Linux the Kernel

When the Kernel went from 2.4.x to 2.6.x we saw a huge jump in the size of the sources. Of course a compiled kernel can be any size the compiler thereof wants it to be beyond the basics that get a usable kernel.

But, not only did the sources size change. Under the hood there were lot of other major tweaks like various sub systems.

As the Linux kernel has grown in size so the complexity of it has grown exponentially. And with complexity comes more and more errors. We can see this, given arbitrary hardware, if we compare kernel 2.4.x against 2.6.x compile time errors or kernel OOPS whilst running.

Given that we _need_ the kernel to grow because 1) We have more users of it now, 2) There is more hardware available now 3) 2*1 = more sources. All well and good. However, as good for Linux based operating systems as that obviously is it comes at a price due to ever increasing code complexity which creates errors/warnings at compile time and kernel OOPSES at run time.

As modular as 2.6.x kernels are, or rather can be, I think we, or rather the kernel coders, can take this modularisation further by splitting off basic kernel functions and hardware functions. I will admit my knowledge in this area is severely limited but that said I am sure this can be done.

What exactly am I talking about? The kernel, which _is_ linux, is the heart of all distributions. Without it distributions would neither function as we know them now nor indeed exist (witness GNU herd) but a radical rethink on how we perceive it is not only due but overdue.

As it stands now to get the latest all singing all dancing kernel sources one must:

1) download a huge archive.
2) build it. Assuming one knows their hardware and general requirements one can then build for oneself a new all dancing all singing kernel.

As far as it goes this is all well and good. But, the first step in that process is were changes could be made. Instead of downloading a huge archive one could:

1) download an archive of just the basics that enable the kernel, once built, to function.
2) Then one could download the various hardware sources, that would be split up into individual archives

As an example. On my main desktop workstation I would:

1) Download an archive of the basic kernel
2) download the driver for my SCSI hardware
3) Download the driver for my VIA hardware
4) Download the driver for my SATA hardware
5) Download the NetFilter sources

And so on.

All archives would extract to linux-2.6.x as they do now. Once everything has been downloaded then the configuration and build process can be done.

I know this all sounds more complex but the idea is to make the main kernel archive smaller and thereby hopefully make gcc spit out less errors/warnings and kernel OOPES less likely.

Oh how laughable!

In other blog posts here and here I have derided new, new as in springing up on the Slackware scene in the last 2-3 years, Slackware users for their attitudes. So, it was with much laughter I saw on one web site "Approved by..." as if that approval means anything to anyone except the person who deems him, or her, self important enough to give it.

I mean. Who gives a rats arse if some web site administrator deems him, or her, self important enough in the Slackware world that he, or she, gives his, or her, approval to something someone else has done.

What the fuck is that all about huh? Approval? Eh?

This particular web site administrator bursts onto the scene a couple of years ago pimping his web site at every given opportunity. Even today he pimps and pimps and pimps o such an extent that it is bloody annoying. Up until the time he came on the scene no-one had heard of the guy and now here he is giving his approval for something that someone has taken from elsewhere, adapted to fit in with this guys way of doing things then he deems it okay by giving his approval to it?

I do not give a flying one and I strongly doubt many long time Slackware users do either for his approval and that is why I will never, ever, submit anything to that web site even though I have hundreds of scripts that would expand his web site a hundred times over. I will never, ever submit anything because:

1) I don't like the way the his scripts are designed and I will not change my scripts to suit.
2) I don't need anyones approval to know my scripts are suitable for the purpose they were written and
3) I seek approval from no-one.

Damn Americans, not all of them obviously, seeking to dominate everything they touch. It does my head in it does and it is totally, utterly and laughably laughable in the extreme.

Emulation.

Or as the wine blarb goes "Wine Is Not (an) Emulator". Which is certainly true of both the Wine project and the Qemu project. Both can use full blown MS Windows installs and that is where the problem lays. As GNU/Linux adoption via a collection of a kernel and several applications commonly known as distributions continues almost unabated we have wine and Qemu which in my honest opinion are fighting the good fight and slowing down Linux adoption in several areas. How and why are wine and Qemu slowing down Linux adoption? Well, the answer to both questions are basically the same. Because people who install a distribution then go on to install wine and/or qemu and then go on to install MS Windows within one or both of those and then go on to install whatever MS Windows application they see as being unavailable under their choice of Linux based distribution are simply stopping the improvement of applications in the same field from growing as the good folks who create these programs don't see the need to create further as people who use wine and qemu are settled on whatever application they used under the MS Windows operating system. And that is the why and how answered. If only these people who dogmatically insist on using applications designed for another operating system on their distribution of choice used whatever search engine they prefer, or look at web sites like this one which shows The table of equivalents / replacements / analogs of Windows software in Linux they would see that in almost every single case there is an equivalent application that runs natively under a Linux based operating system. They could then remove the baggage known as wine and qemu and quite possibly at the same time have a more stable system.

A.O.L.S. (and Usenet in general).

