A Better Installer for Debian?
F1re writes "Linux User mag in Germany has decided to include Debian on the mag and wants to make a more user friendly installer. They are looking for help from Debian developers. More info here Linux User"
← Back to Stories (view on slashdot.org)
The Debian installer is already plenty user friendly, just not beginner friendly. Quite a difference if you ask me (and sometimes even opposites!)
First Windows removes its real-mode command line, now Debian loses its undecipherable installation syntax? What will all the esoteric-knowledge gurus do?
Jouster
Mandrake's installer is GPL. It shouldn't be too exceptionally difficult to port it to Debian...
I was able to install Debian after only few months of Mandrake, and damn I was glad when I had my linux-box fully configured and running. (After a week of configuring with vim.) :)
The installer isn't 'hard' it's rather "are you good enough to use debian?".
Userfriendly installer doesn't fit with debian's style.
Progeny had a good one, I use it and then apt-get dist upgrade to woody
Progeny Debian Linux was a GREAT distribution, when it existed. It had great hardware detection, a very simple installer and some other features. Some of the best parts of it are now available.
You can use Progeny Graphical Installer (PGI) to install a nice Woody release, or download the package and create your OWN customized installer with it. This thing is GREAT. Check it out - they are pushing hard towards the 1.0 release.
http://hackers.progeny.com/pgi/
It is very nice, and has a text-mode and X-based installer (you can even do the X install remotely on another machine). This thing is great, and I use it for all of my installs right now. Thanks, Branden Robinson and team for keeping this great part of Progeny Debian Linux alive...
I think that the basic point is being missed.
From the article
Right now what we'd be most interested in is some feedback by Debian developers and users out there
The golden rule in HCI is "Developers are not target users". Sad as it may seem for some people Linux Developers are no longer the same people as Linux Users. This means that, by and large, interface designers should IGNORE THE DEVELOPERS!
Users are the ones that matter here. As a first time Linux installer I don't really care about most of the things a developer cares about.
I haven't installed Debian, but let me compare my last Linux install (Mandrake 8.something) to WinXP...
All WinXP asked me was, essentially, "What is your Country and TimeZone".
Mandrake wanted to know the intimate details of my network card, how much swap space I wanted, what make of scroll-mouse I had, what sound card I had, what video cards I had (and don't get me started on XFree's Multimon support!). All this does is serve to scare and confuse a Linux Virgin. And if you want Linux on the desktop you can either make the world smarter, or make your products smarter.
Debian should not be soliciting people in the know - they know far more than the average first time user and are, consequently, useless for developing interfaces for newbies.
Sorry for the rant/misspellling/smell.
T
If a square is really a rhombus, why aren't all triangles purple?
Still using the same install system it had 7 years ago! (And its quite good too as long as
you don't freak out if you can't use a mouse)
This will give Debian quite a lot of publicity. Maybe enough for it to take off, at least in Germany ;-)
Personally I really wonder why people still use RPM based distributions, is it just because of the installer or the publicity? I mean, *everybody* who's heard of Linux must have heard of Redhat, but no beginner would have heard of Debian. Yeah like this "I'm using Linux 7.2, what are YOU using?"
Seriously I don't see anything potentially bad about making a user friendly installer, the one Debian uses now really could be improved. It's nice that they asked the Debian guys about it though. I wonder if it will get back to the main distribution of Debian if the installer is really as good as it sounds?
Don't quote me on this.
I recently switched from Redhat to Debian linux. I used the network install (because i have a fast connection) and i found that the most obscure part of the install was finding which NIC model i had (because they went by manufacturer code instead of human-readable names) If a prepackaged installer simply had something that detected your NIC automatically, with some simple instructions to read along with each install stage (easy ones found at www.linuxnewbie.org), then it would be a much less painful install. As a seperate note, something must be added to automatically configure USB optical mice, because as it is they are not (a huge pain for a user with limited skills).
dselect is already a good tool for choosing packages to install and seeing what is out there to install. Its interface could be improved somewhat (always going past help screens becomes a pain, and collapsable trees should be in to reduce clutter(and if they are already, why aren't they obvious))
This should put user-friendliness in, while maintaining most of the customization available in the regular install (after all, you could always ignore the advice...)
Free as in *BUUURP!*
I'm sorry, but when did "I can click on it" equate user-friendlyness?
Debian's current default installer consists of a group of sub-menus with descriptive explainations of what task each menu item would perform.
If this is an arguement of asthetic or practicality, then it should be thrown out. The only way this arguement for a better Debian installer could possibly hold water is if we're talking about the detection of hardware, which is marginally frustrating (I had no problems the first time I installed debian).
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
fp
Storm tried and failed.
Progeny tried and failed.
Corel tried and failed.
The Debian community is simply to anal to take on these "user-friendly" additions. I really wish they would do something about this!
Debian--'cos life's too long :-/
At least anon posting still works ;-)
Propz to all non logged in Trolls. Logged in Trolls suck azz.
- Moomin da farker
The Debian installer used to be awful, and you really needed to know your way around dselect to get it installed properly (or you can just install nothing and apt-get it all once you've installed).
