Zentyal 4.0 Roadmap Published!

[Miklos]

I want to add my 2 cents on this.

First I support the narrowing of Zentyal's scope. There is nothing that is stopping them from doing another distribution down the line if they want to make a GW product etc. - but for now to be successful they need a scope that is reasonable so it's smart to focus on what they are the only ones doing right - the Exchange drop-in solution.

Now it's no secret that for the GW part there are other free solutions that are much better than what Zentyal is/was in that area - pfSense comes to mind. If you can't make 2 machines then make a VMware solution and run pfSense as your GW/IDS/SDI/Whatever solution and run Zentyal for the Exchange - seems like the smart way to go.

The only thing in the Roadmap that confused me was this: Removal of the Free Zentyal Account - What does that mean exactly? No configuration backup for community?

[StuartNaylor]

First of all, about the dropped modules, as the roadmap announcement says: "Documentation will be provided on configuring some of these services with specific software and naturally, community members interested in maintaining any of these modules will be warmly welcome."

Also, for people interested in keeping maintaining the modules in the contrib section, we've written already some quick documentation about it:

https://wiki.zentyal.org/wiki/Building_and_maintaining_a_contrib_Zentyal_module

Maybe is not enough but it may be a good starting point and we will be commited to improve it and help anyone with real interest in helping with that.

Please remind that (as some of you were also wondering), Zentyal has been always a modular platform, and even with less officially maintained module, this has not changed at all. Nothing prevents you to build and install existing modules in contrib, or even create your own ones (and of course share them with the rest of the community!).


Hope this clarifies things a bit. Don't hesitate to ask any other question you may have.

Thought I would kickstart things a bit I am going to have a go at the Zarafa Module because I think Openchange is a great idea but until it has bedded in a bit I think Zarafa the free version is a good basic groupware package.

Going to move to a new post in contribution Zarafa community module & development.

[netwarrior]

Hi Everyone!
Straight to the point. I've been a Zentyal user for just some months but I also came to this ship because it was a UTM/Gateway and a fully Linux modular solution. Having said that, as many of you have pointed it was a little disappointment when I saw the 3.5 roadmap and I was hoping that some module were coming back (BWMonitor) but now I with the 4.0 roadmap I see more modules being dropped. I have deployed 3 zentyal servers as UTM/Gateway solution for a government organization in my country and they were very pleased because it cut them lots of cost, but now I'll have to think how to maintain the dropped modules.
I'm new to the forum but I've been reading posts since my first day using Zentyal and I decided to join the conversation because I feel like we can together find a feasible way to maintain some of those modules and maybe add new ones focusing on a UTM/Gateway Zentyal version as someone pointed.
I would like to see some kind of documentation on how to modify the GUI since I couldn't find one yet, I have already made some modifications to fit my needs but it took me sometime to figure them out and I'm sure modifying other module may be harder.

Greetings from Peru!

[StuartNaylor]

I am trying and because I am unsure about openchange my first port of call was zarafa but if you are up for the many heads approach you decide on the module and I will give it a go in a shared effort.

https://forum.zentyal.org/index.php/topic,22799.msg87788.html#msg87788

I can give you a few tips that I have found so far.

https://wiki.zentyal.org/wiki/Building_and_maintaining_a_contrib_Zentyal_module

First of the wiki needs an edit change the line apt-get install zbuildtools to apt-get install zbuildtools build-essential fakeroot

Also deb http://archive.zentyal.org/zentyal 3.5 main extra already exists in apt sources.list

So doesn't need to be added

From there follow the url and actually if you get a file manager it starts becoming apparent

I think it would be very possible to hack an existing module in the contribs

You choose the module

Do the above and have a browse and maybe say

PS I am much better with applications than the network side

Also because of the move between openldap and samba4 I think actually Zarafa might be a bad one to pick the is a huge transition needed with the ldap.

I get the feeling ftp might be a good dev noobs pick

[Gary Schiltz]

Thought I would kickstart things a bit I am going to have a go at the Zarafa Module because I think Openchange is a great idea but until it has bedded in a bit I think Zarafa the free version is a good basic groupware package.
Going to move to a new post in contribution Zarafa community module & development.

I'm glad to see someone stepping up to get started on maintaining a dropped module. I have never used Zafara, and I may or may not use it in the future, but I applaud your stepping up to the plate and look forward to hearing how it goes. One of the reasons I got interested in Zentyal is as a way to learn how to do a modern single-page web application. Zentyal is cool in that it uses some very mature (practically ancient by today's standards) technology (Perl) on the backend and modern JavaScript/HTML/CSS on the frontend.

