https://www.omgubuntu.co.uk/2026/02/linux-mint-plans-longer-development- cycle
Sounds like a plan to me. Too many distros release on schedule whether the upgrade is ready or not. The only time I had trouble with Fedora KDE,
Plasma, and Qt weren't quite there.
To a degree, Linux has been taken in by the M$ "update fever"
bullshit.
A push for too-frequent updates can mean that not enough time to
find/squash BUGS will be there.
Take it a bit slower, get it RIGHT.
On 2/12/26 00:45, rbowman wrote:
https://www.omgubuntu.co.uk/2026/02/linux-mint-plans-longer-development-
cycle
Sounds like a plan to me. Too many distros release on schedule whether the >> upgrade is ready or not. The only time I had trouble with Fedora KDE,
Plasma, and Qt weren't quite there.
Good plan.
To a degree, Linux has been taken in by
the M$ "update fever" bullshit.
A push for too-frequent updates can mean
that not enough time to find/squash BUGS
will be there.
Take it a bit slower, get it RIGHT.
Obviously, there's cases and reasons to hustle critical fixes out
into the field as quick as you reasonably can - but the endless
churn of new releases under the Cult of the Constant Update is
deeply annoying.
On Thu, 12 Feb 2026 02:05:42 -0500
c186282 <c186282@nnada.net> wrote:
To a degree, Linux has been taken in by the M$ "update fever"
bullshit.
A push for too-frequent updates can mean that not enough time to
find/squash BUGS will be there.
Take it a bit slower, get it RIGHT.
"Release early, release often" is an ESR-ism, initially, and has deeper
roots in the FOSS world than MS-land (Redmond only adopted it in the
Win7 era) - but I agree, it's taken on the quality of a monomania in
the last decade-plus.
(This seems to be the way of things, with anything in the realm of soft-
ware development methodology; first it's a truism, then it's received
wisdom, then it's a religion, then it's an Industry, then something
else comes along and supplants it. Just look at what "agile" mutated
into, compared to the original manifesto.)
Obviously, there's cases and reasons to hustle critical fixes out into
the field as quick as you reasonably can - but the endless churn of new releases under the Cult of the Constant Update is deeply annoying. At
least you can *generally* roll your eyes and disable notifications -
unless you're stuck with the Filezilla fascists.
Hmmm ... my remaining Manjaro box WON'T update any more.
There's an Intel Firmware bit that won't load and that damns updates
for EVERYTHING. There may be a PacMan switch ... I'll have to look.
The Manjaro isn't THAT damned old and I have it set up nicely so I
don't wanna flush everything. Last update was about three months ago,
so I should be good for awhile.
On Thu, 12 Feb 2026 21:09:48 -0500, c186282 wrote:
Hmmm ... my remaining Manjaro box WON'T update any more.
There's an Intel Firmware bit that won't load and that damns updates
for EVERYTHING. There may be a PacMan switch ... I'll have to look.
The Manjaro isn't THAT damned old and I have it set up nicely so I
don't wanna flush everything. Last update was about three months ago,
so I should be good for awhile.
/etc/pacman.conf IgnorePkg
Best tact ... wait a few weeks and see if the Arch repos straighten
themselves out.
On Fri, 13 Feb 2026 02:19:47 -0500, c186282 wrote:
Best tact ... wait a few weeks and see if the Arch repos straighten
themselves out.
Is it Arch or Manjaro's secret sauce? I'm not having a problem with EndeavourOS.
Sorry, but I'm a Deb-o-Phile .....
Anyway, my first distro was the old RedHat 6. Then I bought a
no-name laptop with Windows NT on it, and used that for awhile.
Then I tried to install RedHat. Partway through the screen would
blank and a weird glow would appear. (I later learned it was some
issue with the AMD K6 CPU.)
So I download Debian and burned an install CD. We were on a road
trip, so I spent a lot of time in the passenger seat getting it
installed, getting familiar with dselect, and getting the GUI
running. It was very absorbing.
