Re: [GTALUG] Red Hat Paywall...

2023-07-18 Thread Evan Leibovitch via talk
On Thu, Jul 6, 2023 at 7:04 AM BCLUG via talk  wrote:

> Evan Leibovitch via talk wrote on 2023-07-04 03:39:
>


> >  4. I see an opportunity for SUSE which maintains both an
> > enterprise-Linux focus and good community relations. Are they up to
> > it? As a longshot maybe even Oracle could try to seize the moment
> > and try a charm offensive to attract a community... but that's
> > unlikely considered its many burned bridges (Solaris, OpenOffice,
> Java)
>
> That's an interesting idea - the rise of SUSE / OpenSuse. I kinda like the
> idea - I don't have personal experience with it, but it seems
> well-regarded.
>
> Seems the whole English speaking world (of podcasters and YouTubers) has a
> bit of a blind spot to SUSE and they probably deserve more attention.
>
> Good call, hope it pans out.
>

This path just got WAY more interesting. SUSE has committed to creating and
maintaining an open source fork of RHEL:

https://news.itsfoss.com/suse-rhel-fork/

- Evan
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-06 Thread BCLUG via talk

Evan Leibovitch via talk wrote on 2023-07-04 03:39:


My take:

 2. IBM doesn't really give a damn about Alma and Rocky, they're just
incidental casualties. The #1 and maybe only target of the
subscription-wall action is IBM's longtime arch-enemy Oracle, which
may now be forced to actually maintain its own distro


This was one of my first thoughts too - Oracle will no longer be able to 
"leech" off of RedHat, which is rather delicious considering Oracle's 
reputation.




 4. I see an opportunity for SUSE which maintains both an
enterprise-Linux focus and good community relations. Are they up to
it? As a longshot maybe even Oracle could try to seize the moment
and try a charm offensive to attract a community... but that's
unlikely considered its many burned bridges (Solaris, OpenOffice, Java)


That's an interesting idea - the rise of SUSE / OpenSuse. I kinda like 
the idea - I don't have personal experience with it, but it seems 
well-regarded.


Seems the whole English speaking world (of podcasters and YouTubers) has 
a bit of a blind spot to SUSE and they probably deserve more attention.


Good call, hope it pans out.


As for Oracle, they have free / cheap cloud hosting for tiny instances, 
so maybe they'll try a charm offensive, however even with a free 
tiny-VPS, they don't seem to have much mind share.


I can't really see this.  Charm & Oracle don't belong in the same sentence.


rb

---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-06 Thread BCLUG via talk

Giles Orr via talk wrote on 2023-07-04 18:09:


But for the last several years systemd has been rock-steady, and once
you wrap your head around the basics, it's a LOT easier to use than
maintaining those damn /etc/rc.N/  folders.  At least that's been my
experience.


My experience too - systemd is fine and is much easier to maintain, 
better documented, and has more features.



And, since it's used in all the major Linux distros, I feel reasonably
comfortable that it's thoroughly tested.



If I reject systemd via a "boutique" distro, what else am I giving up?

Devuan - is the team there large enough to keep up with the 
ever-changing security issues, features, etc?


If systemd were anathema to someone, shouldn't they move to BSD, where a 
large and experienced team exists?




Thought I'd add my 2c


rb

---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-05 Thread Anthony de Boer via talk

On 2023-07-05 08:57, Giles Orr via talk wrote:

...  Makes sense in a depressing
way: Debian wants a _stable_ version for stable, and testing is for
testing things that will become the stable distro ...

Huh - testing is called "trixie."  I always wonder which toy is coming next ...


Debian testing runs in a cycle with their releases. Just after the 
preceding one (bookworm) came out, it opened to a lot of heavy-lifting 
upgrades and you're more likely to see glitches. Then they start calling 
code freezes and stabilize things when they get closer to releasing the 
next one (trixie) in about two years and you'll see fine-tuning of bugs 
in the pre-release era and testing will be a lot more stable for any 
innocent users, then comes the grand release announcement and the next 
new cycle.


