What's the best lightweight Linux distro?

[bmarkus]

There is an interesting review of small distros in Tuxradar:

http://www.tuxradar.com/content/whats-best-lightweight-linux-distro

[jur]

Well they didn't test deeper and seem to have gotten the answer wrong.

I will supply the correct answer now:

Tiny Core Linux: 10/10

[bmarkus]

It is not a deep technical review but summarize experience facing thes distros from average customer point of view.

[bigpcman]

It is not a deep technical review but summarize experience facing thes distros from average customer point of view.

I wonder how  tc with all the applications included in Slitaz installed (or equivalents) compares to the stock Slitaz.

Simple things like boot time, general performance, resources required, and package management. Would there be much difference?

[tclfan]

I had been with SliTaz since beginning until I found TinyCore. SliTaz is an excellent distro but Tiny Core is better:
1. Architecture - more modular
2. More flexible. E.g. Any WM can be easily included.
3. Better implementation of Xorg, etc. SliTaz is still using HAL I think and does not always render proper resolution. TinyCore went a step further here and on the same hardware works properly.
4. Networking is better in TC according to my testing
5. Development cycle has been faster so far.

What TinyCore is missing to be rated better in such reviews is a showcase distro. Preserving the core and modular architecture a showcase distro should be composed such as proposed earlier or including more software so it is ready to go and reach vast population of users. Currently, easy as it is, still needs quite a bit of work for the user to configure a useful system and few spend the time to learn and do it.

[lucky13]

The details offered about the demise of DSL are inaccurate. It's ridiculous, too, that DSL(no longer under active development and which is going to be increasingly marginalized due to lack of modern hardware support) gets a higher score than TinyCore and that Puppy (a single-user -- root -- system) gets the same score as TinyCore. What? At least the review didn't blather on and on about aesthetics.

It's a shame the reviewer didn't add in a comparable mix of TinyCore extensions and use that as the baseline from which to judge -- more apples to apples. After all, that's part of what makes TinyCore so unique, you get to choose how and what rather than have a lot of stuff given to you that you may not want -- a problem hinted at with using some of the repositories of the others in the review. The "problem" of bloat when adding things from repositories encountered with Ubuntu-based light remixes (CrunchBang is currently moving to Debian, fwiw) isn't such a problem in TC. Speaking of that, why is CrunchBang's score higher than and Lubuntu's (an alpha version!) the same as TinyCore's if the two Ubuntu-based distros have such an issue with bloated repo packaging? Most users are going to switch some applications around. Once you start doing that and start to make your system more Ubuntu and less CrunchBang, you're losing every bleeping advantage of using the "lighter" sub-distro.

My verdict: The article is complete crap. 2/10.

[danielibarnes]

The article states:
"The important things that we'll look at here are the amount of space needed, how much processing power is required to get the distro running at an acceptable level, and the effort required to get it to work." It would be nice to know the weight given to each metric and how it contributes to the overall score. The absence of quantifiable scoring leads one to consider this article lacking.

SliTax, with its select bundle of applications, scored higher. Sure, you can load the same applications in Tiny Core with very little effort using the AppBrowser, but the author is evaluating the out-of-the-box experience. We place more importance on "the amount of space needed" than the author does.

So how do we address this? Remember, even if the author appears misguided, it means Tiny Core is perceived as requiring effort. The best step we as a community can take is to address the final words of the author: "it may have made things easier to aim for a slightly higher target to begin with."

The recent TinyCore for all - Remastering and Flavors thread is a step in the right direction. Identify a SliTaz-like bundle for Tiny Core, remaster it, and provide it as a download for new users. Then, there can be no complaint!

[bmarkus]

The article states:
"The important things that we'll look at here are the amount of space needed, how much processing power is required to get the distro running at an acceptable level, and the effort required to get it to work."

IMHO in this term article is correct. Remember, how much time it took for you to understand different installation modes, cloud, PPI, backup, etc.

[sandras]

True. It seems simple when you know it. but it's very difficult when you don't. I was aware of TC almost from the start of my Linux quest, but only after living some time with Ubuntu and actively reading about Linux did I understood more things about TC.

Now I don't want to be one of those "let's attract Windows users" guy, but Windows users, who try to shift to Linux often have a hard time understanding even the concept of mount which is a commodity in Linux.

[tclfan]

Now I don't want to be one of those "let's attract Windows users" guy, but Windows users, who try to shift to Linux often have a hard time understanding even the concept of mount which is a commodity in Linux.

It is already too late and too little has been done to attract Windows users to Linux. It was perhaps possible early on but then Linux went to the dark side of bloat, slow and inefficient, not speaking of user-friendly.

Some bad strategic decisions were made such as Gnome and KDE huge cumulative libraries instead of keeping linux modular, fast and efficient.