I didn't try any other distro until Gentoo years later.
Anyway, my first distro was the old RedHat 6. Then I bought a no-name
laptop with Windows NT on it, and used that for awhile. Then I tried to install RedHat. Partway through the screen would blank and a weird glow
would appear. (I later learned it was some issue with the AMD K6 CPU.)
When I first decided to set up a Linux machine, I went to the local
bookstore and perused the various Linux books which had an installation
CD included. The book I liked best happened to be by Patrick
Volkerding, so my first distro was Slackware 3.5, which ran happily on a laptop with 48MB of memory and a 1.3G hard drive.
On 2026-02-15, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:I started with RedHat but Richard K had always raved about Debian so I started with that, but got fed up with the fact that Stable was years
Anyway, my first distro was the old RedHat 6. Then I bought a
no-name laptop with Windows NT on it, and used that for awhile.
Then I tried to install RedHat. Partway through the screen would
blank and a weird glow would appear. (I later learned it was some
issue with the AMD K6 CPU.)
So I download Debian and burned an install CD. We were on a road
trip, so I spent a lot of time in the passenger seat getting it
installed, getting familiar with dselect, and getting the GUI
running. It was very absorbing.
I didn't try any other distro until Gentoo years later.
When I first decided to set up a Linux machine, I went to the
local bookstore and perused the various Linux books which had
an installation CD included. The book I liked best happened
to be by Patrick Volkerding, so my first distro was Slackware
3.5, which ran happily on a laptop with 48MB of memory and a
1.3G hard drive.
KDE looked lovely, but was far too heavyweight ...
On 2/15/26 14:30, rbowman wrote:
On Sun, 15 Feb 2026 11:59:16 -0800, Bobbie Sellers wrote:
-a-a-a-aI personally feel that Mandriva was way ahead of any Ubuntu I have >>> -a-a-a-aseen.bankrupt.
-a-a-a-aI only used Mandriva for a few years before the company went
I don't remember the exact timeline but I think Ubuntu was released about
the time Mandrake responded to the trademark lawsuit by changing the
name.
Ubuntu gained popularity by mailing out free CDs. At the time I couldn't
figure out Canonical's game plan.
-a-a-a-aMaybe they just want to be the Windows-< of Linux.
-a-a-a-aI encountered Ubuntu trying to help people who had screwed up the
-ainstall of of Ubuntu that someone from the LUG had done for them.-a Not a pleasant time for me though I was in better shape then than more recently. But consider that Ubuntu installs in a screwed up way to my way of
thinking.
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt framework, which is optimized to work efficiently on a wide range of hardware configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects quite comfortably on a single-processor 900MHz Celeron machine -- my Asus
Eee 701 netbook.
I don't knock the gurus ... we need them and appreciate
-a their deep skills ... but Linux shouldn't be, doesn't
-a have to be, THAT hard most of the time.
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt framework, which is optimized to work efficiently on a wide range of hardware configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects quite comfortably on a single-processor 900MHz Celeron machine -- my Asus
Eee 701 netbook.
On 2026-02-15, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Anyway, my first distro was the old RedHat 6. Then I bought a
no-name laptop with Windows NT on it, and used that for awhile.
Then I tried to install RedHat. Partway through the screen would
blank and a weird glow would appear. (I later learned it was some
issue with the AMD K6 CPU.)
So I download Debian and burned an install CD. We were on a road
trip, so I spent a lot of time in the passenger seat getting it
installed, getting familiar with dselect, and getting the GUI
running. It was very absorbing.
I didn't try any other distro until Gentoo years later.
When I first decided to set up a Linux machine, I went to the
local bookstore and perused the various Linux books which had
an installation CD included. The book I liked best happened
to be by Patrick Volkerding, so my first distro was Slackware
3.5, which ran happily on a laptop with 48MB of memory and a
1.3G hard drive.
