I use lubuntu (32bits) in virtualbox for many security and testing purpose and also on netbooks (Acer Aspire One - three different models), which LXDE is a must!
Trusty was fantastic!, Xenial acceptable and Bionic a disaster!!
18.04 (32bits) has many problems, boot time, video, performance, and the list goes for ever.
I am still using xenial and everyone I know that has a netbook uses it because of the LXDE.
Dropping development for 32bits and LXDE is a bad idea, believe me, specially for Lubuntu users (ie. many users all over the world).
Remenber the “L” stands for LIGHT weight, buy what I have learned, and the word Ubuntu means “humanity towards others” in the Zulu and Xhosa language, so the “…needs of the many outweigh the needs of the few or the one…”
I updated trusty to xenial with some problems, but at the end I got a stable system, after many hours of tuning (ie. fixing and reinstalling packages).
I tried to upgrade xenial to bionic and the final boot time and strange errors I get after many hours of upgrade or a fresh install (also with many hours for tuning) is a disaster!
This type of forced upgrades is what makes newbies avoid Linux.
The developers should consider this when stopping to support a release and “pushing” a new version out (in a “take it or leave it” way).
Then perhaps you should get the LXDE developers to, well, develop LXDE. It’s been stale for years.
LXDE didn’t change much during this time, so the fundamental concerns you’re describing likely have more to do with Ubuntu-wide core system pieces which ultimately Lubuntu has little control over. Have you tried using the latest kernels? Have you run systemd-analyze? Have you looked at which kernel modules you’re using, what alternatives there are, and what configuration options they may have? There are likely things you could do to improve all of that for your specific system.
I’ve already explained why we dropped LXDE (it’s dead) but 32 bit got dropped because no one in Ubuntu was supporting it and despite multiple calls for help, no one stepped up to the plate in Lubuntu to help with testing it. We were the last flavor to support 32 bit.
And, fast forward a bit: now we have most 32 bit packages removed from the Ubuntu archive, so it wouldn’t have mattered anyways. And there again, just as we reached out to the community about dropping 32 bit images, Ubuntu reached out to the community about dropping 32 bit packages. They received a lot of flack, but really only from the gaming community. No one really complained because they had 32 bit machines.
I’m sure if you elaborated further on this someone might be able to help you.
As opposed to, what? Windows? OS X? Android? iOS? I mean, what operating system out there does not have upgrades? What operating system guarantees that it will support all hardware for all time regardless of the version of the software? Even something like NetBSD which nearly runs on everything requires you to update.
I’m sorry to say that there is only one thing certain in life: change.
At least with Linux you have something you can’t get elsewhere: choice. There’s a whole plethora of distros still supporting 32 bit, all of which certainly have LXDE available. To be fair, you can install LXDE in newer versions of *buntu, too. Most of us have suggested Debian to folks.
I agree with you about LXDE. I liked it better. But, as they say: “easy come, easy go.” I didn’t create it. I enjoyed it while it lasted. That’s better than never having it at all.
Regarding lightweight, that seems to be an evolving definition for all the distros. Xfce is larger than a year ago. I assume it’s due to recent gtk3 support. That makes LXQt’s expansion (compared to LXDE) look less significant. (KDE used to be synonymous with gigantic. Now it’s practically as small as Lubuntu.). There are some things about Lubuntu which seem inconsistent being lightweight (screensavers as a default that make my fan run. Four pre-defined virtual desktops when even heavy distros default to 2.). But, those aren’t big problems. Just a different way of doing things.
Regarding your invocation of Spock’s “needs of the many:” if there were just one 32-bit computer in the world, would you expect a team of volunteers to beat themselves up maintaining and testing all the libraries, etc?
It seems like you’re implying moral failure, leaving people behind, not caring. That seems unfair when we’re talking about people who volunteer an enormous amount of time to help others, provide something free. If you’d maintain the infrastructure for 32-bit even when there’s just one last person with an operable machine, I’d have to ask why you’re not doing it when there are more than one? (If you wouldn’t maintain that infrastructure with just 1, then I’d ask where your threshold is. 10? 100? It’s not really an absolute principle, just a different opinion about pragmatics, right?).
