Slashdot Mirror
Connell Replies to "Grok" Comments
Charles Connell writes "Before heading home for the weekend (and tearing myself away from Slashdot), I thought I would respond to some of the comments made about my recent Linux article.
First, thanks to everyone who posted or e-mailed a response. The vast majority of comments were friendly and thoughtful. I especially liked the reader who wanted to paste the article in every programmer's cubicle. *grin* More substantially, here are some specific responses to major points raised by Slashdot readers.
"
I'm a relatively bright guy with a learning disability. I've been trying to learn to use Linux for the PAST TWO YEARS... and I'm still, PAINFULLY, in the "newbie" category.
Why? Because while it might be worth taking the time to learn everything, that time DOES NOT EXIST for me.
By day, I am a technical writer who must devote his time to his job. By night, I have a wife, compose and distribute music, do a comic strip on the net, and write for another web magazine.
Recently I tried to use Linux as a Samba file/print server for my OS/2-Linux-Win98 machine. I eventually gave up, because while I was making progress, IT WAS TAKING TOO LONG and all of the other things in my life were suffering as a result.
Because I was trying to use the Linux file/print server ALSO as an internet gateway (so I could take out the modem on my client and save an IRQ) I couldn't connect to the net, couldn't publish my strip. I spent four days straight trying to configure this, tweak that -- my wife was pissed. And let's not forget exactly how painfully difficult it is to find the documentation you're looking for on how to configure Samba, then throw in the fact that it's written with the expectation that the reader already has some knowledge of networking (I didn't).
I had plenty of Linux people willing to give me advice, but none of them were right in front of me. It was all very frustrating, and I gave up because I was very, very far behind in the rest of my work and I had to get cracking.
I personally feel that Linux is an OS worth learning... but I only have so much time I can devote to it, and I can't afford to devote more. Until Linux can be presented in an "easier" format, or until I get significantly more free time, I'm afraid I'll be an eternal newbie.
And this is from someone whose primary job is technology based. I don't want a Doctor taking time to learn Linux that he or she should be using peforming a triple bypass or something like that. And I'm pretty sure someone who just wants to get their checking account balanced in time to do taxes won't want to spend a month getting up to speed on using Linux, either.
In other words, yes, it might be worth the time, but is the time THERE?
Eviscerati.Org: All Hail the Eviscerati
I'm not sure I understand the Linux "is hard to install" critique -- even when I was complete Unix/Linux novice. I've never thought it was hard to install nor did I ever have it fail except when using junk hardware -- and by junk, I mean hardware that I couldn't get to work with DOS but didn't have the heart to throw in the trash. Success my first time around on a misconfigured EISA 486-33, with weird, OEM SCSI adapters from floppies, and later FTP, NFS and SMB-based installations from home-made CDs and downloaded distros on all kinds of hardware. Current distros installed from CD are so simple my dad could do it, and he thinks sending mail from AOL is confusing.
I think what screws people up when installing is not that installation process itself, but the desire to "customize" installations and then later growing frustrated when you try to do something that you don't have the right packages installed for.
More generally what I think makes Linux hard is that Linux isn't a complete system, it's a kernel and a few utilities with a lot of ancillary packages added on. There's a lack of coherency to the complete picture, coupled with the naturally high learning curve of Unix generally, that makes it so maddening.
It's not so much hard to install as it is get configured afterwards.
This is what I totally agree with. I was writing a response to his previous article along the same lines, but unfortunately netscape then crashed .. Anyway, as there were many very good and valid points in his article, this is one he still misses - he still doesn't grok Linux.
Linux as a community is not a single entity with only one single goal and single thought. The community - if I may paraphrase ESR - speaks with a multitude of voices, each with individual needs and ideas. There are people who don't want Linux for the masses, but on the other hand , there are people who do - and I dare to suggest that the latter are more populous.
This is also has an impact on the humility note he had on his previous article, that is, if we don't watch out, Linux will be toppled. Well, there is no we to speak of. There are a lot of people who are already happy with Linux (me included), but there are also a lot of developers and companies who are making Linux go forward very fast. Linux was originally intended as a low-end unix server for i386 (a blatant over-simplification, I know), but it since warped itself to a multitude of architectures and a high variety of purposes, morphing (a shameless Transmeta plug ;) itself to embedded computers, PDAs and so on. So, there is no single goal.
So, I would claim that based on the unique and diverse abilities of the Linux community, Linux cannot only not fail, it also cannot be killed.
Look again at your numbers. According to you, only 60 million people (1% of 6billion) even know what a computer is. The population of The United States is 250 million, Japan has 120 million people, and Germany has 90 million.
.01% of the people in the world know how to use Linux, or only 600,000 people. About twice the population of Des Moines, Iowa.
.1%, 90% dont know how linux works.