I stayed with Slack for some time, but the lack of package
management tools finally got to be too much. Ubuntu 10 was
much easier to set up and maintain, but then they switched
to the Unity desktop so I bid it farewell. I tried a few
other distros (e.g. Mint, CrunchBang) and desktops.
Blackbox was nicely lean and mean - perhaps a bit too much so.
KDE--
looked lovely, but was far too heavyweight; even worse, it
was constantly spitting messages out in a console window,
indicating it was doing too much behind my back for comfort.
Eventually I settled on Debian and Xfce.
On 15/02/2026 17:27, Charlie Gibbs wrote:
On 2026-02-15, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:I started with RedHat but Richard K had always raved about Debian so I started with that, but got fed up with the fact that Stable was years
Anyway, my first distro was the old RedHat 6. Then I bought a
no-name laptop with Windows NT on it, and used that for awhile.
Then I tried to install RedHat. Partway through the screen would
blank and a weird glow would appear. (I later learned it was some
issue with the AMD K6 CPU.)
So I download Debian and burned an install CD. We were on a road
trip, so I spent a lot of time in the passenger seat getting it
installed, getting familiar with dselect, and getting the GUI
running. It was very absorbing.
I didn't try any other distro until Gentoo years later.
When I first decided to set up a Linux machine, I went to the
local bookstore and perused the various Linux books which had
an installation CD included. The book I liked best happened
to be by Patrick Volkerding, so my first distro was Slackware
3.5, which ran happily on a laptop with 48MB of memory and a
1.3G hard drive.
behind the curve...so tried Ubuntu and didn't really like it and every
was raving about MINT so I installed it and mostly it Just Worked.
That was about 13 years ago now
On 2026-02-15 22:09, Lawrence DrCOOliveiro wrote:
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt framework,
which is optimized to work efficiently on a wide range of hardware
configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects quite
comfortably on a single-processor 900MHz Celeron machine -- my Asus
Eee 701 netbook.
That was relatively recent. Further past, KDE was heavy weight. At some point they dedicated a collective effort to trim it, and suddenly it
became lean.
ChatGPT says it happened with the transition from KDE 4.x to KDE Plasma
5, starting in mid-2014.
On 2026-02-15, Lawrence DrCOOliveiro wrote:
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt framework,
which is optimized to work efficiently on a wide range of hardware
configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects quite
comfortably on a single-processor 900MHz Celeron machine -- my Asus Eee
701 netbook.
How much memory was used, and did it have a, let's say,
not-bleeding-edge and not-fully-accelerated GPU?
On 2026-02-15 22:09, Lawrence DrCOOliveiro wrote:
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt
framework, which is optimized to work efficiently on a wide range
of hardware configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects
quite comfortably on a single-processor 900MHz Celeron machine --
my Asus Eee 701 netbook.
That was relatively recent.
On 2026-02-15 22:09, Lawrence DrCOOliveiro wrote:
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt framework,
which is optimized to work efficiently on a wide range of hardware
configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects quite
comfortably on a single-processor 900MHz Celeron machine -- my Asus
Eee 701 netbook.
That was relatively recent. Further past, KDE was heavy weight. At some point they dedicated a collective effort to trim it, and suddenly it
became lean.
ChatGPT says it happened with the transition from KDE 4.x to KDE Plasma
5, starting in mid-2014.
On Mon, 16 Feb 2026 12:14:18 +0100, Carlos E. R. wrote:
On 2026-02-15 22:09, Lawrence DrCOOliveiro wrote:
On Sun, 15 Feb 2026 17:27:46 GMT, Charlie Gibbs wrote:
KDE looked lovely, but was far too heavyweight ...
ItrCOs not, actually; remember, itrCOs built on top of the Qt
framework, which is optimized to work efficiently on a wide range
of hardware configurations, including embedded ones.
When KDE 4 came out, I was able to run it with full 3D effects
quite comfortably on a single-processor 900MHz Celeron machine --
my Asus Eee 701 netbook.
