Forum OpenACS Q&A: ]project-open[ V3.2.Beta4 Released (V3.2 RC 1)

Hi,

The ]project-open[ development team is proud to announce the availability of ]project-open[ V3.2.Beta4. This is the first release candidate for V3.2 and includes a vast amount of extensions and improvements.

The most important new feature in this version is the integration with GanttProject, an open-source project Gantt editor (http://sourceforge.net/projects/ganttproject/). This integration allows project managers to define and modify ]po[ projects using the graphical GanttProject editor.

Links:

Major Enhancements:

  • GanttProject Integration: Project Managers can now conveniently define and modify project structures with GanttProject and then upload the result into ]po[ ("round-trip integration"). GanttProject is an open-source MS-Project clone written in Java. The combination of GanttProject and ]po[ offers an open alternative to MS-Project and MS-Project-Server.
  • "Web 1.5" GUI Overhaul: We have improved the GUI to allow users to move their GUI components and to select/ disable components. Also, we have added new icons and a new CSS.
  • Resource Planning and Resource Assignments: You can now add resources to projects and task and determine resource overload across projects.
  • Calendar: The new calendar integration module shows project and task deadlines per day, week and month. Calendars can be either "shared" or "individual". Support for group calendars will follow soon.
  • Bug-Tracker: ]po[ now includes an integrated bug-tracker module that compares to industry leading bug trackers.
  • Expenses and Travel Costs: The new modules allows you to capture consultant's travel costs and other expenses per project.
  • Reporting Engine: ]po[ now includes its own reporting engine, similar to Crystal Reports or Jaspar Reports. A "reporting tutorial" explains step-by-step how to write new reports. A graphical editor is not (yet) available.
  • System Configuration Wizard: A new "SysConfig" wizard allows the first-time user to tailor the ]project-open[ installation for business sectors and company size, reducing the overall complexity.
  • "Petri-Net Workflow Engine": ]po[ now includes a integrated workflow engine with graphical process editor. The WF can be used for example to set and confirm project budgets by senior executives, to priorize projects in a portfolio etc.
  • Project Templates: Project templates (creating new projects based on an existing project template) allows you to standardize and accelerate the generation of projects with similar schedules.
  • Dynamic Fields for Project, Companies and Users: We have greatly extended the "DynField" module to include DynFields in reports and CVS exports. Also, DynFields are now specific to object subtypes (for example: Type of projects).

New Commercial "Enterprise Extension Modules":

The following modules are closed-source extensions of ]po[ covering the needs of large organizations. These modules are available separately from the ]po[ product at http://www.project-open.com/shop/.

  • GAAP/IAS compliant Auditability: This module captures all changes to financial objects and projects.
  • One-Time-Passwords: Manages a list of TANs (Transaction Numbers) for one-time authentication of users with special permissions (i.e. senior mannagers, accounting,...).
  • Profit Center Management & Permissions: The modules allows corporations to define fine-grain financial permissions across departments (who can see/modify the invoices of each department) and to calculate profit/loss per department.
  • SAP Integration: Import and export interfaces to SAP FI.

Minor Enhancements:

There were more then 500 minor enhancements in the last 6 months improving usability and fixing bugs.

About ]project-open[:

]project-open[ (http://www.project-open.com/) is a project management and PSA (Professional Services Automatization) system for companies in the consulting, engineering, advertising and localization industries. It covers the entire project life cycle from sales (CRM-light), staffing, execution (timesheet, controlling, incidents, discussions, and file storage) to invoicing and payment. The P/O architecture is designed for mission-critical applications with a rock-solid infrastructure and a sophisticated role-based permission system.

Collapse
Posted by Malte Sussdorff on
This sounds really interesting and I look forward to see how it works. How much of the new enhancements is OpenSource? The reason I am asking is that we are now developing a pure OpenSource solution for collaborative project management across organizations and OpenACS will be the part used for CRM, Project Management and other core functions (there are a lot of other modules).

To go ahead with it I see in general three options:

a) Use your code, only possible if it is GPL. This can only work if the components can really be displayed independently from each other.
b) Take a look what you did (not the code) and move that into Project Manager (at least for the modules that are not GPL). This has the benefit of residing on OpenACS and work with the OpenACS community as well as benefit all current customers, on the other hand it means that we have two project management solutions and we are not really collaborating on this.
c) Ignore your work and just do it anew.

Therefore my question to you: Is there a CVS tag which allows me to download all modules released under the GPL ( so I dont have to check every single file)?

If I go with option B), would you be interested in knowing what we did and also where we made enhancements so you could see how they fit into your core system?

Last but not least, if we were to include your dynfields work in OpenACS, would you be interested in collaboratively maintaining it there or is your preference to keep the source code close? This lack of clarity with regards to DynFields in particular has so far prevented me from looking at the options it offers.

P.S.: Does setting up a demo server with PO constitute a violation of the license rules for the open source / non GPL modules?

