TinyCore: just a few steps left to the ultimate OS

[tinypoodle]

Probably this is about the issue?

I wouldn't know, but that's more than 5 and half years old, so I would expect it to be fixed by now...

[curaga]

And when the source is needed, it's much more likely to find the source on the authors' pages. For instance...if someone decides to change the Linux kernel used in TC base, where will he look the sources for? Obviously not on tinycorelinux.com but on kernel.org.

Many upstreams remove older releases. And if one wants whatever non-upstream patches used in TC, there's no other source generally than our site.

[Monobit.Fortrix]

And if one wants whatever non-upstream patches used in TC, there's no other source generally than our site.

I didn't mention TC-specific releases. It's obvious that TC-specific software has to be hosted on TC sites.

Since the discussion vector here has changed to another direction, I'd like to ask you about exactly what the topic name says: what steps should the TC distro perform to become an ultimate OS that anyone would want to move to?

[curaga]

I don't believe there's a universal solution anyone and everyone would want, actually.

[Monobit.Fortrix]

I don't believe there's a universal solution anyone and everyone would want, actually.

Let me split the answer into the paragraphs, although you can read it as a whole:
1) I don't believe such a crappy OS as Window$ can even be such a soultion but it still does have lots of users and lots of newbies still want it, actually. This is due to the aggressive marketing policy. Free software community really _can_ do something to override it, but there are problems, they and their solution suggestions will be shown here later.
2) The goal of Cassandra project will be drawing the masses attention to TC technology. Unlike Windows's, the Cassandra's spreading policy will in ideal case be self-advertising - to reach the distribution quality to the stage when most users will want it not because someone "abstract and big" said "it's good" but because they will know it really works flawlessly and does what they want perfectly, and they will really get the practical confirmation of that. The effort to reach this is really worth its effect.
3) Existing Linux distros developers can already do such an effort but they can do it together only. I don't see any sense in the derivatives of the distributions that are already newbie-intended. Moreover, for example, I hate all the Ubuntu derivatives because of this fact (not Ubuntu itself but its derivatives), and note that 20% of them don't even have the reminder they're Ubuntu-based and I find it out when looking through the system more carefully. That reminds me of the "famous" and ill-rumored BolgenOS. Btw, Jolicloud is such a distro. If only all these developers got concerned on improving their upstream projects which are already newbie-intended, these projects would develop faster to reach the above aim. But for now, we do have what we have.
4) To the opposite, TinyCore, while having a perfect modular achitecture, doesn't have any single newbie-intended derivative distro. If such a distro existed, I'd try to join its development but since it doesn't, I decided at least to start the process. But you know that the creation of an ultimate distro can't be possible without the community feedback or contribution, so feel free at these things, that's why I'm here - Cassandra will be an open project.
5) Returning to the quoted post at last, one of the project aims is to make Cassandra not the absolute universal solution which of course isn't possible to create, but at least we'd try to make it as universal as popular OS'es, just outperforming them thanks to the unique TC technology.

P.S. I also have thought about build version numbering system. The version will be a 4-digit (in the future - 5-digit) number, the first 2 or 3 digits will represent the version of base TC system (e.g. 32 for TC 3.2 or 159 for TC 15.9 if it will be released anytime ), and the last 2 digits will represent the current build number (from 00 to 99) within the same version Cassandra is built upon. For instance, Cassandra 3211 will mean the 11th build upon TC 3.2. So, the first build ever released will bear the name Cassandra 3200. Sounds futuristic, doesn't it?

[roberts]

I think what most everyone does get and you don't seem to get is that the goal of our project is:

-=[ TinyCore: is just a few clicks to your ultimate OS ]=-

Forcing application decisions on others is not the goal of this project. Your choice is not mine.
The very definition of a toolkit is to build what you want.

I have stated, ad nauseam,  the above.

You surely don't think that everyone else is only running the base system. Each is running their ultimate OS setup, configured, and running as they wish.

Again, there are hundreds of turnkey Linux system for those that desire that type of system.

An analogy...
Some kids get erector sets for Christmas, some build really awesome things just by plugging some pieces together.  Some will cry in disgust because it ts too difficult. And some kids say "Wow look what  I just built".

A pre-built Tiny Core remaster does not show the potential of Tiny Core, the same way a pre-built erector set toy does not show the potential of an erector set.

[tinypoodle]

Nice analogy! 

[gerald_clark]

Where do you think he got the picture of the screw head?

[danielibarnes]

The goal of Cassandra project will be drawing the masses attention to TC technology.

I support this. It is a good thing. A simple derivative distro can simply pluck a few apps from the repository, bundle them into a cpio.gz, and add it to an existing tinycore_3.x.iso. No base modification needed to get started.

TinyCore ... doesn't have any single newbie-intended derivative distro.

Thus your project. There are numerous distributions which target specific markets, like audio-video editing, cluster computing, etc. They also have to base their distribution on something, and I think Tiny Core is as good a base as anything (like Robert said, that's what it's for). Debian is common because of its large package base, but I think Tiny Core is adequate. When it isn't, developers are easy to reach via this forum.

I also have thought about build version numbering system.

Here is another suggestion: x.y.z.build where x.y.z is the Tiny Core version (e.g., 1.4.3) and build is the build number. That sort of system is widely used.

There are numerous comments about which app is best for browsing, playing music, etc., so you'll need clearly defined criteria like "supports flash, plays mp3 files, memory usage, stability" which can be quantified. For example, flash support means more web sites work, but the browser could crash. Chromium might use more memory (no idea, really) but it's more stable. Identify these criteria, the various solutions, the tradeoffs, and how important they are to your target audience. It's called a QFD (Quality Functional Deployment).

