Re: [gpsd-dev] Baud barf problem on Pi3 SOLVED

[Gary E. Miller]

Yo Bernd!

On Tue, 3 May 2016 21:01:56 +0200
Bernd Zeimetz <address@hidden> wrote:

> On 05/03/2016 08:31 PM, Gary E. Miller wrote:
> > Yo Bernd!
> > 
> > On Tue, 3 May 2016 20:01:12 +0200
> > Bernd Zeimetz <address@hidden> wrote:
> >   
> >> So far you did not even say which problem you have with systemd.  
> > 
> > I will admit one major problem with systemd is lack of user
> > education, so partly the users fault, and partly the fault of bad
> > documentation.  
> 
> Oh, I think there is no project better documented than systemd.
> https://www.freedesktop.org/wiki/Software/systemd/
> Even gpsd lacks documentation compared to systemd.

I'll check it out, when I'm finally forced to.

> > The main gpsd related symptom I see of this is that when people try
> > to experiment with systemd they find it difficult, maybe
> > impossible, to get systemd to stop trying to manage the gpsd
> > daemon, and the gpsd port.
> > 
> > Coupled with no clear feedback to clue the user into why his
> > attempts are being siliently thwarted.
> > 
> > Noting the problems other have with it, I have avoided systemd.  It
> > is clearly not ready for prime time.  
> 
> I think the main problem is that people fail to read the documentation
> and do not want to learn new things. If you want to work with systemd
> and want to debug issues, you have to understand it first. And systemd
> is something you will not understand by reading some random
> shellscripts.

Thus systemd is unsuitable for those that are not toally hannd-off or 
gurus.

> > They fail to know, much less understand, the corners of the complex
> > problems they are trying to solve, then come up with inflexible
> > solutions, user hostile solutions, that work for the mainstream user
> > that never tweaks his distro.  Being inflexible they violate core
> > UNIX priciples and make customization impossible.  
> 
> Oh well, I think YOU also never read the documentation.

I readily admit I have not read the systemd doc.  The screams of others
are all I need to wait.  Those that live on the bleeding edge are showing
more blood stains than usual.

> So far I did not find a single I was not able to handle with systemd.
> Actually it is much more flexible than sysvinit or other init systems.

I'm an openRC fan myself.  If systemd had stuck to being a launchd clone
it would have been interesting.

> The major pain at the moment is that too many programs are still stuck
> with sysv init scripts.

Thus the newbie mistake of failing to ensure back compatibility.  Programmers
that say 'my way or the highway' are not my friends.


> You might want to read
> http://0pointer.de/blog/projects/the-biggest-myths.html - it sounds a
> bit  like a blog post to play bullshit bingo with, but Lennart is
> actually right.

Just read it.  Wow, that was depressing...

"The lady doth protest too much, methinks"

I have personal experience that several things he says are true.

> > I can see why for support reasons Ubuntu wants to inflict
> > inflexibility on its users, but any hacker will feel trappped.
> > 
> > Newbies gotta start somewhere, but not in the init system.  
> 
> Lennart is definitely not a newbie. He might have strong opions about
> how things need to be done, but he is creating and/or contributing to
> open-source-projects for years.

OK, so maybe more Dunning-Kruger?

> https://codesearch.debian.net/results/poettering/page_0

Bad link?

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

pgpdSzHrsu1wa.pgp
Description: OpenPGP digital signature