny, so that you can browse your tasks directly in a calendar view. (still in prototype) And, to better organise priorities, we implemented drag & drop of tasks to change their priorities/sequences.

A system of delegation of tasks and dependencies between tasks has also been added for a better control of the delegation chain.

The unit of measure is now fully configurable per company so that you can work in hours/days or weeks. Screens change automatically based on the configuration you put in the company.

Of course, your project management module remains fully integrated with the others components of the ERP:

Complete integration with timesheets and analytic accounting for financial and cost control,
Integration with the company planning system,
Dashboards and statistics for project managers,
Human Ressource system for availabilities,
We plan to integrate an OpenObject Wiki for description of needs in the task,
Getting Things Done module system for personal time organisation,
Scrum methodology for IT companies, ...

We also improved the integration with sales order that can now:

triggers tasks automatically,
invoice task based on sale order quantities or work on tasks,
email to customer and request to project manager,
planify the dates and responsibles on the task.

Ergonomy Improvement

2008-08-24T16:40:00.000-07:00

We continue our ergonomy improvement process. Here are the latest news:

We have a completly reviewed our wizards to help users to configure their OpenERP system. We have wizards for all profiles: association mgt, auction houses, crm only, accounting company, services mgt, manufacturing industries. Here are two screenshots:

After the new date widget, released a few days ago, we implemented a new widget in the GTK and Web application: progressbar. They are available in form and in tree views. Here is how it looks like in sale orders list.

We put Question Mark at the left of labels for fields that have a tooltip.

We made some layout modifications in GTK:

1. We moved the tab position of all notebooks to top by default in the GTK application. The idea is to be compliant with the web client and allow a better to rendering of views where you put some data above the notebook, see attachment for example of the new product form.
2. Grey background for toolbar to emphasize it
3. Separator and O2M titles in Bold

We are reviewing all views according to the one we made with products, bellow. Note that with the new attrs feature, all fields related to Weights are readonly because we selected a product of type service. (it changes dynamicly if you change the type of the product).

We are looking for ideas to improve the current internal request system of OpenERP. Please fill in a blueprint on launchpad if you have good suggestions. Any other suggestion is welcome too :)

Modules for next version

2008-08-24T07:47:00.000-07:00

Hello,

We made some changes in modules that will be delivered with the next version. Some modules where in addons (automatically included in the main package) and will be set as extra_addons (to download separetly): multi_company_account, product_extended, product_variant, productivity_analysis, purchase_discount, sale_category, sale_rebate, sandwich, travel

The following module will be in the base distribution (addons) instead of extra_addons: account_analytic_plans, account_balance,, account_budget_crossover, account_date_check, account_invoice_layout, account_reporting, association_vertical, auction, base_contact,, base_module_record, base_report_creator, base_report_designer, board_association, board_auction, board_crm_configuration, board_document, crm_configuration, crm_profiling, crm_vertical, event, google_map, hr_holidays, idea, membership, mrp_jit, mrp_operations, point_of_sale, profile_association, profile_auction, profile_crm, project_gtd, project_retro_planning, project_timesheet, purchase_analytic_plans, sale_analytic_plans, sale_crm, sale_journal, invoice_payment_tab, project_mrp, stock_back_order, stock_invoice_directly

The base modules are growing mainly because we are adding new profiles by default, which are association management, crm profile and auction profile.

Latest Changes in the Framework

2008-08-21T06:18:00.001-07:00

A quick note to list some changes made in the framework in the trunk version:

1. To import big files, you can use the option -P when running the server:
tinyerp-server.py -Pa.pickle -dtrunk -i module
It will commit every 100 lines to not slow down the system when importing very big files. If it crash in a big .CSV or .XML file, you can rerun the query above and it will continue after the crash. Very practical to make tests on big imports and fix problems in your initial data.

2. Properties
Properties are not anymore added automatically in views. You have to inherit forms and add the properties in the form like any other field. The tag in views does no exist anymore. I changed all addons module, but you will probably have to modify your own module in this direction.

