Looking for alternatives to Simple Groupware?

[Patrick Pliessnig]

Simple Groupware is based on a very old version of PHP and is no more maintained since 2012. Time might come when you need to look for alternatives (the following is by no means a recommendation of any sort, just information to help you plan your future ).

One such alternative could be Nuclos ERP-Toolbox. The main purpose of Nuclos is the configuration of business systems (enterprise resource planning). It allows you to freely configure business objects of all sorts (assets) and attach status models (business processes) and rules ( programmed in Java ) to them:

1. site: http://www.nuclos.de/
   https://de.wikipedia.org/wiki/Nuclos
   
2. licence model: open source
3. based on Java (Spring Framework)
4. maintained by Novabit GmbH

negativ:

documentation in german (forum, wiki)

quality of the wiki is quite questionable

positiv:

configure your ERP yourself

multilingual software

flexible desktop and online client

process engine (status model)

easy to extend with Java code

lots of interfaces to communicate with the outside world

complete reporting engine

in contrast to SGS the initial installation has no business objects (assets) by default, but is a complete framework for cross cutting concerns like user management, etc. You need to add configuration for what you want or download business templates ( for 99€ ).

as with SGS you don't need to be a nerd to use Nuclos the way you want. If you know how to customize modules in SGS ( and php functions ), you can learn how to Nuclos ( and custom rules ).

I hope this helps

Patrick

NB: I keep using SGS actively.

[Massimo Arnaudo]

Hi Patrick,

Thanks for your point of view about SGS competitors, if i will find valide alternative i will mail it.

Massimo

--
You received this message because you are subscribed to the Google Groups "Simple Groupware" group.
To unsubscribe from this group and stop receiving emails from it, send an email to simple-groupware+unsubscribe@googlegroups.com.
To post to this group, send email to simple-groupware@googlegroups.com.
Visit this group at https://groups.google.com/group/simple-groupware.
For more options, visit https://groups.google.com/d/optout.

[Vlad 0304]

Hi all,

Just for information..
We also using ERP/CRM/SCM Adempiere (iDempiere). This huge client-server system which will be maintained without code programming.

It has Java based code and Oracle/PostgreSQL database. Adempiere is not so clear to understanding but has long "success story" and many documentation.

Some things of Adempiere looks like SAP. Based on my experience, you need 1-6 months for training of Adempiere, depending on the depth of the dive.

It is completely open source and free of charge but flexible.

But it seems to me, it's easier to port the SGS to a new version of PHP, than to study any other system from scratch.

Regards,

Vlad

[Patrick Pliessnig]

Hi Vlad

I fully agree that a port of SGS to a new version of PHP would be the preferable way. The overall concept is amazing.

- For the port we would need seasoned PHP developers.
- For commercial use we would also need a maintaining unit (preferably a team)

- And last but not least restructuring the software architecture using current PHP capabilities could make SGS more flexible, easier to adapt to business needs and with enhanced quality built into the source code.

I guess a lot of users would prefer this approach.
I am just not aware of people ready to take over.

Patrick

To unsubscribe from this group and stop receiving emails from it, send an email to simple-groupwa...@googlegroups.com.
To post to this group, send email to simple-g...@googlegroups.com.

Visit this group at https://groups.google.com/group/simple-groupware.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Simple Groupware" group.

To unsubscribe from this group and stop receiving emails from it, send an email to simple-groupwa...@googlegroups.com.
To post to this group, send email to simple-g...@googlegroups.com.

[Vlad 0304]

Hi Patrick,

Unfortunately, I do not have time to take any more projects too.

Now I have to launch five of small projects for Kickstarter, at the same time there is a process of developing a set of several controllers to manage the water purification system for our local corporation Ecosoft. At the same time, I must ensure that our GPS server does not fall down. It is good that the projects are at different stages and I will delegate some of the tasks to my colleagues, but the number of requests is only growing. In general, if someone had given a $ half of million to build a powerful engineering team, I would probably be able to deal only with code development. While it is only possible to wait, when I will become Ilon Mask If I have enough life of course :)

Just now we have to work in the fire brigade mode.

I have contacts with good web developers, one of them is my brother, but the firm as a whole works for US contracts with rate of $33 per hour .. They do everything in the web, from casinos to e-shops, but first a Contract.Next a Development task and approval of the Budget.

I'm sorry that I can not really help, but I'll follow SGS project from time to time, maybe something will change.

Regards,

Vlad

To unsubscribe from this group and stop receiving emails from it, send an email to simple-groupware+unsubscribe@googlegroups.com.
To post to this group, send email to simple-groupware@googlegroups.com.

[Patrick Pliessnig]

Hi Vlad

Thank you for your thoughts. We will see what happens.
In the meantime as long as we can take along old PHP to new operation systems, the status quo can live.

