Slashdot Mirror
Should Aunt Tillie Build Her Own Kernels?
DeadBugs writes: "Linux Weekly News is reporting on a new linux controversy. The inclusion of a Kernel Autoconfiguration program that would make it easy for almost anybody to build a custom Kernel on their computer. Eric Raymond supports this idea saying that this will bring Linux to a wider market. Those that oppose this idea mainly think that only those educated few should custom build their own Kernels. I for one hope this gets included if only to make standard installations and upgrades faster."
Why the heck is this a controversy? It seems to me that anything that makes good technology accessible to more people is a good thing.
I'd like to hear good arguments in the other camp, though.
I can see it now from the vendors "Compiling your own kernel voids any software support." Can you imagine trying to keep up with all the changes as a software vendor? So maybe as a test system, yes. Supported? No
"If you are on fire you can just stop, drop, and roll. If you fall into Lava you are just dead." - my 5yr old daughter
Am I reading correctly? Is this a debate over limiting vs. allowing certain behavior? What part of the Open Source philosophy got suspended while I was at lunch?
Let some distribution try this. It may take off, it may fail-- that's what it's all about...
davejenkins.com |
Expect to see a lot of "This software only supported under the standard Red Hat v7.2 Kernel."
I don't blame the software companies one bit either.
Most Linux users are already familiar with the caveats and reprecussions of customizing your kernel. This kind of tool would just make it easier to get to.
There aren't all that many "casual" Linux users. That market is dominated by Microsoft. And if you've deployed Linux to a work environment, chances are you won't allow a tool like this to be used, because you'll probably want to lock down the configurations (making your life as a sys admin a lot easier).
Assuming Linux continues to proliferate to the consumer market, I still wouldn't be worried about people tinkering with their kernel too much. Most people, especially at the "average Joe" level, don't understand the inner workings of their OS. Heck, most of them fear their OS and assume that they'll break something if they tinker with the OS's inner settings. I wouldn't conclude that simply because the tool is there that most people would be interested in using it.
My sigs always suck.
I really can't emphasize strongly enough that I believe that if Aunt Tille has to build her own kernel, we have much bigger problems that Eric's autoconfigurator will solve.
Most of the kernel configuration is simply a matter of which drivers to include in the kernel instead of as modules. Distributions put most of the stuff out as modules, so all that a kernel autoconfigurer would do is notice which modules are loaded and build them as part of the native kernel. The advantages of this are minor--slightly better memory utilization, no need for initrd.
On the other hand, there are some areas where an autoconfigurator would be handy. That's when determining which chipset features/bugs to compile for. Hopefully this project will focus on the areas of configuring that are more complicated than (y/M/n).
While 2.4's module support is excellent, and modularisation is become more and more prolific throughout the Linux architecture, there are still several important features which need to be excised from the kernel core and made available as runtime modules. Trivial features such as APM support, SMP and Unix sockets shouldn't require a full recompile to activate. Why do we insist on prolonging the life of "make config" and its brethren when we could very well do without it altogether?
Ponder:
Should Aunt Tillie Build Her Own Kernels?
Should Aunt Tillie Install Her Own OS?
Should Aunt Tillie Install Her Own Applications?
Should Aunt Tillie Run Her Own Applications?
Should Aunt Tillie Produce Her Own Documents?
Should Aunt Tillie Think Her Own Thoughts?
*****WARNING. USING THIS TOOL CAN SCREW UP YOUR COMPUTER, BIG TIME. IF YOU WANT TO USE IT, DONT BE MAD AT ANYONE BUT YOURSELF IF YOU FUDGE THINGS UP*****
.. I mean, seriously. The only thing you could end up with is some fucked kernels (who should get along just fine with the fucked registries) and some users who will learn and be cautious, and end up having a better understanding their computers.
If MS can include regedit, you cant tell me that we can't inlcude autoconf
"Old man yells at systemd"
...doesn't everyone build a custom kernel? I've been using Linux for years, and I always assumed that the prebuilt, "Christmas Tree" kernels were just to make installation easier. People actually run with those things? Heh..