I may just dust off the old Camel Book and see how it goes with the UPS (nut) module. Seems that would be a lot simpler as a starter. If that goes okay, then maybe the virtualization manager would be a next step - it is something I never got a chance to use, as it was moved to contrib before I started with Zentyal.

[CNServices]

I'm glad to see someone stepping up to get started on maintaining a dropped module. I have never used Zafara, and I may or may not use it in the future, but I applaud your stepping up to the plate and look forward to hearing how it goes.

While I agree with congratulating initiative, where will this effort take the community? If the goal (and there should be a goal) is to salvage Zentyal as a gateway product, then J. A. Calvo’s suggestion to fork Zentyal makes sense more than retrofitting future versions of Zentyal.

If the goal is to improve Zentyal 4.0, then perhaps that should be clearly identified as the purpose of the community effort.

[StuartNaylor]

Good point and I missed the fork suggestion. Which really makes me wonder that actually we should fork Zentyal 2.0 but place it on the new ubuntu LTS.

Just wondered what others thought?

PS I know that sounds weird but just an idea that maybe an more linux pure with AD connectors and 3.0 would be better for Zentyal.

I dunno as just as happy with 3.0 but if any-one would like to kick start a thread in contributions it would be great if the community discussed.

[Panzerfather]

First of all, about the dropped modules, as the roadmap announcement says: "Documentation will be provided on configuring some of these services with specific software and naturally, community members interested in maintaining any of these modules will be warmly welcome."

First of all I am a bit disappointed that Zentyal is cut down this way but I can understand the background behind. For me it's a little bit too widely cut down in some points and I want to say why and the thoughts behind in detail and MAYBE a solution to get the best out of it. Maybe a synapse between free and payed modules for community edition will lead a way out of this scenario (like in ClearOS).

Now to the points I thought of. I have no problem with the removal of:
- IPS module
- RADIUS module
- IPsec module
- Free Zentyal Account
- Webmail module (Roundcube)

For a most secured network upset Firewall + IPS + RADIUS should always be outsourced to another (hardware / virtual) machine. There are enough (free/open source/payed) distributions like IPFire, pfsense, smoothwall that can be used for this propose and has same or better functionality in this areas than Zentyal.
What I don't understand when removing these security modules is the point that the Squid module won't be removed also because running squid on a system without an IPS can be very dangerous. All of the mentioned distros also support setting up a web proxy in an easy way (at least for IPFire). Also the VPN Service could be removed because it can be run on these distros.
The point that the free Zentyal Account is removed points to the thing that this is either not used or that it is too expensive to provide it. If it is the second one, than it could be converted into a payed module.
The Webmail module (Roundcube) won't be a great loss when it will be replaced with Sogo-Webmail.

But I'm a bit concerned about the removal of:
- UPS module (NUT)
- Backup module (Duplicity)
- Monitor module

When people have to splitting up security and their mail + AD system because of the removal of the security modules, you must have in mind that they need to handle at least two machines. So they have more work to stay updated and needed to be informed if something is getting wrong.
Therefor the Monitor module is the right thing and makes a real difference in competition to other AD/Exchange Servers which don't have these functionality even Windows SBS and 2012 Servers don't have such reporting functionality.
The UPS module (NUT) made it so easy to connect and setup an USV. Nowadays it is essential to have such a module to avoid data-loss and get a clean server shutdown before running out of power.
A backup module is also essential for such an environment Zentyal wants to play in. Administrators should have an easy way to backup and/or restore files when they need to and therefor Duplicity is the best solution for now. Bacula is to complicated to be run in a small business environment because of the too powerfull interface and bash. It has no GUI interface at all and so your target audience won't be reached at all.
Correct me if I'm wrong, but which of your supported companies have their own linux administrator? So all this 3 modules are essential for your audience you want to reach and were you have your niche you should focus on.

Now to my SOLUTION I thought of:

Why don't Zentyal integrate a "payed" module solution like ClearOS foundation has one?

As example [because there are a lot more modules like these in ClearOS]:
For the ClearOS community edition there exists a payed Zarafa module with outlook integration which costs about 10 EUR which is supported by an freelancer and not by ClearOS directly.