3. Nested Trees
For objects with tree structure, you can define _parent_store='M2NFIELD' on the object. It will compute parent_left and parent_right automatically so that you can compute on a whole tree in one SQL query instead of doing very slow recursion algorythms. More infos here:
http://www.intelligententerprise.com/001020/celko.jhtml

Use this to speed up all computations on tree structure. For instance, to compute the recursive debit and recursive credit of the account having a parent_left=3,parent_left=22 and all his childs, you can do that in only one query: select sum(l.debit),sum(l.credit) from account_move_line l left join account_account a on (a.id=l.account_id) where a.parent_left>=3 and a.parent_left<22;

4. Function fields
Sometimes it's interresting to compute several function fields using one call and not one call per field. For instance, for speed improvement, we should be able to compute fields debit,credit and balance at the same time (same SQL query), which was impossible before. Now, we can do that with the multi argument on fields.function. So several functions fields will be computed at once if they share the same multi argument value. Check how debit,credit and balance are computed with only one function in addons/account.py

GTK Ergonomy

2008-08-20T14:43:00.000-07:00

Lots of improvements have been made on OpenERP to improve the ergonomy of both our Web and GTK client. Here is a small post on the latest one: date, time and datetime widgets improvement.

In the trunk branch of the GTK client, these new widgets uses masks that are set according to the locale of the connected user. Thanks to these masks, the user better knows how to encode these fields and can not encode something false. You don't have to write the separator character, OpenERP does it for you.

You can also fill in only a part of the widget and OpenERP completes automatically the rest of the field according to the current date and time.

But the best new feature of this widget is the ability to use operations on the date. That will be very helpful for accounting entries encoding or planning. (thanks Camp2Camp for having suggested and worked on this feature). When you press '-', '+' or '=', the system enter in the operation mode and what you write appears on the right until you press tab, enter or escape.

Some examples of operations:

+2w : adds 2 weeks to the selected date
= : set to current date/time
=12w : set to monday from the 12th week of the year
-3d : decrease selected date from 3 days
- : clear the date field
+3m : adds 3 months to the selected date

Reviewed Documentation Process

2008-08-18T05:48:00.000-07:00

The documentation of OpenERP is one of our biggest priorities. We currently have the best documentation amongst all open source ERP but it's not enough and I am sure we can do much more better.

We already invested a lot in writing documentation and the result is not as good as we were expected: the software is growing very quickly and writers have lots of difficulties to track changes and write quality documents.

Writing a Good Documentation is Complex

For Years, we have try to make a community based documentation on a wiki where everyone contribute. With OpenERP, we have a strong community but, after years, the community documentation does not reaches our expectations:

Deprecated compared to new versions of the software,
Original authors do no maintains what they write / contribute,
Seems like if we assembled different pieces of unrelated documentations,
Sometimes contributions comes from people that do not understand perfectly the software,
Not end-user oriented,
Some wants to write in English, others in French, Spanish, eso.
Some have experience only in food industries, others services companies, others manufacturing, eso.

We also tried to make documentations written by our own developpers. The idea was that each time a developer finish a module, he writes the documentation about it in the wiki. It did not work too because:

Developers have lots of priorities and sometimes a hurry customer is more important than writting the documentation in the wiki,
Writting a good documentation on a part of the software can easily take 100 hours, most of the developers do not have such a time,
Some developers do not like to write documentation so the work does not reaches a very high level of quality,
Most of the time, a developer develop a module for one customer only, so the documentation will be oriented based on one case only.
A good documentation has to be written by functionnal people that have a strong experience on how they use a particular module in production,

Having said that, we decided to review the way we will write and maintain futur documentations about OpenERP and OpenObject.

The New Process

I had a look at the open source software with the best documentations. Most of these documentations have been fully written in one try by only one or two people / author. After that, the community improved the content with small improvements from here to there.

So I think we'd better have to focus our ressources on some good individuals rather than a global community. But writting a complete documentation (or a complete chapter) can take a long time (we took about 200 hours to write the OpenERP book in french, 280 pages). So we need incentives as to motivate these contributors / authors.

So, we decided to manage all documentations about OpenERP through real books. We will setup a real editing chain with the following actors:

The editors: will work with authors to review the quality of the book and will distribute in a range of countries through webshops and real bookstores. The editor also define authors contracts and the graphical charts and style. He will manage the collections (set of books).
The authors: everyone can write his own chapter(s) and propose them to the editor. The editor will make a selection of chapters to publish complete books.
Re-Readers: will re-read books before publication and work with the editor to achieve a high level of quality.
Translators: will translate a particular book to their mother tongue based on the english version.