But I don't think "Aunt Tillie" should accidentally come anywhere near a kernel: Users should not care about kernels because they have to, but because they can. That means that most hardware configuration tasks should be accessible without touching the kernel, including installation of new drivers. So include lots of warning signs -- optimally a normal user will never have to log into his box as "root" except for installing new software with a graphical apt-get like tool.
I mean honestly, what claim can linux hold over windows if not that the availabillity of the source code allots the user more freedom? This is, as far as I'm concerned, what linux is all about. I am totally unable to understand any argument against making one of the most important benefits of linux more accessible to a wider market.
I suppose I'd have no trouble believing this. I'd still like to know what the requests are and why they are ridiculous.
Aunt Tillie shouldn't have to
But.. she does. We don't have runtime autoconfiguration that works in every case. If an autoconfigurator is easy to build, and won't impact the people working on runtime configuration, then why stop them from doing it? My computer should read my mind (or at the least, the pointer should move to the thing my eyes are looking at) but I'm not going to tell people to stop working on improving mouse support.
The autoconfigurator is bound to be an imperfect job
True enough, but this is true for runtime autoconfig as well.
The kernel people are already drowning in bogus bug reports
Kernel bugs are reported via email on the mailing list. This is described in marginal detail in /usr/src/linux/REPORTING-BUGS. Furthermore, it begins with the following dubious paragraph:
What this document highlights more than anything is that kernel developers are drowning in bug reports because linux kernel bugs are reported in an informal format on the mailing list. Get a proper bug tracking system and it will be much easier to keep track of real bugs. This should be done regardless of whether or not we make "kernels for the masses". I hadn't heard about the bug report problem until you brought it up, and it's frankly amazing that it hasn't been addressed in this manner already.It's rare that you're presented with a knob whose only two positions are Make History and Flee Your Glorious Destiny.
It would be great if anyone could build a custom kernel.
Imagine this... Let's say it's 5 years down the road, and the hot new computer is the 72 ghz Apple Pentium G7 with 64 gigs of on-chip ram. Hard drives have been totally eliminated because new, memory based permanent storage technology has been invented and proven over the past 2 years. An entire meg can be recorded in under 1 microsecond. The only remaining mechanical component of a computer is the standard Glass-RW drive (the 2 terabyte recordable successor to DVD), so the whole computer is now a small single board, and most of the electronic hardware is inside the main processor, an inch square in size. In fact, the plugs on this board take more room and cost more than the computation hardware.
Now imagine that a build world takes 4 minutes to complete. Here's how installation of FreeBSD 9.8-RELEASE takes place. (Yeah, I know this was a Linux thread.) You pop the Glassdisk in the drive, choose a few options, and all your software is configured, optimized, checked for security vulnerabilities, compiled and installed within 2 or 3 minutes.
In order for that to happen in 5 years, Granny needs the ability to custom configure her own kernel right now.
...when elite hackers said that only elite hackers should have Linux, and all of these "Red Hat" guys are polluting the user base. They are, of course, full of shit.
The whole point of Linux is having a stable and friendlier version of UNIX that is GNU and doesn't have any ties to MS. We now have Average Joe User with their own copy of Linux/X and they are using it just fine. Why should we limit ourselves because we need to do it the "old fashioned way"? Let them (and us) have a easy-to-run auto-config script for building kernels. Are we going to delete our "make menuconfig" scripts and tell everybody to replace it with "vi Makefile", just for being elitists?
Personally, I think these are the "10 miles in the snow, both ways" people, who still believe that the best way to configure PPP on Mandrake is rolling your own scripts. (Uhh..."netconf"...duh!)
Zodiac Survey
Easier to use tools are great.
I just hope we don't start designing things such that people say "oh, to do that just reconfigure your kernel with the foobar option". Feature sets should generally not require kernel recompile imho. For a long time, this was a UNIX weakness.
If we can avoid this (which is after all worse than the old "reboot NT to configure something"), I'm for it 100%. I'm not saying that you have to recompile the kernel much nowdays (I had to once to get an unsupported Ethernet driver working), but kernel recompile gets really easy, I'm nervous that people would start to rely on that way of doing things. Which would be bad.
