[ale] Gearing up for the future (wuz: boot speed, systemd, vi vs emacs, etc)

[Steve Litt]

On Mon, 19 Feb 2018 11:24:26 -0500
Leam Hall via Ale <ale at ale.org> wrote:

> On 02/19/2018 10:19 AM, Jerald Sheets via Ale wrote:
> > So let me interject into this conversational process a “norm”
> > that’s evolving out in silicon valley.  
> 
> 
> Let me render an opinion based on Jerald's comments. Containerization 
> will change "things" as much as VMWare. SMBs will use a cloud
> provider (AWS, Linode). Large enterprises will use thin clients or
> portable workstations to connect to their environment. We've probably
> all seen the signs.
> 
> Griping about systemd feels good but doesn't prepare me for the next 
> career challenges. So, what to do?
> 
> We have, individually and collectively, at least a few choices.
> 
> 	1. Change careers so systemd/boot times/containers don't
> matter. 2. Hang out with the legacy systems.
> 	3. Join those moving to the "new thing".
> 	4. Help enable the "even newer thing".
> 
> I'd say the "new thing" (#3) is cloud and the "even newer thing" is 
> containers (#4). There are multiple container technologies and the 
> market is likely to settle on one.
> 
> As I look at my career and this list the path forward isn't clear. I
> get paid for #2 while #1 has been a long standing option. Would need
> energy and a team to do #4 while #3 is a safe forward option.
> However, "safe" is relative; they can find younger and cheaper cloud
> admins.
> 
> Thoughts?

One thought. Systemd was not the result of a meritocracy: It was
financed by Redhat, who, as a purveyor of training and consultants, has
everything to gain by a universally more complex GNU/Linux. During the
coup, Redhat must finance the development and troubleshooting of
systemd, and it doesn't come cheap.

I'm pretty sure they expected everyone to accept systemd by late 2015.
I think they might abandon systemd pretty soon, for yet another "new
thing". If my suspicion is true, you can take your systemd learning and
throw it in the trashcan, because it stops having relevance.

I'm a purveyor of books and courses about troubleshooting, so to a
certain extent I can afford to ignore what's going with one little
section of one operating system. You're not in that position: I suggest
you learn a lot about systemd, make money with systemd, but be ready to
jump to the next thing (which might be even worse).

I'd also suggest that, just to keep your perspective on how a computer
works at the lower levels, you make your home computer init with runit
or s6. The combination of systemd knowledge and runit or s6 knowledge
will make you rare, and will frequently enable you to solve problems
others can't.

SteveT