Re: [dev] Never Ending Systemd Chronicles

From: Eon S. Jeon <>
Date: Thu, 18 Aug 2016 01:53:14 +0900

Hello there,

> On Aug 8, 2016, at 11:57, Alex Pilon <> wrote:
>> while a shell script can be understood by anybody.
> Bullshit. How often have I had to check people's lockfile code, or
> manual isolation?

So, you did understand, eh?

Personally, I hate definitive DSLs, because their use is very limited while it requires much time understanding internal states and details that are prone to change.

Learn New Technology¢â is the usual rhetoric here, but new isn't a synonym of better, and no one is obliged to learn every new technology. systemd is merely newer and does not directly benefit most people, so I don't want to know what it *precisely* does.

> Having policy built on top of mechanism is a Good Thing¢â ... Having mechanism
> alone is not a good thing.

Claims without basis.

Having no rigid policy other than function itself is a way to open up possibilities. This makes system more flexible, and that's where happy "hacking" happens.

You have heard that Linux is about choices, right? That doesn't mean there are whole bunch of cost-free software you can choose from. It means you can choose any software to build up an organic system. Flexibility is what being assumed, but systemd isn't flexible at all.

Other than these, well...

Too big. I don't mind it using dozens of MB of memory, but I can't stand its monstrous LOC. Many procedures are just translated shell scripts, pushing up LOC pointlessly. Also, too many modules in one pot, which encourages tight coupling b/w them.

Breaking too much. That's what communists did in both Soviet and China. I'm serious. It's a way of establishing a strong dictatorship. Rather than communicating and discussing with stakeholders, boom, "I know better, so you listen to me."... Let's hope we don't get any Stalin popping up.

Received on Wed Aug 17 2016 - 18:53:14 CEST

This archive was generated by hypermail 2.3.0 : Wed Aug 17 2016 - 19:00:14 CEST