That was relatively recent.
2008??
I don't always use Debian, but when I do it's Debian Sid
(unstable).
Actually Plasma 5 was very light but Plasma 6 added back a lot of volume
in disk and memory space. Despite that it contains useful improvements especially to the clipboard. Sadly KDE has decided to beome dependent
on systemd so that it will become much heavier and I will have to give it
up as PCLinuxOS is specifically anti-systemd or Poettering's folly as we
call it. I started using KDE about 20 years ago on version 3.57 for
Mandriva because it gave me a Desktop Environment on which I could
maintain the workflow I had become accustomed to on AmigaOS.
I have run it on lots of laptops including a thrift shop Inspiron 4000
with a 700 MHz coppermine with <1 GB of memory. It came with
Windows XP and KDE was as fast as XP once I cut back Virtual
Desktops to only 1.
KAOS is also trying to kick the Plasma habit as well because they
are unhappy with the increased dependency on systemd. They have used
systemd up to now and so must be waking up to the idea that it is a anti-GNU/Linux kludge.
Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
I don't always use Debian, but when I do it's Debian Sid
(unstable).
There is nothing wrong with Debian sid. Someone needs to test it so
that we can put out a stable release with less bugs.
But if it breaks, you need to be able to fix it. And you shouldn't
complain if it breaks, it is supposed to. And it seldomly does.
On 2026-02-16, Bobbie Sellers wrote:
Actually Plasma 5 was very light but Plasma 6 added back a lot of volume
in disk and memory space. Despite that it contains useful improvements
especially to the clipboard. Sadly KDE has decided to beome dependent
on systemd so that it will become much heavier and I will have to give it
up as PCLinuxOS is specifically anti-systemd or Poettering's folly as we
call it. I started using KDE about 20 years ago on version 3.57 for
Mandriva because it gave me a Desktop Environment on which I could
maintain the workflow I had become accustomed to on AmigaOS.
Not that this says anything about whether it will depend more on systemd
or not, but some posts I saw on the fediverse suggest that currently
only their graphical login manager depends/will depend on systemd.
If the KDE sources are not misleading and really mean it by saying it's
just the login manager, then the desktop itself ought to be still usable without systemd.
On 2026-02-17, Marc Haber wrote:
Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
I don't always use Debian, but when I do it's Debian Sid
(unstable).
There is nothing wrong with Debian sid. Someone needs to test it so
that we can put out a stable release with less bugs.
But if it breaks, you need to be able to fix it. And you shouldn't
complain if it breaks, it is supposed to. And it seldomly does.
I can't help but think there's a significant problem with your
affirmation - how can you want it to be used for testing in order to fix >issues before the software hits stable if you then tell its users not to >complain!?
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave
much to run other not-so-light applications like OOo.)
Possibly it uses systemd login features to know what user has access to >resources such as the USB, audio, DVD.... Who has the seat.
"Carlos E. R." <robin_listas@es.invalid> wrote:
Possibly it uses systemd login features to know what user has access to
resources such as the USB, audio, DVD.... Who has the seat.
Isnt that policykit? At least it is not necessary any more to put
shell users in that long list of groups.
Nuno Silva <nunojsilva@invalid.invalid> wrote:
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave
much to run other not-so-light applications like OOo.)
In these days the memory footprint of the DE stops counting when the
browser is started.
On 2026-02-17 14:17, Marc Haber wrote:
Nuno Silva <nunojsilva@invalid.invalid> wrote:
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave >>> much to run other not-so-light applications like OOo.)
In these days the memory footprint of the DE stops counting when the
browser is started.
Right.
I have seen Thunderbird at 15 GB resident size in my desktop machine. In this laptop, it is no 1.7 GiB. FFx is more difficult to evaluate, it is dozens of processes.
