This is essentially just old history - maybe the most interesting part is that it means we are one of the oldest distributions still alive today.
When Mandriva (previously Mandrake) went out of business, the community didn’t want to let the distribution die, so it was turned over to a team consisting of previous contributors, and people from related similar projects (Unity Linux, Ark Linux) joined forces to form OpenMandriva.
We agreed with what remained of Mandriva on terms for all further development:
The original source code;
The initial team, or part of it;
The idea of building an operating system that is simple enough for someone who has never seen Linux to get productive with, without dumbing it down to the point that it stops being useful to experts.
OpenMandriva philosophy is inspired by the Open Source principles philosophy.
Given we are an Open Source project with quite a few mirrors and bittorrent downloads, and we have no idea how many people share their download with others, it is impossible to get accurate numbers.
We can get documented stats only from SourceForge mirror. Maybe worth to mention that many users and/or newcomers are invited to download and test the latest ISO images snapshots, during development cycle in-between the officially announced releases, directly from our build server, ABF (cf. Forum topics Most recent Cooker ISO and Most recent ROME (rolling) ISO which we keep constantly up-to-date).
Of course we have more accurate numbers about developers and commits.
There are 7 main developers, and a few people who submit a patch once in a while.
There have been 82350 commits in the last year, out of which 7323 were in the last month.
The commits go to our repositories OpenMandriva Association and OpenMandriva Software and the packages are built on ABF:  
OpenMandriva has 2 main entities:
Council for what concerning the legal/paperworks, PR and organization side;
Technical Committee for what concerning the products’ technical development side.
The decisions are made depending on the specific subject however more often than not they are virtually identical pertaining to both sides.
When at all possible we aim to reach consensus for final decisions. Crucial help is provided by the shared target and common sense.
People who have been contributing consistently over some time are invited into the relevant entities.
- Main website https://www.openmandriva.org/
- Forum https://forum.openmandriva.org/
(Users’ support and communication notices)
Forums are also bridged to mailing lists
- Matrix rooms
- Internet Relay Chat (libera.chat IRC network)
- Telegram (bridged)
- Social media
Fediverse (Mastodon), Twitter, Facebook
That really depends on what kind of user you are.
For all types of users, the concept of having both a traditional point release (rock – soon reaching the 5.0 milestone) and a rolling release (ROME – using time based versioning, with 22.12 expected any day now) is a good thing, since users get to choose if they want the latest and greatest features at all times (while still getting some testing - the “cooker” development tree [also available to the public, but recommended only for developers since it may break at any time] is ahead of ROME), or if they want a user experience that is essentially guaranteed not to change until they choose to update to a new point release.
For desktop users, we think we have one of the easiest to use desktops, and we get the user there without having to edit config files, open a command line, or figure out what thousands of packages are all about - the default install comes with pretty much everything we think a typical desktop user will need. You will also find a system that performs well, even on older hardware.
For developers and experienced users, there’s the up to date clang based toolchain (with current gcc also available), availability of many tools you won’t find in most other distributions’ repositories (e.g. crosscompilers to all relevant architectures, fiptool, github-cli are just a “dnf install” away), picking the best tool even if it’s not the one traditionally used in Linux systems (e.g. our default “tar” implementation is libarchive tar - allowing the tar command to work on zip files etc. as well as tarballs), and generally up to date set of libraries.
Server users may well appreciate current versions of the PHP and Java stacks and always up to date PostgreSQL and MariaDB, our docker containers, and the performance (booting our Ampere Altra server with the preloaded CentOS took more than 7 minutes - which we cut down to less than 4 by installing an OpenMandriva snapshot).
There are multiple target audiences – each supported by a different set of features.
Our traditional target audience has always been desktop users, and that is still one of the most important topics. For a typical desktop user, we provide an installation image with everything they would need that is easy to set up, and a tool that will make the user feel right at home by customizing the desktop to resemble something they are familiar with already - be it Windows or MacOS.
Another important audience is developers - of course preferably developers willing to help us work on our projects. Developers are rewarded with a state-of-the-art toolchain, current kernels (including an effort to keep the latest -rc kernel packaged as well as the mainline version), current libraries, and loads of development tools in the repositories. One feature currently under construction will make life much easier for developers who have to/want to work on multiple CPU architectures.
A more recent addition is server users - we have been working with Ampere to get OMLx running on their aarch64 servers and inside cloud services running on them, such as Oracle Cloud. Here our aim is to prove that stable is spelled with a “b” in the middle - and doesn’t have to imply “stale”. We believe a system can be stable without relying on technologies that are a few years behind the time. Of course we also support x86 servers.
