First major release of 2nd generation of OpenWISP

[Federico Capoano]

Hi everyone,

I'm imagining the next major release of OpenWISP:

- core modules merged to avoid double maintenance (django-config / openwisp-controller, django-freeradius / openwisp-radius, etc)

- switch to docker-openwisp for installation and provide migration instructions

- may be a good idea to update ansible-openwisp2 to automatically install docker-openwisp2 and migrate the contents to it

- publicly release the following modules, including some basic screenshots and documentation:

  - openwisp-radius

  - openwisp-monitoring

  - openwisp-frimware-upgrader

  - openwisp-wifi-login-pages

- ensure those modules will be available options in the default installer, monitoring and firmware upgrade could be active by default

- provide an official OpenWRT image (and image-builder aka image-generator) for OpenWISP

- users looking for a solution for public wifi, should be able to get a basic public wifi service working out of the box (captive portal coova-chilli, freeradius TLS, captive page with registration and login)

- implement the automated setup of the development environment mentioned here: https://groups.google.com/d/msg/openwisp/kXnmasof_qs/in5N48xjBQAJ

- boost the user documentation: have each module provide its own user-documentation and openwisp2-docs will aggregate those documents depending on the branch/version (so we'll be able to provide user docs for each major version of openwisp)

This could be the first point release of the 2nd generation of OpenWISP, we may give a codename to it (or use the year of release like Ubuntu and OpenWRT), I think it would also be a good idea to issue point releases of all the modules, for example:

OpenWISP 21.02

- openwisp-controller 1.0.0

- openwisp-radius 1.0.0 

- other modules, all having 1.0.0

OpenWISP 22.02

- openwisp-controller 2.0.0

- openwisp-radius 2.0.0 

- other modules, all having 2.0.0

If we accomplish to achieve all these goals, OpenWISP will make giant leaps forwards, because I'm pretty sure most of the users now get stuck in the beginning: they install it, they try it out and they give up shortly after, because to get to a working installation, you have to dedicate a lot of time only to understand how to plug all the dependencies all together.

I'm also aware that there's is no open source alternative to OpenWISP which provides what I described above.

If we do this, we'll be the first, and it's pretty ironic this will happen in 2020 / 2021, after these technologies have become common like electricity and running water.
Considering how important networks have become in our daily life, building and maintaining networks is still pretty damn hard and doesn't generate as much money as less basic services do (instagram?). Yet, there's probably more people around knowing how to build an instagram clone, than how to build a network which delivers working connectivity.
This is WRONG and we as a species will soon realize it.
Instagram clones won't allow you to do remote work, send payments, study, stream documentaries.

The COVID19 epidemic has already helped to enlighten us about other wrong things in our world. Suddenly everybody realized that paying million of dollars to football players while paying pennies to nurses and other workers who do work which is essential for our survival is deeply wrong.

Those are my main core motivations to work on OpenWISP. Whatever is yours, let's join forces and get this done. I commit now to put effort into building this vision and ask your feedback. I'm genuinely interested to know what you think about the idea expressed above. Especially if you used OpenWISP in the past or if you're still using it.

Best regards
Federico Capoano

[Noumbissi Valere]

I strongly by with these ideas and new goals for OpenWISP. With the appearance of 5G, networks will be more indispensable even to areas that have not yet accepted networks. Because the capacities which this technology (5G) provides, will make those who aren't network familiar look several decades behind compare to the rest of the world. Thus, they will undoubtly turn to a tool like OpenWISP (if these goals are successfully achieved).
I am available to give in skills and time to see we can achieve these goals.

[Ajay Tripathi]

Hi,

The goals look good to me.

Some of the points like:

> implement the automated setup of the development environment mentioned here: https://groups.google.com/d/msg/openwisp/kXnmasof_qs/in5N48xjBQAJ

Can be divided into small tasks and be a part of Google Code in as well, which will help us achieve the goal of adding more tasks important for the project in GCI! :-)

Best,

Ajay Tripathi

[Federico Capoano]

You can bet on it.

I am still waiting for more feedback from the users of openwisp, I know there are some which need more time to reply to this thread.

Once there's some more discussion and we get some more ideas, over time I will be creating issues in the various github repository and I will link them to a github project board.

Some of those issues will also end up in the dedicated Google Code-In project board. We could also consider issues that are both in the Google Code-In plan and the Major release plan to have more weight in the Google Code-In evaluation, so that students will feel motivated to push us towards the release.

BTW: I forgot to mention the new module openwisp-notifications is also part of the release plan (notifications are now embedded in the alpha monitoring module but it won't be that way in the near future).
 

[Federico Capoano]

I've created a dedicated project board on github for this plan: https://github.com/orgs/openwisp/projects/13.

Over the next weeks I will be linking issues to it, as well as create new issues and feature plans.

Clarification: most probably there will be several intermediate releases before we get there.

But that release will mark a huge difference between before and after.

[Federico Capoano]

Hi everyone, I wanted to give an update on this subject.

On Friday, April 24, 2020 at 9:47:38 AM UTC-5, Federico Capoano wrote:

Hi everyone,

I'm imagining the next major release of OpenWISP:

- core modules merged to avoid double maintenance (django-config / openwisp-controller, django-freeradius / openwisp-radius, etc)

This is progressing well.

 

- switch to docker-openwisp for installation and provide migration instructions

We haven't started with this yet.

 

- may be a good idea to update ansible-openwisp2 to automatically install docker-openwisp2 and migrate the contents to it

We haven't started with this yet.

 

- publicly release the following modules, including some basic screenshots and documentation:

  - openwisp-radius

We're progressing.

 

  - openwisp-monitoring

We're progressing.

I also forgot to mention openwisp-notifications, which is also progressing well.

 

  - openwisp-frimware-upgrader

We're progressing.

 

  - openwisp-wifi-login-pages

Not progressing here at the moment.

 

- ensure those modules will be available options in the default installer, monitoring and firmware upgrade could be active by default

- provide an official OpenWRT image (and image-builder aka image-generator) for OpenWISP

- users looking for a solution for public wifi, should be able to get a basic public wifi service working out of the box (captive portal coova-chilli, freeradius TLS, captive page with registration and login)

- implement the automated setup of the development environment mentioned here: https://groups.google.com/d/msg/openwisp/kXnmasof_qs/in5N48xjBQAJ

We haven't started with these points yet.

 

- boost the user documentation: have each module provide its own user-documentation and openwisp2-docs will aggregate those documents depending on the branch/version (so we'll be able to provide user docs for each major version of openwisp)

I started drafting a plan:

Comprehensive documentation: https://github.com/openwisp/openwisp2-docs/issues/107

Named release and release notes: https://github.com/openwisp/openwisp2-docs/issues/108

Federico

[Federico Capoano]

Hey everyone,

here's a quick update on this subject.

A lot of work has been done on different modules in the last 4 months.

Some minor details have to be addressed and the installers need to be upgraded so that everyone will be available to start using these new features.
Documentation is also still an ongoing process and we need help to improve it.

I prepared a project board on github which is dedicated to pushing OpenWISP to the goals stated at the beginning of this thread.

I encourage everyone involved in this community to help us to push towards these goals!

I ordered the items in the board so that the issues needed for the very next release are at the top.

Best regards

Federico Capoano