There are many developemente that depend on DSL. Remember what happened to Feather (my favourite distro until they decided to change its orientation): They decide to increase the size of the distro to include some soft more.
The result was to lose its identity; now they have fallen in the Distrowatch hit parade.
DSL have to remain as 50 Mb power penguin. We do give ideas how to deal with different developements. But the main though is to keep SDL as usual.
My (humble) contribution is to suggest how to name different sizes of DSL, but I'd like to keep finding DSL as a 50 Mb distro sized.
*JT.TTK,
I basically agree with you, but since DSL-team's major aim is to keep linux 'damn small', a lite version of DSL at ~34 MB won't damage the identity of the project. In my opinion, it even just underlines the identity of the project more.
DSL works basically with a kernel, some useful apps, and a set of other apps (.dsl and .uci apps) that you don't have to install in order to work with (a kind of 'klick' before the age). That's the big difference compared with Feather and other near-to-DSL linux distros.
Thus, it seems to me that conceiving a lite version of DSL (and why not in both kernel version 2.4.x and 2.6.x? but ok, let's begin with kernel 2.4.x, it would be already a great challenge) is just pushing the DSL-philosophy and culture more in the future of computing, and, I guess, it would just consolidate the DSL identity and its unix roots ('small is beautiful').
You basically need a kernel with necessary libs and limited apps in order to start in x; other apps have to be downloaded and used regarding your needs, without the constraint to be installed on one of your devices. With backup/restore, you can save your preferences and settings. And you are just good to go.
From this point of view, I find the concept of Robert and John a (if not THE) visionary one. In my opinion, a DSL lite/near apps-less version of DSL would just be the next step in the development of DSL. I also agree with other members of the forum that a DSL lite don't have to replace DSL or DSL-N. It would be just the necessary third brother/sister.
You'll have then (according to my speculations):
1. DSL-S near app-less; you have a recognition of old hardware and a minimal profile at the software level, you can personalized it just as you want in order to fit your needs regarding your use of DSL in several contexts (= I don't need all apps when I use DSL as FTP server only, or when I just want to check my emails, etc.); DSL-S is then a very adaptable linux, which you can personalized depending on where you are, when your want to use it, and what you want to do with it;
2. DSL with the standard apps = you have a recognition of old hardware and an already defined profile at the software level, which you can enrich with other apps to fit your needs (remastering included); as more apps has been already given in the .iso, you won't have the same flexibility as in DSL-S, eventhough DSL remains one of the most flexible distros on the market compared to other linux at nearly this size, because of the conept of DSL; for me, it is the ideal step in order to begin with DSL and more generally with Linux;
3. DSL-N for new machines = you have a better recognition of new hardware and an already defined profile at the software level, with quiete the same issues as in the second point; with DSL-N, you could even have more soft as in DSL (the support of .deb should be better than in DSL-S and DSL); it is just the fun to have a little distros with a new concept of computing, that you can enrich and install on brand new machines (more 'bling bling', as Robert and John said).
DSL-N as an up-to-date version of DSL for new hardware remains quiete similar to DSL. It is not the same with DSL-S in my opinion, which would deepen the concept of computing of Robert and John.
yours zOne bit of concern...
Quote
You basically need a kernel with necessary libs and limited apps in order to start in x; other apps have to be downloaded and used regarding your needs
That may sound like a straightforward task, but I think this is one of the biggest issues with stripping DSL any further than it is. The more apps and libs removed from DSL, the less compatibile with existing myDSL extensions it becomes, and myDSL is one thing that makes DSL a revolutionary system.Thank you mikshaw for your reaction.
My formulation was brief in order to express the general idea more than the methode. On that last point, it's just not my part, I don't have the knowledge for.
Anyway, what you say proof to me that a DSL-lite would be really a great challenge, just because you have to keep it smaller and compatible with extensions. Again, I don't know if it is technically possible. But if it is, it would be just great to try.
This would be also a possibility to give the use of DSL more flexibility according to the specific context of this use, without to restrict it if more extensions are needed.
This last point, the 'specific context of the use' of DSL, as I call it, is a little bit different concept compared to DSL and DSL-N, which in my opinion are closer related to hardware specifications and, this given, software needs to enable tasks in several areas of the use of computers' OS.
As you observe, DSL-lite is surely a more problematic one to construct. But it also well supports the identity of the DSL-project, and that was the major point I would like to underline in my previous post.
yours zIn my attempts to see just how minimal it will get while retaining compatibility with existing extensions, and also allowing the use of all of the DSL system tools (cpanel, mydsl gui, etc), so far I've gotten the KNOPPIX file down to just above 30mb. This still isn't ideal for a super small DSL, but i'm only working on it a little bit a couple times a week. So....yeah, it's definitely a challenge.Next Page...
original here.