Anthony

---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-05 Thread Giles Orr via talk
On Tue, 4 Jul 2023 at 23:01, D. Hugh Redelmeier via talk
 wrote:
>
> | From: Giles Orr via talk 
>
> Thank you and others for telling us about your experience.
>
> |   If I need something newer, I can "pin" a
> | package, getting it either from testing or backports.  This is
> | admittedly a PITA to set up, and I hardly ever use it because of that
> | ... but I _have_ used it, and it proved reasonably reliable in
> | practice.  Although getting your head around pinning weights is ...
> | nasty.
>
> Can you use that to get the newer FireFox that you need?

Uhhh ...  I had never considered it, which seems more than a little
blind.  I haven't used pinning in a couple years, and when I did it
was usually to get a piece of software that wasn't in the current
distro (but was in the upcoming release).

However, on closer examination (I used this page:
https://www.debian.org/distrib/packages - you should become close
friends with this if using Debian) "testing" also includes Firefox
ESR, so the answer would appear to be no.  Makes sense in a depressing
way: Debian wants a _stable_ version for stable, and testing is for
testing things that will become the stable distro ...

Huh - testing is called "trixie."  I always wonder which toy is coming next ...

-- 
Giles
https://www.gilesorr.com/
giles...@gmail.com
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-05 Thread o1bigtenor via talk
On Tue, Jul 4, 2023 at 10:01 PM D. Hugh Redelmeier via talk
 wrote:
>
> | From: Giles Orr via talk 
>
> Thank you and others for telling us about your experience.
>
> |   If I need something newer, I can "pin" a
> | package, getting it either from testing or backports.  This is
> | admittedly a PITA to set up, and I hardly ever use it because of that
> | ... but I _have_ used it, and it proved reasonably reliable in
> | practice.  Although getting your head around pinning weights is ...
> | nasty.
>
> Can you use that to get the newer FireFox that you need?
>

I've been using Firefox-esr for years from debian testing.
Seems to work reasonably. At present its my predominant use
browser.

HTH
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-04 Thread D. Hugh Redelmeier via talk
| From: Giles Orr via talk 

Thank you and others for telling us about your experience.

|   If I need something newer, I can "pin" a
| package, getting it either from testing or backports.  This is
| admittedly a PITA to set up, and I hardly ever use it because of that
| ... but I _have_ used it, and it proved reasonably reliable in
| practice.  Although getting your head around pinning weights is ...
| nasty.

Can you use that to get the newer FireFox that you need?
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-04 Thread Giles Orr via talk
A couple of points following on to 01bigtenor's reply:

According to Wikipedia: "Devuan is a fork of the Debian Linux
distribution that uses sysvinit, runit or OpenRC instead of systemd.
Devuan aims to avoid 'lock-in' by projects like systemd and aims to
maintain compatibility with other init systems to avoid detaching
Linux from other Unix systems."  When systemd was initially a horrible
mess, I was very much in favour of this.  But for the last several
years systemd has been rock-steady, and once you wrap your head around
the basics, it's a LOT easier to use than maintaining those damn
/etc/rc.N/ folders.  At least that's been my experience.

As for Debian stable vs. testing: 20 years ago Debian testing was
great ... about a decade ago I found it to be noticeably less stable,
and the constant volume of updates got on my nerves.  So I switched to
stable.  Even if the stability of testing has improved, I don't want
to put up with the updates.  If I need something newer, I can "pin" a
package, getting it either from testing or backports.  This is
admittedly a PITA to set up, and I hardly ever use it because of that
... but I _have_ used it, and it proved reasonably reliable in
practice.  Although getting your head around pinning weights is ...
nasty.

