Will Open Source Ever Become Mainstream?

Prabhu Ramachandran asks: "I am a graduate student at the University of California at Berkeley and as part of a course project I am trying to gather comments on the following question: Will the Open Source and Free Software communities develop software that will find widespread adoption amongst the mainstream, or is such software, by its nature, suitable only for sophisticated users? As part of my literature survey I found an academic perspective that seemed to indicate that open source projects do not reach the mainstream because the developers tend to listen only to their smartest customers. There also seems to be a lack of detailed documentation and an easy-to-use interface which normally attract the not-so-sophisticated users. I would like to hear the thoughts of Open Source developers and others on this issue. If you would like to view my references or the comments posted on a website hosted for this purpose, please visit my website." There have already been some interesting comments posted on his website. What is your take on this issue?

Comments

Well, let's look at the list

[Reality Master 101]

It sounds like this guy is talking about end-user applications that would be used by "normal people".

How many open source success stories are there, where the open-source solution is so clearly superior that it's used by everyone? Uh, zero.

Well, how about open source application that are good enough to compete with proprietary software? Uh, one. Mozilla, perhaps.

How many are "up-and-comers" that just need good word-of-mouth to take over from a proprietary solution? Uh, zero. (IE is already free-money)

The only one that I can think of MAYBE for the latter category is Gimp, and the user interface on that thing is so horrible as to be useless for anyone but a true geek (at least, the last time I used it which was admittedly a while ago).

Bottom line, I don't think proprietary software has much to worry about at this point.

Re:Well, let's look at the list

[Procyon101]

>How many open source success stories are there,
>where the open-source solution is so clearly
>superior that it's used by everyone? Uh, zero.

Well, as a developer, I see cygwin or mingW on just about every windows platform because nothing beats a goot GNU toolkit for development. As seperate projects that have each stomped all the competition, bash, grep, sed, etc are clearly dominant. Yes, most of these tools are clones of proprietary solutions, but most of them have come to completely replace their older cousins for reasons of quality, standardization or availability.

>Well, how about open source application that are
>good enough to compete with proprietary
>software? Uh, one. Mozilla, perhaps.

Same answer as above. Add apache to this list which is still the dominate web server. MySQL has a nice following and scares the poop out of SQL Server.. a huge Massively Multiplayer game (Dark Age of Camelot) runs it as it's backend because SQL Server and Oracle seemed to be just to expensive for the feature set they had beyond the Open Source solution. Back Orifice 2K WAS a popular remote admin tool for windows boxes, but the virus scanners kill it now .

>How many are "up-and-comers" that just need good
>word-of-mouth to take over from a proprietary
>solution? Uh, zero. (IE is already free-money)

If they need more word of mouth, then you probably haven't heard of them yet. There are some open source graphics engines which are up and coming and near ready to compete with their proprietary cousins.. gnutella has been a very nice architecture for one of the dominant P2P networks, which are very big with end users.

>The only one that I can think of MAYBE for the
>latter category is Gimp, and the user interface
>on that thing is so horrible as to be useless
>for anyone but a true geek (at least, the last
>time I used it which was admittedly a while ago).

uhh.. it's proprietary cousin's interfaces are just as bad. Face it, anything more complex then MSPaint is a geek tool and needs a geek interface.

Open source solutions fill a huge niche that needs filling. Many things in computer science need to be done, but there just isn't a nice $$ stream in it, so it gets neglected, or are simple enough that why pay for the solution when you can write it yourself, as is the case with alot of smaller tools (cat for example). Open Source projects typically don't dominate on the client side because 1) this is the biggest market and competition is fierce, and massive $$ goes into funding development for these platforms, and $=RAD. 2) Open Source projects don't advertise, since there is no profit motivation, hence the uninformed user remains uninformed. Therefore, there are few client side apps that are open source, because why support a project for free that is doomed? But don't knock the development model which gives us an excellent and protected public resource of shared pre-developed systems that don't need to be purchased or developed from scratch.