I recently installed unstable, using the testing installer, and I was surprised to see that it doesn't seem to use dselect anymore! There's a much more friendly (although still text-based) utility it uses to select packages. Honestly, the installer really is pretty easy now (on par with RedHat, anyway, only a bit less pretty). It could be better for non-computer-literate users, but only stuff like the partitioning utility.
This was on alpha, btw. Things may be different on the x86 side of things.
Debian is the first Linux distro I have ever installed and had no problem with it on ancient p200 with mainly ISA slots. (NOV 2000) :)
Except I had some grief with the SB AWE32 but thats another story
ERR 411[Max number of witty sigs reached]
I've installed Debian dozens of times on many PCs. What's wrong with the installer ? Ok, the only tricky part I'd say, is the package installer. That DSelect thing can be painful at times, but the rest of the installation process is pretty straightforward and I actually enjoy how they present all the actions on the menu so you can jump ahead/back anything you like.
-Billco, Fnarg.com
dselect sucks. It's the hardest thing about getting a working debian install, akin to a purity or intelligence test. This is exclusionary, and the only way to fix it is to streamline the way a base debian gets installed. And to me, that means dselect must go. It's too hard and takes too long to get right. I've always found it much easier and faster to completely ignore dselect and add the packages I need later using apt, which is far more friendly (and actually works).
In HCI terms, you *must* understand your users. If your user base is educated professionals who have done hundreds of debian installs and can compile their own kernel without assistance, then the current installer is probably okay, but it's not where Debian needs to go. It has the developer Linux user sown up; Debian needs to add to the collection other types of users.
So we pick another user set - the Linux newbie and/or Windows refugee. These people don't want to know about installers, and you must make the interface hard for them to screw up. Remember in HCI terms, allowing the user to screw up might be powerful, but it's wrong. I'm not talking about GUIs here (even though I like 'em), I'm talking HCI and interface. You can have a very decent text installer.
Moving along... You describe to the potential newbie users why you need an installer in very basic non-prejudiced terms, so they understand the problem space but without suggesting to them potential solutions. Grab their suggestions and recommendations and experiences and write them all down. This is your specification to a certain extent. Users have a keen insight on what they like and they don't like. Ignore their advice at your own peril.
You create a first cut at an installer, constantly second guessing the users: "will my mum be able to do this?" "Do I have to do this now?" "Is this a reasonable set of defaults that don't need to be adjusted?" You want the user to make as few decisions as possible, whilst postponing as many decisions as possible to allow experienced users to customize it if they wish.
Once the first cut of the installer is done, you must get a bunch of new users, and watch them use it without assistance. Learn from the mistakes or missteps they make, and learn if there's steps you can eliminate. And of course, eliminate any bugs the users find.
Repeat ad nauseam until it's hard to get a bodged unrecoverable install.
Developers are truly the worst people to ask to do this. They *know* the right answers, and will not even think that there might be other possibilities.
A good OS installer is like the old A/UX 3.0 installer - it literally was a one button install if you had a disk ready for it.
Other OS's with decent installers are NetBSD (with the possible exception of the very confusing disk partitioner) or WinXP (very few questions indeed).
Andrew van der Stock
Why hasn't Debian project adopted the Corel Linux (nowadays Xandros Linux) installer? It's absolutely best Linux installer there is; much better than Red Hat, Mandrake, or SuSE.
Is the installer non-free software or what is the reason?
IMHO, using the Corel installer would give Debian a big jump forward. Debian's installation, especially the awkward dselect, is definitely its weakest point.
In my (university) environment I noticed that most start with Mandrake, Red Hat or SuSE and sooner or later realize that RPM is a nightmare for keeping a system up-to-date. Then they try Debian and are blown away by its ease of use (me included).
We dont want a better install! how shall i bost to my friends under wintendo that I INSTALLED IT! Piff leave my debian alone.
I wonder why they haven't posted anything on the Debian mailing lists...
The Debian people are by the way already working on a better installer. Woody will be the last release that uses the current one. This new installer will use aptitude instead of dselect for example.
Also, Linux User only plans for i386 support. Check this page. Debian supports many architectures, and the installer should work on all of them. Also, remember that Debian is being ported to non-Linux kernels. The Hurd is coming along nicely, and will probably released in Woody+1, and people are starting on a port to NetBSD. Again, the installer should support these kernels.
I would like to see a package manager for KDE (and those who do Gnome, probably want one for Gnome;)), and one package system for *BSD/Linux. Is there a single good reason to have a bunch of different ones? Especially knowing how good Red Hat is at doing anything (need I remind you of gcc 2.96.x for example...).
It's the same kernel, there are no need for special packages for different distributions. How come the linux distributions can't cooperate worth shit? If they want to add value to their distributions (such as is done with MacOS X and Windows XP, it's the same idea) that's great, but don't add different solutions to the same problem without _really_ adding anything. I am sure ever single linux user would love to be able to download the same package regardless of what distro you are on. And belive me, all those who try to support linux (iNTEL with their great compiler, nVidia with drivers, etc) will have a much easier time.
Then you can have your own package manager, like debian apt-get or a full blown bloated "want to be netscape and do everything by ourselves" super GUI app with IM, mail and a word processor.
First time when i installed a Debian distro i chose the Corel distro.
And that was a time when i only had 16 megs of ram on that machine (yes i know wery well).
The Corel installer vent on all night long and died at 99%.
