As I’ve been working on it for the past few weeks I thought I’d bring you some news and information about LMDE 2 aka “Betsy”.
What is LMDE
As you might know, our project maintains two distributions: Linux Mint and LMDE.
If we were to briefly compare Linux Mint and LMDE we’d probably say this:
- What is similar:
- They look and work the same, they’re both “Mint”, they provide the same user experience.
- They meet the same standards and expectations. They’re released and tested the same way.
- They both ship a MATE and a Cinnamon edition.
- What is different:
- Under the hood Linux Mint uses Ubuntu as a package base. LMDE doesn’t. In LMDE the package base is Debian.
- LMDE is usually a bit leaner, faster and more responsive than Linux Mint.
- LMDE is simpler in terms of architecture, and usually for the better. This empowers the dev. team on particular projects (the installer, locale management etc..).
- Linux Mint is more mainstream. It’s easier to find help, documentation or other people who use Linux Mint.
- Linux Mint benefits from more packages/components. Some aspects of Linux Mint aren’t available in LMDE (as we speak I can think of the Driver Manager for instance, Compiz, PPA repositories..etc).
- Within our project Linux Mint has a lot more users than LMDE, thus it gets more focus, more resources and it’s better maintained (the move to Debian Stable, in Betsy, should improve this greatly, but it’s still true that resources are allocated on projects with the most demand).
What is Betsy
LMDE was a rolling distribution for a while and eventually turned into a semi-rolling distribution. This was good at the time (among other things it made it possible for LMDE users to enjoy Cinnamon while it was still using a GNOME backend) but it also presented challenges and caused issues. The biggest issue in LMDE was the fact that it required a lot more maintenance than Linux Mint but that it had far less users. This hurt the frequency of updates it received but also the quality of the distribution.. since so much efforts were put in the base, very few LMDE-specific developments were made.
In an effort to boost both development and quality Linux Mint recently switched to LTS bases, thus keeping the same base for 2+ years and iterating releases (17, 17.1, 17.2..etc) which brought desktop innovations and new releases of user applications and desktop environments on top of the very same base. This change was hugely beneficial to Linux Mint and made wonders not only for 17.1 but also for 17 before it.
Similarly, it was decided that LMDE would adopt a frozen release cycle (just like Linux Mint) based on Debian Jessie. This new LMDE was given version “2” and codename “Betsy”. Just like Linux Mint, it will keep the same base for a couple of years until one day we start talking about “LMDE 3”.
Linux Mint 17.x is a family of releases. Although they share the same base and it’s trivial to upgrade from one to the next, each release brings a distinct set of innovations and new versions of packages. The main advantage of this is that users have the choice to upgrade or to keep the release they’re currently running, it’s an opt-in.
In LMDE 2, we’re considering skipping that. First, it has a cost, and LMDE has a much smaller user base. Second, the LMDE audience is less mainstream and usually more experienced than the one in Linux Mint. For years new versions of software got into Linux Mint first and were then ported to LMDE. We’re thinking of changing that. New software would get into LMDE first and then into Linux Mint. Getting into LMDE would mean updating Betsy, whereas getting into Linux Mint would mean being part of the next 17.x release to come.
In other words, whereas Mint 17.x would see distinct releases (Qiana, Rebecca..etc..) with opt-in upgrade paths, LMDE 2 would be continuously upgraded (it would just be Betsy basically) and simply receive updated ISO images now and then.
In line with what LMDE is to Linux Mint, this would make LMDE a tiny bit more risky than Linux Mint and a tiny bit more exciting for experienced users.
As I said in the previous chapter, we’re “considering” doing that… it’s not set in stone yet. We’ve Romeo and we’re based on Debian Stable.. but the big big thing here is that we’re on the same base for 2+ years: that means quality because it’s worth for us to invest time in tackling way more things and being more demanding than before.
This is the last big thing to iron out for Betsy. A couple of months ago we had very important decisions to make (skipping systemd and keeping sysvinit), and critical issues to fix (upgrade to syslinux 6, initrd compression, multi-hdd installations, gtk 3.14 compatibility). This is all in the past now. We’ve a Betsy which boots and installs like a breeze, no matter how many HDDs you have, whether you’re using EFI or BIOS or which even gets you online if you’re using one of these broadcom wireless chipsets :)
Many bugs were fixed in the system itself, a lot of work was done on the installer (credits to Kernc for his help on this) and the vast majority of what it’s in Mint 17.x is there too.
You can follow the Roadmap to see the remaining issues. As the quality of Betsy is getting higher and higher we’re getting close to QA testing. Of course during QA we’ll probably uncover more bugs so it’s hard to give an estimation for a release at this stage. I’d like to think we’ll see a public release in March, possibly even sooner.. we’ll see. As usual we’ll take our time, and we’ll release when ready.