On Tue, 4 Jul 2023 at 08:24, o1bigtenor via talk  wrote:
>
> On Tue, Jun 27, 2023 at 10:22 AM D. Hugh Redelmeier via talk
>  wrote:
> >
> > This Red Hat change concerns me.
> >snip
>
> >
> >   It feels as if RH steers the future of Linux by making so many
> >   contributions.
> >
> > - Ubuntu LTS + fresh Ubuntu has been pretty good.  I've had more
> >   problems with package updates on Ubuntu than of Fedora, but it has
> >   been pretty good.  Distro version upgrades have been good but not
> >   perfect in my modest experience.
> >
> >   Canonical has repeatedly acted in ways that offend or scare me.  So
> >   Ubuntu, although easy, feels like a potential trap.
>
> Found this to be true - - - - spent a lot of time a number of years ago
> looking into LXD  - - - the snap environment is one that NEVER will
> be seen here again if I can help it.
>
> Canonical is, imo, desperately looking for ways to monetize their
> brand - - - - - someone's just have to have more $.
> >
> > - debian Stable + Testing + Unstable.  I don't have much experience
> >   with debian.  I fear that the lack of full-time paid engineers might
> >   reduce the safety relative to RH (that could easily just be FUD).
> >   debian's goals are good by me.
> >
> > So: I'm thinking of switching to debian.
>
> I was with Debian for over 10 years - - - have now switched to
> Devuan - - - thereby getting rid of another item of 'control'.
> >
> > I'd like to learn from others.  How do you choose to solve these
> > problems?  Maybe some of them are non-problems.
>
> The problem is that the small encroachments don't tend to isolate
> - - they tend to grow - - - somewhat like microbes! (With similar results
> in my experience!).
> >
> > 
> >
> > Giles has a problem with needing a stable distro with a more recent
> > FireFox.  I suggested, against my preferences, that this might be a
> > perfect use for Snaps/Flatpacks.
>
> Snaps are a system controlling adventure - - - be aware of this BEFORE
> starting down that road. Haven't used flatpacks.
>
> I find that browsers are needing updates almost on a daily basis.
> Am wondering if there is a way of reducing the 'encroachment' of
> the nefarious bits of cruft adhering to all browsers (at least as far
> as I see).
> >
> > I wonder if I should be using a stable distro everywhere but with
> > containerized upgraded packages where they matter.  I yet don't think
> > so.
> >
> > The rest of my family uses Fedora on their workstations.  But they
> > hate applying updates (even when I do the work).  They are way behind
> > most of the time.  Maybe a stable distro + a fresh FireFox would be
> > best for them too.
>
> I've already been informed that if I weren't doing updates windows would be
> applied post haste - - - something about the devil they know (Work systems
> are all M$ Win!)
> >
> > How many other packages would I need to have fresher-than-stable?
> >
> > - support for newer hardware
> >
> > - compilers etc.
> >
> > - more pain-points would be discovered.
> >
> > 
> >
> > A fundamental problem is that feature changes and bug fixes are
> > usually mingled in upstream.  In some cases, it is a false
> > distinction.  Few developers want to maintain a bunch of old releases.
> > It is very hard for a distro to correctly separate these two, and yet
> > that is required to maintain a stable distro.
>
> I tend to run in debian's equivalent of 'testing'. Found over the last more
> than 10 years that that was a reasonable compromise to stability  and
> currentness.
>
> HTH
> ---
> Post to this mailing list talk@gtalug.org
> Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk



-- 
Giles

Re: [GTALUG] Red Hat Paywall...

2023-07-04 Thread o1bigtenor via talk
On Tue, Jun 27, 2023 at 10:22 AM D. Hugh Redelmeier via talk
 wrote:
>
> This Red Hat change concerns me.
>snip

>
>   It feels as if RH steers the future of Linux by making so many
>   contributions.
>
> - Ubuntu LTS + fresh Ubuntu has been pretty good.  I've had more
>   problems with package updates on Ubuntu than of Fedora, but it has
>   been pretty good.  Distro version upgrades have been good but not
>   perfect in my modest experience.
>
>   Canonical has repeatedly acted in ways that offend or scare me.  So
>   Ubuntu, although easy, feels like a potential trap.

Found this to be true - - - - spent a lot of time a number of years ago
looking into LXD  - - - the snap environment is one that NEVER will
be seen here again if I can help it.