Then, after that, when i got a taste for Debian i installed the real thing.
It only took about 1-2 hours compared to the corel 6-8 houers.
Debian does not need a new installer. I, as a newbie, read the installation guide, sat in #debian on irc.openprojects.net and asked questions when i got stuck, and installed it and compiled a kernel in about 6 hours. (RedHat took me 1 hour, but I didn't understand a thing about the system.) What Debian needs a well indexed book of all the typical problems that people go to #debian with. It should have a list of common and not so common hardware and their chipsets and what kernel modules are needed. It should explain what packages what users might want, and why. It should explain Debian's init script setup. It should have a chapter each dedicated to apt/apt-get/dpkg/dselect, kpkg, networking, modems, cd burning, sound, printing, and XFree86. It should have some examples of files likely found in /etc. It should explain every option in the kernel configuration and suggest why you would or would not need it. It should be sure not to go over the head of newbies. If it is downright boring and unreadable to experts, fine, its not for them. I'd have gladly paid $50 for a book like that.
A web site where people could post exactly what hardware they had and exactly what they did to set up their system would be great. Do a search for your hardware, read what other people did, get yours set up, and post your list.
The masses are the crack whores of religion.
What I'd really like to be able to do is to sit down at a fully installed Linux system and run a program that lets me specify out the hardware configuration of my system and the packages that I want to install on it. It should allow me to choose whether to install stable, testing or unstable. It should inquire whether I have a network connection I'm willing to install the system over. It should then compile a static kernel from my /usr/src/linux directory to my specification and build a bootable ISO image that I just need to boot on the target system to run the entire install. Ideally it would be robust enough that all I'd have to do is hand it to a user with the instructions, "Just boot this. It'll solve all your problems."
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
For reasons unknown to any living man since the dawn of time the debian installer has been the gates to the distro of the geek. It has protected us from the suits, script kiddies and those evil "home users" who legends say may even have regular sex. If these people bridge the gap between us and society the results could be disasterous, what if the world sees inside the geek safe-haven that is debian! KEEP THE NON GEEKS OUT!
"Madness is something rare in individuals - but in groups, parties, peoples, ages it is the rule." -- Nietzsche
Finally someone who doesn't want to re-invent the wheel! For all the inherent benefits to Open Source and code reuse, the amount of code-duplication (and therefore time and effort wasted) in the Open Source world amazes (and disappoints) me.
And no I am not talking about Gnome vs. KDE. I am talking about things like having 10 different ICQ clients, all with different implimentations of the protocol. Sure, a different GUI and different features is worth making a new program for. But why not borrow the code for the network stack from someone else who already has that part tackled? Same with filters for MS Office. What is the big deal about KOffice, Abiword, and OpenOffice coming together and making some nice libs that translate .DOC into an XML format they can all interchange?
Simmilar things can be said about other softwares as well. Let's work together people! No need to re-invent the wheel!
In most of the cases you are right: the target audience should be Joe User and not developers, but I've made an interesting observation on a few web services where I've had the chance to have a look behind the scenes.
especially one of the systems was particularly interesting. the service they produced was one that is supposed to make life easier for you by providing a web interface to certain tasks that you would otherwise spend a lot of time on. the problem was that it didn't.
the user interface was so full special gags that it would only render reliably on IE 5 with the browser stretched to exactly the right proportions. for all other browsers it would be unreliable and often not work at all. also the service was incredibly sluggish because of all the UI sugar.
now, it turns out that the developers of the system are pretty heavy users of their own product. they spend several hours using it every day. since they too were annoyed with the utter crap that the web designer morons had slapped on top of their system, they made their own UI. and lo and behold: their UI was much better.
not only was it faster since it didn't make excessive use of graphics and UI sugar, but it worked in a wide range of browsers and for most sensible browser geometries -- plus, it had some new ways to handle certain very common operations with only one page load. big wins all around. it looked tidy and neat, and the fonts were actually readable on my screen.
why didn't they offer this UI to the customers? well, because in the opinion of the worthless people who are supposed to know about web design it didn't look "professional".
what I am trying to say is that when designing an UI, the people you should listen to are the people who use the application a lot. the case described above is not unique; I have seen this pattern many times before: developers make their own interfaces because the official interface is usually not practical.
but since we're talking about an installer here you would be absolutely right: most people will run across an OS installer very seldomly and it is absolutely critical that novice users are shielded from complexities. but it is also an important point that OS installers provide flexibility so users who know what they are doing have some choice as to which tasks they want to perform differently. (for instance I might have certain preferences for how I want my filesystems, but I might be less interested in having to answer 29365 questions about what software I want to install).
Libranet home page
This Debian distro seems to be growing rapidly in popularity. Good installer and thoroughly equipped with GUI-based administration tools.
The only complaint I have is that the CD does not boot with a SCSI-capable kernel.
I've read a number of good recomendations here from make it scalable to make it pointy-clicky to make it feature rich, etc. But it seems to me that anyone reading Linux User magazine would be a different set of users that those who might pick it up at best buy.
I think anyone reading the magazine probably has already tried Linux and have probably run it for awhile (perhaps not regularly, but at least as their desktop for a short amount of time). Do these people need the most user friendly installing?