On 2026-02-16, Bobbie Sellers wrote:
Actually Plasma 5 was very light but Plasma 6 added back a lot of volume
in disk and memory space. Despite that it contains useful improvements
especially to the clipboard. Sadly KDE has decided to beome dependent
on systemd so that it will become much heavier and I will have to give it
up as PCLinuxOS is specifically anti-systemd or Poettering's folly as we
call it. I started using KDE about 20 years ago on version 3.57 for
Mandriva because it gave me a Desktop Environment on which I could
maintain the workflow I had become accustomed to on AmigaOS.
Not that this says anything about whether it will depend more on systemd
or not, but some posts I saw on the fediverse suggest that currently
only their graphical login manager depends/will depend on systemd.
If the KDE sources are not misleading and really mean it by saying it's
just the login manager, then the desktop itself ought to be still usable without systemd.
I have run it on lots of laptops including a thrift shop Inspiron 4000
with a 700 MHz coppermine with <1 GB of memory. It came with
Windows XP and KDE was as fast as XP once I cut back Virtual
Desktops to only 1.
KAOS is also trying to kick the Plasma habit as well because they
are unhappy with the increased dependency on systemd. They have used
systemd up to now and so must be waking up to the idea that it is a
anti-GNU/Linux kludge.
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave
much to run other not-so-light applications like OOo.)
On 2026-02-17 11:15, Nuno Silva wrote:
On 2026-02-16, Bobbie Sellers wrote:
Actually Plasma 5 was very light but Plasma 6 added back a lot of volume >>> in disk and-a memory space. Despite that it contains useful improvements >>> especially to the clipboard.-a Sadly KDE has decided to beome dependent
on systemd so that it will become much heavier and I will have to
give it
up as PCLinuxOS is specifically anti-systemd or Poettering's folly as we >>> call it.-a I started using KDE about 20 years ago on version 3.57 for
Mandriva because it gave me a Desktop Environment on which I could
maintain the workflow I had become accustomed to on AmigaOS.
Not that this says anything about whether it will depend more on systemd
or not, but some posts I saw on the fediverse suggest that currently
only their graphical login manager depends/will depend on systemd.
If the KDE sources are not misleading and really mean it by saying it's
just the login manager, then the desktop itself ought to be still usable
without systemd.
Possibly it uses systemd login features to know what user has access to resources such as the USB, audio, DVD.... Who has the seat.
...
KDE Says Plasma Desktop Will Never Force Users to Use systemd
Only the Plasma Login Manager will be dependent on systemd, but future Plasma releases arenrCOt
dependent on Plasma Login Manager, nor systemd.
by Marcus Nestor February 16th, 2026 -7 Comments
policykit did that, but I think that systemd does it currently in my
distro, openSUSE. I'm not 100% certain.
I have seen Thunderbird at 15 GB resident size in my desktop
machine. In this laptop, it is no 1.7 GiB. FFx is more difficult to
evaluate, it is dozens of processes.
On Tue, 17 Feb 2026 15:00:34 +0100, Carlos E. R. wrote:
policykit did that, but I think that systemd does it currently in my
distro, openSUSE. I'm not 100% certain.
On my Debian Unstable system:
root@theon:~ # systemctl status '*pol*'
ruA polkit-agent-helper.socket - Authorization Manager Agent Helper
Loaded: loaded (/usr/lib/systemd/system/polkit-agent-helper.socket; enabled; preset: enabled)
Active: active (listening) since Tue 2026-01-27 14:27:17 NZDT; 3 weeks 0 days ago
Invocation: 7d33176310b9436f87a5f8b590fa1d2f
Docs: man:polkit(8)
Listen: /run/polkit/agent-helper.socket (Stream)
Accepted: 0; Connected: 0;
Jan 27 14:27:17 theon systemd[1]: Listening on polkit-agent-helper.socket - Authorization Manager Agent Helper.
ruA polkit.service - Authorization Manager
Loaded: loaded (/usr/lib/systemd/system/polkit.service; static)
Active: active (running) since Tue 2026-01-27 14:27:17 NZDT; 3 weeks 0 days ago
Invocation: a745987d0eed4984851d69b4b6f6cbc7
Docs: man:polkit(8)
Main PID: 1138 (polkitd)
Status: "Processing requests..."