Editors will be different according to the coutries. For all french countries, we already contracted our preceeding books with Eyrolles. For the english version, we will create our own edition company as to keep the rights on the text to be able to publish it online for free.

Now, we have our incentives for authors and translators:

Getting a monthly revenue through author rights based on monthly sales,
Being published in bookstores with our worldwide distribution channel.

Having established that, this should leverage individuals effort and promote contributions on writing good documentation. As people will be published and paid for their good work, this should promote new authors and translators.

Books are sold in bookstores in different countries and languages, this allow us to remunerate the full edition chain through author rights. It will help us to motivate authors and translators to write more chapters, and improve existing ones.

We had to find a good middle point between selling through bookstores to remunerate authors and publishing online to provide a free version. So, most of our books (depending on the contract we have with the editor in the related country), will be published online some months after their official publication.

Our goal is to create lots of different books for different domains: technical books, marketing management, services companies management, openerp for accountants, ...

Books writing will be our main documentation process. At each new version of OpenERP, we will release new versions of the different books. The editor will organise this and will get revenues for this.

In addition to this, we use a community wiki for all others documentations that are not directly related to the software or for documentations that are "per module". For our main knowledge management system, we use Launchpad Answers, where you can ask questions or reply answers.

Current Status

Lots of things have to be set up to be efficient in this editing chain. As a proof of concept, we wrote the first book with Eyrolles: OpenERP, pour une gestion d'entreprise efficace et intégrée. It was a great success, they expected to sell their printed stock in 18 months and they had to reprint after only 2 months !

This book is already translated to english and some chapters are being added. We plan to release it within maximum 2 months, at the same time of the next OpenERP major version (5.0). We also find translators to translate it in 6 others languages. We are looking for editors in countries we don't have a distribution chanel yet.

Some others books have started being written or are in the pipeline:

OpenObject: build or customize application quickly, without developments
Office Productivity: Business Intelligence, Document Management System, Email Integration, Word/Excel plugins, ...
Pre-Sales & Marketing: based on our new CRM and direct marketing modules

These very firsts books allowed us to setup the rules for clean and systematic processes: defining stylesheet and guidelines, reviewing process, role of the editor, contracts for author rights, findind printing and distribution channels, ...

So the next phases are:

Release the english book
Create the edition company and hire people within it
Define all the processes in details
Translate the english book and contracts with authors/translators
Work on futur books

The Role of The Community

This documentation writing and maintaining process tends to emphatize individuals efforts from authors and translators. This does not means we think the community is less important for writting documentations. The community remains very important for different reasons:

The community provides authors and translators because everyone can request to be an author for one or several chapters, no need to write a full book.
The editor needs readers that will review the book before his official release.
These readers will be choosed amongst the different contributors in the community.

To improve the quality of books versions after versions, we will extensively work on erratas in the wiki. Community will be able to propose improvement through the errata section.

You can get detailed information about the process in the wiki.

Contact

Contact me if you are interested to contribute in one of these aspects: being an author, being a translator, setting an edition company, ...

Accounting Speed Improvement

2008-08-18T00:17:00.000-07:00

We had some requests to improve the speed of OpenERP in the accounting system so that it can easily work with more than one million of entries.

So we did some improvements in the trunk, to speed up the computing process. Here are the result of computations in the current trunk, it's about 20 times faster than the current stable version.

Computing 539k entries, spread in 915 accounts and 24 open periods:

Number of accounts: 915
Number of account.move.line: 539096
Computing Balance for Root: 5.83 sec
Computing Debit, Credit and Balance for Root: 5.82 sec
Computing Balance for all accounts at once: 7.07 sec
Computing Debit, Credit and Balance for all accounts: 7.31 sec
Avg computing Debit, Credit and Balance for one account: 0.22 sec

1078k entries with half reconciled, filtering on non reconciled:

Number of accounts: 915
Number of account.move.line: 539096
Computing Balance for Root: 9.72 sec
Computing Debit, Credit and Balance for Root: 9.73 sec
Computing Balance for all accounts at once: 11.17 sec
Computing Debit, Credit and Balance for all accounts: 11.19 sec
Avg computing Debit, Credit and Balance for one account: 0.37 sec

1078k entries (=1 million), no filter:

Number of accounts: 915
Number of account.move.line: 1078192
Computing Balance for Root: 29.56 sec
Computing Debit, Credit and Balance for Root: 29.07 sec
Computing Balance for all accounts at once: 30.70 sec
Computing Debit, Credit and Balance for all accounts: 30.59 sec
Avg computing Debit, Credit and Balance for one account: 0.67 sec

It has been tested on my personal laptop: Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz. This is the time without any pre-computation, cache or stored values. Some partners made modules that pre-computes some data (for instance balance by account/period/state) to speed up this process.

