Selling Linux to the Masses

[br073n]

I've been doing some informal polling, lots of research and some writing on just what it would take to improve the mass adoption of Linux. I won't pretend I'm any sort of Linux Messiah, but I believe I'm getting some traction.

I found some reinforcement in the previous thread by gankhuu about how it felt to deal with the differences. First, I'm thrilled this user found a sympathetic ear here. I can't count how often I've seen something altogether different, particularly in my early days with Linux some 15 years ago. We've come a long way in polishing the image of the Linux community. I've written often about that in articles and on many forums. Linux is far better and easier to use in itself, but the people make all the difference in the world.

In my mind, people are the reason for computers. While plenty of us come dangerously close to a religion in our devotion to these tools we love, the vast majority of those who own and use computers do not see it as a hobby. Even if it's essential to their routine, I'd say the majority of people I've helped in my computer ministry don't share 10% of my enthusiasm for exploring things I'll probably never really understand. If I can't make these things serve their needs, I might as well go away and bug someone else. It's like preaching in a graveyard, as I'm sure you know if you've ever advocated Linux to others.

Indeed, I can't continue to look at Linux from the hobby mindset if I expect to help anyone who isn't a hobbyist. Being able to jump from one distro, and among releases of that distro, and to BSD, and back again, etc. -- that puts me in a different world from all my clients. They don't have time or energy. It has to work pretty much "out of the box" or it goes back to the store. Over the last few years, and especially the last couple of months, I've tried to get outside the hobby-head and find out whether there is any way to make Linux come close to what they experienced with Windows all this time, but make it better.

On the way, I wrote a bunch of articles for my ezine (OfB.biz) and even more on my blog. As noted here, I've finally arrived at a solution based on what I honestly believe I can "sell." Perhaps you'll catch the hint I'm settling in for a long haul as if there's no escape. I have to quit giving myself any escape hatches, because common users won't go there. I have to experience it from their perspective.

At the same time, I have a couple of friends (developers) who are plunging into a completely new distro aimed at that very audience. I'm not sure it will get anywhere, but they find it worth researching. With only a couple of guys involved, I figure it will be a long time coming, and I have to have something I can use in the meantime. I'm hoping I can stir some interest and support here, not simply for specific issues, but with an empathy for the vision. I know I've certainly found here some folks with a fine attitude about helping others.

[Magic31]

Wow Ed.. That is quite a story you are making us read.

I like the way you're thinking on this as also how you try to find the right middle ground and compromise from an end users point of view.

I do think you have a good chance there.. your idea of finding one right formula that can be sold as a total product is what end users will like about it.

I personally think openSUSE is a good choice. Especially if you back up openSUSE with good knowledge and support that the end user can turn to. That's the thing making it harder for starters to find their way though it... it can be a hard long learning curve.

This was also true once for people having to use Windows.. It's a common use thing now, Windows, and many safeguards have been built in to keep things runing and stuff added to mask the bad and dangerous bits.
Anyway, enough of my years have been filled with frustrated users calling me because the PC (Windows) won't do what they want them to.
Linux might be facing a similar thing.

During it's growing pains, MS had a couple of things going for them.

A) the market was still fresh... sure Apple was a big brand name, but Windows was easy to get by and inexpensive (also read the words illegal installs here )

B) With the OEM bundling and support, systems c(a)(o)me preinstalled.. ready for work and you only have to run a couple of setups to get going.

We've come to a point where 90 percent of the people just want to be able to Internet, have multimedia, do standard type processing.
The OS is less important as (almost) all consumer OS'es are able to deliver this.

So.. If you can find a/the right formula that fits
'cheep & easy handling' along with 'great -affordable- support' .. then you will have something holding much potential.

If this desktop can also handle all the common formats thats being thrown at it.. i think you should have a winner.

Make enough waves and others will join you.

Ps. I'm also wondering... what would be the strong points of Linux over Windows? Is that enough to convince an enduser comfortably tucked into Windows?

[br073n]

Ps. I'm also wondering... what would be the strong points of Linux over Windows? Is that enough to convince an enduser comfortably tucked into Windows?