Tasks: 4 (limit: 37349)
Memory: 4.3M (max: 32M, swap max: 32M, available: 27.6M, peak: 11.5M, swap: 3.7M, swap peak: 4.6M)
CPU: 26.253s
CGroup: /system.slice/polkit.service
rooroC1138 /usr/lib/polkit-1/polkitd --no-debug --log-level=notice
Jan 27 14:27:17 theon systemd[1]: Starting polkit.service - Authorization Manager...
Jan 27 14:27:17 theon polkitd[1138]: Started polkitd version 127
Jan 27 14:27:17 theon systemd[1]: Started polkit.service - Authorization Manager.
The two are not mutually exclusive ...
Did not occur to me to check this way.
On Wed, 18 Feb 2026 02:32:12 +0100, Carlos E. R. wrote:
Did not occur to me to check this way.
I first did
ps -ef | fgrep pol
(looking for rCLpolkitrCY or rCLpolicykitrCY) and found processes with rCLpolkitrCY in their name.
The next logical thing to check was for systemd services with
similar names.
This is all part of find-your-way-around-an-unfamiliar-Linux-system
101.
Nuno Silva <nunojsilva@invalid.invalid> wrote:
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave >>much to run other not-so-light applications like OOo.)
In these days the memory footprint of the DE stops counting when the
browser is started.
On 2026-02-17, Marc Haber wrote:
Nuno Silva <nunojsilva@invalid.invalid> wrote:
I haven't used desktop environments in a long time, I remember the days >>>KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave >>>much to run other not-so-light applications like OOo.)
In these days the memory footprint of the DE stops counting when the
browser is started.
And that's another thing that needs to be improved - how did it become
so acceptable for some web pages to require so many resources, let alone >browsers themselves.
I remember the days when the desktop, which had
just a bit more of that (384 MiB, perhaps?) could keep hundreds of tabs
open (might have been in Firefox back then,
In these days the memory footprint of the DE stops counting when theAnd that's another thing that needs to be improved - how did it become
browser is started.
so acceptable for some web pages to require so many resources, let alone browsers themselves. I remember the days when the desktop, which had
just a bit more of that (384 MiB, perhaps?) could keep hundreds of tabs
open (might have been in Firefox back then, in the time shortly
afterwards Mozilla tried to give up on the suite (which is what I use nowadays)).
I know some examples of websites that become so unusable because of how
many content they load or because of CSS animations and effects, and
that's not to mention the ones which don't work because some underlying framework requires a shiny new feature and is not backwards-compatible
when it's not present. Also, at least one site has required webgl in
order to work, which I find especially amusing...
And that's another thing that needs to be improved - how did it becomeIrrelevant. It's reality now, and noone is working on that
so acceptable for some web pages to require so many resources, let alone
browsers themselves.
"improvement". You're free to do so, but you're riding a head horse.
I remember the days when the desktop, which hadI am not sure whether Firefox had tabs back in the times when we had
just a bit more of that (384 MiB, perhaps?) could keep hundreds of tabs
open (might have been in Firefox back then,
less than 1 Gig of RAM.
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly >improve its performance.
The Natural Philosopher <tnp@invalid.invalid> wrote:Agree, sadly...
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly
improve its performance.
Maybe, but since noone is working on that, it's moot to think about
it. The opposite is reality: We had more browser engines in the market
five years ago than we had today.
Greetings
Marc
Nuno Silva <nunojsilva@invalid.invalid> wrote:[...]
I remember the days when the desktop, which had
just a bit more of that (384 MiB, perhaps?) could keep hundreds of tabs >>open (might have been in Firefox back then,
I am not sure whether Firefox had tabs back in the times when we had
less than 1 Gig of RAM.