And what do you mean by 'know how a computer works'? You mean like knowing how to use one? How to program one? How to build one out of transistors? You only need to know how to use a computer in order to use Linux. I would think that more then 6 million people know how to use a computer. I'd put that number at a several hundred million at least (The majority populations of the US, Western Europe, Japan and Taiwan. Plus large numbers in India and Eastern Europe)
Whats the problem?
The problem is in the idea that only 6 million people even know how to use a computer. By the way, if you look back at your post, you're saying that only
The percentage of the population that is illiterate and lives in mud huts is probably around 90%
(literate: 600million)
Of the remaining 10%, 90% dont know what a computer is.
(Know what a computer is: 60million)
Of the remaining 1%, 90% dont know how a computer works.
(Know how a computer works: 6million)
Of the remaining
(Know how linux works: 600,000)
[ c h a d o k e r e ]
ReadThe ReflectionEngine, a cyberpunk style n
To address the comment about people not wanting Linux to be used by the masses. I can imagine the type of induhviduals that would say something like that and presume to talk for the Linux community. I would guess that they are just some nerd who feels cool because he can use linux and be on the edge, whatever. I can understand the contempt strong technical users have for ordinary and clueless users. I however think of it a little differently.
I think there are different types of people who require different things from a computer and therefor from the OS. I generally put people into 3 groups. Those that just need a computer to word process, check email, browse the web, and a few other minor tasks. They probably aren't very computer litterate and dont have patience for it. The second group of users is also focused on using the computer as a tool. They would be power users, the kind that actually learn to use the the shortcuts and advanced features in programs even if they dont care about how the computer works. This second group would probably want more control over what they can customize and would use addition applications such as high end games and graphics programs. The third class of users would be the experts that like to be able to customize everything. They can appreciate how GUI makes things easier but aren't afraid to use a console at times when the text is more important than the pretty decorations around it. They enjoy learning how things function on all levels and want almost complete control. They would use another complex superset of applications including compilers and server-type apps.
This is where I get to the bit about why someone would say they don't want the masses using linux. The 3 classes of users require different things from an OS, so that if you made a linux that the first class of users could use, the third likely wouldn't want to, so actually I believe those people saying I dont dont want class 1 users causing *my* linux to get watered down. To quote Murphy's Laws on Technology, "Make a computer that even an fool can use and only a fool will want to use it."
I've typically used Mac, Window, *nix as a metaphor for the three types of users. This isn't an entirely accurate picture of things, but it isnt so off. Mac is a very simple OS, doesn't give you much low level access but does a lot of stuff for you easily and quietly. Windows allows more control through more setting in the control panel and the registry if you dare, and generally the programs are more feature rich (or bloated depending on your point of view). Linux is definately difficult but I enjoy putting up with the difficulty because I am rewarded for my time and effort by the level of control I get and by how much I learn. However there are things that make Linux better than windows and macOS other than level of control (reliability perhaps?). I also think that there are plenty of improvements to linux that experts could benefit from. I definately think that regular users need a better OS because teaching them that software is supposed to be buggy and your OS is supposed to crash once a day will only serve to make them more computer wary.
I think a linux for Joe Computer-User is possible and still retaining the power expert users have because of the modularity of linux. I dont think that the current developement environment is capable of doing that. It is basically just a set experts programming for experts. The great thing about linux for me is that it has all the features I want and none that I don't (well not really but in theory...). This is the strength of the community developement environment, one not driven by requests from marketing for new features, be they frivolous, gimicks, or just plain useless. The people get what the people want. What would have to happen to the way linux is developed for an OS with a user scalable level of complexity/control? I dont know. But defining the problem is step 1.
PS. The note about require a linux expert friend to help you configure it, this is something that while I admit wouldn't work for a mainstream linux is important in an expert linux. Essential the arrangement is a mentorship. That is the way it worked with me, a friend told me about linux, helped me install it, and taught me some unix basics real quick. Whenver I had a problem I'd call him and just about the time he started getting sick of me I started to know enough to find the answers to my own problems. Now I have a couple of friend who turn to me for linux help and I'm happy to give it.
Ben "JonKatz" Buchwald
Oh yeah? This is probably true, if population == all the people on Earth.
If the population means all the people currently using computers, I'd say it's more like forty per cent, even more.
Of course, most people wouldn't even try to install Linux, just as they wouldn't install Win98 or NT. But using a preconfigured Red Hat 6x, Suse 6x, Caldera or Corel distribution is in no way more difficult than using Windows. Assuming, of course, that the system boots up to a graphical login, but that's assumed of the Windowses also.
If up comes Gnome or KDE, a user most likely clicks on the menus a few times and soon is happily browsing the web or typing a letter or whatever he or she does with a computer.