Maybe that's a way Zentyal could also inspire the community to do more work and get a "little" payment back for their time and work they spent in providing and supporting modules to Zentyal. I know that there is much work to do to integrate such a workflow, but I think this could be the way to get through without cutting down the potential Zentyal has too much.
For my point of view I wouldn't have a problem to spent 5-10 EUR per module/year when I have an UPS module (NUT), Backup module (Duplicity), Monitor module back again. Directly from Zentyal or someone in the community.

These are my thoughts about the future of the Zentyal development.

[Gary Schiltz]

Good point and I missed the fork suggestion. Which really makes me wonder that actually we should fork Zentyal 2.0 but place it on the new ubuntu LTS.

Just wondered what others thought?

PS I know that sounds weird but just an idea that maybe an more linux pure with AD connectors and 3.0 would be better for Zentyal.

I dunno as just as happy with 3.0 but if any-one would like to kick start a thread in contributions it would be great if the community discussed.

I think it would be better in the long run (for the Zentyal ecosystem), to try to keep "resurrected modules" in sync with the Roadmap builds. But, I doubt this would work in practice, with the volunteer developers not being part of the planning and design of the Roadmap version. I can imagine very fundamental changes just showing up in HEAD and having to try to figure out how to deal with them. Maybe the best would be to fork the current release, then work on adding modules from the previous release. Then, when the Roadmap release is out, start updating the modules for that release. If the modules work as advertised, perhaps Zentyal the company would be interested in bringing them back into the following Roadmap version. Is there a list of all modules that have ever existed, along with the versions that they were included in?

Perhaps we need a "Developers" forum.

[StuartNaylor]

Gary thanks for your input as your right.

The current version I think we know just from install is a rolling road. I wouldn't like to develop anything on it, maintenance could be high.

I guess its the call from the community for a separation of stable and bleeding edge versions.

Which I guess is good as at one time Zentyal are going to have to stop changing the fundamental base of Zentyal if they ever want to be taken seriously.

With 4.0 maybe we have a chance of this as the differences between 3.5 & 4.0 are pretty big and I guess this will mean working an upgrade path is pretty big.

I think maybe Zentyal needs to have a look and a big think if they have finished  making major changes and maybe even more needs to come out of 4.0?

Also if the community does start adding modules are Zentyal going to cherry pick them and place them in the main repository?

[shadowfire]

There are so many good ideas and thought processes here.  It would be ashame if Zentyal  does a Micro$oft  Windows 8 to 8.1 on its users.   Let me explain.  Micro$oft decided to move in to the touch experience head on... running, jumping and pointing the way as if to lead (which I know there has to be direction - but not at the cost of your soldiers and supporters)   Saddly to say this is not the first time that Micro$oft has done this and had to eat crow.

Back to the point and I will dumb it down to two main feature and not a list for time preservation.   Micro$oft went through Alpha and BETA testing  with their audiance they getting appal amounts of feed back from users - for dislikes and likes, keeps and throw aways.   And at first it seemed that MicroSoft was listen with the new Windows 8 request and thoughts from the test community and users.  Then slowly it became apperant that Micro$oft was vearing of course from the feed back they were getting from their community that was testing Window$ 8.  Things like making Metro the new desktop... almost completely stripping the old faithful desktop away. Taking away the Window$ Start button and all its menu items.

For some reason Micro$oft forgot that the users were the ones who were commanding their experience on the desktop and still had first hand knowledge of what happens or doesn't happen on it.   Micro$oft also miss the memo that the users that would using Window$ 8 didn't all have touch screens and they all didn't want to use a new desktop.  Luckly they at least kept the desktop and made it an app on Metro but wouldn't for the life of them make it possible for the user to select their enviroment they wanted - desktop or Metro.

To make matters worse, even with strong feedback from the users and external test contributors Micro$oft stongly bucked and ignored the Start button/menu suggestions and feedback and desided to make the new side start menu.

On October 25, 2012 they release it got not mixed feelings, but a great number of strong feelings against it's assumptions of how the users wants to work on their already established desktop.  Yes - some people went with the flow and accepted the awkward changes made to their desktop, but many more were disappointed at the direction Micro$oft pushed on them.  I doing so Micro$oft had to rethink it decision in its  public beta in June 2013 that it rolled out to manufactures in August 27, 2013.  They also decided to make the upgrade free, because they knew they screwed up and had to gain the confidence back of their audience with the blunder of not listening to their customers and testers communities feedback.