I hope nothing I said will sound unwelcoming to the forum. I haven’t used Lubuntu as my desktop for almost a year. I’m not a fanboy. It just seems like some of that commentary could be invalidating to people who do far more good than harm.
Why do you thing having 4 different virtual desktops isn’t consistent with leightweight?
For example, for people that use netbooks, very limited resources with small screens, having multiple virtual environments is great.
Do you think there is a lot of overhead in using multiple virtual environments? Have you measured it? Openbox manage them and openbox as a whole only uses 7Mb of RAM.
Yes, if I had a company with a budget for R&D I would gladly participate with the financial needs to pay developers. And be ascertained, although not important, that I have donated for many projects FOSS/GNU/others as much I yearly can (by the way, that has been more than what I spent on my son’s desktop gaming system with M$W, updates and hardware upgrades), but I have family budget limit to keep in mind, and also, at the end, noone can actually choose/know where donnation goes (what is the maintainers goals).
Please don’t get me wrong but I have postponed to comment on the Lubuntu forum for some time, because “carbon units” don’t like negative feedback’s, as per the responses I see on the forums.
But this did not stop Lubuntu from using it for years. So way not continue ?
As the saying goes, “never change a winning team” but is that really the best possible advice? I beg the difference and argue; always challenge a winning team if you have a proven better alternative . As with most things, it all starts with your strategy and hard work testing all possibilities to find faults (from high to very low end systems; from new laptops and old netbooks), this is what the Linux community is all about - there is always someone that has that particular system, and will help to test the code. One alone does not have it all, but many will have about everything we need.
Yes, that’s what “tuning” also means in my first post.
Taking time to read blogs, forums and share feedback is one of them. And by the way, thank you for your quick replies. I really do appreciate it.
Please also note that I do read many misleading answers (on many forums, blogs and similar boards), and some can do much harm to a system. And I do help many local friends recover “dead” systems after these wrong doings.
For example: when you try to install lubuntu 18.04 on virtualbox or a netbook (or try to upgrade from xenial) you will get the bad screen after the boot. Many answers say it’s hardware issue. Hold your horses! Not so fast! It is a software bug! xrandr chooses the wrong device for the screen (for exemple: VGA-0 instead or VGA-1), in all the tests I did here. It could have many coding solutions, you will need to debug source code to find the correct patch.
Changes must for for better, not for the worst (usually saying).
Check this:
The past week, Canonical announced the latest version of Ubuntu, which caused quite some buzz (unfortunately, in the wrong way). According to the statement, the company had plans to discontinue 32-bit support starting from Ubuntu 19.10. This news wasn’t received quite well by Ubuntu-enthusiasts as they showed their disapproval of this decision on various online forums so much that even Canonical couldn’t ignore it.
However, the company demonstrated its genius and made the right decision by listening to the positive criticism of their community (including gamers, Ubuntu Studio, and WINE users) and announcing that this significant change can wait if the users aren’t fully prepared for it. Accordingly, Ubuntu users will get selected 32-bit i386 packages when they update to Ubuntu 19.10 and 20.04 LTS.
I haven’t measured it. To me, it suggest having a lot of apps running, windows open. When I think of resource-constrained laptops, I think of doing less multitasking. Obviously, if someone needed 4 (or 8) they could create that. Larger distros default to 2. You’d think there’d be even more need for more desktops in that case.
It’s not important. Perceptually it stands out to me.
I’m not trying to be antagonizing. But, “selected” implies someone’s not going to be happy with what’s not selected. I don’t believe this topic is as absolute as it’s presented.
Please don’t get me wrong, but where I live, and in most countries in the world, the rule is to have a old desktop/laptop/netboot for many years (very expensive for most to upgrade).
I my self use a i7 laptop but, most schools, friends and communities where I am volunteer to help, uses, at most, a core2duo 1.8Ghz and many types of 32bit Intel and AMD processors (netbooks). Much more than you can imagine.
I definitely respect trying to get the most use out of still-functional hardware. I’m almost OCD about that myself. But, shouldn’t you be rallying all the people in your area to step up and volunteer to keep the 32-bit infrastructure alive? It sounds like there is an abundance of people with a vested interest in keeping that going. But, you’re implicitly criticizing others who already volunteer countless hours, but who draw their line (of capacity for caring) differently.