This trunk version does more computation than the stable one because we compute the recursive debit and credit for each account ! Means the debit of an account is equal to his debit+debit of all his childs.

On this new version, you can note:

computing only the debit of an account or the debit+credit+balance is the same computation time !
computing the root account or computing all the accounts (including the root) takes nearly the same time !
I don't understand why, with postgresql, when I double the number of entry, it takes more than double of the time. I still have to investigate.

I find another way to win some seconds in some cases but did not had the time to implement it yet.

The conclusion:

With the new system, all data can be computed for all accounts in one query if the report is well structured. As we don't use pre-computed data, you can use any filter available in the indexes without decreasing performance. So the slowest report for 500k entries will take 7.31 seconds to compute the whole report. On a professionnal server with a tweaked postgresql, I suppose you can decrease this up to 2 seconds for 500k and 10 seconds for a complete report with one million of entries. This is quite fast and the delay is acceptable for a report that computes on all entries.

For browsing the account chart in the tree, it's acceptable too. The first one is slow (up to 10 seconds for the first node with one million of entries) but all others are quite fast. We plan to improve speed of loading the account chart by storing some values like the balance by account, so that it's immeditate, either with very big databases.

It's commited in the trunk on launchpad.

We plan to do the same review and work on the stock management system and production system.

KTiny - Announcing New Features

2008-08-14T08:57:00.001-07:00

As a reminder OpenERP has three available clients you can use depending on your needs:

The Official GTK Applicative Client
The Official Web Client
KTiny : A quite new but promising one based on QT

Here is a news I rely from NAN-Tic about the new features of KTiny:

Although they haven't made an official release yet, application stability and features are progressing steadily. Here are some of the features you can find in trunk in the subversion repository (note: I hope they will join us and use launchpad for the developments of KTiny) :

Stylesheets: style your application however you like or put your customer's logo everywhere Wink
Full text search
Store client behaviour settings in the server, on a per role basis. Some of the settings include: number of preloaded rows in list views, possibility of loading rows on demand on those views, that means no more LIMIT / OFFSET with reasonable performance (though it has the drawback of not being able to sort by function fields)., Configuring tabs positions, Show/hide system tray icon,Show/hide toolbar

Store application state information. This means it remembers (per user) column order and width and the sorted column in tree views, and sliders width in forms (dashboards typically).
Faster form loading by use of a client cache.
System tray icon with a popup menu giving access to most frequent actions.
New charts with better design and faster loading (hbar still missing though).
New (very preliminar/proof of concept) calendar view.
New web widget that allows embedding maps in contacts, for example.
Indexable attachment information, including OCRing images.
Extensible attachments window
Right clicking in an attachment allows opening with default desktop application or open image (if it's an image, of course).
Context menu actions for setting current date / time for those widgets.
Input of formulas in numeric integer/float fields

Note that it's still a development version so it has some known issues but give it a try! We're open to comments and contributions!

You can see some screenshots at http://www.nan-tic.com

PS: Take a look at the shortcuts dialog, you'll find some useful tricks there!

We also advanced a lot on the GTK client, some new features:

Improved binary widget
Improved ergonomy in editable lists
Web Browser widget (thanks to Eduard Carreras i Nadal)
Configurable launcher per application type
Attachment visible
Preformated date format

End-Users Workflow : Processes

2008-07-14T16:04:00.001-07:00

For the new version of OpenERP, we are working on user-friendly and dynamic workflows. These processus will serve as an help system available on any object to understand the dynamic of the object. These workflows will be configurable by the integrator of the system for any database.

The following screenshot is the very first draft of this new processus system. On each node, the user will be able to click on "Open Form", "Print Document", "Contextual Help". And, on arrows, he will see available buttons for this status and rôles required to continue in the flow.

The goal is to help people understanding how to use OpenERP by showing the document which is being used in his global context and processus. They are quite different from the current workflow system that are used by developpers and that are very technicals.