Canonical is, imo, desperately looking for ways to monetize their
brand - - - - - someone's just have to have more $.
>
> - debian Stable + Testing + Unstable.  I don't have much experience
>   with debian.  I fear that the lack of full-time paid engineers might
>   reduce the safety relative to RH (that could easily just be FUD).
>   debian's goals are good by me.
>
> So: I'm thinking of switching to debian.

I was with Debian for over 10 years - - - have now switched to
Devuan - - - thereby getting rid of another item of 'control'.
>
> I'd like to learn from others.  How do you choose to solve these
> problems?  Maybe some of them are non-problems.

The problem is that the small encroachments don't tend to isolate
- - they tend to grow - - - somewhat like microbes! (With similar results
in my experience!).
>
> 
>
> Giles has a problem with needing a stable distro with a more recent
> FireFox.  I suggested, against my preferences, that this might be a
> perfect use for Snaps/Flatpacks.

Snaps are a system controlling adventure - - - be aware of this BEFORE
starting down that road. Haven't used flatpacks.

I find that browsers are needing updates almost on a daily basis.
Am wondering if there is a way of reducing the 'encroachment' of
the nefarious bits of cruft adhering to all browsers (at least as far
as I see).
>
> I wonder if I should be using a stable distro everywhere but with
> containerized upgraded packages where they matter.  I yet don't think
> so.
>
> The rest of my family uses Fedora on their workstations.  But they
> hate applying updates (even when I do the work).  They are way behind
> most of the time.  Maybe a stable distro + a fresh FireFox would be
> best for them too.

I've already been informed that if I weren't doing updates windows would be
applied post haste - - - something about the devil they know (Work systems
are all M$ Win!)
>
> How many other packages would I need to have fresher-than-stable?
>
> - support for newer hardware
>
> - compilers etc.
>
> - more pain-points would be discovered.
>
> 
>
> A fundamental problem is that feature changes and bug fixes are
> usually mingled in upstream.  In some cases, it is a false
> distinction.  Few developers want to maintain a bunch of old releases.
> It is very hard for a distro to correctly separate these two, and yet
> that is required to maintain a stable distro.

I tend to run in debian's equivalent of 'testing'. Found over the last more
than 10 years that that was a reasonable compromise to stability  and
currentness.

HTH
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-07-04 Thread Giles Orr via talk
I was just reading about it - this is an interesting (depressing) take
from Bradley Kuhn:

https://sfconservancy.org/blog/2023/jun/23/rhel-gpl-analysis/

On Tue, 4 Jul 2023 at 06:40, Evan Leibovitch via talk  wrote:
>
> My take:
>
> I wouldn't go as far as "I saw this coming", but I have long suspected that 
> the IBMification of Red Hat was far from complete -- layoffs, CentOS Stream, 
> now this. RH employees that I know describe a hard shift in corporate 
> culture. And I don't think they're done. It wouldn't surprise me the least if 
> they even change the name of the product to "IBM Red Hat Enterprise Linux" or 
> even just "IBM Linux". Now that they've effectively (and knowingly) destroyed 
> the community goodwill that was formed over more than a decade of Linux Expos 
> and Bob Young roadshows, I don't even see much added value in the RH brand to 
> IBM; the Red Hat we've known for decades no longer exists. Come to think of 
> it, the IBM that helped start LPI and championed Open Source against the SCO 
> and Java assaults of a decade ago is also long gone.
>
> IBM doesn't really give a damn about Alma and Rocky, they're just incidental 
> casualties. The #1 and maybe only target of the subscription-wall action is 
> IBM's longtime arch-enemy Oracle, which may now be forced to actually 
> maintain its own distro and will no longer be able to claim bug-for-bug 
> compatibility with RHEL (or whatever it will be called). They've calculated 
> that the value of the harm this causes Oracle exceeds the lost value of 
> community rejection.
>
> This unfortunate momentum could be stopped (or at least slowed) by a Fedora 
> developer revolt but I don't see that happening.
>
> I see an opportunity for SUSE which maintains both an enterprise-Linux focus 
> and good community relations. Are they up to it? As a longshot maybe even 
> Oracle could try to seize the moment and try a charm offensive to attract a 
> community... but that's unlikely considered its many burned bridges (Solaris, 
> OpenOffice, Java)
>
> Evan Leibovitch, Toronto Canada
> @evanleibovitch / @el56
>
>
> On Tue, Jun 27, 2023 at 11:23 AM D. Hugh Redelmeier via talk 
>  wrote:
>>
>> This Red Hat change concerns me.
>>
>> LONG: Some thoughts on what my "go to" distro pair should be.
>>
>> | From: Alvin Starr via talk 
>>
>> | On 2023-06-27 08:19, Lennart Sorensen via talk wrote:
>>
>> | > Yeah I am happy I switched to debian 25 years ago because Red Hat's
>> | > quality was so poor at the time.  Debian having a better designed
>> | > packaging system was a bonus.
>> | >
>> | Strangely around the same time I switched to RedHat because I got tired of
>> | having to apply my own security patches to the kernel and applications 
>> because
>> | the distribution was shipping with largely unmodified sources.
>> | Like many things in life "your mileage may vary".
>>
>> The RHEL / CentOS / clones drama is certainly unsettling.
>>
>> I don't think that Fedora is directly affected but it is hard to judge
>> whether there will be secondary effects.
>>
>> One upcoming GTALUG talk will be from a Rocky Linux guy.  That should
>> be interesting.
>>
>> I've already been struggling with where I want to go for a stable
>> system.  Besides the drama, I just don't think that RHEL's pinning
>> versions for 5 years is a good strategy.  Backporting for that long
>> feels like a wasted effort, prone to errors.
>>
>> Why do I care about the effort RH puts into backporting?
>>
>> - it creates a RH "moat": it prevents others from competing with them.
>>   Rocky, Alma, Oracle feel like clones, not creative competitors.
>>   That may be unfair to Oracle but that's not a company that I want
>>   a relationship with.
>>
>> - it is labour that feels wasted.  Perhaps that labour could be used
>>   for more constructive purposes.
>>
>> On the other hand RH has added a lot to the community and does do a
>> good job of beating back bugs.
>>
>> I do think that I need a pair of distros: one that is up to date, and
>> one that is low-drama.  If they are in the same family, that cuts down
>> on redundant learning on my part.
>>
>> - CentOS + Fedora has been a good pair for me.  TBH, CentOS has left
>>   me with technical debt: I get stuck on obsolete versions because the
>>   upgrade paths have been disrupted (twice!).  Fedora release updates
>>   have been good for some years.
>>
>>   For my workstation (desktop and laptop) use, I've been very happy
>>   with Fedora, but it sure has a firehose of updates.  I don't think
>>   that it is affected directly by any of this.  But if a lot of people
>>   migrate away from Red Hat stuff, it won't likely be good for Fedora.
>>
>>   It feels as if RH steers the future of Linux by making so many
>>   contributions.
>>
>> - Ubuntu LTS + fresh Ubuntu has been pretty good.  I've had more
>>   problems with package updates on Ubuntu than of Fedora, but it has
>>   been pretty good.  Distro version upgrades have been good but not
>>   perfect in 

Re: [GTALUG] Red Hat Paywall...

2023-07-04 Thread Evan Leibovitch via talk
My take:

   1. I wouldn't go as far as "I saw this coming", but I have long
   suspected that the IBMification of Red Hat was far from complete --
   layoffs, CentOS Stream, now this. RH employees that I know describe a hard
   shift in corporate culture. And I don't think they're done. It wouldn't
   surprise me the least if they even change the name of the product to "IBM
   Red Hat Enterprise Linux" or even just "IBM Linux". Now that they've
   effectively (and knowingly) destroyed the community goodwill that was
   formed over more than a decade of Linux Expos and Bob Young roadshows, I
   don't even see much added value in the RH brand to IBM; the Red Hat we've
   known for decades no longer exists. Come to think of it, the IBM that
   helped start LPI and championed Open Source against the SCO and Java
   assaults of a decade ago is also long gone.