[Guy]

I agree

I think, when other distros are based on Tiny Core, Tiny Core will become more well known and more widely used.

Look what Ubuntu has done for Debian. Since Ubuntu, other distros, such as Fedora and Suse, are trying to encourage people to make distros based on them.

The concepts in Tiny Core are brilliant. Tiny Core has many advantages over conventional operating systems. I think this is the next generation of Linux. A large proportion of Linux users have not tried Tiny Core, and don't understand. As more people understand, expect more distros to be based on Tiny Core.

I think the future of Tiny Core is going to be bigger than most people realize. Expect many more distros to come out based on Tiny Core.

[Guy]

Just my opinion.

The name Cassandra does not sound like an ideal name for a Linux distro.

I have not thought of a better one.

Why not get the suggestions of others. See who can suggest the best name.

Just my opinion. You may disagree.

[Monobit.Fortrix]

Thanks for the constructive answers, anyway.

Again, there are hundreds of turnkey Linux system for those that desire that type of system.

You see, no one of that turnkey systems is based on TC technology and thus hasn't got its main advantages.

A pre-built Tiny Core remaster does not show the potential of Tiny Core, the same way a pre-built erector set toy does not show the potential of an erector set.

At least more people will know about TC itself and if they want its full flexibility potential (and really know how to use it) they will either get a bare TC version or will remove the userspace level from Cassandra, in fact getting the same bare TC version. I know that mostly anyone here (including myself) is ready to do plenty of reconfiguration work to get the simple tasks be done seamlessly but Cassandra would be intended just for those who aren't ready for this, showing some optimal functionality from the start and architecture advantages of TC at the same time. It's a pity but the world isn't limited by computer geeks only. Why must the soultion for others (those hundreds of distros mentioned above) be worse than geeky solutions such as TC?

I support this. It is a good thing. A simple derivative distro can simply pluck a few apps from the repository, bundle them into a cpio.gz, and add it to an existing tinycore_3.x.iso. No base modification needed to get started.

You see, I had to do some reconfiguration in the userdata scripts in order to:
1) get actual mounting information for Monto;
2) let the user configure the keyboard layout and wallpaper by editing the files (yet) keyboard.set and wallpaper.set in the MySystemSettings folder without any intrusion to session start scripts;
3) start Conky, Tint2, Monto and restore the layouts after starting an X session.

Thus your project. There are numerous distributions which target specific markets, like audio-video editing, cluster computing, etc. They also have to base their distribution on something, and I think Tiny Core is as good a base as anything (like Robert said, that's what it's for). Debian is common because of its large package base, but I think Tiny Core is adequate. When it isn't, developers are easy to reach via this forum.

That's the point. When more people, including developers, notice that TC is a good base for everything, probably it will become the main platform technology of the further Linux distributions.

Here is another suggestion: x.y.z.build where x.y.z is the Tiny Core version (e.g., 1.4.3) and build is the build number. That sort of system is widely used.

And that's why it looks not so pretty. Thousands of newbies use Windows XP and don't even know that in fact it's NT 5.1.2600, for example (I don't remember exactly but 2600 build was AFAIK for OEM XP versions without service pack). Most of them are too lazy to remember it. So I look for a numbering system that would reflect the actual state of the things but looked not so complicated, so any dotless, slashless and dashless system would suit. Moreover, 4-digit system is also proven to be handy as everyone can easily remember Nokia cellphone model numbers. Any more suggestions?

There are numerous comments about which app is best for browsing, playing music, etc., so you'll need clearly defined criteria like "supports flash, plays mp3 files, memory usage, stability" which can be quantified. For example, flash support means more web sites work, but the browser could crash. Chromium might use more memory (no idea, really) but it's more stable. Identify these criteria, the various solutions, the tradeoffs, and how important they are to your target audience. It's called a QFD (Quality Functional Deployment).

I support the idea. The criteria priority now is: (from highest to lowest)
1. Stability (with different hardware).
2. Usability (convenience, design, operating speed).
3. Up to date versions.
4. Functionality.
5. Resource usage.

I think this priority guideline would help choosing the alternatives when doing some contribution to the project.

Just my opinion.
The name Cassandra does not sound like an ideal name for a Linux distro.
I have not thought of a better one.
Why not get the suggestions of others. See who can suggest the best name.

Of course, for the current stage, Cassandra is rather a project codename than a finally fixed title. The first idea (after the UltimateTC) was Sulphur but I found out that some Fedora release already had such a codename. I decided to choose the current name Cassandra because it would be the name embodying the beauty of the distro look and the perfection of its base architecture, while giving everyone the sign that this is the architecture of the future.
Any other suggestions are welcome, though. While the community base of the project and users amount aren't that large it doesn't matter in fact, and sometimes we can run a great vote but for now, I'd suggest to concern on more important things.

P.S. The 3200 build (according to the current version numbering) is going to be tested tomorrow on a brand new Lenovo Thinkpad where any Ubuntu version just refuses to run X due to an unknown issue. The Thinkpad is not mine so possibly it will be the first Cassandra fieldtest on a non-common (as far as I see) hardware.

[roberts]

OK. I think we have all made our respective points of view known.
I have let this go on long enough and no one was censured.
It is now becoming off topic, not about Tiny Core but about a remix.

This topic is not going to become free forum hosting for a remix as per our forurm rules.
If you wish to announce your remix and if it meets the GPL requirements then...

As per forum rules we have established an appropriate topic.
http://forum.tinycorelinux.net/index.php?board=43.0