Or to give it its full title Alt.OS.Linux.Slackware (where OS stands from Operating System). Since switching from BSD to Linux during 1993 one of the things on the Internet that sure was an excellent place o get help and/or glean some excellent information was on UseNet. In those early days the Usenet was a place of gentlemen where every question about a given subject was, providing it was asked in the correct place, answered with a skill level so sublime as to be a joy to witness. Sure, even during those heady days there was those whose only premise was to disrupt proceedings. These were quickly identified and given the derogatory mantel of Troll. Overall though the eloquence and excelence of the vast majority of these early adoptors shone through. Because of this the Usenet was one of the first places any one would visit if in need of an answer. Back in the early 1990's I was something of a prolific poster to the Usenet reading as I did hundreds of messages, known as posts, a day and replying to as many of those as I wanted to. Sure, I got the odd thing arse backwards and got hung out to dry, known as being flamed, because of that but the overall experience was a good one. But as a whole new generation of posters came along so the quality of Usenet declined. Just as in everyday life English language skills declined so the quality of Usenet posts. Until today the situation is so bad the it could be said that the Usenet is a shadow of its former self. Nowadays I rarely post to the Usenet proper (to define Usenet proper one has to know the basics of how Usenet posts propagate. Basically, very basically, Usenet can be defined as being like motorway stopping points. Throughout all the motorways here in the U.K. you can get from one stopping point to another via the motorway road network. Usenet is similar to this methodology) though I do still read several Usenet groups added to this is the fact I run a subscription only news server of which I take part by posting and replying to articles posted. Why do I rarely post now? Well, the standard is generally so low now that I have dropped myself to the disinterested level. And so to the subject. A.O.L.S. The standard in that particular Usenet group is no different to any other Usenet group. But it does seem to have suffered more than most. As the Usenet more or less mirrors the outside world so the standard of the posts contents have dropped along with the general demeanor of the outside world. People are more arrogant. poorly, by comparison to the early/middle 1990's, educated and this reflects in the text of the Usenet articles. A.O.L.S. suffers as all Usenet has suffered from this blight. Oh and let us not forget the role in this decline google groups has and is playing. Remember that google groups is not Usenet but even with that fact known and understood the quality of google group users posts is so bad that it is dragging the Usenet ever further down. Why is this? I have one theory as to why this is. Permit me to espouse it here. Along the straight line of time we have taken the following route. Usenet, fidonet (another olde worlde but wonderful article based network), mobile phone texted'ing. Now that may look like a harmless route to have taken but consider how piss poor mobile phone texted'ing is quality wise is then apply that to what is left of Usenet and you get a huge drop in quality. Added to this is, as I said, the general decline in other areas and of course as the wider worlds populace has got more and more arrogant in their general demeanour and we have the stakes right there that are embedded deep in the carcase of Usenet.

Phoronix

That is a nice web site you have there with Linux users well catered for with some interesting articles. But, I have to ask one question regarding your choice of distributions.

Ubuntu? I mean why?

On my wifes machine it was as unstable as an unstable thing can be in a land called unstable. At least it was until I replaced it with another distribution.

If you don't want to lose the head of steam you have made for yourselves then please lose Ubuntu.

Use a distribution that carries little to no patches in its wake. Ubuntu, being Debian derived is, just like Debian, patched to Hell and back and carries all the baggage as well. Lose it. Seriously.

Dependancies

All long time Slackware users know its package management tools do not calculate dependencies. It never has and possibly never will. At least not officially.

The author of Slackware, a certain Mr Volkerding, is on record as saying Slackware will never do dependacy checking due to the inherent risks of doing such. Now, it is perfectly feasible that as he gets older and married life takes a hold and the probability of children being added to the fold, that he may change his mind. However, at this time in Slackware's development cycle, it is at version 12 with -current (the play area that will become the next Slackware version when the author deems it fit to be so) in full swing, Slackware's package tools do not support dependancy checking.

There are a couple of packages in /extra that do do dependency checking but as is witnessed all over the Internet these programs sometimes wreak havok on users systems. It is not for me to say whether these issues are caused by user or program errors but having played with these programs on my test desktop machine I can only state what happened to me. I am by no means a novice Slackware, let alone Linux, user but both Slaptget and Slacktrack (the programs deemed worthy of a place in /extra) both bit me and totally hosed my test desktop machine. Luckily, having used Slackware since forever I recovered that desktop but as sure as night follows day there are many poor tortured souls out there that do not have me experience who would be left with but one option and that is to reinstall. I would hazard a guess that Mr Volkerding knows about these oopsies that occasionally happen and until such a time as these reports and complaints stop he will never incorporate such programs into Slackware proper.

And so it should forever be so.

So, why then do various third party outfits add in extra tags to their package sets that are for dependency checking? These outfits add slack-required files into their packages which are then used by Slaptget. Linuxpackages.net started this rot and others such as slacky.it and slackbuilds.org are doing it too. It is wrong and flies in the face of what Mr Volkerding himself has stated.

I ask these package repository owners to clean up their act and remove these slack-required files from their packages.

A while ago I joined a party whose aim was to properly explain how Slackware's package tools worked. The grand plan was to document every aspect of those tools. A noble aim I am sure you will agree. However, it soon became apparent that what they wanted to do was expand on what Slackware's package management tools already did by adding in various extra bits. As soon as this became apparent I and my partner in crime for this very blog, left.

Slackware is lean and clean. Long may it remain so.

Stealing

That is the only word I can think of that hits home how I feel about this particular thing.

In the latest slackware-current changelog there is a reference to someone who is not averse to stealing scripts by other people and calling them his own. In this particular case it is the build script for XFCE4.

I put up my own build script for XFCE4 when XFCE went from the older XFCE to the newer XFCE and labled it XFCE4. In fact, I was building every new XFCE4 as and when they came out upstream. I would build it, put up the package, build script and slack-desc. For one or two releases I was prompted by users as I was slow in catching the upstream release. However, as soon as I was notified I went and downloaded it, built it etc etc.

The similarities between my own script and the script in slackware-current attributed to a particular person who has been noticed by others for stealing other peoples scripts is disappointing at worst and downright annoying at best.