On 24/02/2026 12:51, Marc Haber wrote:
The Natural Philosopher <tnp@invalid.invalid> wrote:
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly
improve its performance.
Maybe, but since noone is working on that, it's moot to think about
it. The opposite is reality: We had more browser engines in the market
five years ago than we had today.
Agree, sadly...
The Natural Philosopher <tnp@invalid.invalid> wrote:
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly
improve its performance.
Maybe, but since noone is working on that, it's moot to think about
it. The opposite is reality: We had more browser engines in the market
five years ago than we had today.
The Natural Philosopher <tnp@invalid.invalid> wrote:
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly >>improve its performance.
Maybe, but since noone is working on that, it's moot to think about
it. The opposite is reality: We had more browser engines in the market
five years ago than we had today.
Marc Haber <mh+usenetspam1118@zugschl.us> writes:
The Natural Philosopher <tnp@invalid.invalid> wrote:
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly
improve its performance.
The most performance-critical part (i.e. JavaScript) has already been reimplemented multiple times, as compiler technology improves.
Maybe, but since noone is working on that, it's moot to think about
it. The opposite is reality: We had more browser engines in the market
five years ago than we had today.
Ground-up rewrites are usually a mistake.
Ground-up rewrites are usually a mistake.
On 2026-02-25, Richard Kettlewell <invalid@invalid.invalid> wrote:
Ground-up rewrites are usually a mistake.
Either that or they're a sign that a bad mistake was made
back in the beginning.
"There's never time to do it right, but always time to do it over."
In these days the memory footprint of the DE stops counting when the
browser is started.
Abundance justifies waste. :-(
The Natural Philosopher <tnp@invalid.invalid> wrote:
Indeed. But I suspect a ground up rewrite of Firefox et al would vastly >>improve its performance.
Maybe, but since noone is working on that, it's moot to think about
it.
The opposite is reality: We had more browser engines in the market
five years ago than we had today.
Frankly *every* aspect of modern computing is in need of industrial liposuction ...
On 2026-02-25, The Natural Philosopher <tnp@invalid.invalid> wrote:[...]
On 25/02/2026 16:43, Charlie Gibbs wrote:
On 2026-02-25, Richard Kettlewell <invalid@invalid.invalid> wrote:
Ground-up rewrites are usually a mistake.
"There's never time to do it right, but always time to do it over."
I ported a lot of programs that had gotten into that state,
and enjoyed doing a lot of cleaning in the process, reducing
line count by 30% or more and greatly improving readability.
My favourite was when a feature the customer had tried to add
never worked, but suddenly started working after my clean-up.
(Yes, I know that some spoilsport is now going to give
a lecture on bug-compatibility. Cut me some slack...)
Nuno Silva <nunojsilva@invalid.invalid> wrote:
On 2026-02-17, Marc Haber wrote:
Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
I don't always use Debian, but when I do it's Debian Sid
(unstable).
There is nothing wrong with Debian sid. Someone needs to test it so
that we can put out a stable release with less bugs.
But if it breaks, you need to be able to fix it. And you shouldn't
complain if it breaks, it is supposed to. And it seldomly does.
I can't help but think there's a significant problem with your
affirmation - how can you want it to be used for testing in order to fix >>issues before the software hits stable if you then tell its users not to >>complain!?
Maybe a language issue. I don't consider a bug report as a complaint.
A complaint is like "YOU broke my system and it took me two days and
my backup to fix it". A bug report is like "/usr/bin/foo crashes if
invoked with --bamboozle and --frobnicate, core dump attached".
On Wed, 25 Feb 2026 20:00:17 GMT
Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
Abundance justifies waste. :-(
Might be truer to say that abundance provides a *pretext* for waste.
Seems like the bill always comes due, sooner or later...
On Wed, 25 Feb 2026 20:00:17 GMT
Charlie Gibbs <cgibbs@kltpzyxm.invalid> wrote:
Abundance justifies waste. :-(
Might be truer to say that abundance provides a *pretext* for waste.