Re:Well, let's look at the list

[EZmagz]

I hate to say it, but you're right. Currently, open source software isn't ready to be adapted by the mainstream. That's the problem most of the /. folks have...not EVERYONE is using the open source software that they're using.

I recently asked a friend of mine (non-techie) ask me why I fired up Moz to show him something instead of IE. I told him it was open source, and that I liked that (and that IE wasn't). "Well shit, I can't read the code anyway...why would I care if it's open source?" he came with. Only after I showed him how tabbed browsing worked (and how insecure IE can be) did he even think about trying it.

My mom sure as hell isn't gonna care if she can see how linked-lists are implemented in IE! She just wants to check her email, and if it works, then that's all that matters. For her the term "Open Source" is just one more buzzword she has to ask me about.

Re:Well, let's look at the list

[binaryDigit]

He was referring to software for "common folk", he says so in the first line of his comment. Apache, MySQL don't count in this regard. There are five things people do most often:

1. Surf - IE dominates, hard to knock off when the thing is sitting right there on the desktop of every machine sold. The common user would have absolutely no reason to bother to download anything else.

2. Email - Again, Outlook/Express, dominates for the same reason.

3. IM - Here AOL rules the roost, but MSN Messenger is making some headway.

4. Office Productivity - MSOffice is the king here. Potential for OS inroads since MS does not bundle/force OEM's to put it on every desktop. Licensing can get quite expensive so OS could have a compelling value proposition. Plus only about 10% of MSO features are utilized by the average user anyway.

5. Games - well the funny thing is that no one seems to be crying for OS games. Funny that, as it's the one area that the supergeek is similar to the typical end user, they just want to use the software (i.e. play the game) and don't give a flip about having any source available to "fix bugs" or to "improve the software by having a million eyes looking over it".

What does this question even mean?

[Carnage4Life]

Open Source software is already mainstream both for regular users (look at Apple's OS X) and developers (look at all the work IBM's been doing in this area). What else is there to be done before it is considered mainstream? Grandma submitting a kernel patch by sending in a diff? W

Definition of "mainstream"

[boaworm]

This of course depends on the definition of mainstream..

1) Mainstream = "The biggest/largest/greatest".. then probably no

2) Mainstream = "Widely accepted and used amongst normal people" then yes.. this is today.

Look at companies like IBM and Dell.. would you call them mainstream ?.. most likely.. So if they offer PDAs/Servers/Workstations with Linux or any other OSS product on.. then it is mainstream already.

Open Source Target User

[Blindman]

First off, the motivation of a an open source developer tends to differ from that of a closed source developer. The closed-source developer is doing it at least partially for the money, so they have a great incentive to make it easy for customers to use. An open source developer generally creates software for some other reason. It is not that an open source developer wants to make things hard for people to use it, but since it isn't a goal it tends to be overlooked.

My understanding of the general flow of open source projects is that somebody writes some code for their own needs. They think it is cool, so they show it to some of their friends who may also be developers. The friends have some suggestions and pass it on to some of their friends. Soon you have a project written by developers for developers. If somebody else wants to use it, that's fine too.

Obviously, not every open source project starts this way, but the enduser generally isn't the first consideration.

So many things wrong with open source

[Uhh_Duh]

Open source -- as we know it today, has so many things wrong with it I can't even begin to tell you..

1> Documentation is usually 2nd priority. In my world, if there's no documentation, there's no product.

2> The product is usually 2nd rate. Because there's often no money on the line, my experience has been that the programmers take less accountability for their efforts. Big bug? Guess you have to wait until the programmer (or someone else) gets around to it. Big bug in a program you paid thousands of dollars for? My experience is that enough screaming can get you a patch in very little time.

3> The user interace is lacking severely. Bigger companies hire people who specialize in usability to the design the UI. Open-source projects have HORRID user interfaces (A perfect example of this would be Request Tracker -- the software rocks.. the documentation sucks, and the awkward user interface effectively makes the product useless for large-scale deployments).