Collapse
Posted by Frank Bergmann on
Hi Malte,

Thanks for your consideration. It's an interesting idea to develop a completely open-source ERP system.

We tried that in the beginning. However, the idea didn't really work out for us, because we've found that ERP-like stuff is 1.) too boring for any volunteer to help, 2.) mostly too complex for a truely distributed OSS development process and 3.) OSS dynamics don't work well amongst companies that compete against each other.

Concerning your collaboration options: I'd be happy to collaborate with you on our GPLed modules, that's why we've GPLed them... In particular "DynFields" could be a good candidate for that. However, knowing about Cognovis' past track record and the fact that we're going to compete directly with each other (please see futher below) I ask you to choose option c) for everything else.

Here is the list of our GPLed packages:
- intranet-bug-tracker
- intranet-core
- intranet-dynfield
- intranet-crm-tracking
- intranet-spam
- intranet-payments
- intranet-timesheet
- intranet-timesheet2
- intranet-hr
- intranet-wiki
- intranet-search-pg

if we were to include your dynfields work in OpenACS,
would you be interested in collaboratively maintaining
it there or is your preference to keep the source code
close?

We'd be interested to see "DynFields" as part of the OpenACS core. This is where it actually should be, and how it has been planned right from the the beginning. We basicly tried to complete the partially developed OpenACS "SQL metadata system" (acs_attributes table etc.) and to make it useful for real-world applications.

But yes, DynFields is kind of obscure (we had to implement a lot of stuff in very short time) so that I doubt whether it has the quality to go into the OpenACS kernel right now. It's really a complex area, and there are some good reasons why the ACS 4.0 developers didn't finish the package...

P.S.: Does setting up a demo server with PO constitute
a violation of the license rules for the open source /
non GPL modules?

I guess you're fine to setup a demo server...

In http://www.cognovis.de/referenzen/successstories/e4 you say that you have been granted the right to commercially exploit ("vermarkten") the result of the "E4" project, so I have to admit that it appears kind of "funny" to me to see a competior asking us for permission to setup a system basicly in order to show his customers what exactly he's going to copy and reimplement for them.

Wouldn't it be better to accept that ERP software isn't very suitable for the "pure" OSS model, and to go together with us for a sustainable solution? There are many examples that show that service companies (and all "pure" OSS companies are service companies...) are not suitable to maintain a "product" (because a service company can't eliminate their own revenue stream if they'd really streamline their product to reduce training, customization and installation services).

Bests,
Frank

Collapse
Posted by Nima Mazloumi on
Just a thought :)

Do you really compete? Is the market that small that you would consider yourself as competitors instead of collaborators?

Collapse
Posted by Frank Bergmann on
Unfortunately, yes. Cognovis and ]po[ have already been stuggeling over what would have been our biggest customer at that time. So yes, I guess that's what is called "competition"...

Cheers,
Frank

Collapse
Posted by Malte Sussdorff on
A little bit of background on this story (take a look at http://e4.cognovis.de to see what the project is all about).

The E4 project aims to provide a collaboration platform for clusters and virtual organizations with a focus on finding the right partners (take your freelancer module and enhance it by automatically filled in Key Performance Indicators) and allowing them to coordinate using a collaborative project management system which will feed into a multitude of PM systems actually installed at the clients site. It is explicitly *NOT* an ERP solution, and definitely not target towards service companies. Meaning, any of your products and customers you are targeting is nothing the E4 system will deal with.

For a complete open source ERP solution, look at http://www.tooleast.org/. Again, focus is on manufacturing, not services. Still they need project management and suppliers.

As written, our goal is exploitation of the final result of the product. Due to the fact that none of the development partners had a solution for CRM and project management, I got OpenACS based CRM and PM into it, as I knew it well and know that it meets the criteria. The question is now only, should we really use this or use something else for the final product and who will maintain the CRM and PM part taking into account that we are not a full development partner in this project.

With regards to the demo server: I could not find one of the 3.2 version on your site and I am going to talk about process oriented project management at an ERP conference tomorrow in Switzerland where I will show what we are doing in the E4 project and what OpenACS has to offer with that regard. I assumed it would be nice to mention and maybe show the latest version of ProjectOpen there as well. So it is not at all related to the E4 project and although an understandable assumption (we demo something we will "lend" from projectopen), this is neither the intention nor the case here. But well, too late anyway now...

For the record, I don't see PO as a competitor as they are selling a product. Sadly reselling this product is costly (20% from the total revenue generated with the client, Frank, correct me if something changed from our discussions), especially if you need to do a lot of modifications.

Collapse
Posted by Frank Bergmann on
Hi,

I don't see PO as a competitor

Sorry Malte, you must have a funny understanding of "competition". Looking at your website you easily find dotTrans (http://www.cognovis.de/xowiki/dotrans). This product is basicly a 1:1 copy of our specific ]project-translation[ system (http://www.project-open.com/solution/translation/), and I believe there is little doubt who was first in the translation market, right?