Even more importantly, fragmentation of Linux into 207 distros makes attracting Windows users not likely. Such potential convert would have to do quite a research to determine which distro to pick. This is unlikely to happen and user will prefer to stay with Windows.

If there was a unified Linux strategy and effort was not wasted on 207 distro but rather to focus on one - efficient and user-friendly, then an OS would emerge far superior to Windows and provide a clear choice to users.

This time and opportunity has been lost.  Now the only chance is a new revolution, such as system completely componentized, modular, user friendly such as new trend emerging in Linux world (TinyCore+, Igelle), but this needs to reject the past bloat legacy, such as bloated libraries and applications need to be self-contained modules...
I think TinyCore and Igelle are going in the right direction, each focusing on separate parts of this strategy...
This is unless it is too late for this too and such revolution will be Web OS, which will make underlying core system core irrelevant...

[curaga]

Now I don't want to be one of those "let's attract Windows users" guy, but Windows users, who try to shift to Linux often have a hard time understanding even the concept of mount which is a commodity in Linux.

Oh, the pirates are well aware of Daemon Tools What surprises them in linux is that loop mounts are considered a part of the core os, not some payware.

[lucky13]

So how do we address this? Remember, even if the author appears misguided, it means Tiny Core is perceived as requiring effort. The best step we as a community can take is to address the final words of the author: "it may have made things easier to aim for a slightly higher target to begin with."

Easier is debatable, and so is effort; I agree that TC's modularity is extreme and to some users (particularly those who don't care to learn enough about the system to use it) that's not desirable. It would've behooved the reviewer to first understand the underlying concepts and what's required to set up a system based on TinyCore and then set it up as desired.

The recent TinyCore for all - Remastering and Flavors thread is a step in the right direction. Identify a SliTaz-like bundle for Tiny Core, remaster it, and provide it as a download for new users. Then, there can be no complaint!

I don't know what the real difference would be between slitaz and tc-built-more-like-slitaz. All this talk about a "showcase" image with lots of stuff thrown into it, but you're likely going to run into similar complaints about "why does it contain this {application,driver} instead of that one?" as you already have now. IMO, the TinyCore project shouldn't worry about such a "showcase."  Anyone with enough desire can do that and host it separately, but the fact that it contains preselected apps and some group of drivers and a variety of services by definition removes the whole issue of modularity. The users attracted to such a thing already have similar (even if less desirable in some cases) options; they'll also come to see the "showcase" as the "base" and expect more rather than less, particularly when it comes to which sets of default apps, which drivers, full ssh instead of dropbear, more aesthetic improvements, etc.

@Sandras: Windows mounts devices, too (users are supposed to "safely remove" USB devices, which is the same as unmounting). Many Linux distros have opted to use Windows-like auto-mounting by default. The concepts are the same regardless of operating system, even if the terminology is different and if one is a bit more automatic (and possibly more forgiving of ignorant user practices, such as not "safely removing" attached media) than another.

[Guy]

Probably the most difficult thing for people learning Tiny Core for the first time, is installing it.

I see an opportunity for someone to make an install script, which prompts for input from the user, downloads the extensions needed, and installs Tiny Core. I have mentioned this in another post. If someone starts, others may improve on it.

This will not work for people with wireless internet connections, as they can't connect to the internet until appropriate extensions are installed. This creates an opportunity for someone to make a remastered cd.

The Tiny Core team are doing a great job of developing Tiny Core. I don't expect them to do more.

If someone wants to create and host a remastered cd, it would be a benefit to new users. Anyone doing this needs long term commitment, creating a new remastered cd with each new release. It would not be so great if it is done once, then the person gives up.

Just my opinion. I know some people have different opinions.

[loserguy]

Hi, I'm new here and I think TC is great, if not I wouldn't be using it or posting here  Most modular (bar LFS) of any alternative out there.

But I also think there is slightly more effort required compared to some of the other "works-out-of-the-box" distros out there. To each their own. I have used puppy for a couple of years now, and not having to sudo 10+ times a day isn't really that bad, just got to be careful not to rm -Rf something to your system.

For what its worth, the order of my grub menu.lst (yes, I distro-hop, who doesn't  )
1) Puppy
2) MicroCore
3) ...
.
.
6) Slitaz
.
13) Windows XP (came with the laptop)

I've also got CAElinux (Ubuntu  ) running on my workstation. I mainly use my laptop to remote login to it, hence the lightweight distros.

[thane]

"My verdict: The article is complete crap. 2/10."

Thank you lucky13 for my best laugh of the year.

Yes, TCL takes some effort to learn, but to create the setup you want (rather than what somebody else thinks you should want) there's no Linux better.