I'm no expert at Linux either, but if I was a subscriber to the magazine, I would look for something different, not necessarily easy. I'm not afraid to try new and potentially difficult things, but I don't want the same crap over again.
I think Debian is a good choice, since it is different and very handy to use. If I'm joe-blow SuSE in Germany who never tried another distribution, I might be enticed by the CD that came with my magazine (1000 free hours of Linux!).
If I was currently a Potato user, I would probably be excited if my magazine came with a fresh copy of Woody when it's released. Then again, does the typical reader of a Linux mag really wait for a new copy of their OS to come with a magazine?
I dunno, I think y'all might be targetting the wrong user demographic.
Straightforwardness is not what the deb installer is. I had 7.1, it was awful, so awful I never finished the install and waited for my slack 8 CD. The M-68K version installs with pretty much the same installer as OBSD/68K which is a good installer. The only way to make a good text install is to have it be straightforward and simple to use, not to mention putting all options in one spot so the user can decide. Help and extended info should be avalible for all packages, and Slackware has failed in my opinion on one thing, SuSE lets you pick from all the packages at once, which is why I believe YaST is king. Debian should look at Slackware's and SuSE's setup tools for inspiration, as a blend between the two would be nearly as perfect as one can get.
Don't call my crazy, that's what they called me back in the home!
Debian is a uncommercial distri that is developed like free software. It's very stable and provides the better installation concept. Therefore Debian is quit strong in Europe.
The different RPM packages, that's a real problem.
Let's smash Red Hat and Gnome...
I use Debian, I absolutely love it. I don't think the installation process is that bad, but I should not fail to mention that I didn't get it right the first time I tried it.
Of course, it wasn't the first Linux distro that I installed. I started with Redhat 6.2. I got frusterated with that because I couldn't figure out how to compile a kernel in Redhat, and all of the docs I found that were Redhat specific said don't recompile your kernel unless you ABSOLUTELY know what you're doing.
Then I tried Mandrake, and after using that for awhile I managed to compile my first kernel, but I still wasn't completely happy because I found it hard to configure.
I tried Slackware, and oh did I like that. The config files in /etc were super easy to modify, but the package management system left a little to be desired.
This whole time I was learning, and becoming better at using a Linux OS. I was seeing the different types of packages out there, rpm, tgz, and the one I hadn't used yet, deb. So I had to give debian a try. Like I said, I didn't get it right the first time, although the installation is easier than Slackware. Once I did though, and I discovered apt-get, I was hooked. I now have three machines running woody and one running potato and I'll never switch to anything else.
It's not the distro for beginners. It doesn't have to be. It's a good distro, perhaps the best, and it's not for beginners. There is nothing wrong with that.
My Karma was at 49, then they switched to words. All that work for nothing!
As both a long time Linux enthusiast (1996, Slackware) and as an specialist in HCI I can say with confidence that Debian installation is the less user frienldy of all the Linux distros I have seen (I have tried RH, Suse, Mandrake, Slackware, Progeny, Corel, SmoothWall and maybe one or two more).
dselect is the antithesis of what user friendliness means, I congratulate the people in the Debian project for getting rid of it to hopefully use a piece of software more fitting of the best distribution currently available.
IANAL but write like a drunk one.
...but can I install it on my Dreamcast? ;)
Well, this is great to hear.
:), I would have tried a different class of installation, been given a more helpful error message...something. I can appreciate the concept of Debian being less-than-user-friendly. I can see how some people would like the inaccessibilty, to keep out the riff-raff. Maybe, on the basis of the fact that I couldn't properly work the installer, I am the riff-raff.
I built a new computer on Saturday, and I'd hoped to finally make the switch to Debian. Starting in 1994, I was a Slackware kind of guy. Somewhere in there I made the Red Hat transition. Starting about 8 months ago, I switched to Mandrake. Saturday, I was going to switch to Debian.
At least, that was the idea. The installer was less than descriptive. It failed to recognize my IBM Deskstar 40GB on a Promise RAID IDE controller -- both parts that are reportedly fine. At least, I think that it failed -- the error message was brief and undescriptive, without further recourse or details available. No problem, I thought, I'll do a net install. No such luck: it wouldn't recognize my 3Com Fast Etherlink. Not exactly a crazy off-brand of NIC. Not having any way to dump the terse error messages to a file, I did my best to memorize/scrawl the messages and Google for them, but that yielded no useful results.
With another installer (well, not Slack
But, hey, Mandrake sure does work nice on this shiny new system.
-Waldo Jaquith
The only people that think Debian is hard to install are people that can't read. The hardest part of it is the fdisk (easy to me). Other than that it is just hitting the enter key and your done.
The above is not worth reading.
Anyone who utters the phrase "user-friendly" ought to be required to define the word "user". Most of the stuff I've dealt with which was called "user-friendly" was actively hostile to the kind of user who's been herding computers for a quarter of a century and expects them to just do as they're told with no back-talk. I'm always asking vendors to make their software less user-friendly and more usable.
Let's hope that this doesn't lead Debian in the wrong direction.
We remember Storm Linux, right? While I do not use it today, I still find it the most user-friendly installer for Debian --- select usual preferences (keyboard, language, blah blah) and have the installed figure out the 'hardware stuff.'