Seems like the bill always comes due, sooner or later...
On 2026-02-17 14:17, Marc Haber wrote:
Nuno Silva <nunojsilva@invalid.invalid> wrote:
I haven't used desktop environments in a long time, I remember the days
KDE was heavy enough to be a problem with, what was it, 64 MiB RAM? :-)
(The problem might have been not the DE itself, but that it didn't leave >>> much to run other not-so-light applications like OOo.)
In these days the memory footprint of the DE stops counting when the
browser is started.
Right.
I have seen Thunderbird at 15 GB resident size in my desktop machine. In this laptop, it is no 1.7 GiB. FFx is more difficult to evaluate, it is dozens of processes.
While I was manually running a filter in alt.comp.os.windows-10, it went
to 6 gigns temporarily (and 100% cpu), for several minutes. Now I have
been doing email, not usenet, and it's gone down to 2.5 gigs.
On Sat, 28 Feb 2026 20:48:08 +0100, Carlos E.R. wrote:
While I was manually running a filter in alt.comp.os.windows-10, it went
to 6 gigns temporarily (and 100% cpu), for several minutes. Now I have
been doing email, not usenet, and it's gone down to 2.5 gigs.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
465515 rbowman 20 0 3588388 340332 161492 S 2.3 2.2 49:53.57 thunderbird-bin
A few things might be relevant. I don't use Thunderbird for news. Pan is using 9.6% or 1.4 G. This is the Ubuntu box and T-Bird is a snap. My
inbox only has 4 items. Most are sorted to local folders on receipt and I delete junk aggressively.
Slrn takes up a litte more than 30M :-)
On 2026-03-01, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Slrn takes up a litte more than 30M :-)
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
8519 rbowman 20 0 242964 8936 5640 S 0.0 0.1 0:00.05 slrn
rbowman wrote this screed in ALL-CAPS:
On 2026-03-01, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Slrn takes up a litte more than 30M :-)
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+
COMMAND
8519 rbowman 20 0 242964 8936 5640 S 0.0 0.1 0:00.05
slrn
2991674 shyster 20 0 30.8m 20.0m 6.1m S 0.0 0.1 0:01.68
slrn
On Mon, 2 Mar 2026 07:22:22 -0500, Chris Ahlstrom wrote:
rbowman wrote this screed in ALL-CAPS:
On 2026-03-01, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Slrn takes up a litte more than 30M :-)
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+
COMMAND
8519 rbowman 20 0 242964 8936 5640 S 0.0 0.1 0:00.05
slrn
2991674 shyster 20 0 30.8m 20.0m 6.1m S 0.0 0.1 0:01.68
slrn
Your user name is 'shyster'? Explains a lot. Are you related to Slippin' Jimmy?
rbowman wrote this screed in ALL-CAPS:
On Mon, 2 Mar 2026 07:22:22 -0500, Chris Ahlstrom wrote:
rbowman wrote this screed in ALL-CAPS:
On 2026-03-01, Chris Ahlstrom <OFeem1987@teleworm.us> wrote:
Slrn takes up a litte more than 30M :-)
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ >>>> COMMAND
8519 rbowman 20 0 242964 8936 5640 S 0.0 0.1 0:00.05 >>>> slrn
2991674 shyster 20 0 30.8m 20.0m 6.1m S 0.0 0.1 0:01.68 >>> slrn
Your user name is 'shyster'? Explains a lot. Are you related to Slippin'
Jimmy?
Nah, not shyster. The Devil made me do it.
| Sysop: | Amessyroom |
|---|---|
| Location: | Fayetteville, NC |
| Users: | 59 |
| Nodes: | 6 (0 / 6) |
| Uptime: | 03:10:40 |
| Calls: | 810 |
| Files: | 1,287 |
| D/L today: |
4 files (10,048K bytes) |
| Messages: | 200,610 |