If you survive in NT environment, you'll be able to use Linux. Or just aboot any other Unix with a decent desktop. Many Windows users don't understand anything about the Windows GUI, but are using it anyway. The problem for both home and corporate environments is the lack of applications and the learning curve in existing software.
If the free software community replicated MS Office as faithfully as they did Minesweeper and Solitaire, even the most clueless people would now be buying cheap computers with Linux preinstalled. And for a hudred bucks less price.
So what about the benefit to the community? We all do our typing in Emacs or sometimes StarOffice. Why should we care about having more people use Linux?
For me, the answer is simple. Drivers and games. I would buy a DVD drive and an mpeg card right away, if it had Linux drivers. I'm not going to software decode movies. I'd buy a good fast USB scanner and a digital camera, if I'd know I could use them with Linux. I'd probably even buy games, if I could play them on Linux. All this would be so much better, were there only 25 per cent or so Linux users.
NOSPAM@REMOVETHIS.NO.SPAM - you'll find the real address somewhere
Why should we make the average user install Linux at all. Computers for average users should ship like appliances (as most windows computers already do). That is to say that joe simple user should never have to run the installation program. It should be installed on his/her computer at the OEM, along with the basic office style apps. I mean can you imagine if you bought a VCR and someone said "OK now you have to install the OS before you can play a tape."
Sooner or later soemone at a major computer maker is going to figure out that they can build a very slick GNOME startup screen, license Star Office or Applixware, drop most of the workstation and server packages out of Linux, and distribute a slick internet enabled PC without paying microsoft. My bet is that Sony comes along with this in about 6-9 months.
>>I'm one of the ones who wonders if we really want one click installs and easy setup.>>
From your complaints(below) it seems that you choose to avoid easy installers in favor of greater control and precision. I commend you for that. I sit firmly in that camp myself, but I can't expect everyone to feel the same way.
>>Seems every dist that has gone that route causes me to spend hours afterwords cleaning up and optimizing the "simple install." I'd rather spend more time to begin with than have to deal with "expert" installs such as Mandrake 7.0 or Corel that sure don't seem to offer "expert" level choices.>>
I'm with you all the way when it comes to my own choices. However, since we have other distros that meet our needs, why criticize 'friendly' distros for trying to meet the needs of others?
I want linux to become accessible to a broad range of computer users. I want to see linux succeed in the broad market. Because I believe that it can. The easy-to-use variants are necessary to achieve this goal. Are they there yet? No, just as Mr. Connell points out, there's still a great deal of work to be done. The question I must ask is: why should they stop?
When you believe your opinion is absolutely correct, pause to identify the logical fallacy inherent in that belief.
>>Linux was designed by geeks for geeks...>>
This line of thought establishes a false dichotomy: that linux cannot be for geeks and end-users. The truth of the matter is that it can. The two are not mutually exclusive in any way.
Linux has the flexibility to be 'by geeks for users' while remaining 'by geeks for geeks.' It seems to me a waste of that potential to not pursue the potential for linux as a desktop alternative. And a bit snobbish.
>>I don't think our objective should be "Total World Domination." After all, aren't we all rather upset at Microsoft for having the same goal?>>
I'm not upset with them for having that goal; I'm upset with the way they choose to pursue it. I think many others feel the similarly.
When you believe your opinion is absolutely correct, pause to identify the logical fallacy inherent in that belief.
I didn't read the original article the first time the Grok article was posted, but now I checked the second article, and it looked like this:
I heard the Linux hype probably reading most mass media, and thought it was probably not worth it for most users to install, so I went in with this opinion to go and try Linux.
I create a scorecard, which makes it more a game of numbers, even though there really is no reason. It is mostly a matter of opinion. In addition, this is compared to a *perfect* OS with full score that doesn't really exist?
I install RedHat... probably knowing Unix find out that some things aren't recognized properly. Complain about it..etc. No comparison to installation of Windows. Note that if you are installing on a bare drive, you would have to partition for Windows as well.
I add a stupid category that I cannot test with my period of time. Give it a rating just by word of mouth. Fail to notice that almost all Linux apps cannot crash the OS? Comparison to Windows?
Again, probably having exposure to Unix? I laud Linux for the obvious fact that it is command based and isn't wrapped up in a nice GUI.
Make another category that I do not test, note anecdotal evidence, pick a number, give a score. Did you _try_ LyX? Did you _try_ StarOffice, which is quite similar to MS Office? WordPerfect?!
Another thing he says in his reply is that changing office suites is hard. Well, that automatically means anything non-MS has a permanent defect?! Give me a break.
This guy knew the "flaws" of Linux, probably from mass media, went in with a bias, and just took a cheap shot. He used this "test" and "scorecard" just to move his own opinion that he started with from the beginning.