*NOT* an ERP

That is not the point. We're also working with project-oriented departments of non-service companies now.

take your freelancer module and enhance it by
automatically filled in Key Performance Indicators

But why would you need our new PM extensions for that, as you asked in your first posting?

Sadly reselling this product is costly (20% from the
total revenue generated with the client, Frank, correct
me if something changed from our discussions), especially
if you need to do a lot of modifications.

We have changed that. There is no such percentage anymore. Instead, we offer partners to share revenues from our new "Enterprise Extension Modules".

To summarize:

I agree with with Nima that the market for _services_ around OpenACS is huge and (geographically) fragmented, so that most OpenACS companies can collaborate on the code without fear of competition (even though I know quite a number of cases where companies are holding back particulary valuable extensions to maintain a certain competitive edge...).

However, ]po[ is a "product company", and we face a very knowledgable and resourceful competitor following our path, so we can't just share our code, atleast not in the usual manner.

I hope that is reasonable or atleast comprehensible...

Cheers,
Frank

Collapse
Posted by Malte Sussdorff on
Nice to see us appreciated as "knowledgable and resourceful". 😊. And even better to know that you dropped the 20% "penalty" on using ]po[.

As for your question. The reason I am contemplating using ]po[ instead of project manager is simple. You are going to maintain it as a product. So I can be fairly certain that I could have a reliable Open Source project management solution which will be maintained and where someone has an interest in keeping it maintained even if cognovís is going in a different direction.

Collapse
Posted by Robert Taylor on
Frank, as someone running I.T. and having looked at your product, may i humbly suggest that the market is so large that long term benefits of co-operative competition outweigh the american interpretation of capitalism you are applying here?

Unless I'm misunderstanding your interpretation, your approach only applies to to companies in mature markets that are at the end of their growth cycle. Only then is the benefit to any one company greater in a kill or be killed scenario than a co-operative competition scenario.

Thus, your perception of Cognovis or really any other entity being strictly a competitor is a patently unfair label. One should look at them as partners as well.

At this stage of the game, I believe it is wiser to put wager money on trying to grow the marketplace together rather than try to establish some sort of quasi dominance in a particular niche as that is a stragedy that really only works with a pure closed source play.

With that being said, I do appreciate where you are comming from and your contributions to the community are warmly welcomed by us. We look forward to working with you in the future to the degree possible as well as everyone else in the greater community.

Collapse
Posted by Malte Sussdorff on
Frank and myself talked off site about the issue and resolved in a friendly manner following your's and Nima's advise. I see the major value of ]po[ of being a highly integrated suite for running a whole agency and consultancy. Contacts is complementing this suite by offering CRM functionality that is lacking.

Project Manager on the other hand is suitable for organizations interested in project management based on OpenACS that do not have a need for sophisticated invoicing and complete integration, but use it as a more sophisticated task tracker. We will use and maintain project manager for our existing clients and anyone who is interested to use PM as an addon for their existing site, but work with Frank together based on ]po[ to expand the market for this integrated solution.

Collapse
Posted by Caroline Meeks on
Robert,

Nice point and this analysis applies to all of the OpenACS companies.

Thanks
Caroline

Collapse
Posted by Frank Bergmann on
Hi,

Just to let you know that V3.2.Beta5 is out, the next V3.2 iteration. Beta4 had an issue with the installation of TSearch2 in certain environments. We already had first feedback from Beta5 that invoice templates are not well preconfigured and that the Petri-Net workflow graphics ("dot") doesn't work. But apart from that it's getting there...

Robert Taylor wrote:

co-operative competition scenario

I agree with Caroline, you've written a great comment about the OpenACS "ecosystem" and its competitive/collaboration dynamics.

As Malte wrote, we're back on the collaboration track, and I completely agree that collaboration in general makes much more sense then competition. The competitive situation in the past happened in the "Translation Sector", which basicly consists of some ~1000 target companies worldwide so that we had really different competitive dynamics.

pure closed source play.

I know, I know... ]po[ is just a contradition in itself: Trying to be a good member of the community (giving back not only installers and code but also marketing etc.) and preparing the ground to become a scalable product company (otherwise we would never have obtained the resources for the "integrated solution" effort that Malte has mentioned). There are a lot of conflicts. So yes, we basicly need to play quite a pure closed-source game to maintain the economic counterweight to our investments into this seriously boring ERP solution stuff...

This strategy fortunately doesn't keep us from sharing quite a lot of our stuff with the community. It's more technical barriers that slow down the integration of our code into OpenACS such as the different permission philosophies and GUI frameworks. I'll just post an update to the Bug-Tracker anouncement (http://www.openacs.org/forums/message-view?message_id=494700)
explaining our current issues.

Cheers,
Frank

Collapse
Posted by Robert Taylor on
Superb.

Onward and upward!

😊