Please Ignacio Correas and Zentyal Developement - You are at a pivitol point right now, and making a decision like the one your about to make can have very lasting effects on your clients, customers, users, and community.  Don't do a Micro$oft on your users.   Don't loose the respect of those who enjoy this product by alienating those people who use your product on a daily basis.  The ones who get you in the door to the companies would would have never know of you, except by the IT person who has trusted in your product and plans on rolling it out to many other client/cuistomers in the future because of its strengths.   You may not see or understand that these modules create a strengths for your server offering to your clients and the community - but I suggest at the very least, you should give your customers and community a chance to help you with the ideas that may make you, from the idea that you have decided that may break you.

If their is one thing that I know about, it is the willness of your followers and fans of your product to want to help you make it better!  To help where they can.   The reason for all these posts is that these are users who enjoy your product and want to keep using it and keep enjoying it.   Even I would like to see how I could help with allowing backup module to move forward - as I feel this is extremely important in setting your self apart from other packages out their that are in your niche.  As you know and I won't list them... there is more then just Micro$oft in the SBS market.  We are willing to work with you, yet at the sametime, we don't want to be shaffed and ignored.   Reading through the forums we know that this has happened in the past.  I believe you have lost enough people in the past.  I would guess being to proud and arrogant is an additude towards those that help you, that you can not afford to make.  Zentyal is not the only one who has gone through these growing pains, but they can make a concious effort to stand with their cutomers/user/community instead of letting them stand alone and having to go at it alone.  So reach out to us and help us help you! 

Let me suggest this.  Why not hold a gathering of Zentyal/user/community on a medium like mumble (voip server) / screen connect meeting, on a specific day, where parties who are interested in letting Zentyal Company Reps/Devs, know why these modules or lack of would be good or bad.  What Zentyal has help them and where it needs help.   You can gain alot of user feedback, but at the same time allow participation from your user/community base to help you find a road to a solution.  I think some great ideas have been stated here on this thread - but I beleive we can build upon them.  Due to time costraints and life, people are not able to always convey everything they want to in a post.  Let do it via conversation, let find a solution to Zentyal module dropping together.  We all love what Zentyal has been and we all want to see Zentyal succeed and be the best it can be. 

I realize Zentyal Company members are busy and have things to do like us all.   I would be glad to help with setting up a day and the availibilty of a mumble or screenconnect meeting with the proper Zentyal Rep/Devs.  (helping Zentyal setup a server if they have never done it - also putting together instructions for those customers/user/community that want to be apart of it)  Let me know what I can do to facilitate this event with Zentyal.

Thanks,

-SF-

[StuartNaylor]

Maybe that's a way Zentyal could also inspire the community to do more work and get a "little" payment back for their time and work they spent in providing and supporting modules to Zentyal. I know that there is much work to do to integrate such a workflow, but I think this could be the way to get through without cutting down the potential Zentyal has too much.
For my point of view I wouldn't have a problem to spent 5-10 EUR per module/year when I have an UPS module (NUT), Backup module (Duplicity), Monitor module back again. Directly from Zentyal or someone in the community.

+1 again in fact your karma might considerably grow just for one post.

Its been mentioned so many times that Zentyal is a small team with limited resources. Please allow a marketplace and follow the huge success that free product can garner from third party transactions. Start looking at alternative forms of revenue so you can increase the availability of resources.

There is a huge potential for high volume low cost purchases and the only way to do this cost effectively is that we can all share the same mechanism.
Provide the mechanism, take a percent at least it will pay for itself.

It might take a fiscal carrot to really kickstart community modules, but don't limit the offering to modules, make this a true marketplace of offerings from documentation to services.
The market place is a purchase point with a community feedback network and the community can start to gauge the quality of offerings by community review.

If zentyal had this in place we might even see diversity of function, a specific module say FTP might have several offerings increasing the choice to the community and making Zentyal a more valuable product.

Before I go to far though leave the marketplace open and allow free product to be offered in the same manner.

Also this doesn't mean we are all going to be solely out to increase wealth as we could have community modules that create revenue to fund activity.

There is a faction in opensource that take the free model too literally which is detrimental to opensource. There are ways to help projects grow and this takes resources.
If we lived in a free world then the free model would work. Unfortunately this isn't true but don't hinder innovation in business models even if they are not of the free model.
Allow them to coincide and gather resources and leave things Open.

There are so many things you can do by creating a totally open market place where the activity, delivery, review and licensing are centralised and part of the product.
Unless the activity, delivery, review and licensing are centralised and part of the product then this creates a huge amount of work for the offering and offerer.

