cancel
Showing results for 
Search instead for 
Did you mean: 

Alfresco application automatic generation : a MDA approach

jck
Champ in-the-making
Champ in-the-making
BlueXML just released a free (gratis) version of its graphical editor to configure Alfresco, based on Eclipse. Go and download it at http://www.bluexml.com/en/product-download.html

This tool permits to generate automatically Alfresco applications from UML models.

UML diagrams are converted in the following way :
* class diagram => type and aspect definition, all template and configuration files are generated
* workflow diagram => advanced jBPM workflows generated
* security diagrams => access controls generated for each group according context

These automatic generations permits to be very, very, more productive than configuration file edition by hand for types, aspects, workflows, templates, …

Moreover, the use of UML permits to increase reuse and make evolution a lot easier.

Jean-Christophe Kermagoret
_________________
Ingénieur BlueXML
MDA Alfresco Générateur
18 REPLIES 18

jpfi
Champ in-the-making
Champ in-the-making
Hi Jean-Christophe,

MDA is a very nice approach to speed up the configuration process of a dms.
I'd like to try your product but regrettably I cannot find any helping docs or tutorial…

Cheers,
Jan

bch
Champ in-the-making
Champ in-the-making
Hi,

You can find few tutorial on this wiki :
http://www.bluexml.com/wiki/index.php/How_to_create_a_new_diagram_%3F
Or here :
http://forums.openbluelab.org/forums/index.php

You can see a little demonstration here :
http://www.bluexml.com/en/popup/quickstart.htm

Our project migrate actually from OpenBlueLab to BlueXML. If you need help, you can post message on our forums :
http://www.bluexml.com/forums/

We work actually to provide a set of models for an ERP and to write a quick start to simplify the use of BlueXML Developer Studio.

Cheers
Benjamin Chevallereau

bch
Champ in-the-making
Champ in-the-making
Hi,

I send you a link for a quick start :
http://www.bluexml.com/wiki/index.php/Quick_Start

jpfi
Champ in-the-making
Champ in-the-making
Hi Benjamin,
thanks for your kind help. I tried out some basic examples. I'm very impressed.
From my point of view Alfresco is currently missing an administration client. I think this should base on the eclipse platform and its miscellaneous frameworks.
Perhaps there is the possibility to open your product, make it open source and make it the starting point of this admin client.
What I mean by saying admin client:
- Generating configuration files by your MDA approach
- Node Browser
- Perhaps a groovy srcript editor for quick administration tasks
- CQL Editor (Alfresco Content Query Language)
- Report Generation
- Monitoring issues
- User & Group Management
-…

What do you think?
Cheers,
Jan

jck
Champ in-the-making
Champ in-the-making
Hi,

I think you're completely right.

We could be interested in such an approach but our business model is a little different. Our tool is not GPL because we think we are not located at the right place of the "software map", contrary to Alfresco who says to its customers he sells compatibility, robustness, …

I think we cannot tell the same because we are as compatible or robust as Alfresco is… but we can't provide more. So, once used, user could throw away our software and should not buy any guarantees, contrary once again to Alfresco 😞

WDYT ?


Jean-Christophe Kermagoret
BlueXML CEO

rdanner
Champ in-the-making
Champ in-the-making
Hi,

I think you're completely right.

We could be interested in such an approach but our business model is a little different. Our tool is not GPL because we think we are not located at the right place of the "software map", contrary to Alfresco who says to its customers he sells compatibility, robustness, …

I think we cannot tell the same because we are as compatible or robust as Alfresco is… but we can't provide more. So, once used, user could throw away our software and should not buy any guarantees, contrary once again to Alfresco 😞

WDYT ?




Jean-Christophe Kermagoret
BlueXML CEO

My thoughts on license is that it's a function of motivation.  If you want adoption on a mass scale and you don't care about return on investment (that is – if ROI is adoption rather then cash or code) then MIT or ASF is the way to go.  If your a business who has to feed families than GPL makes a lot of sense: don't be anyones free lunch – Cash or Code required.

The value really isnt in the bits.  You may need to get the product to critical mass in order to see service around it but once you do, any software works with the model.  The community can help you get to the point of critical mass.  The question is how long do you have to ramp up and how much of your own resources do you have to bootstrap the thing to critical mass?  

Open source is good for business - that is what i think.

rdanner
Champ in-the-making
Champ in-the-making
Hi,

I think you're completely right.