With Storm long gone, would it not be legit to utilize the Storm installer for another product?
"There ought to be limits to freedom"
Its really a toss up between the 2, well installer wise anyway
IMHO you missed the point. As Debian is not a commercial Distribution there is no reason why it can't have an installer for beginners and one for developers. If someone is interested in writing and maintaining a it, it will be there -- and the PGI installer is already a quite advanced example of an alternet installer.
GUI's can allow end users to make better and more informed choices iff the person designing the GUI knows what the hell they are doing, which for most linux desktop developers is a pretty damned big "if". There seems to be an attitude of "If we simply make it into a GUI, it'll instantly be easy". If a set of GUI widgets are laid out in way that makes the relationship between the actions they perform extremely ambiguous, the GUI'ized interface will not be any better than the current command-line crap that proceeded it. In fact, it can be worse and mislead the user into making a very destructive choice that they thought that wouldn't be making. One of the biggest problems the linux interface design world is currently facing, as you pointed out, is the idea that "pretty == usable". All the anti-aliased text in the world won't ameliorate a menu selection with the caption "Save" that deletes everything on your hard drive.
I once talked to someone who designed a linux installer for a major linux distribution. This installer had many widgets laid out in a confusing and ambiguous manner, and in same cases widgets that are really meant for conveying one type of information were used to convey a totally different and wrong kind type information (for the widgets that were being used).
I mentioned some of these problems, and he didn't see what was the big deal. "You don't think it's pretty enough?" he asked me.
Ergonomica Auctorita Illico!
Am I the only one here who actually likes dselct?
People bash it for it's key commands, but they don't
seem to appreciate what it *does*. dselect is an incredibly useful and
powerful system maintenance tool.
If they had done there research, they would know that there is a new installer for debian all written, it will be included with the next release of debian, woody, which should happen real soon now(tm).
-- free as in swatantryam - not soujanyam.
Computer: "Press Any Key to continue."
Homer: "Press any key to continue. Well, where's the Any key? I see Esc, Ctrl, and PgUp, but no Any key! This complex computer hacking is making me thristy! I think i'll order a tab."
Homer presses Tab.
Homer: "Not now, the computer's starting up! Let's see. 'Vent Radioactive Gas?' N-O."
Homer types N-O.
Computer: "Venting prevents explosion."
Homer: "Well, give me a Y!"
Homer types Y. The computer beeps.
Homer: "Hey, I only need to type in Y! Hey Marge, I just tripled my productivity!"
I really don't understand what all of the hubbub is about the installer program. If it works, and gives you a workable system with a shell prompt, who cares how 'pretty' it is?
Granted, Debian's (current) installer isn't very user-friendly. However, it wasn't an issue for me, really. I've been through numerous RH installs (pre 6.2) and IIRC the installer wasn't much different from Debian's. RH6.2+ might have changed, but I've never used them, so I don't know.
But hell, I installed an OLD version of Debian - 2.0 - and apt'ed up to Sid without a hitch. The installer is only a very miniscule part of the picture. After the system is working, do you REALLY tell yourself, "Boy, that was a really froody installer," or do you amaze your friends and family with apt-get?
GIR: I'm going to sing the Doom song now. Doom doom doom doom doom doom de-doom doom doom doom doom doom doom...
http://ftp.us.debian.org/debian/dists/woody/main/d isks-i386/3.0.21-2002-03-19/images-1.44/bf2.4/ -- 2 floppies, network install, 10 minutes (depending on your bandwidth).
If you ask me, it can't get much better than that.
. . . take over your job, for which they are vastly overqualified. If someone is a "guru" of some sort of "esoteric-knowledge", that probably means they are smart in other ways as well.
Anyway, mod this down as flame bait. I don't mean it as such but package format is as much a religious issue as distro or editors or anything else so there's no hope of having a resionable discussion on it.
If this is true, then Linux will start faltering soon. One of the reasons that open source is so good is that generally its developers are its users. If there's a bug in it or something they don't like, it gets fixed quickly. Also, advice from users with no knowledge of HCI can often be more harmful than ignoring users altogether.
On the other hand you had some good and valid points: this does eventually need to be tested by the target audience, newbies. Feedback on what actually works, regardless of HCI theory, is also essential. However, eschewing the developers from the process will only alienate them and make the software worse, not better. I might also add that it probably helps the developer understand and relate to the user if the developer is a user.
Users are the ones that matter here.
Yes, but remember, with (good) open source, the developers are users too.
A story: Back way back when Debian was still using a 2.0.36 kernel (they did finally move to 2.2, right? ;) ), and I was a Linux virgin, my family decided they wanted to have all the networked computers in the house on the net at once. We were using an absolutely horrible piece of software called WinGate, may it die a thousand deaths, to do that and it just wasn't very satisfying. More to the point, I couldn't play Half-Life over it.
Fortunately, we had a spare 486 lying around. My dad mentioned he heard of something called Linux that could do what WinGate was(n't) doing, so I downloaded some Debian install floppy disks. I was somewhat computer-savvy, so I wasn't terribly upset at having to manually specify what was in my computer during the install, or play with partitions - I mean, that stuff you've gotta do in Windows, so what's the big deal about doing it with a console interface?