I just have this gut feeling that all communities throughout history have a certain level of protection and if they are not self sufficient there is only a singular outcome.
I feel that the Zentyal marketplace should offer a years free access to at least all zentyal community modules.
This is the hello period where you can evaluate but also get to know the community and the community to know you.

Then allow a licensing scheme for active modules of a time period but leave the percentage of free vs subscription modules open to the community.
Even allow your own developers to have side projects or zentyal projects of specific function that are not main stream that create revenue and are self supporting.

Then last of all because from experience of quite a number of opensource projects there is a norm of a huge difference in the size of the active community and the leacher.
The leacher  isn't a bad term its just a community perspective for the large number of users who don't have the time or energy or even inkling to join in community effort.
Contribution licensing can reward community activity that doesn't really fall into the standard offerings of product. It helps promote community activity and also can create revenue for the project.

I could quote a load of clichés from "give a man a fish and he will feed for a day, show him how to fish..." I know the above is different, like what the hell is contribution licensing. Free year subscriptions!!! WTF!!!

Non of this has any effect on Zentyals core support subscription model and actually please believe me it would be really easy to accomplish if we all collaborated.
 
     

[netwarrior]

IMHO  I can say that looking to Zentyal website and the features they present the North they are taking is to push the Communications and Office Servers while dropping modules from System, Infrastructure and Gateway/UTM.

So as Gary pointed lets make a list of all dropped module since lets say version 2.2 and after having a full list make some polls to see what modules are the ones the community wanna see back, maybe only 50% of them will get many votes, and after those results we gotta make one last poll to see how many of us will be committed to maintain those "wishlist" modules, in the end we may have only 25% of the modules from our original list but we'll be sure those are the modules the community voted for and that we will have people involved to maintain them.

Greetings,

[StuartNaylor]

Thats fine and its a community.

I have opinion don't know if its humble

Been part of the Zentyal community for a long time now and one thing I am sure of is this vast disparity between usage and usage with support subscription.
The difference in downloads user base and support subscriptions its a guess but its massive, infact huge with two Gs.

So this is the scenario from me what I could do is the Zarafa module, In fact I am going to have a go at the Zarafa module. Done quite a bit already.

I could offer this free and provide support services on my module.

Thing is firstly I know it runs as is if correct its got a plethora of people who know Zarafa and actually quite frankly doubt even if I might get a single dime.
There is no marketplace and I am still going to do it, but wow from experience this is the exception.

Even if I did get a support subscription it still doesn't benefit Zentyal.

Really in terms of numbers the only people who are really grinning are those who just leach.

I don't like that and I am quite passionate about value networks and paying my due.

What I am going to do is say... Hey run this for a year free, after that I don't think anyone will begrudge $5 a year for my Zarafa module.
Or go and develope it yourself!

But that is from all people running it and quite probably actually that has a bit of potential for a few beers.
MMMMMmmm beer Stuart happy and he is thinking maybe another module rather than thinking bloody leachers.


Also with that little bit of revenue there is 10% for Zentyal. Being like I am prob there will be a 10% donation to Zarafa.

My 80% just keeps Robbs moderators chores very pertinent.

Also though if I am not being a pessimist and you do want support subscriptions well here is by offering them through a Zentyal market place we might be talking revenue that is of use to Zentyal. Again zentyal the offer the software and the payment gateway where my reviews are good yeah hell please take my 10%

Its a mutual value network.

Also being me for tireless people like robb and a few others I want to say I value your input as you are a contributor. For you its free.
Hopefully they will do the same for me.
This way contribution starts to get value and creates another type of value exchange.

Currently people who are active will get me doing a Zarafa module and if it works everybody will leach it and we are all the poorer.
At one point I will be too busy or lazy and there is the chance it will just lie stale.

Its just reality from my opinion and there are other ways.
 

[jgould]

Rather disappointed. Radius removal makes no sense to me. They are removing features that windows servers provide which is what they are trying to replace/compete with.

I don't get the big interest in an exchange replacement. Most companies especially SMBs that I see are moving away from on premise email anyway. Easier and more reliable using hosted email and the cost actually ends of cheaper (unless you don't value your email data and don't require any redundancy, HA, fail over).

At this point Samba 4, openvpn, and CUPS are the things that Zentyal does that interest me. But honestly at this point setting samba 4 up is easy via cmd line and then handle with RSAT. OpenVPN I've been doing with cmd line for years. And CUPS has its own admin tool but I was hoping the permissions and sharing would be simplified with zentyal, however at this point the module is a mess with various bugs. So probably easier to just install Ubuntu 14.04 and set it all up without the GUI.