--LP
She doesn't, and the original poster doesn't understand what ESR's system is for. The current build system misses some dependancies and has some other flaws that I can't remember at the moment. Basically, this discussion began a *long* time ago (in linux time at least)...something like a year ago it was coming across lkml. This has nothing to do with granny compiling a kernel, it's about making the build process *better*.
The best way to accelerate a windows box is at 9.8 meters per second square.
Aunt Tillie doesn't need this. But, as a computer consultant and VAR, I need the ability to easily make these kinds of changes based on what my customers need.
Sure, I can do this myself the old-fashioned way. But this is the kind of thing I prefer to delegate to someone with a lower billing rate so I can focus on the things that really bring in the bucks. It is easier to train someone to use Eric's AutoConfigurator than it is to explain make files and such...
Jack William Bell, who likes the KISS method in most things.
- -
Are you an SF Fan? Are you a Tru-Fan?
I have multiple engineering degrees and several years' experience building and using Linux and BSD, and *I* have trouble configuring, building, and installing the Linux kernel. Forget Aunt Millie -- I want a good kernel autoconfig tool for myself!
Programming is the act of creating automations of complexities that are made up of simpler things.
Does the programmer re-write open() every time they need to open a file?
There is not only nothing wrong with making it easier to build a custom kernel, but in fact there should be a growing interest in doing this sort of simplifying, given the GNU Hurd is about not only modularity but about servers/transltors and creating such, even custom as is needed.
This can be taken even further in that autocoding tools can be and should be built for the GNU users.
In a hundred years from now, how do you suppose programming will be done (given programming today is only about 50 years young)?
As things are being done today, it is not possible to do such a program of complexity as can be imagined of what would be a holodeck program (And we do have such virtual reality cudes today in university labs).
It won't be untill the general programming field realized the need to genuinely and honestly address and do the automation of the field of programming. Certainly everything else can be automated, including human balance and movement (segway).
It's fooling to continue the illusion that programming is not itself automatable. And to begin making it happen, where better than on higher level like autoconfiguration system that allow custom kernels to be done? (Or at least one place for it to begin)
A recent research paper on autocoding presents the current/recent mindset on autocoding. It's worth reading to see how young and admitedly immature the field is. Open system and Open Source Software such as the GNU efforts (Linux, the Hurd, etc..) with their open community has far better ability to do what needs to be done than any private effort which will be biased away from doing the things that need to be done.
Soooo, anything that automates computers and their use is inherently a good thing, for iot will allow us all to reach and achieve much more advanced systems and the benefits of.
I want a distribution that has a similar GUI installer that RH and Mandrake have, but instead of invoking "rpm -i" for each package, it would build all the install packages from source drops. The "installer script" could be a large XML file that describes how to compile each package, what its dependencies are, and provides a mechanism for tweaking the packages configuration. Most of the packages out there can have their runtime configuration configured via their 'configure' script (wow that's a lot of "configures"), making it a fairly uniform approach. In addition, at the beginning of the install, it would be neat to see controls for your *exact* hardware configuation that get turned into CFLAGS like -march=i[my]86 and -O3, etc.
The only drawback I can see is that it would increase install times by a *lot*. However, in the end you would end up with a *highly* optimized distribution.
The idea came to me while building my own.
std::disclaimer<std::legalese> sig=new std::disclaimer; sig->dump(); delete sig;
I don't really know if there are any tools that gather it all into one place. However, there certainly can be, so if anything, your claim has nothing to do with the kernel itself, and rather with some simple userspace tools that may/may not be written at this time.
How can we continue to believe in a just universe and freedom to eat crackers if we have no ale?
I think an easy to use, Joe User can make his own kernel is a great step as far as making linux more usable for the masses, i.e in a desktop environment. You can make a more compact kernel without knowing ALL the minute system details and random hardware that most people don't have or know what it is. Definitely a step in the right direction.
~.Evanrude
Happiness.
...". Needs to describe how to have multiple kernels, in case something wasn't quite 100% safe. Needs to be concise enough so that a total newbie WILL read it.
Aunt Tillie does not need to build a kernel, ever.
Aunt Tillie can easily build a kernel if she feels like it.
Misery.
Aunt Tillie needs to build a kernel.
Aunt Tillie cannot build the kernel she needs.