   2. IBM doesn't really give a damn about Alma and Rocky, they're just
   incidental casualties. The #1 and maybe only target of the
   subscription-wall action is IBM's longtime arch-enemy Oracle, which may now
   be forced to actually maintain its own distro and will no longer be able to
   claim bug-for-bug compatibility with RHEL (or whatever it will be called).
   They've calculated that the value of the harm this causes Oracle exceeds
   the lost value of community rejection.

   3. This unfortunate momentum could be stopped (or at least slowed) by a
   Fedora developer revolt but I don't see that happening.

   4. I see an opportunity for SUSE which maintains both an
   enterprise-Linux focus and good community relations. Are they up to it? As
   a longshot maybe even Oracle could try to seize the moment and try a charm
   offensive to attract a community... but that's unlikely considered its many
   burned bridges (Solaris, OpenOffice, Java)

Evan Leibovitch, Toronto Canada
@evanleibovitch / @el56


On Tue, Jun 27, 2023 at 11:23 AM D. Hugh Redelmeier via talk <
talk@gtalug.org> wrote:

> This Red Hat change concerns me.
>
> LONG: Some thoughts on what my "go to" distro pair should be.
>
> | From: Alvin Starr via talk 
>
> | On 2023-06-27 08:19, Lennart Sorensen via talk wrote:
>
> | > Yeah I am happy I switched to debian 25 years ago because Red Hat's
> | > quality was so poor at the time.  Debian having a better designed
> | > packaging system was a bonus.
> | >
> | Strangely around the same time I switched to RedHat because I got tired
> of
> | having to apply my own security patches to the kernel and applications
> because
> | the distribution was shipping with largely unmodified sources.
> | Like many things in life "your mileage may vary".
>
> The RHEL / CentOS / clones drama is certainly unsettling.
>
> I don't think that Fedora is directly affected but it is hard to judge
> whether there will be secondary effects.
>
> One upcoming GTALUG talk will be from a Rocky Linux guy.  That should
> be interesting.
>
> I've already been struggling with where I want to go for a stable
> system.  Besides the drama, I just don't think that RHEL's pinning
> versions for 5 years is a good strategy.  Backporting for that long
> feels like a wasted effort, prone to errors.
>
> Why do I care about the effort RH puts into backporting?
>
> - it creates a RH "moat": it prevents others from competing with them.
>   Rocky, Alma, Oracle feel like clones, not creative competitors.
>   That may be unfair to Oracle but that's not a company that I want
>   a relationship with.
>
> - it is labour that feels wasted.  Perhaps that labour could be used
>   for more constructive purposes.
>
> On the other hand RH has added a lot to the community and does do a
> good job of beating back bugs.
>
> I do think that I need a pair of distros: one that is up to date, and
> one that is low-drama.  If they are in the same family, that cuts down
> on redundant learning on my part.
>
> - CentOS + Fedora has been a good pair for me.  TBH, CentOS has left
>   me with technical debt: I get stuck on obsolete versions because the
>   upgrade paths have been disrupted (twice!).  Fedora release updates
>   have been good for some years.
>
>   For my workstation (desktop and laptop) use, I've been very happy
>   with Fedora, but it sure has a firehose of updates.  I don't think
>   that it is affected directly by any of this.  But if a lot of people
>   migrate away from Red Hat stuff, it won't likely be good for Fedora.
>
>   It feels as if RH steers the future of Linux by making so many
>   contributions.
>
> - Ubuntu LTS + fresh Ubuntu has been pretty good.  I've had more
>   problems with package updates on Ubuntu than of Fedora, but it has
>   been pretty good.  Distro version upgrades have been good but not
>   perfect in my modest experience.
>
>   Canonical has repeatedly acted in ways that offend or scare me.  So
>   Ubuntu, although easy, feels like a potential trap.
>
> - debian Stable + Testing + Unstable.  I don't have much experience
>   with debian.  I fear that the lack of 

Re: [GTALUG] Red Hat Paywall...

2023-06-27 Thread D. Hugh Redelmeier via talk
This Red Hat change concerns me.