For example, in this new processus system, you will find things like "Phone to Marc", "Write in the Agenda", "Request a Validation from Your project Manager", ... And, as the current workflow engine of OpenERP is per object, these processus will be transversal to the whole system: lead -> quotation -> sale order -> delivery -> invoice. You will be able to make documents evolves, simply by clicking on arrows.

We are also working on different modules to show the processus according to different enterprise sectors. For example, in industries, you will see processus like: SO -> Delivery -> Invoice from Delivery, and in services the processus will look like SO -> Advance Invoice -> Invoice From SO. The configuration wizards at installation will help you to configure the system (for instance: invoice based on delivery or sale order) and it will install the right processus according to your needs.

Of course, to be efficient, these processus are linked to the workflow engine of OpenERP, so that both are dynamic and evolves at the same time.

The great thing is that above graph is generated automatically by OpenERP. So, you don't need to design them, they are fully integrated in the system and fully dynamic. Currently it produces PNG with tooltips and imagemap but we are thinking of producing a Flash or SVG for a better ergonomy.

One note more: we designed a graph rendering routine in pure python that will be able to replace graphviz. So in a near future, we will be able to remove our dependency against graphviz and pydot. That's not a bad thing... We used the same code for our workflow editor in etiny.

Any idea on how to improve help is welcome...

Community Days' Feedback

2008-07-02T07:33:00.000-07:00

Last friday we organised our annual community days. The goal was to prepare OpenERP for the next stable release. These two days where extremely productive and lots of very good things have been done. I'd like to thanks all people that attended the meeting and contributed to build the best enterprise management software ever made.

Lots of decisions and actions have been made during these two days. Here is a summary of the biggest by workshop.

Community & Quality Processes

We reviewed all our community and quality processes for an efficient work
to leverage and benefit from the best of each actors (community, partners, editor). A summary page has been set up here This page will be used when we will announce the transition.

The biggest changes:

Setting Bazaar to allow contributions
Selection of commiters
A new bugtracker and community portal
A new system to manage modules through the wiki

You can get more details in the new wiki, section community.

Our system admininstrator is installing the new servers and we are trying to set this new process as soon as possible (2-3 weeks). Things that have already
been done:

Writing pages in the wiki: processes (contribute, bug tracker, version releases, quality checks, mailing lists) and guidelines
Preparing pages that explain the new process to announce the big change
Semantic Media Wiki is being installed on our http://openerp.com/wiki
Installing Bazaar

For the quality processes we worked also on guidelines and automated tests.
We implemented a XML validator using Relax-NG on the trunk.

Reporting Workshop

A workshop has been set up to improve reports on OpenERP. All reports have been reviewed and will be improved within the next days. You can find guidelines on how to write reports here.

Already done:

We developped the extension on our report engine for a better control on
the report: date and float format according to lang of user/partner, repeat table header on new pages, styles in tables, ...
Two reports have been made in the trunk as an example:

The balance sheet: an internal report with "internal header"

The sale order: with the external header

I am waiting for your feedback on the forum about these two reports. If people are satisfied with these, we will improve all repors with the following style.

Ergonomy Workshop

The ergonomy workshop made lots of works for improving the interface:

Ease of use
Improving daily usage

Already done:

A simplified/extended view system (some views have already been implemented)
Small changes in the clients, especially visibility/readonly/required on fields based on the content of any other field. Example: if you set a bank account of type IBAN, the swift code becomes required, ...
All menus reviewed for a better structure (not finnished)
Configuration wizards at installation and module configuration
Translation system by module
Lots of Small changes: sort & rename shortcuts,
A translation system by module, ...

To be done: lots of things, too much to explain here.

Security Workshop

In the new version of OpenERP, we reviewed the security to don't have access on objects by default. Then, all groups/menus/rules are defined by default for most common usage. You just have to assign groups on people when you configure the software and everyting is preconfigured.

A security on fields has been setup too: read/write on views or objects.

Already done: part of the work, reviewing all menu with a better and more flexible declaration. instead of . The name come from the action so that we can change a full menu tree with only one change.

Accounting

Lots of improvements have been decided on the accounting modules. A team of 5 financial experts reviewed the software in all ways. Trust me, the next version of the OpenERP accounting will really rocks ! A team of 3 full-time developpers is working on all these aspects.