To be honest, I'm not sure there's a straight line to that goal. Going back into history, Windows came to the fore by first gaining control of the business market. People used computers at work, found them useful, and wanted one for home use. Familiarity with DOS/Windows at work meant it got installed at home.

Now that the issue is one of "comes with the hardware," at least part of our hope is the Linux inroads into the business market. Another item is the degree to which Windows fails. That's the sad part, because I don't rejoice in human suffering, but every time someone can't get Windows to work for them, that's one more vote for Linux. But even then, we have no case if what we offer isn't a whole lot better.

You and I are already convinced, and to us it's a silly question. Whiners, the truly lazy, will never be happy. They make the most noise, but don't represent the majority. Sadly, we get suckered into helping whiners by convincing them to try Linux, only to have them demand their Windows back next week. That a few of them have legitimate gripes does not help.

The bulk of my first approach to this issue was calling for the Linux community to slow the race to new features, at least in some areas. A primary example is the big hiccup we are experiencing over the "new and improved GVFS" -- it'll be a long time before that quits raising alarms. It's fine to innovate, but the constant "never quite finished and working properly" flavor of the entire Open Source field of endeavor is a real put-off.

The question I have to ask is: Do we not yet have something solid, something so useful and effective we can afford to devote some time and resources to maintaining it for a couple of years? Can we not find a waypoint worthy of support and put some effort into supporting it with backward compatibility and back-porting to it our new toys? I find the mere mention of backward compatibility is like heresy in Open Source.

Example of the good: I can get a current copy of Seamonkey to run on SUSE 7.3. Example of the bad: I can't get a single modern distro to run WP8 for Linux. People in the Windows world can appreciate the former, but they just don't understand how they can run WP8 for Windows on XP, which in the minds of many Win-users is still "current." WP8 for Linux is in our Dark Ages, just barely functional on anything after the introduction of XFree86 4.0. Yet I am not the only one who thinks OO.org hardly does word processing any better.

I think we abandon good stuff too quickly, and common users don't think they can trust Linux -- whatever "Linux" means to them -- to support them tomorrow. Assume for a minute we have a die-hard running SUSE 7.3. It's not too risky for the home user in terms of security threats, but would there be anyone here who even remembers how to work with it? I thought it was one of the better SUSE releases, and I realize there's no money in supporting something that old with such a tiny user base, but for someone who operates in the realm of "it came with the hardware," there's not even a volunteer support base for them. It's not a lack of knowledge, but a complete lack of interest at the most fundamental level of how we think.

Please forgive me if it seems I open old wounds, or a can of worms. I'm just thinking outloud.

[Magic31]

Originally Posted by br073n

Please forgive me if it seems I open old wounds, or a can of worms. I'm just thinking outloud.

I think many can relate to what you are saying.
As it is a good idea.. in my opinion what you are expressing is an ideal and a goal in itself.
In this I can't speak for the community, but only give my view.

With the current openSUSE project I think that what you are trying to fit in is a difficult task at the least.

There have been suggestions of creating a release or fork (whatever one wants to call it) that is more suited for ' the simple' end user.

In my view Linux has had a lot of catching up to do to be able to present itself as a full and worthy desktop (vs Windows and OSX). The last four years have been crucial in this goal... and with it I think the pressure and instability on some points followed. The relation openSUSE has to the SLEx products also plays a role.

As I'm quite new to Linux I say this with caution, and again, this is how I'm viewing the developments over the last 5 years, as others will have different views.

The notion that developing should (an can) slow down a bit to bring back the stability and polish is one I share. But to truly make openSUSE the 'fit for all' .. I do think a second version is needed.
I would rather see the ' cutting edge ' version being re-released once the biggest bugs and annoyances have been solved (3 to 6 months) and see a longer support cycle for the re-released version. That could then be the stable base.