It installed and worked fine _the first time_. I didn't have X11 (that didn't really fit on the floppies), but I had been fine with DOS, so command lines were fine. I will grant you that I probably knew more about computers than the average joe at that point in time, but I really think that reasonably intelligent person who knows what's in their computer could install Debian with no problems back then. It's probably more true nowadays.
And, yes, I did make it into a router with a simple kernel compile and some patches. :) My GPA went straight to hell after that, too much Counter-Strike...
-Erwos
Plausible conjecture should not be misrepresented as proof positive.
You know what I'd like to see? A CD that you pop into a computer, boot the computer up, the CD comes up with a screen saying "Hi, do you want to install Linux on this computer (y/n)?" and when the user clicks (or presses, or whatever) "yes", it installs Linux.
That's it. No partition muddling. No tricky hardware questions. One question and you're done. Just do it.
I realize that Linux has a long way to go before it gets there, and it's hard to anticipate and handle every possible thing that might go wrong. But can you imagine having a Linux distro like this? Automatically resizing existing partitions, installing common software and setting up a boot loader to load Linux and whatever other OSes are on the machine? It can be done, it's just going to take a lot of work.
Everyone is complaining about that Debians installer is to hard. But there has been several attempts on creating easy installers for Debian such as Stormix, Corel, and Progeny. Now there is Xendros (former Corel) and Libranet. Libranet is an uptodate debian with an easy installer.
If you want them to continue and improve there work, support them!
While we're on the topic of improving Debian's installer, the only thing (and I do honestly mean the *only* reason) that kept me from installing it this weekend was the fact that there is no option for a journalled filesystem "out of the box." I don't care if it's ext3, JFS, XFS, ReiserFS, or whatever. And no, installing ext2 on one of two large partitions, placing the OS on the ext2 partition, recompiling the kernel, formatting the other partition with a journalled filesystem, moving all of the info from one partition to the other, editing GRUB's config, etc. does not count as "Debian supports journalling filesystems already."
If people are serious about making Debian userfriendly, we need to avoid things like this after a power outage (or bumped powercord):
Enter root password to run fsck:
%
Now what? I know, I'm sure many of you know, but what's a casual user to think of this? Add to this the fact that if they run fsck without the correct parameter, they'll be answering yes/no questions until they decide that it just isn't worth it and they install a different distribution or (quite likely) give up on Linux altogether and grab that Win2K CD.
I would love to help out here, but I have no real experience with Debian and its installer. I have some free time though and a programming background; If someone wants some help, reply to this post with some project info.
My $0.02
- I don't need to go outside, my CRT tan'll do me just fine.
So I must ask... 'ta heck are you talking about?!
You just need to grab the bf-something Woody install. It's the one with the 2.4.18 kernel. During the partitioning part of the install you select Linux extended. Then when configuring each partition, the installation asks you if you want it to be ext2 or ext3. Just select ext3. It's really that simple.
I almost kick myself everytime I read one of these "re-inventing the wheel with open source" tidbits. I'm not sure what your experience is but it doesn't sound like you have a foot in the free software community.
:).
Because it is a community and the community isn't hiring people from Universities with CS degrees, rather within the community people are learning to code. They are getting experience. Some of us are rather new to coding so you will see "Hello World" re-implemented thousands of times. You'll see hundreds of the most routine shell scripts. And you'll see dozens of IRC clients all from a different code base. Why?
Its part of the fun of computing. Honestly, it is less fun (IMO) to start from someone else's program than from starting from scratch--especially when someone else's program already has all the features you want. The beginning stages of a software project are probably the most exciting.
Of course there are other things. Like it is more difficult to grok a large code base than a smaller one. And sometimes more experienced coders pull tricks that newbies don't quite grok yet--so decide to use more simpler and apparent methods.
Free Software isn't going away and I think you're going to see a lot more of this. Programmers going through different stages of experience and writing software that demonstrates different levels of skill.
One thing I've noticed is that software is becomming more and more complex. We may see what I call generational programming. Basically, instead of one programmer understanding a code base or even an entire community understanding a code base we may get to the point where several generations are needed to understand and contribute to a code base (or it may be a conveniant excuse for the TUNES project
You do know that a lot of rpm-based distros have good tools that are under the GPL or other open licenses...maybe if Debian would consider using these or similar tools I'd think about giving it a try again at home...after all, there is apt for rpm now, so maybe there is hope that the different distros could play nicer with each other :).
I agree with what you're saying, and I understand fully that the Open Source community is a dynamic one that is largely based both on developers scratching an itch, and developers learning new skills. But the number of large-scale community driven projects that are there to serve a specific goal are growing rapidly. Look at OpenOffice, Mozilla, KOffice, Evolution, Apache for examples. All of these are fairly large, complex pieces of software. While some might have initially started out scratching an itch, they are now vital pieces of software we all depend on.
These are the kinds of projects that can and should benefit the most from code reuse and collaboration, but in my experience utilize it the least. Take my example about MSOffice filters. Or the fact that the Mozilla project wrote a whole new GUI system when several excellent cross platform toolkits already existed.
You missed the point.
Ask users what they want, because they will use the installer for installing the OS.