Too much things have been decided during these 2 days, so I can't detal them all here. We will announce big improvement on the planet when it's commited.

Summary

You can also have more information on what have been done and said during the community days here. Check the section Workshops.

Two questions to close this topic:

Do we have to organise community days once a year or twice a year ?
Where will we organise the next ones ?

Roadmap next Stable Version

2008-06-18T06:01:00.001-07:00

As you may have noticed, lots of changes are being applied in the trunk branch these past weeks, on eTiny, the server or the GTK client. We are on the way to freeze the the trunk branch in about one month to prepare the new stable release.

If you have patches, bugs, new module suggestions or anything, please contact us asap so that we can integrate them in the future version of Tiny ERP. We plan to release the next stable version for the beginning of september. This is just an estimation, stability matters more than delay.

If you want to discuss on some of these points, please join us at the Open ERP community days in Belgium. We will apply or planify everything needed in the next 4.4 version during these 2 days.

Relations avec le réseau Certa rattaché à l'Education Nationale

2008-05-22T04:32:00.001-07:00

Les travaux du réseau Certa continuent et on les en remercie. Voici un compte rendu de l'avancement pour les intéressés.

My First Book On OpenERP

2008-05-06T13:24:00.000-07:00

The first book on OpenERP is now available in most common french bookstore. It has been co-written by Geoff Gardiner (Seath Solutions) and me. I am really excited by this book that will bring new possibilities to enterprizes that plans to analyse some integrated management system.

More information here:

You can also check the detailed information on the Eyrolles website. You can order it on amazon.

If you read it, please tell me what you think about it...

The english version is being written. I hope we will be able to publish it in a few month.

RAD with Tiny ERP

2008-03-13T13:57:00.000-07:00

Here is how you can design a complete application in a few minutes with Tiny ERP, without writting any line of code.

I just released a module for DIA that is able to export a UML diagram to a Tiny ERP module. (uml_dia in extra_addons). It automatically: create the objects (with fields, defaults, functions, and inheritancies), the views, the actions and the menu based on a UML schema in dia. So that you just have to design your UML graph and it saves this as a .ZIP Tiny ERP module.

Then you will be able to use our view designer in eTiny (first release in april) to design the form/tree/calendar/graph views and adapt them to your needs. Here is an image that present how this will looks like:

Then, you will be able to use our DIA module to design a workflow for the different object directly in dia and export it as a Tiny ERP module.

After this, you define your custom reports using our OpenOffice report designer. See the following demonstration: http://tinyerp.com/images/flash/report_designer.swf

Then you can design your reports and statistics directly in the Tiny ERP client using the base_report_creator module, available in the extra_addons svn. Moreover, the 4.3 version of Tiny ERP already includes a dashboard designer.

We are also working on a workflow editor in eTiny that will be based on; http://js-graph-it.sourceforge.net/ But this is not expected before 3 months. (we have one full time developper on this project).

And because everyone needs good reporting, we are launching our own BI application, independant of Tiny ERP but also fully integrated. So that you can design your own cube within a few minutes. Here are the very first screenshots (the graphist has not started his work:) :

The first release of our BI application is expected for April. That includes a cube designer (in eTiny), the cube browser, a full featured MDX query engine (mdxalchemy). Everything is web based and full ajax (drag&drop, eso...). Our engine currently support: Oracle, PostgreSQL, MySQL, MsSQL, dBASE, Sybase, DBII, BDB.

Of course, you do all these operations using the module recorder so that you get a complete module for a custom application within a few minutes. The module includes everything: our objects, views (forms, calendar, lists, menus, ...), actions, reports, statistics, workflows and dashboards. And as all these methods support inheritancies. So, you can easily benefits from current modules to map on them and upgrade according to your needs.

Once you developped your own modules/applications, don't forget to use the module publisher (base_module_publish) to share your developpments with others users. And others users will be
able to build on top of your own modules to have a module catalog that will growth extremely fastly. The community can install these modules but also contribute with the help of our new (base_translate) module that allows to decentralise translations effort to benefit from the
community workforce. Check within this email for details about this new module.

I am preparing a complete demonstration on these development methods in Tiny ERP to show how you can build a full application for university management in less than 30 minuts. And it gives you directly the Ajax/web application and the GTK rich client !