LONG: Some thoughts on what my "go to" distro pair should be.

| From: Alvin Starr via talk 

| On 2023-06-27 08:19, Lennart Sorensen via talk wrote:

| > Yeah I am happy I switched to debian 25 years ago because Red Hat's
| > quality was so poor at the time.  Debian having a better designed
| > packaging system was a bonus.
| >
| Strangely around the same time I switched to RedHat because I got tired of
| having to apply my own security patches to the kernel and applications because
| the distribution was shipping with largely unmodified sources.
| Like many things in life "your mileage may vary".

The RHEL / CentOS / clones drama is certainly unsettling.

I don't think that Fedora is directly affected but it is hard to judge
whether there will be secondary effects.

One upcoming GTALUG talk will be from a Rocky Linux guy.  That should
be interesting.

I've already been struggling with where I want to go for a stable
system.  Besides the drama, I just don't think that RHEL's pinning
versions for 5 years is a good strategy.  Backporting for that long
feels like a wasted effort, prone to errors.

Why do I care about the effort RH puts into backporting?

- it creates a RH "moat": it prevents others from competing with them.
  Rocky, Alma, Oracle feel like clones, not creative competitors.
  That may be unfair to Oracle but that's not a company that I want
  a relationship with.

- it is labour that feels wasted.  Perhaps that labour could be used
  for more constructive purposes.

On the other hand RH has added a lot to the community and does do a
good job of beating back bugs.

I do think that I need a pair of distros: one that is up to date, and
one that is low-drama.  If they are in the same family, that cuts down
on redundant learning on my part.

- CentOS + Fedora has been a good pair for me.  TBH, CentOS has left
  me with technical debt: I get stuck on obsolete versions because the
  upgrade paths have been disrupted (twice!).  Fedora release updates
  have been good for some years.

  For my workstation (desktop and laptop) use, I've been very happy
  with Fedora, but it sure has a firehose of updates.  I don't think
  that it is affected directly by any of this.  But if a lot of people
  migrate away from Red Hat stuff, it won't likely be good for Fedora.

  It feels as if RH steers the future of Linux by making so many
  contributions.

- Ubuntu LTS + fresh Ubuntu has been pretty good.  I've had more
  problems with package updates on Ubuntu than of Fedora, but it has
  been pretty good.  Distro version upgrades have been good but not
  perfect in my modest experience.

  Canonical has repeatedly acted in ways that offend or scare me.  So
  Ubuntu, although easy, feels like a potential trap.

- debian Stable + Testing + Unstable.  I don't have much experience
  with debian.  I fear that the lack of full-time paid engineers might
  reduce the safety relative to RH (that could easily just be FUD).
  debian's goals are good by me.

So: I'm thinking of switching to debian.

I'd like to learn from others.  How do you choose to solve these
problems?  Maybe some of them are non-problems.



Giles has a problem with needing a stable distro with a more recent
FireFox.  I suggested, against my preferences, that this might be a
perfect use for Snaps/Flatpacks.

I wonder if I should be using a stable distro everywhere but with
containerized upgraded packages where they matter.  I yet don't think
so.

The rest of my family uses Fedora on their workstations.  But they
hate applying updates (even when I do the work).  They are way behind
most of the time.  Maybe a stable distro + a fresh FireFox would be
best for them too.

How many other packages would I need to have fresher-than-stable?

- support for newer hardware

- compilers etc.

- more pain-points would be discovered.



A fundamental problem is that feature changes and bug fixes are
usually mingled in upstream.  In some cases, it is a false
distinction.  Few developers want to maintain a bunch of old releases.
It is very hard for a distro to correctly separate these two, and yet
that is required to maintain a stable distro.
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-06-27 Thread Alvin Starr via talk

On 2023-06-27 08:19, Lennart Sorensen via talk wrote:

On Mon, Jun 26, 2023 at 07:29:23PM -0400, Colin McGregor via talk wrote:

Let's see if I understand this correctly, Red Hat has now put a whole
lot of open source / GPL software behind a paywall, where you have to
pay $$ for a subscription in order to access the source code. Then if
anyone uses that subscription to produce an Red Hat style distribution
(ie: Red Hat minus any support services) their subscription will be
cancelled... Is the above what I am reading here? :

https://www.jeffgeerling.com/blog/2023/im-done-red-hat-enterprise-linux

If this is true then it confirms my decision to use Debian GNU/Linux
as for all its' faults (and it does have several) Debian GNU/Linux
just doesn't pull this sort of @#$% .