Ask developers what they want, because they will be writing code, modules, extensions, and keeping the installer up-to-date with newer hardware. If you don't take developers past, present, and future into consideration, then you will end up with an installer that noone wants to maintain, extend, tweak, polish, etc... In addition, hardware developers should also be consulted, because they will be making the hardware that is to be autodetected and configured. Companies that support Linux by providing specs, drivers, and such, should be companies whose hardware is correctly autodetected and autoconfigured by the installer.
Ignoring developers would be one of the most ignorant and just plain stupid things that you could do when designing a Linux installer. My point is that, in fact, the developers will be dealing with the installer far more than the users because they will be writing code or fabricating hardware, all of which must correctly and easily integrate with the installer.
Furhermore, asking users should mean asking and supporting different types of users: Joe Six-pack users, power users, and most importantly, system administrators and system integrators. How many MS Windows users installed their copy of windows? Yeah, very little, so why ask Joe Six-pack what he wants when he most likely won't ever install an OS? Its more important to ask the target groups that will be installing operating systems: OEMs (aka system integrators), system administrators, and power users.
Having an installer that can be used by any idiot off the street might seem like a good idea, but you will be ignoring your target audience.
Truely, for Linux to "make it on the desktop", requires that OEMs start shipping PDAs, laptops, desktop PCs, and workstations with Linux on them. Otherwise, you will be making a Linux-for-Dummies installer for a group of people who will never use it. You want Linux to end up on more desktops? Ask Dell, Compaq, and HP what they want from a Debian Linux installer. Also, note that an automated patch application utility such as apt-get makes support oh so much easier, as long as it is setup to run at 2AM every night or something like that. OEMs should think about that. No more workstations 3 service packs behind means far fewer people with rooted web servers, desktops, etc... (ala Code Red and Nimda).
Finally, some flamebait, but it must be said: Most HCI or CHI people are the wrost types of Software Engineers and Computer Scientists... even worse than those AI guys (most of whom don't even understand the basic limitations of formal, i.e. computational systems: incompleteness, undecidability, diagnalization). I mean, most things that the HCI research guys find out is common sense, and when its not, its most likely misleading, "dumb down the world" advice, like the parent post. Users can mean developers in one situation (OS installers, if OEM system integrators and sys-admins are considered developers) and other times it can mean Joe-sixpack (Web Browsers, IM, Media Players). But not so if you ask the HCI croud... to them, everything should be dumbed down. Any retard should be able to carry out any task. I am still waiting for the HCI croud to start writing papers on making open heart surgery more user-friendly and easier to carry out by Joe-sixpack. We have all seen what happens when you dumb down system administration. You dumb down the security of the internet. I mean, any idiot can admin a server right? Microsoft preaches so, but is it a good idea?
Choice, options, control... they should all be taken away from the user, no matter who that user is. For according to HCI types, we should pander to the lowest common denominator: Joe sixpack and Soccer Mom. They LOVE installing operating systems. Its part of or at least could and should be part of their everyday life.
While the HCI research community has a few nice apples, most are bad apples that just couldn't cut it in their Computational Theory 101 class or Algorithms 101 class. So they spend their entire life writing papers on common sense topics and idiotic topics. Whoever gives them funding is wasting their money.
What ever happened to the very effective Progeny installer ? Like the post above says, it has very good hardware detection and a useful graphical user interface. The last I heard about that installer, it was said that it would be absorbed into the main debian distribution. Does anyone know what has happened to it ?
No real problems, though, other than the fact that it takes forever to load dselect.
fnord
For the install process, XWindows is overkill. Use Picogui instead - it is designed to run on everything, it is really cool looking, and should run lighter than an X installer.
Why not using Conectiva Linux's Modular Instaler (MI) ? http://distro.conectiva.com.br
Conectiva has one of the best GPL instalers.
It is:
- modular, so you can easily create new modules,
- light (the ncurses version needs less than 8MB of RAM)
- apt centric, it should be really easy to adapt it to use deb instead of
rpm. and apt can be the hook here.
Fuck that!
I think the point is, you should try to find some well documented free code to possibly use in your project before attempting something that has already been done. A good coder uses resourcefulness as well as innovation to produce excellent work. Remember, code can be taken out or trimmed down to get better results.
"And even experts shouldn't have to do stupid things like manually set up network cards when they can be easily auto-detected. "
True, however one shouldn't forget a windows lesson.
You know the one were it auto-detects a piece of hardware. Says it's a particular brand which you know is untrue. Spend the rest of the day fighting with windows trying to convince it otherwise. Do we want to bring that kind of "auto-detect" to Linux?
" The thing about Debian vs. Red Hat is that Debian's package-management KICKS ASS. "
So isn't this more the reason to fix RPM than it is to "fix" Debian's installer?
"as for the dep problem, you would still need the same dependencies on another distribution so IMO that arguement is invalid"
The difference between dependencies and Debian vs RPM and others is who handles the dependencies. You or the computer? IMHO I'd rather have the computer handle it with the possability for me to overide were needed.
Debian is one of the few distros that is nearly so portable and consistent as NetBSD (my other favorite distro).
;-)
Remember, some of us (particularly Alpha and SPARC users) do _NOT_ have big, nice X and/or VGA graphics hardware. We use DEC VT and WYSE terminals on our servers and even some of our workstations
Anything that would detract from the consistent multiplatform installer to appease the x86 PC and PPC Mac users would cause me (and other Alpha/SPARC developers) to switch to NetBSD full-time.