In short of what you saying: It's more efficient to to make the application fit the OS rather than having the OS fit the application. Agreed on that!
openSUSE build service is a brilliant tool that I think can be utilized for this. But there is need of more standardization for this to work over a greater scale. I for one would not mind seeing some consolidation of all the serious distros , projects, etc.. bring it back to the holy 10 or something

As another way to approach the solution to your goal:
Have you seen the possibilities of Virtual Applications (VMWare Thinapp / ZenWorks Virtual applications / etc )? I think this is a very promising approach to getting applications working on many different platforms.

Creating a new version of the openSUSE distribution does seem like a good idea. But it calls for more resources and more helping hands to make it to a success.

Another stable alternative for the 'fit all' would be SLED. but then you are steering into the enterprise world... Witch I don't think is bad.. but maybe very different from what you and others would want. Still.. SLED is OSS for the most part!

I'll stop rambling now

[Axeia]

I don't think not being able to run 'old' software on a new operating system is something (ex-)windows users would be too suprised about, there is plenty of software that ran on win95/98/me but wouldn't on XP, and Vista has it's own share of incompatible software as well.

I agree with you that openOffice is probably a whole lot more than most users need, I however don't think that the way to go is supporting old software.

I think these users are simply not provided with the right software for what they're trying to archieve. They should be using software better suited for the task.
OpenOffice is aimed at a target audience that wants a 'full' office suite on par with microsoft office.

If someone doesn't need more than WP8 then that someone should be using a wordprocessor who doesn't try to provide all possible features but instead focusses on being simple.

[br073n]

I believe it's less important we allow running old software on a new OS, as it is we find a way to support older OSes, or at least older essential elements. I'm told the fundamental issue of incompatibility is the often radical changes in the underlying Glibc core. We need not burden new versions of Glibc with so much backward compatibility, as perhaps enable compat by enabling the system to run different versions at the same time. I know we've done some of that before, but not so much lately.

I doubt VM is the way to go just yet. Something short of that, such as a built-in emulator, might cover it. Wine is already making great strides to solving the old, "you can't run Winware on Linux" issue. Indeed, for a brief period our DOSemu was a better DOS than you found on XP, but the latest release breaks some of my old DOS apps. To our shame, it's easier to emulate older Linux on FreeBSD than it is on Linux, and do it seamlessly.

As for word processing, I now do it all in XHTML, and link to a printer CSS. It's the universal format, more portable than plain text for printing. For really long printed stuff, I use Lyx, because those documents are never requested for sharing on anything but paper. Word Processing is already about dead in some areas of business communications.

Some new ideas really are better than old ones.

[br073n]

Well, I sadly must report this isn't working as well as I had hoped. Without entering into a technical discussion, I don't agree with the choices made by the community regarding one-click multimedia upgrading for 64-bit. What a mess! Given the one reason any ordinary user might want to run 64-bit is the improved sound and video, this could be a deal-breaker.

The reason it seems worthy of remark is that I had so little trouble with the 32-bit version. At any rate, I'll be testing and checking more.

[oilpaint]

The future's not about os stuff. The toaster, a moment to check email, markets, weather, off it is. All on a iphone like thing or blackberry. The big in the way box on the floor, forget it. Does it matter how the toaster's programmed? No, just works. A work system, same thing, the computer as appliance, end of story.

[br073n]

Maybe so, oilpaint, but not today.

I'm 52 this year. I've used computers since DOS. You can sell the new toys to my son, maybe, and probably his kids. Meanwhile, I'm taking care of my health and plan to live a good long time, as do most of my clients, many already older. We aren't dead yet. We still like our old clunky computers and would like them to work well.

There are an awful lot of computers left in homes across the world, and I want to make them as useful as possible while they still work. The whole point of this inquiry was founded on the observation from my many years in serving clients -- the vast majority don't want the latest new toys. They consider their old toys a major investment. The problem is, far too few people in the world of Open Source realize this, because they can't imagine wanting what common users want. Maybe I can get lucky and find a way to mediate between the two.

[Axeia]

I just realized your way of thinking seems to match that of the Haiku OS, or at least from what I understand that they're trying to archieve. It's pretty much "BeOS continued".

The site can be found at Haiku Project it doesn't seem like it's going anywhere fast though..