We believe in eating our own dog food (of course our mascot Chwido enjoys that in particular): we are running OMLx everywhere. ABF (https://abf.openmandriva.org), our build server, is an ARM box running ROME.
Pretty much all hardware you’ll find in a traditional PC is supported - with the obvious limitations to supporting Nvidia GPUs if you want to be true to Open Source. (And we definitely want to be true to Open Source - fortunately the Nouveau driver is good enough to be workable. But people getting new hardware should avoid Nvidia.)
On top of that, we support a number of AArch64 (ARM) devices - be it Raspberry Pi 3, 4 or 400, Rock 4B (Rock5 is in the works and may well be ready by the time the article is published), Pinebook Pro, or a high end Ampere Altra server. We are also working on a PinePhone and PinePhone Pro port - and it has the basics working, but for most users, it isn’t ready to replace an AOSP based phone. Unlike many other distributions, we have decided not to require UEFI support on AArch64 devices - while that is one option we are providing, we don’t want to leave out interesting hardware that doesn’t implement UEFI. To make that easier, we have developed os-image-builder, which essentially builds a custom bootloader (and possibly kernel), grabs all the OMLx packages for the userland, and then builds a custom image. It’s essentially a way between the Yocto-like “build it all from source so we can target even the most obscure hardware” and the traditional “here’s one binary image, if your hardware doesn’t support what it needs, go get something else” approaches.
Maybe the most interesting topic in new hardware support is that we are trying to make OMLx available on some super cheap hardware - this port is still in early stages, but we expect to have a ful fledged OMLx system running on $25 hardware soon.
We have started a port to RISC-V when the first SiFive Unlimited boards became available, but had to interrupt it because our board broke down and qemu alone didn’t cut it - but more recently we have added 2 SiFive Unmatched boards and 2 VisionFive 2 boards are on their way. We expect to fully resume the RISC-V port and bring it up to the same state as aarch64 and x86 after the 5.0 release.
Even though there is currently little “real world” hardware available, RISC-V is a very interesting target to us, primarily because of its open nature.
A big thing is that we are fully independent, not based on any other distribution (which of course doesn’t mean we don’t borrow patches from multiple others where we like them).
This gives us more room to be innovative and go new ways.
For example, on the technical side, we have been free to decide to go with Clang as our main compiler. OpenMandriva is the first distro ever built with Clang, starting with OMLx 3.x development cycle in early 2016 (Beta1 is here) - even before Android switched its compilers.
We are using a different implementation of the tar command, we are exploring a more cross-compiler friendly filesystem layout making sysroots more useful than ever, and we have started experimenting with builds using alternative libc implementations.
We have also developed a tool that automates the process of updating packages to some extent - making sure we have up-to-date versions even of most packages core developers can’t pay much attention to.
On the user experience side: Fast updates are an important thing. Cooker usually gets things like KDE Frameworks updates or new kernels the day they are released, and ROME isn’t far behind.
We also have some unique OpenMandriva brand name tools: OM Welcome, OM Control Center, Repository Selector (repo-picker), Update Configuration (om-update-config), and Desktop Presets (om-feeling-like) – a tool that helps users to configure their desktop to resemble something they are already familiar with, be that Windows, Mac OS or a different Linux system.
Of course, the official ROME 22.12 and 5.0 point release are in the immediate future. Not much more will change before they are released.
As for plans beyond that, a lot has already been mentioned in previous answers - more hardware support, filesystem layout tweaks, alternative libcs and possibly libc++ to build faster, more memory efficient variants, and more.
One more thing that we are excited about is the upcoming release of Plasma 6 – which of course we will pick up and integrate as quickly as possible. We want to do this right - we are also porting our distro specific tools to Qt 6 so we don’t end up pulling in older library versions for our own tools. We have already added Qt 6 support to many other 3rd party packages like LibreOffice.
We have a lot of plans, and not a whole lot of people - and nobody is working on OMLx as their day job. While we have shown over the last decade that we can get things done, we can always use additional manpower. Drop by our Matrix channels if you are willing to help out - regardless if you are a developer, a tester, a translator, a graphic designer, or a PR/marketing person.
We have a mascot, Chwido, who is also assisted by his ‘sister’ (some would argue ‘wife’), Laska. Chwido and Laska are my real world puppies and they would be happy to say “Woof woof!” to the readers!