PA
I lives in a country where the moslems are majority, and they will kill you if you utter anything close to "piggy".
I am afraid http://hackers.progeny.com/pgi/ may not be kosher for the moslems.
To the moslems, Osama Bin Laden is kosher, but "piggy" is not.
Blowing up others is kosher to the moslems, but when others retaliate, it's not.
Muchas Gracias, Señor Edward Snowden !
I started off with redhat. I nearly gave up linux altogether because of it. Now I use exclusivley slackware & bsd. heres why:
/etc/ppp. Redhat has conf files in /etc/sysconfig/networks/local/hidden/special/ut ter-stupid-deep/path/can/u/find/it/yet?/ppp. need i say more.
:) but then i thought the same about BSD and its by far the best server O.S. I've ever used.
1) unfamiliarity. redhat has things laid out like no other distribution so despite me rtfm'ing like mad I couldn't learn anything about it.
2) inconsistancy between work & home. Due to the excessive security problems my work won't touch redhat with a bargepole.
3)There is lots of software as RPM packages. unfortunately netcat.rpm was the only one that ever worked due to dependancy problems. Now I use the source, because dependancies are listed in the README and once you install those (from source) you do NOT get any problems.
4)Utter Shit Documentation. Net-HOWTO says look in
5) I'll agree with your bit about debian users. They do seem a bit dauntingly like a cult from the outside
It seems that, in the push for "Linux on the desktop", everyone assumes that every distribution should be aimed at the "newbie" Linux user. From my perspective, this is the wrong approach.
The strength of the distribution approach is that each can be a tool that is used in different situations. I can use Mandrake or RedHat for my desktop and Slack or Debian for my servers. That puts the responsibility for user friendliness on people who WANT to specialize in it, while putting the responsibility for speed/performance/stability/etc on those who want to deal with those issues.
It's all a matter of what the developers decide is important to their retaining user base. After all, if there's one thing we should learn from Microsoft, it's that it's very difficult to be all things to all people, and a product that tries to do everything does very little well. That's no way to win popularity.
I'll use the age-old cliche of my toolbox. I know that to tighten a bolt, I use a wrench. I know that I need a hammer to handle a nail. In the same respect, I know that Mandrake and RedHat have put together very newbie-friendly distributions. On the other hand, I know that if I need a minimalist or tightly controlled install, I know that I can turn to Debian or Slackware.
I guess all I'm saying is that it's a very difficult thing to try to go from expert-oriented to newbie-oriented, and it's not necessarily a step that SHOULD be taken in all regards. Consider your user base and type of tool they consider your product to be, whether you'll lose the ability to use it for that purpose (and therefore lose that user base), and whether the risk and sacrifice is really a good idea overall. Consider that there is a lot of time and money ahead of you in the new market you'll be joining, and that you'll be sacrificing all of the time you've already spent getting to where you are in the market you're now leaving.
Maybe the secret to world domination is diversification. Windows' approach is unification, which gives them an AWEFUL large set of tasks to master in a single product. Things that you and I know should never really go together - like solitaire on a router. Pick a market and dominate it. If the distributions all pick different markets and dominate them, then the umbrella - Linux - will dominate all markets.
Hmmm. I think that the average user has a tendency to want to browse to see what is available, which is best served by a dselect-like program. For people wanting to set up server boxes or boxes for a particular task (CPU farms, storage servers, etc), your approach is probably preferable, though.
My point is that Jo Shmoe and Linux Guru are two different people. There are huge numbers of window managers to choose from, to suit different kinds of user - why can't there be different levels of package manager?
I have been patient. I have been persistent. I come bearing the news to Debian webmasters everywhere that the "bf-something Woody install" is not obvious. Not only in name obscurity when a Debian newbie would only know to look for 2.2 or 3.0 disc images, but also in placement on the website.
I have gone searching in vain for this bf-something install. I have looked in all of the obvious places on the website under such topics as "Getting Debian," "Debian on CD," and ""Download with FTP." This is bullshit. If this is everyone's definition of publically available, I must have missed that day of class. I even download some of those potentially nifty netinstall CD images in the hopes that they simply weren't labelled correctly with the magical bf- prefix.
Believe me, I have gone through a lot more effort than most casual visitors to the Debian site would have gone through. Unfortunately this is one area that Debian could learn from RedHat, Mandrake, SuSe, et al in that the others provide an iso image, you download it, you burn it, and off you go. If the newer install with the updated kernel works so well, why hasn't the old installer been mothballed? Why would the old installer be offered? If the new installer has problems that preclude its replacement of the old installer, then the appropriate answer to my previous post would have been "they're working on it and it should be ready when 3.0 is released."
By all means, prove me wrong. By all means, show me an obvious link that demonstrates me to be a dullard who cannot read a web page. I am not above humility. Otherwise I will assume that a clean and complete Debian install is bullshit, must first be excavated by a Debian veteran who knows how to find it, and/or is of no use to the general public. Debian may be a great distribution, but that's pointless if most people can't install easily without sacrificing popular features (like journalled filesystems) or hunting through mailing list archives without really knowing for what they search.
- I don't need to go outside, my CRT tan'll do me just fine.