It's been a long time since I've compiled a kernel except. The last kernel I compiled was to get an NTFS read-only module so I could ftp it to a "rescue". I wish any other configuration was as easy and straightforward. Need to get the "right" starting point and extremely explicit directions, including all the "remember to
Besides, when Aunt Tillie has reconfigured her kernel, she knows the "My" of "My Computer" now really means Aunt Tillie's computer.
I think if Aunt Tillie can create a swap partition during installation, pick a window manager, download compile + install the latest mozilla browser update (or maybe she prefers Opera), configure her firewall, and set up lpr for her printer, she can recompile her kernel. I just don't want to be around when she starts looking for "Freecell."
_______
2B1ASK1
That was way back in the 2.1.x days. Then, I knew all the caveats of the minor revisions, and I knew which particular revisions were more stable than others. Now I'm nearly the opposite. I'm happy to leave my system running for months on end without checking the status of the kernels. I actually have to "cat /proc/version" to see what revision I have fired up.
That attitude is only reinforced with the 2.4.x tree. Pondering a kernel upgrade is like pondering if I want to step into a minefield.
Reading the comments in "2.4, The Kernel of Pain", I know there's still Version Whores out there. They know the obvious stuff like "don't use 2.4.15". And I'm sure there's less obvious stuff, too. Aunt Tillie or whomever isn't going to keep up. If she steps onto the 2.4.15 mine (or its equivalent in the future), she could do damage to her system.
To that end, we could use short, digestible ranking/summary system of the kernel revisions. (Or does one already exist?) Which kernels in the stable branch are really unstable? Which are the most robust? Many, Aunt Tillie and myself included, would find value in such a system, regardless of a Kernel Autoconfiguration program.
I don't think that anyone genuinely interested in linux/open source/what have you, and who doesn't have their head stuck up their own asses in conceit would honestly say that an autoconfiguration tool for the kernel is bad. Let's look at this objectively.
First, such a tool would only make linux easier for people that are not knowledgeable with computer workings, and make it a more viable option for those who don't want to mess with, or aren't knoledgeable of, the inner workings of the computer. I've run into many people (online) who don't have support for xy device with #.#.# kernel, don't want to install another distro, and need to compile a kernel.
Second, (as far as I know) this would be something fairly easy to do, provided that the device that wants to be used is already attached to the system - the kernel seems to have a decent detection system already, just have, say, a 'kernel compilation disk' which would have the kernel you want to compile, with all the possible modules compiled in, which would use your system. it'd have it's own initscript, which would have a step-by-step process, walking you through the configuration (eg., Is the kernel source tree untarred already?, Is the kernel source tree in a location other than the standard location? etc)
Just some ponderings.
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
This thread's funny.
I put together a kernel rebuild guide a few years ago ( Kernel Rebuild Guide ). I'd guess that for perhaps 95% of Linux users, there's absolutely no need to rebuild a kernel. For those that do, it's usually to enable a feature or to tweak just an iota more performance from the system.
Sure, anything that makes the system easier to use is good. It would be wonderful if guides such as mine were obviated. At the same time, should we really be wasting time on what's essentially a band-aid? By this I don't mean that Aunt Tillie shouldn't re-compile her kernel, only that if Aunt Tillie (a regular user) requires the feature then the distribution should already support it through other tools.
The main problem I see is that no matter the frontend, a kernel recompile will invariably ask a lot of questions that Aunt Tillie may be unprepared to answer. And if she can answer them I strongly believe that she would have absolutely no problem with the current configuration tools such as xconfig/menuconfig.
that people understand precisely by messing with things they don't understand. In fact only for very unimportant things does anyone stand a chance of understanding prior to messing with.
This is perhaps the dumbest flame war I've yet seen on discussion. This is one of the reasons Windows is leading Linux by FAR in the OS wars - if you even want to call it a war.
The answer is very simple. Of course allow this autoconf. Autoconf's are great, people should be able to run a program that makes life easier.
BUT
If you're building a Distro of Linux for end users like your fictional Aunt, don't include the feature. Just don't include it. There isn't enough of a performance increase that you'll see from a kernel optimization in almost every case. Truly if Linux wants to make it onto the mainstream, they are for all intensive purposes going to have to dumb it down a bit. People who just want a simple environment to write their reports, file their taxes, surf the web, and email friends are not going to give a crap about optimizing their kernel. That is best left to hackers. Why not create a distro that speaks to the masses? So don't put it on your 'enduser' distros. That's why distros exist, isn't it?