Red Hat charges for access to their software and support.  If you get
their software that way, they of course are required to let you have
the source code.

But then they do in fact tell you that if you share that source code
I have looked a bit at various statements but nowhere near enough to 
comment authoritatively.
Where do they actually lay out the restrictions on the use of the source 
code for their subscription?


Also curious. What is Canonical's(Ubuntu) take on cloning Ubuntu and 
selling it as a competing distribution?

(which you are of course permitted to do by the license), they are allowed
to cancel your contract and decide to never do business with you again.
Well I think they say they can cancel your contract for any reason they
want, but that sharing the source code would very likely make them do so.

So similar to what the grsecurity people were trying to do, but at IBM
scale instead.  Red Hat is far from the company it was 30 years ago.
Yeah I am happy I switched to debian 25 years ago because Red Hat's
quality was so poor at the time.  Debian having a better designed
packaging system was a bonus.

Strangely around the same time I switched to RedHat because I got tired 
of having to apply my own security patches to the kernel and 
applications because the distribution was shipping with largely 
unmodified sources.

Like many things in life "your mileage may vary".

--
Alvin Starr   ||   land:  (647)478-6285
Netvel Inc.   ||   Cell:  (416)806-0133
al...@netvel.net  ||

---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


Re: [GTALUG] Red Hat Paywall...

2023-06-27 Thread Lennart Sorensen via talk
On Mon, Jun 26, 2023 at 07:29:23PM -0400, Colin McGregor via talk wrote:
> Let's see if I understand this correctly, Red Hat has now put a whole
> lot of open source / GPL software behind a paywall, where you have to
> pay $$ for a subscription in order to access the source code. Then if
> anyone uses that subscription to produce an Red Hat style distribution
> (ie: Red Hat minus any support services) their subscription will be
> cancelled... Is the above what I am reading here? :
> 
> https://www.jeffgeerling.com/blog/2023/im-done-red-hat-enterprise-linux
> 
> If this is true then it confirms my decision to use Debian GNU/Linux
> as for all its' faults (and it does have several) Debian GNU/Linux
> just doesn't pull this sort of @#$% .

Red Hat charges for access to their software and support.  If you get
their software that way, they of course are required to let you have
the source code.

But then they do in fact tell you that if you share that source code
(which you are of course permitted to do by the license), they are allowed
to cancel your contract and decide to never do business with you again.
Well I think they say they can cancel your contract for any reason they
want, but that sharing the source code would very likely make them do so.

So similar to what the grsecurity people were trying to do, but at IBM
scale instead.  Red Hat is far from the company it was 30 years ago.

Yeah I am happy I switched to debian 25 years ago because Red Hat's
quality was so poor at the time.  Debian having a better designed
packaging system was a bonus.

-- 
Len Sorensen
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk


[GTALUG] Red Hat Paywall...

2023-06-26 Thread Colin McGregor via talk
Let's see if I understand this correctly, Red Hat has now put a whole
lot of open source / GPL software behind a paywall, where you have to
pay $$ for a subscription in order to access the source code. Then if
anyone uses that subscription to produce an Red Hat style distribution
(ie: Red Hat minus any support services) their subscription will be
cancelled... Is the above what I am reading here? :

https://www.jeffgeerling.com/blog/2023/im-done-red-hat-enterprise-linux

If this is true then it confirms my decision to use Debian GNU/Linux
as for all its' faults (and it does have several) Debian GNU/Linux
just doesn't pull this sort of @#$% .
---
Post to this mailing list talk@gtalug.org
Unsubscribe from this mailing list https://gtalug.org/mailman/listinfo/talk