[e7620]

systemd is not an init system, that's how it started, but they state they want to comprise a whole OS.

Debian may seem too big to die, but they are slowly being commoditized away, because RedHat is known to offer better sales force, better service and support:

   How do you differentiate your product if your core mission is to ensure
   that your product operates exactly as your competition?  The bottom
   line is that you don't

[toyg]

> because RedHat is known to offer better sales force, better service and support

When did Debian ever offer support or sales? Debian is not a corporation, it works under a different paradigm. Debian will be the last Linux distribution to die, and it might even survive as a non-Linux OS. As long as there are people who can benefit from a free open-source operating system, and people willing to dedicate their time to make it possible, Debian will live on.

[e7620]

That's precisely the point, it's part of the quote, the more Debian standardizes on the upcoming systemd distro, it'll get erased by RedHat real fast, who offers a differentiating advantage.

   How do you differentiate your product if your core mission is to ensure
   that your product operates exactly as your competition?  The bottom
   line is that you don't .... Theoretically, you could have a better
   sales force or better service and support .... Yet these are the assets
   of the larger, entrenched companies.

> As long as there are people who can benefit from a free open-source operating system

As if Debian were the only free open-source OS... Not even considering only GNU/Linux, where Slackware was first (why not also last?).

> and people willing to dedicate their time

A complete OS will take much more than that, if Debian loses relevance, people will leave. What differentiating advantage to choose Debian over CentOS or Fedora? Debian will have to fight that battle, being or not a corporation is irrelevant.

I hope Debian endures, but you have to understand that systemd standardization is not going to be positive to Debian relevance.

[toyg]

> As if Debian were the only free open-source OS

It's the only free open-source OS backed by an explicitly democratic approach, enshrined in its constitution. It's "the GNU System that works, will always be Free, and will give an equal say to any developer" (at least in theory). All the other distributions are owned by a specific group of people (or in the RedHat case, by shareholders) and/or are not Free. That's the Debian differentiator that RedHat will never be able to match, no matter how many "community editions" they sponsor.

> What differentiating advantage to choose Debian over CentOS or Fedora?

Those are both owned by a corporation.

> being or not a corporation is irrelevant

I respectfully disagree there. You just have to look at the evolution of the Linux ecosystem to see how "being or not a corporation" makes a huge difference. History is littered with corpses of Linux vendors. In fact, there is an argument for big Linux projects being naturally incapable of making money as corporations in the long run, a concept that was seriously challenged only by RedHat and Ubuntu at this point.

> you have to understand that systemd standardization is not going to be positive to Debian relevance.

My point is, if that's the case, the project will likely have the necessary strength to reconsider and correct this choice later on. It's not like they can "run out of money" or something like that; they have such a huge mindshare that it would take ages to dissipate, and votes seem to indicate that most developers don't really care about the init system that much. If things come to worst, Jesse will just go down in history as a terrible release (wouldn't be the first...) and the project will move on.

If anything, it's downstream projects that have to worry (i.e. Ubuntu) since they have to differentiate in a competitive market, but they seem to have already adopted systemd, so...

[e7620]

> "being or not a corporation" makes a huge difference

Take a look at Mozilla, they implemented DRM, otherwise they could have lost relevance, they explained.

Now Debian must commit resources just making everything work with every new "innovation" brought by systemd. RedHat will dictate the pace and the terms, Debian will follow, and once the future "systemd + linux OS" integration has been declared standard, they can't correct the decision, they will be stuck with "systemd OS" forever.

[toyg]

That's as much a guess as mine. We'll see in time.

[hga]

"because RedHat is known to offer better sales force, better service and support"

And tremendously longer support periods.

Right now I'm helping a non-profit move from squeeze and XCP/XenCenter (which is CentOS), which I initially set up for them, to wheezy. And we're looking rather enviously at the RHEL/CentOS support periods. Even before this systemd debacle, something else was looking likely for when wheezy support ends.

[e7620]

Don't quote me on this, but I recall reading from official channels that the last Debian version without mandatory systemd is going to see its support period indefinitely extended for a long time. Good luck! :)

[hga]

If it's anything like the squeeze LTS experiment it's not going to be suitable for a number of people. There are a number of things they've been forced to abandon for various reasons.

And per the above experiment, it's not something you can depend on, not something you can build your plans around.