:: Re: [Dng] What's new in Systemd
Page principale
Supprimer ce message
Répondre à ce message
Auteur: Nate Bargmann
Date:  
À: dng
Sujet: Re: [Dng] What's new in Systemd
* On 2015 02 Feb 07:53 -0600, Martijn Dekkers wrote:
> I found this an interesting read:
>
> * read ahead implementation dropped: in the age of SSDs the benefit is not
> big enough to have this. All systemd developers have SSDs and no more
> spinning disks, nobody could/wanted to support this anymore. The idea was
> to read-ahead the bits needed during the boot process and remember it next
> time, for faster boots. But with SSDs, this support is dropped.


I am now decloaking. I have lurked this list since its inception and
this latest revelation has motivated me to finally post.

The decision referenced above would be appropriate about two months from
now, unless someone spoiled the surprise two months early. Sadly, I
trust that they are serious. Really, just who do they think they are to
dictate what hardware a target system can use to this level? All of the
arrogance of Apple without the market capitalization.

These people are incredible.

Support dropped. Just like that. No intention of compatibility. No
desire for inclusive development. We decide, you accept. Should you
question our decisions we'll call you a "hater" (personally, I have no
hatred toward systemd or its developers as that is an emotion I reserve
for very few select things in my life). Just that easy. Wow!

> Linus's approach is "as long as *somebody* is using it, we don't drop
> support. Systemd's approach is "we now all have SSD's so we are not going
> to bother with this anymore" - not a very mature or inclusive approach. am
> I supposed to trust my server fleet to a bunch of dev's that don't really
> care about the hardware I am running? I **WISH** all my servers had only
> SSD's, unfortunately, they are still too expensive for the amount of data
> we manage.


At this point I don't have any SSDs and I suspect they are nice, but
spinning platters are serving me well at the moment at a fraction of the
price. I'm simply too conservative to be one of the cool kids.

> very unprofessional.


Not very wise, either, IMO.

Was that thud I just heard the collective jaw dropping of Debian
developers? Perhaps not, some seem genuinely excited for the Embrace,
Extend, and Extinguish of Debian by systemd et. al.

Right now I am using systemd in Sid and have not had any problems with
it. My systems are reasonably simple and are home desktops running
Xfce. Once I sorted logging to syslogd and a couple of other minor
nits, all has been well. I'm no fan of the pile of scripts in
/etc/init.d either, but I can work with them to some degree. For all
the ease it has brought to my desktops, udev has probably been the most
vexing thing to control and now that it has been assimilated by systemd,
future customizations are probably hopeless.

Now to change the subject a bit.

IMO, it is time for a Devuan Alpha release, developer preview, or
something for this project to show the world it is serious and
developing a distribution. I do understand the desire to get things
reasonably stable prior to a beta release but remember Linus' mantra
from the early '90s, "Release early; release often." I also understand
scheduled plans and "When it's ready", but it is time for this project
to start putting its cards on the table.

It is time.

- Nate

--

"The optimist proclaims that we live in the best of all
possible worlds. The pessimist fears this is true."

Ham radio, Linux, bikes, and more: http://www.n0nb.us