This demonstration will be published in the beginning of April.

April will be a great month for Tiny ERP:
* First prototype of the BI engine,
* New RAD methods and modules,
* View designer, dashboard designer,
* Lots of new things in the CRM,
* The books on Tiny ERP will be published,
* And the SaaS offer !

A growth at 3 numbers !

2008-03-12T16:48:00.000-07:00

Since 4 years, the Tiny company had a very fast and organic growth. We went from 1 to 65 employees on the group in a few years, without any loan or investor.
And the company is still self financed and profitable.

The 10th of March 2008, we already made sales equals to our total turnover of 2007. So we know that this year, like the preceeding one, we will have a growth of at least 100%. I expect to continue being profitable in 2008 and having about 100 employees before the end of the year.

I have to say, I could never have made this growth without open source.

If you want to contribute in this very exciting project, we are hiring 4 developpers in Belgium:
* The open jobs
* Working at Tiny

Tiny ERP Mind Mapping

2007-12-28T01:18:00.000-08:00

I played a little bit with vym during my spare time and we made a mind mapping of all components of Tiny ERP. This is still a draft version, you can click on the image for the full view.

We will improve this map so that it will serves like a table of content on the architecture help of the software. Vym is able to add URL to each node and to save this as a web page.

Modularity: The Fatal Weapon Against a Complex Market

2007-11-04T02:12:00.000-08:00

This post is about the new 'base_module_publish' and 'base_module_record' modules and how they will speed up the growth of Tiny ERP by giving more control to the Open Source community.

The ERP market is quite complex because it is heavily segmented into two axes:

The business of the customer: the software is not the same for an association, a manufacturing industry, a service company, a car reseller, ...
The size of the customer: companies of different sizes (1, 5, 20, 50, 200, 1000, ...) are managed in a completly different way.

The verticalisation task is not easy because both of these two axes implies serious changes in the whole software and in the configuration process.

Either the most complete ERP (SAP) has only 26 verticalisations of their software. That's ridiculous if you think there is thousands of different businesses multiplied by dozens of different sizes.

So, for years, companies had to choose between two approaches for their management system: 'best of breed' or a 'generic ERP'.

That's where Tiny ERP is breaking the averages. Due to his open source nature and his module system, you get all the advantages of the 'Best of Breed' approach, and thanks to it's hundreds of generic modules, you get everything you find in a "All in One" ERP.

Tiny ERP is known to fit the needs of lots of different markets (auction houses, services companies, manufacturing industries, food-industries, textile industrices, marketing, etc.) and different sizes of companies; we have customers from a self-employed man to a company with 2000 employees.
This diversity is mainly due to our module system which one of the best you will find in any application framework.

A module can be nearly everything: some menus more, some actions, change workflows, define new reports, inherits and add some fields on objects, inherits on views, etc. But the great thing about modules is that:

They automatically migrate from one version to another,
They can change everything with inheritancy; your custom developments can be packaged in separate module so that you will benefit from new features versions after versions,
They contains demo and data that auto-test and configure the software at each installation,
They can add features or change parts of others modules,
They can be used separetly or connected to each others,
They are per database, so you can manage different database on the same server with different modules
They can be published in repositories; so that each partner can maintain his own repository for his own customers.

Tiny ERP is the only one ERP in the world where you can install only the accounting, the purchase or the project management system, according to your needs.

This is extremely flexible, a new company can build his own system by choosing between our 200 modules 'a la carte':

Electronic Product,
CRM,
Quality ISO9001,
HR and Contracts,
Timesheets but not Sign In/Out,
System to automate the lunch order, ...

In the new 4.2.0 version, we released two new modules for managing modules: 'base_module_publish' and 'base_module_record'.

The first one allows you to automatically publish your own modules in the module management system of Tiny ERP. When you install Tiny ERP, it is able to download and install new modules on the fly. So, if you publish your own modules, they could be installed by others people within a few minutes, through the official Tiny ERP repository. Of course, if you install non official modules it's at your own risk. For production environment, it's better to wait that the editor validate the module.

The 'base_module_record' module allows you to record all operations you do in the Tiny ERP client. So that you can configure your ERP system from the Tiny ERP client and save everything you made in a module. That's extremely interesting to develop pre-configured or vertical packaging of Tiny ERP. By doing this, every end-user can save his own configuration and modification through a module. Then this module can be published and people can deploy the same system simply by installing this module.

