zentyal and the road passed 3.0

[robb]

last year we seamed to have a bit of trouble finding our way in how to discuss the new version,  how to determin included features and other sutch things.

in my view this lead to a somewhat unclear roadmap,  in the short term,  but a nearly non-existent one over a longer time;
-  do we need to change this,
- and if so,  how are we going to manage it.
- what is our current releasing scedule about and is that sufficient or even on par with reality.
- how to grade new idea's, technologies  requests,   how to keep balance
- do we need big changes and if (or when) so,  how do we plan them in a way so that they dont slow down the entire product and its development.

but also non- technical questions:

-  do we want to use the zentyal summit to talk about it future, future development  and other sutch issues.
-  how can we get people to prepare for sutch a debate,  whould there be enought people interested in sutch a discussion.


i know that there are some really dificult questions to be answered here, but i strongly feal that we need to try to consider them in order to get a more solid view of the feature.  not just about a possible  3.2 release  but also a more clear distinction between released that include a new  core (ubuntu version)  1.0  2.0  3.0   and those that dont    1.4  1,5   2.2  etc.

I would like this discussion to be expanded to the whole community. Feel free to comment and give your thoughts on how the community can chip in, in finding a way to features and additional functionality for future versions of Zentyal. Especially what should the process  look like to come to a useful set of features that can be implemented, considering that:
Zentyal is a Small Business Server
The dev team of the company has limited time, so there is no way that every idea can be implemented
There are (still) no community developers that can do some work on popular features.

I think the main questions are:
- What features should be available in Zentyal Core
- What features should be available through an advanced howto in integrating them in Zentyal (LDAP)
- How do we 'measure' popularity of features
- How (what criteria do we use) do we select from the most popular features

I am curious what you all think.

[J. A. Calvo]

I think the main problem now is that we have several places to collect/store the ideas that came from the community even the ones directly from the dev team. That is: wishlist in the trac wiki, nice to have someday milestone tickets, feature requests forum and any other suggestions coming directly from customers, training students, etc.

Our proposal is to have a single centralized system for this, allowing any community member to submit their ideas, and vote or comment existing ones. With this voting system we can easily measure which are the most popular ideas (and therefore the most likely to be included in the roadmap if they make sense). In this system, we could also track which ideas are currently under development or those which are already implemented.

Of course, there is a lot of initial work to populate this new system with the contents from the existing ones, probably the community could also contribute to this.

Although some things are yet to be defined and implemented, we think this may be a step forward with regard to the current situation. What do you think?

[ichat]

the whole idea about this topic is to encourage the  broader community to look 1 step further than a given feature-request.   

for example how many people have been asking for  the same feature,  or one so simular that it could have easily been added to the other,

from many standpoints your are right to say that a cental more clear approach to feature (and other information) tracker  can be a good idea.  but first lets stop for a minute, and think  what kind of feuture we invision ourselves for zentyal...   should we rather incoporate  10's or dozens of new features, maby ranging from  intergrated (web)aplications to  filling and emptying your washing machine, dishwasher or tubledryer...

or should we take another aproache and maybe even reduce some defaults, but making it easier to intergrate them yourself...

or do any of you have even better idea's

we dont care about how you join the conversation, we just want you to share your  idea's,  explain why, what might be the benefits and what could be the risks...  there are a lot of questions in the opening post,  and we can easily add more.

[stuartiannaylor]

I think this opens an a much needed and unfortunately complex topic.

Personally I think the community site needs a revamp.
I stated a long time ago that we need a peer review based community collaboration platform.

Its not rocket science and it is easily achievable.

I have a solution in mind if anybody is interested and the mechanisms to help contributors and provide preference over leachers.

Stuart.

[robb]

Hi Stuart,

Good to see you back in a constructive way. I second the idea of having a peer reviewed community site/documentation and would love to see your idea.
Bare with me that although you have a good idea, it might be or not be feasible to use your idea. Seeing your history with the project I (personal opinion) think you have been too disappointed the project didn't follow your ideas.

That said, I hope you want to propose your ideas in improving the community tools.

[stuartiannaylor]

I have a personal opinion that Zentyal is trying to be everything for all.

Applications such as Zarafa should be offered via an extention directory or marketplace.
Something like the Joomla extention directory or the Oxwall market place.

Oxwall in my mind would actually provide a much better community site than the current forum.
Saying that the way they lay out their site based on their own product is a good example.

The oxwall store is a collection of extentions by 3rd parties and oxwall. Some free, some not.
I feel offering Zarafa in the way it is, will hinder 3rd parties offering other solutions and so narrowing choice.
Choice is a huge element of open-source philosphy.

I feel Zentyal as it is already presented is currently bloated with features and this should be cut back.

Zentyal should concentrate on a developer API for LDAP and core settings so that developers can supply a plethora of applications without any conflicts.

Zentyal could increase their revenue stream by providing a smaller core and a selection of extentions all with subscription and support options.
This would reduce the standard subscription and allow the market to select what combination of extentions and subscriptions would be used.
Monolithic products have a history of toppling and my opinion is that Zentyal should be slimmed by pushing out certain functionality to a marketplace / extention directory with individual subscription and support options.

Stuart.

[stuartiannaylor]

Rob,

Yeah don't get me going Actually its not that my idea's where not used. I have just had a browse because of the new release and I am dissapointed that we are still stuck with this forum.

Look ignore my temper tantrum as I admit I am at the very left of organic peer review collaborative efforts.
This is often countary to the honest efforts and comercial interests of great open-source products such as Zentyal.

There is a problem with all open-source projects and its a catch-22 situation of providing ROI without closing the doors of standard comercial models.
I must admit I did get very frustrated about the idea of partitioning and weakening a small community into geographical islands. I felt also that I gave up because some seemed very determined to go down a cul-de-sac and not listen to others. This was on the back that ROI for both zentyal and a community in my mind is of a much higher priority.

The offer is there anyway.

Glad your still on board and active

Stuart

[ichat]

Hi stuart, let me seccond Robbs notion in welcomming you back.

for you idea on a more peer-review based, eco system  i do agree,  BUT as the community is small i fear that it will take a lot of effort ... and gets verry few results ... non the less someone needs to brake the line and take a first step...

for the rest of your idea pertaining to for example zarafa intergration, i agree that this eventually could be verry interesting for zentyal development.   however this would be a daunting task, and not verry feasible in the near feuture...  therefor i would suggest a middle-way approach,  where we start designing a new module system, that supports packages, services and scripting. and provide the services needed for those packages,  and leave zarafa out of it at least untill the most of the mutch needed other improvements have been addressed. 

[stuartiannaylor]

Hi Ichat,

Good to here you are still on the Zentyal bandwagon.

I think we are going to have to flesh out what is meant by a new module system...

I am running blind but I guess we need some sort of key system based on system registration and developer registration.
The contributions repository is already there and this allows for free submissions. We should not close this to only that option and need some form of system to allow evaluation, subscriptions and paid for apps.

What are your thoughts on new module systems? New module system?

I am playing catch up with Zentyal and I don't know if or what has been proposed?