Patrick

[Patrick Pliessnig]

Vlad

Just as a side note for your projects in case you want to move away from fire brigade mode, away from reaction to proaction.

If Java is not something to avoid for you have a look a the framework Apache Isis. It is like the formula 1 approach of SGS. Instead of specifying an asset type ( a module ) with a xml configuration, you specify the asset type as a plain old Java class ( a domain model ). from this Java class the persistence and the presentation ( the DB and UI ) are automatically generated. This makes you super fast, but you can still configure the UI with bootstrap. you can develop domain models as complex a you like using all the Java stuff like inheritence, mixins, etc.

But it goes much further than SGS. For example you can automatically deploy the application as a restful webservice (hypermedia interface) with a rest api or even execute it headless on the command line and much more. With the rest api you could effectively add a custom made UI ( eg AngularJS ) if you wish so.

But be aware. The software style is deeply rooted in Domain Driven Design and the main pattern is Naked Objects. If you have never worked that way, it may seem very confusing in the beginning before you can sail with the winds. ;-). Just to give you an impression, if you are experienced with the framework you could do live coding together with the users in the same room without sacrifying quality.

Patrick

PS: if you're interested have a look at this:
https://skillsmatter.com/skillscasts/7892-closing-the-feedback-loop-with-apache-isis
https://isis.apache.org

[Paul Zarucki]

Hi All,

Just another thought to add to the pot.

The problem of keeping a system or application working for the long term is much wider than Simple Groupware. Many small organisations and businesses spend time and money setting up a system to do a job but can't easily afford to rewrite the application each time the platform / operating system version changes. Also, as old hardware fails, there is the problem that an older operating system might not run on modern hardware.

One solution which has worked well for my clients is to use virtual machines to keep old platforms / operating systems running. For Simple Groupware I use a VM running Debian 5.

One of my clients also needs an application running on Windows 98 which they now run on virtual machines. A separate VM provides a Samba server which allows Windows 98 to access the new file server (in reality an NFS server). Samba also provides virtual printers which enable Windows 98 to print to modern printers that don't support Windows 98 directly (a print-to-PDF service is also provided).

Best regards,
Paul Zarucki

[Patrick Pliessnig]

One should add to the equation security issues for those applications that are exposed to the outer world.

In fact each new version of PHP fixes security vulnerabilities which keep to be a threat in older versions. However I am not able to evaluate this in terms of SGS and a particular installation. Maybe someone could shed some more light on this. The bigger the potential damage, the more one should take into account this topic.

Patrick

[Paul Zarucki]

Hi Patrick,

I agree with your points about security and old software. Use should be restricted to internal use on a well protected network. For remote access by mobile or home based workers, we relay all communications via SSH running on up to date servers. We also use VNC-over-SSH for remote desktop access.

Regards
Paul

[Vlad 0304]

Hi Patrick,

Thanks for the information, I know some of this, but unfortunately the structure of our main project was formed long time ago. To make big changes in the structure, we have to do everything anew.
For this reason, I just have to push the train ahead :)

Previously, we had plans to promote GPS-monitoring in conjunction with ERP,, CRM and other processes automation elements (Adempiere based), but users only needed to see their cars on the map in the office PC.
After 13 years, customers are ready to automate business processes, but they want to see it in their smartphones and in the form of "box solutions", "cloud services" and other miracles that do not make them think about installing and maintaining software. So, our priority direction now is mobile applications but the hardware is slowly moving to the IoT: BLE5, ZigBee, Apple Home kit etc.

Regards,

 Vlad

To unsubscribe from this group and stop receiving emails from it, send an email to simple-groupware+unsubscribe@googlegroups.com.
To post to this group, send email to simple-groupware@googlegroups.com.

[Patrick Pliessnig]

Hi Vlad

We plan to integrate IoT products together with the LoRaWAN network into general internet applications. Some applications will be for mobile devices like smartphones, some for office applications and some simply as web-services to integrate with other applications.

Luckily, as I am not bounded to use legacy (ERP) systems, I plan to put the whole application logic on the server with Apache Isis and decide independently which part of the application logic I'll deploy on which user device. An office grade HTML-frontend I'll have for free. also the RESTful web-service interface is for free. For mobile apps I'll probably integrate with Ionic framework over REST. That way I will be able to automatically stream updates directly to the user device App. I recently checked out that I can integrate with SGS as frontend over REST configuring assets as proxy assets like the imap module. That's cool, some of my clients like the SGS interface for its simplicity and versatility. The only thing, I have no idea for now, is a general framework for publicly displaying (marketing grade) application logic like real-time monitoring or similar. I'll need to spend time searching for such a framework.