We could be interested in such an approach but our business model is a little different. Our tool is not GPL because we think we are not located at the right place of the "software map", contrary to Alfresco who says to its customers he sells compatibility, robustness, …

I think we cannot tell the same because we are as compatible or robust as Alfresco is… but we can't provide more. So, once used, user could throw away our software and should not buy any guarantees, contrary once again to Alfresco 😞

WDYT ?




Jean-Christophe Kermagoret
BlueXML CEO

My thoughts on license is that it's a function of motivation.  If you want adoption on a mass scale and you don't care about return on investment (that is – if ROI is adoption rather then cash or code) then MIT or ASF is the way to go.  If your a business who has to feed families than GPL makes a lot of sense: don't be anyones free lunch – Cash or Code required.

The value really isnt in the bits.  You may need to get the product to critical mass in order to see service around it but once you do, any software works with the model.  The community can help you get to the point of critical mass.  The question is how long do you have to ramp up and how much of your own resources do you have to bootstrap the thing to critical mass?  

Open source is good for business - that is what i think.

Sorry, I can't resist here Smiley Happy

One more comment on this line in particular "but we can't provide more. So, once used, user could throw away our software and should not buy any guarantees, contrary once again to Alfresco"

I don't think you should look at your software in this way.  You are all but saying there is no value in it.  Simply not so.  There is opportunity all over the place.  If the user can use the product once and get what they need , then throw it away because it has no further value – then you don't have a real business model to begin with.

Leverage the open source to build out what your product is missing in terms of long range value.  Focus on features that allow the customer to really manage or grok their models.   Maybe branch in to these untapped gap areas in alfresco's own administrative capabilities.

If you are worried that alfresco will begin to encroach on that market at some point with their own tools – you would probably be right, they will but that's competition and its great for customers and although painful at first it is ultimately good for you because it will drive your innovation.

You can certainly be the best at something and as long as that something represents an opportunity in the market you're golden. Over time opportunities will come and then be swallowed up by those who follow your lead, you will have to seek new one and over time, just as alfresco did, you will gather enough feature mass that you will have a base line of functionality that needs support even though it doesn't cover gap in the market – again, to make this cost effective for your company, open source is a good option.

Anyway…  The product sounds very cool … going to check it out… Smiley Happy

loftux
Star Contributor
Star Contributor
I think it is a bit pessimistic saying that this is "once used, user could throw away" software. Looking at the potential customer base, end customers will use is repetitively. An ECM solution will evolve, you may start with one department and adapt to their needs, then comes the next. And after running for maybe a year, users have found new needs, they are usually very creative once they start to see the benefits of ECM. And they reorganize and adapt to a changing world, requiring change of the Alfresco implementation So it will be used over and over again. Maybe not every day, but more often than one might think.

As a partner I see value in this kind of tool. First I would use it to quickly adapt to the customer I'm going to do a demo for, so that they could see metadata fields that they would most likely use. Next it would be useful for live implementetations, saving time for the more challenging tasks. Also quicker turnaround, customer says "what if we do it like…" and I would be able to show them in minutes.

Alfresco is also an evolving product. To take advantage of new features, customers would need to stay in touch with you (because you will follow Alfresco features, right?). Think of configuration now moving into the repository or the DM forms.

As you have clearly identified a market opportunity, you should think of how to capitalize. And for this I think you should really go for establishing your tool as the de facto tool buy going open source. Even if that means maybe a little less revenue to start with, it means revenue in the long term. Because I'm very sure that if you don't create this kind of tool as open source, someone else will. That may even be Alfresco themselves, after all the new collaboration stuff is in place, ease of configuration is the most important feature to add in my opinion. Go talk to them!

Maybe a bit repetitive of what other have said here, but an important discussion. Not only for the BlueXML tool, also for others that develop cool addons to Alfresco .

Peter Löfgren

jpfi
Champ in-the-making
Champ in-the-making
Hi all,
I agree with the previous speakers. But we are currently in the same position with our opsoro prototype (http://wiki.alfresco.com/wiki/Opsoro). From my point of view this statement is the key:
I'm very sure that if you don't create this kind of tool as open source, someone else will.
If you'll do it, you'll have the steering wheel. If someone needs support or some new/custom features he'll first contact you. But this business model is certainly hard to understand, specially if you feel at home with a more closed model…
Cheers,
Jan