Now let's face it, the majority of people who use Linux are using it in server environments. If I'm a sysadmin and I want to setup this new distro of Linux quickly and easily without having to search through lines of what ends up looking like a bunch of code, I'd easily take autoconf. I don't see what the argument is about, really. What it comes down to is, there's a bunch of little Linux brats (no better than 5cr1p7 k1dd13z if you ask me) who are trying to protect their little clique of windows-bashers and Linux advocates (who probably don't use Linux anyway), who would rather dismiss the general public as idiots than work with something innovative and smart that makes life easier. These are Syds of the world who insist that the world was better when people did their programs using punch cards.
Well, if all the Unter-Geeks start flooding the lists with why doesnt my kernel work then it sucks..
UNLESS that leads them to learning to do it themselves.
Ive often wondered why DISTOS didnt have an autocompile script for their kernels so at install it builds one to suit your system
Sig went tro...aahemmm.....fishing........
I have problems with Eric Raymond's scenarios. Forget about if it's a good idea to make it easy for anyone to build a custom kernel, my question is, why should you need to recompile the kernel just to install a device driver ? That's just stupid. Installable drivers, that's the way to go.
According to LWN, Alan Cox said that Linux 2.6 will not have compiled-in modules.
From: Alan Cox
To: babydr@baby-dragons.com (Mr. James W. Laferriere)
Subject: Re: ISA hardware discovery -- the elegant solution
Date: Mon, 14 Jan 2002 18:08:32 +0000 (GMT)
Cc: alan@lxorguk.ukuu.org.uk (Alan Cox), esr@thyrsus.com,
cate@debian.org (Giacomo Catenazzi),
linux-kernel@vger.kernel.org (Linux Kernel List)
> Hello All , And what mechanism is going to be used for an -all-
> compiled in kernel ? Everyone and there brother is so enamoured
> of Modules . Not everyone uses nor will use modules .
For 2.5 if things go to plan there will be no such thing as a "compiled in"
driver. They simply are not needed with initramfs holding what were once the
"compiled in" modules.
Alan
cpeterso
Granted, I've never used the CML2 tools, but I followed along on the discussion on LKM for quite some time. It seems as though every post is way off the mark.
First, ask yourself this. Is 'make xconfig' a bad user interface? Nope, it ain't. What sucks about kernel configuration? The dependency resolution crap. Linus has a nifty little program in place that does a pretty good job of figuring them out too -- but it's ugly, and admittedly a kludgey solution. CML is more "elegant and flexible" which is a damn good thing -- but last I knew the bugger took 2x longer to do it's job than the old system. Kernel developers do probably 99% -or more- of kernel builds so why on God's good green earth would they want a system that's going to slow them down right now? They don't and I can't blame them in the least bit.
CML2 is nice, and it seems like it's a really good little system, nobody on LKM is opposed to it really (that I saw) they just don't want something that's going to suck minutes out of their programming day. "Aunt Millie" can't answer kernel configuration question anyway, period. Heck, most Windows users don't know if they have 95/98/NT/2000/Me/XP some of the time, let alone if their processor is Pentium III, Pentium IV, or K7 based.. unless the sticker is still on it. Shoot, they don't know if their mouse is ps/2 or serial, or what USB is. Do they know if their USB host system is UHCI or OHCI? Hell no.
CML2 is about making kernel configuration easier in terms of expandability -- not usability. The current interface is very usable, just not very flexible. Because of it's inflexibility and complexity it leads to un-bootable systems sometimes when depency stuff get borked up in strange configuration situations. CML2 takes care of -that- and nothing else. It doesn't keep you from having to know your hardware inside and out. End of story.
> Your analogy, too, is false. If someone new to computers, someone without the "respect" for the technology you mentioned, tries something dangerous that could delete all his or her files, don't you think that there should be at least one little modal dialog that pops up asking "Do you REALLY want to do that?" What you are suggesting is nothing more than classic Slashdot eliteism: "The user shouldn't touch anything they don't understand." Where's the problem in that statement? Think about how you learned to use a computer ...