Patrick 

[Steve BOURNE]

Hi Patrick

I've searched for a long time (since 2012) some alternatives to SGS : ADompiere, Concrete5, eGroupware, Drupal, TikiWiki, PIMCore, Zimbra, Horde, and even MS Office365 (and by the hell not Google docs/apps !) ... The last I've planned to evaluate is CODE/Online OpenOffice bit this a more a cathedral business than a simple PHP application, and probably not covering all what my needs.

The truth is that none of them, though they have all their own advantages, let me do what I can do with SGS.

A feature that SGS has and others not is its directory/tree based architecture, allowing copy/paste of a subtree with all its assets ! A real advantage when you need to create a lot of similar projects (in my case some treaning materials with complex scenarii and tree wich repeats period after period as I welcome new groups/sessions of teaching/training). It allows me to create sort of template tree that I can base new ones : it gives me a lot gain of time !

Then, the concept of transforming a simple directory to a specialized application with its own assets is a killer concept, versatile as you need ! And versionning is working transversally in all modules (though in spreadssheet it is not filly ok)

Finally, the possibility to create new custom assets fields (with filtering and other actions on them), and if this not enough, the possibility to create new modules/applications by the mean of simple XML files, and PHP if it is needed, letting me specifying the model ("table", though it could rather/really have been called "model" by Thomas), the view (view and tabs), and if needed the application logic (the "controler" ?), and all this surviving a SGS upgrade by the provided "custom" directory.

All that was a clever great idea by Thomas, as he had promised to him when he was leaving the "MoreGroupWare" team ! Again great Thanks Thomas.

(Great thanks to you all, Patrick and others, who are pursuing the initial goal while the initial mentor has left SGS by 2012).

But yes, SGS is buggy though a great (inestimate) tool. Yes, SGS is probably insecure nowadays ! But I think this great product merits a big lifting to renders bugless, more appealing, and with revamped features. I would prefer a project that refactors SGS, than searching again and again another product which ressembles to it but not based on the great backbone/architecture !

 

What if using the SGS core architecture, and giving it the applications ergonomy and features lifting to go where Horde and Tiki are ?

Some ameliorations I would be pleased to have in SGS :

- more appealing look and feel (an other package of CSS ?) and ergonomy (actions on the directories via contextual menu for example)

- replace the old and yes uggly calendar GANTT view by a new one

- make the category system working everywhere

- unify the assets identity management

-- for example, the identity of a spreadsheet is its name, not an artificial auto-incremented id. So when a copy/paste a spreadsheet in another directory, SGS complains that the spreadsheet asset already exists

- upgrade the (really great) spreadsheet module (a very good spreadsheet module can be the killer app needed in such a great CMS tool, since spreadsheet allow lot of things to be done in a do it yourself context)

-- sort and filtering that really works (buggy)

-- give it a good ergonomy with contextual menus

-- linking to other spreadsheet assets

- webmail bugs and ergonomy improvements

- quizzes with rich text content as a question, including pictures, allowing SGS to be a start of LMS (Learning Management System)

- a lot of other things -)

I know supporting and amelioring SGS is a big challenge, and that none has full time on it. But would a sponsor be interested on it ? Convinving him that the core founadation is exceptionnaly great, and that an effort would drive SGS to a challenging product on the cloud scene ? Perhaps I'm wrong and I'm swimming into an utopy lake :-P

Steve

[Steve BOURNE]

(Sorry for embedding the original post to my previous answer/post)

Hmm, I forgot : another great feature of SGS is the directory fusion with other similar directories, a must. I've already another approach of the same idea in OpenGoo/FengOffice with what they call "contexts" ...

Steve

[Patrick Pliessnig]

Hi Steve

I fully agree with you that SGS offers unchallenged features for the user.

The SGS core architecture however seems to me less appealing. 1st, the software architecture is oriented towards the limited php possibilities of the days when SGS started. 2nd the differential sgsml mechanisms for designing asset models (tables) are hard to deal with when the models grow complex.

All this results in a few problems - like buggy or harder to maintain code than necessary. Hence refactoring doesn't seem a good choice for me. I'd suggest a complete rewrite with more modern approaches.

One such approach ( indeed the one I would choose ) is to take an existing application framework that implements the naked object software pattern together with a hexagonal architecture and to build SGS on top of it. This approach extends the usage philosophy of SGS to what is under the hood.

I would also rethink SGSML to something less differential. One way could be to define a simple OO-DSL ( domain specific language to design modules ). With this you could design things like a general "partner" for your business and then "employee is a kind of a partner" and also "customer is a kind of a partner" which makes it much easier to design real-life complex models.

One such application framework could be 'Apache Isis' from the Apache Foundation. This framework is - so to say - an SGS for application developers but not for end users. To get it end user ready, you would need to develop the out-of-the-box modules, simple installation procedures,  probably some more stuff and of course the aforementionned DSL ( or write a parser for the existing SGSML ). As an added bonus you'd get even better integration strategies into existing business-level-it-systems

Cheers

Patrick