I fully support your goal. But, if it’s so important… where is everyone? People have a finite amount of time, passion, etc. They have to make choices about how to spend that. From what you say, there should be a massive amount of people eager to help keep 32-bit alive. Why aren’t you targeting them?
I hope nothing I’ve said will discourage you from participating here. I think it’s been a good discussion, not too heated. It’s good to talk about things like this. My only concern is that it could be demotivating to people who do give (and give a lot), while not holding those with more “skin in the game” to the same standard.
As a user of x86 (32bit) on actual hardware (x86 only laptops), I don’t have any issues with Lubuntu 18.04 LTS i386.
I don’t stopwatch my boot times, I turn the laptop on & usually do other things as it boots, so I cannot compare the old boot time of Lubuntu 16.04 LTS to what it has been since with Lubuntu 18.04 LTS as I made that upgrade long ago (16.04 reached EOL 2019-April for all flavors except Kylin)
Whilst very subjective, on actual hardware performance is about equal to 16.04 or at least such that I didn’t notice any changes.
I tried Lubuntu 18.10 & 19.04 (with LXQt) on actual x86 hardware and actually I was most impressed. It performed much better in my opinion that MATE or XFCE did since the movement to using GTK3 which has a performance hit.
GTK2 is long dead, so LXDE needed to move itself to GTK3 and suffer that performance hit too, or do what it did by switching to Qt5 instead actually had less of a performance hit in my very subjective opinion (on x86 only hardware).
The porting of LXDE to GTK3 is still progressing, and I’ve not experimented with it, but have little desire to given what I saw ages back when MATE ported itself to GTK3 (used on a pentium M laptops the hit is sizeable), then far more recently XFCE did (again pentium M laptops). GTK3 seems to assume more grunt available in later hardware.
Lubuntu LXQt is I consider light weight.
I’ve also used Lubuntu 19.10 on x86 having forced a 19.04 box to release-upgrade so it worked there too, however with Ubuntu stopping building many x86 packages so it was only an exercise and shouldn’t be trusted by anyone. I may have accidentally wiped that partition in 18.04.4 x86/i386 testing as I’ve not seen it since then
i386/x86 is still supported with Lubuntu 18.04 LTS, and will be until 2021-April. I’m a happy user of it !
I was thinking about this more. You’ve probably seen the distro AntiX? To me, that seems like it would be perfect fit. It’s Debian-based, still has 32-bit support. It strives to be resource-conscious. It’s a little rough (like Puppy). But, it shares your passion for accommodation, leaving nobody behind, justice for all. As a result, it’s very basic in some ways. But, very powerful and customizable. You can respin it to something when you get it just the way you want it, and then distribute that to your colleagues.
Antix is closely aligned with MX, which also continues to provide 32-bit versions of itself. It’s larger than Lubuntu. But, it’s not as large as the ultimate distros (like Ubuntu-gnome, Mint Cinnamon, Zorin Core). It’s mid-range. But, you can use other window managers and dial it down to be smaller. It shares a lot of DNA with Antix; both let you respin your customized install. (You can dress up an Antix install to be more like MX. Or, dress down an MX install to be smaller like Antix.).
Sparky Linux is 32-bit too (directly based on Debian too). It reminds me more of Linux Lite, which is resource-conscious too. But, based upon Ubuntu (and has to follow Ubuntu’s 32-bit direction).
I think the first two would be ideal for you. You could dial in your own distro. Customized to what you think would be ideal for the people you know. Snapshot it, and make that into an installable .iso. Those distros come with those tools.
It seems to me like, even if Ubuntu continues “selected” 32-bit support, everything is getting larger, more resource dependent. Every release. I don’t remember the last 32-bit machine I had. That must have been when 2 or 4gig was the max laptop memory? 32-bit support seems pointless when the desktops built on top of it are getting large enough that they wouldn’t be suitable for those machines. I think the Antix/MX realm would be better. (Especially a better rallying point for continued Debian support for 32-bit.).
I’m not trying to steer you away from Lubuntu. The “conscious uncoupling” has been known for a year or more. Any continued “selected” support is only going to be “supervised visitation.” Why try to make that work when there’s a more viable distro for your demographic?