.. well, how do you know what you know and what you don't? In windows, I don't think you can! What does msgsrv32.exe do? Can I stop it? Wheres the docs on it? Nothing! But imagine it was named the "Explorer Runtime Checker (Do Not Stop!)" .. then, if you had to stop it, presumably, you'd have checked out if it was safe to do so. So I'm not saying everyone should inherently know what they do and what they don't, but rather that peoples furor over the notion of people knowing this hinges on how well the interface describes it's various componants! If every file is named "sdfg34.dll", who knows what you know, but in Linux, we have the opportunity to actually name and label various things properly .. after all, we havn't made the promise to the world that *nix is foolproof like MS has.
.. guess what, I never heard of the thing. So I'm not going to screw with it until someone has told me what the worst case scnerio is. It truely is not my fault that other people do not operate this way, but I charge that they do - except when it comes to computers, because they've been taught that a good piece of software is unbreakable and can't screw the newbies. Dude, I'm not elitist. I'm advocating the opposite of what you think I'm advocating. I say, dangerous things should be the first thing you see when you march through the 'door' of another interface. Once you know what not to touch, you can mess about with other stuff, and graduate to the big red easy to pull lever with the proper signage and warning labels and confirmation step when you feel you're aware of the consequences of your actions and when you're comfortable with the level of knowledge you have regarding the clearly labeled interface. All these people are saying that making interfaces that are available, but kinda outta the way is how to best keep newbies from screwing themselves, but I say put the dangerous stuff upfront with very clear worst case scenario warnings, and then you can seperate the irresponsible people that deserve crashage (ie, those who pull the level before they are ready) from those who heed warnings, accept accountability for their actions, and actually do some reading and thinking before doing. Same with physical objects. How many times, when friends show you some new toy or something, tell you what not to do first? I think it's the natural, time honoured way, and it's frusterating to see so many people who think thats suicide in the computer interface world.
.. System! Would you want to delete the system? Even if you didn't know anything about MacOS or computers?). Linux does it .. fairly well (the boot dir is a good example). Windows is the worst .. look, don't take my word for it. Usability experts the world over bemoan the windows way (ie, the advanced stuff is there, but just hidden, and no real easy way to tell it from the non-advanced stuff), love the macOS way (an extention, the mac equivilent of a group of dlls, has a full, complete descriptive name (generally including the name of the application that depends on it) all in one file .. god bless the resource fork), and think *nix is definately heading in the right direction! I support all this, and thats why I support the disitrbution of this kernel configuration tool! If people are gunna screw with it, as it is a foregone conclusion in my argument, might as well label it well, provide enough warnings and confirmation steps, and put it right out in the open. Don't hide it, or only the newbies who've fallen off the trodden path and don't deserve the greif are gunna get abused by it. I'd rather see the careless uncautious users befall that kind of fate .. they deserve it the most, and will only make the mistake once or twice before learning that there isn't a single garaunteed safe action on this planet, so best to think before you do. :)
.. the frusterating thing is knowing I'm alot closer to how usability experts feel things should be than the average and/or power user.
You're confused. I support putting this tool out in the open, available to the novice, if, and only if:
a) It's labeled clearly
b) There are descriptive, helpful confirmation dialogs that provide a means of finding additional information
c) describe the worse case scenario of the tool in case of gross misuse (which everyone goes through, I know!)
My point in not touching things you dont understand is touchy with Windows Users (of which I am one), because I feel, in my experience
If some part in my computer is labeled clearly "Event Transmographier"
Windows does this very poorly. Nothing is written with a descriptive name, so people screw around because they can't tell if its important or not. What I'm saying is that the labels and signage should indicate this. MacOS does this well (the most important file, System, is called
I know lots of people don't agree with me
"Old man yells at systemd"
I'm with Cox on the matter that I think Aunt Tillie would be better off with the distro's kernel (where she might have lm_sensors, nVidia, TV and Radio drivers), but !
I'l defend Aunt Tillie's *right* to chose !!
That's what freedom is all about, options !
echo '[q]sa[ln0=aln80~Psnlbx]16isb572CCB9AE9DB03273snlbxq' |dc