Of course, modules works by inheritancy, so that you can configure your own Tiny ERP system by using others modules and by applying modifications of them. Then, you can save your modifications as a third module and publish it.

Through this system, I think we will be able to quickly launch verticalisations of Tiny ERP for our futur SaaS offer. And we will maintain them very easily.

Open Source Business Model of Tiny ERP

2007-10-24T01:26:00.000-07:00

"How do you win money with Tiny ERP, as this is an open source software ?"

For my first blog entry, I couldn't get a better idea than replying to the most asked question on Tiny ERP. This also allows me to explain our fully reviewed business model that have been discussed during our last partner meeting.

As an Open Source Editor, our goal is to set up a business model that covers:

All Actors: the partners, the community, the customers and the editor.
All operations: pre-sales, sales, research and development, marketing, etc.

Unlike a proprietary editor, we can not impose our strategy by constraints. So we have to create a contractual win-win relationship between these different actors.

Here are the 4 win-win contracts proposed to our partners and contributors:

The Partner Program
The Shared Funded Development Projects
The Maintenance Packs
The SaaS Service Provider Contract

The "Partner Program" is dedicated to our partners, who are value added services companies that offer all range of services based on Tiny ERP. This contract sets up the context so that:

Our partners get enough 'intensives' to work in a way that benefits to the product: publish their success stories, promote the mark, participate in partners meeting, publish their own modules as open source, etc.

Our partners get enough services from the Editor in return to the activities they generated for the product: support, training, sales leads, etc.

The Editor has a way to value activities that benefit to the partners, like pre-sales.

The Editor can offer more services to the best partners and less services to less active partners.

A win-win context where the cycle of growth auto-powers himself. The more a partner generate activities, the more services he will get/need from the Editor. And the more services he has, the more activities/sales he will generate.

As an Open Source ERP, most of the developments are financed by the one that requested it: "pay this small feature you need and get the rest of the ERP for free". This is a good approach for most of the features but this does not works well for very big or generic developments. The SFD Projects allows the Editor or a Partner to pre-finance the development he needs. Then the module is sold (open source but not distributed) per customer until we reimburse the Partner or the Editor for his investment.

This will sustain the growth of Tiny ERP because every partner can get the developments he needs without paying it, simply by advancing the money and taking the risk at his own charge. It first have to ask the Editor for a budget about the needed feature, then he pays for this development.

After that, the module is sold until we fully reimbursed the partner. That's extremely scalable because the development of these module are paid by the end-user/customer, not by the partner or the Editor. Of course, this system doesn't change anything to the R&D budget of the Editor, it only concerns modules that are not in the roadmap but requested by partners.

The "Maintenance Pack" is a system to finance the maintenance of very complex modules that costs a lot per year. For example, it replies to questions like "Who will pay for maintaining the Switzerland Payroll that evolves every year and need lots of experts works." With these maintenance pack, a small part of the needed budget is paid by each end-user. There are no more maintenance risks and costs for the editor and the partners.

There are lots of things to say about the new "SaaS Service Providers" contract, so this will be the subject of a later entry. But the interesting part for the business model, is that it will solve another problem for the growth of the product: "who will do the verticalisation of the product and the packaging of these verticalisations." It will setup a context where the SaaS provider has advantages to invest on the verticalisation effort to reduce starting costs. In counterpart, the Editor guarantee them an important return on their investment by focusing sales by sector on these partners.

These four contracts will guarantee a win-win environment around Tiny ERP that will sustain the product's growth and will benefits to all actors. The great thing is that they covers all aspects needed for the development of the product.

As a summary, a proprietary editor is characterized by high licences revenues but lots of risks and investments to launch and maintain the product. The Tiny company is at the complete opposite of this common proprietary editor business model:

We don't have licences revenues but,
We don't have any risk nor investment to do because every hour spent by one of our employee is paid by someone: pre-sales, sales, research & development, support, maintenance and packaging efforts.

Moreover, the open source nature of Tiny ERP ensures a fast growing curve of the users, the product and the trademark. That guarantee us a big demand in our different services.

That's all for the basics of the business model. If you want to learn more on the subject, we made a FAQ on our website.

Waiting for your feedback,

Fabien
CEO Tiny