This is dumb. Linux IS ready, if you are willing to learn. Get a book, read the manual, be willing to play around. No one said it didn't take any effort. It is not that hard. Wow, its a command line, so intimidating. Most people used DOS before Windows was invented, and the Linux command line is a lot friendlier(tab completion anyone?). You choose to install Linux, and if you do, you should choose to learn it. I don't think the Linux community was over-reacting, this is the kind of FUD (yes I don't use the term loseley) that comes straight from MS's Linux myths page.
"Oh you can use Linux, but you might have to deal with new and unusual applications. You will also have to deal with a command line that I, a computer expert cannot figure out." fear, uncertainty, doubt?
Although, you have to give this guy credit for replying, unlike most of the writers who just outright bash the flamers and make them an example of the whole Linux community. The thing is though, he is still doing what he did before, picking things out to further his opinion. I don't think he really even admitted any faults. He's like yeah this but blah blah.
The installation is the main bump in the road, after that it is pretty smooth driving. The fact is, if there is a bug in Linux, YOU can fix it. You can pay someone to fix it. Can you do that with MS? They have gotten quicker at responding to bugs, but mainly because of the pressure from Linux. Plus, with open-source, you don't get burned badly, like when Microsoft decided to abandon NT for Alpha.
Linux is what you make it. The fact is, it is more your OS. You can do whatever you want to the internals, mess around totally. No longer are you dependent on MS to do everything. You can choose a different look for your system, you don't need to install things you don't want to. This user empowerment is an important thing, that many people fail to realize. It gives you more power to let you tell the computer what you want to do and make it the way you want it. Some people want to hide that power behind a bunch of dialog boxes, but why?
"To use Linux or not" is a question of this: "You want to trade some hand holding and a bit of work for an OS that is yours to use freely and in whatever way you choose?"
Also, when the CTO of a Fortune 100 company is deciding whether to commit to Linux for 50,000 new computers, he (or she) is certainly going to be demanding.They might be a little uncertain where to direct their demands, but the overall ability of the Linux community to meet her needs is going to influence that purchase decision.
Again, I think this is a mistaken point of view. Don't expect the Linux community to meet your needs - become part of the Linux community, and do what you need to do. This could mean either having your programmers work on features, or throwing money at someone who can. However, expecting "the community to meet your needs" would be disastrous. If this fundamental misunderstanding isn't straightened out, Linux will be a miserable failure in the business environment.
----
In some sense, i have to agree, linux, as it is today, i would not ask my mother to use. And Mom used to code in pascal on macos.
I used to be an OS/2 user, and you might have even called me an OS/2 pundit, but, the mantra i followed in OS/2 advocacy is "OS/2 is Not for Everyone".
people are best served by the tool that is best suited to both the task at hand and their ability to use it.
For many things, OS/2 was, and in some cases, still is, the best tool for the job, and simple enough for most any geek to use.
For many things, Linux is, and will continue to be, the best tool for many jobs, if you have people around who are capable of using it.
for most end-user tasks, the client variant of Windows or MacOS will continue to be the best tool for many. This is unfortunate in several senses, but it isn't the end of the world.
I don't belong to a LUG, and one of the reasons i don't is the same reason i never joined an OS/2 user's group: I don't advocate convincing someone that they should use a tool other than that which is best fitted to them is the tool they should use, and user's groups do things like "install-fests"
when people ask me if they should run linux, i don't automatically say yes, i ask "why do you want to use linux?"
if they don't have a good answer, i don't offer to help them. if they have a good answer, i help them find the right questions to ask, and then help them figure out how to answer them for themselves. I find that's a lot more effective, in the long run, than simply telling them the answer.
This is just like television, only you can see much further.
What you're neglecting to consider, Charles, is that Linux cannot "fail". I do not care whether any interesting applications are *ever* ported to Linux. I was using Linux before it got buzz; I will continue to use it after it loses buzz. So will enough developers that it will continue to be useful to me.
You're thinking of Linux as if it were a business, with venture capitalists, looking to either take it public, sell it to someone, or else shut it down.
-russ
Don't piss off The Angry Economist
As Malcontent said, a significant slice of marketshare is all that is necessary to get drivers. Games I can personally live without, though they're a would-be-nice.
But beyond that what I hate most about Microsoft's system is it's one-size-fits-all nature. Whether you're a professional programmer or a first-time user you get the same wizard happy interface. It's important to me that Linux remain more than just a cheaper and more reliable Windows. I want it to be remain a system that rewards learning with power. And one that does not *force* you to sacrifice performance to interface glitz, or creativity to simplicity.
It's fine with me if people want to make Linux friendly and usable, and the more folks using it the better. But mass acceptance isn't worth the cost of changing the development model or design philosophy that makes Linux worth switching to.
This is not said out of elitism, it is said in the hope that the software market will become a more healthy ecology. I'd like to see programs and systems filling niches rather than bloating up in an attempt to dominate and kill off all other products.