One thing that we have been disappointed by is the lack of true support for Linux players.
We have a regulatory requirement that our players must be installed on operating systems that get security updates whenever new CVEs or other security weaknesses are published. The updates must be installed within 30 days of release. And we need something that has security updates for several years so that we don’t constantly need to install new operating systems on the players. We need an LTS operating system. As a result, our two options are basically Windows and Linux, as Android doesn’t have an LTS release, and few OEMs are committed to security updates for non-phone Android devices.
We currently have Windows licenses, but we are gradually moving away from Windows. Windows has been a huge drain on us for security and productivity reasons. In the next 5 years, it’s possible Windows won’t be an option for us.
So we thought we could go with Linux for our players. Unfortunately, the Linux player is “community supported” and obviously isn’t a first-class citizen in the Xibo universe. In fact, it appears the Linux player barely gets updates, which is a non-starter in enterprise.
We are asking that the Linux player become a fully supported option. We would be fine if only certain Linux distributions were fully supported, so long as the supported distribution is an LTS release.
1st class citizen status for Linux is something we’ve spent a lot of time and resources on internally and also something our community has worked hard to achieve. But unfortunately you are right, it isn’t a solved problem and we no longer have expertise in the team to continue development with the player as it stands now. This is the reason we’ve labelled it as “community supported”.
Some interesting work has been done recently by a contributor who wanted the Windows player to function in WINE, which might be interesting to try, although as it stands I wouldn’t call this enterprise or 1st class due to all of the dependencies.
It isn’t all bad news though, we do have some ideas to explore over the coming months:
WINE as above
SMIL
Porting some of the windows player into .NET core and building something new to cover the gaps
I think the best chance is the 3rd option (or perhaps a mixture of the 1st and 3rd). But to be open and honest, anything we do with Linux Player is a medium to long term time horizon for us (so 6 to 12 months), and is really hard to commit to given all the challenges this sort of development brings.
To anyone else reading this, please like the original post so we have an idea how much demand there is in the community for a Linux Player.
For me personally, I’m not so much married to the notion of Linux as I am married to “not Windows,” at least not in 3 to 5 years.
For example, I’m not diametrically opposed to Android. It’s just that it is difficult to find an LTS release of Android. Unfortunately, almost all Android OEMs either support Android for a very short period or don’t support it all; what version of Android comes with it is the verison it will have the day it is trashed. The issue isn’t lack of features; it’s that the device isn’t getting security updates at the OS level.
If anyone does know of an LTS Android distribution, though, I’d be more than happy to look at it.
In any case, I figured a more supported version of the Linux Xibo player might be more likely than someone creating an LTS version of Android.
Our android hardware partner who produce the DSDevices series of hardware do work hard to make sure relevant security updates, and patches are made available in new firmware. You are right though, fundamentally you’re at the mercy of the chipset manufacturer and even with that effort it isn’t LTS in that sense.
I do think a Linux offering is important for the project and there is clearly demand for it, so we’re certainly keeping it on the table.
Which linux distro were you using? I’ve deployed the xibo player on RHEL Workstation, which is long term supported, provided you purchase their support package. I would think that there aren’t many digital signage companies out there that would want to offer a long term stable product as it essentially locks the feature set, and if your selling perpetual licenses, new versions of the software are the only way to introduce new features, so purchasing new versions every year or so is kind of washes out the need for long term stable. I’d recommend just forking their cms and linux player and managing it internally.
I am starting to test the applications on Linux this month. The Linux player didn’t work well for me (everyone is already aware of the reasons.), and i will start testing the client (Windows) with Wine.
I think this is a good option.
As the days go by, I will let you know what I was able to make work.
Update:
It did not work… i’ve installed the R308 using Wine… the instalation process did run well, and the CMS sync as well. But it can not to open the Xibo Player after all, even when i forced to run the player.