Open-source definitely has it's place. It's fabulous for the "quick fix it" jobs and the "I've got lots of time on my hands to figure it out and fix any problems I find" solutions. Sadly, however, my experience has been that this stuff is only truely free if your time is worthless.

Don't get me wrong.. I love open-source software. I wouldn't be able to do my job without it -- but with these drawbacks, it will never take the place of the mission-critical elements where I can hold someone responsible with I don't get what I need. :) (yes, those things cost money -- sometimes money needs to be spent).

Re:So many things wrong with open source

[nuttyprofessor]

I find OS software bugs to be found and eradicated
much quicker than comercial software (in many
critical cases anyway).

The weakness of OSS, which is also its strength,
is its dynamic nature. You constantly need new
library versions or new kernel versions. Only geeks
like to keep building new kernels to keep up -- this
doesn't fly in the mainstream.

Re:So many things wrong with open source

[b17bmbr]

documentation and the documentation in windows is so stellar? it is mostly non-existent. oh, that's right, go out and buy teach yourself win..., or win... unleashed, or whatever. and try to find help on line. if FOSS docs are 2nd rate, windows docs are 4th rate.

product is second rate which products are you using? hell, if you download and install a 0.6 beta, yeah, you're helping to work out some bugs. but you would rather be a paying beta tester, only to have improvements come out in the next release (um...windows) by the time apps get to 1.0, Mozilla, OO, they are first rate. i wonder what Office betas look like. we just never see them.

user interface i have both a linux desktop and ibook. i love both. windows was never, and still is not "easy". riding a bike easy? no, but you learned it. like windows. it's what people know.

maybe you should give some examples of your "quick fixes, and why it is only worth it if your time if worthless. i'm sure the good folks at ibm, dell, oracle, sun, and many others think it is "mission critical".

Re:So many things wrong with open source

[blincoln]

The user interace is lacking severely.

There are a few exceptions, of course, but in general I couldn't agree with you more.

UI design is one of the most important parts of a project, and the vast majority of OSS projects seem to slap it on as an afterthought.

It's *possible* to use software with a geek-oriented, unfriendly interface (*cough*mainframes*cough*), but it doesn't *invite* users.

Putting a pleasing, intuitive UI on a product is the equivalent to looking nice for a job interview. It's *possible* to get hired if you come in with ripped jeans and scraggly facial hair, but you limit your audience severely.

Yes ... with the USA being last in line.

[IGnatius T Foobar]

If you want to see mainstream adoption of open source, you have to look outside of the USA. If you follow the Linux news sites you'll see lots of foreign organizations, particularly governments, looking to make big switchovers to Linux and other open source software. Bill and Steve have been doing a lot of travelling lately, offering what basically amounts to bribes to keep these organizations on Windows.

So yes, the world has already started the mainstream move to open source, but the United States is the last place you'll see this effect -- because we're too heavily entrenched in Microsoft crap to be among the first.

This parallels other technology shifts. Why do other nations have wireless networks that are so much better than those found in the USA? Because they didn't become heavily entrenched in landlines the way the USA did, so they were able to leapfrog. It's the same way with software: fewer installations of Microsoft crap mean an easier deployment of something else.

Just give it time. Basic economics will work it out.

What motivates open source developers?

[knowbody]

Open Source development is done on free time, except for the lucky few who are sponsored. That makes it a hobby and hobbies are for fun.

Dealing with non computer literate people is not fun; it is work. Given this contradiction I doubt that "pure" Open Source will ever become mainstream.

However, I can see the possiblity of the hybrid open source / commercial groups succeeding in that area. These organizations (such as SuSE) pay people to do the boring stuff like write documentation targeted at non-techies and so forth.

Short list

[carlmenezes]

Mainstream :
Apache, Sendmail, Pine (used in almost every university of the country), GCC.

Potential Mainstream with primary need:
Mozilla - word of mouth and improvement in stability
Ximian Evolution - word of mouth and hands on use.
OpenOffice - word of mouth, universal office document format
Linux -
for the general internet browser :
better GUI, fonts, documentation, games and more applications.
for the new power user :
better GUI, fonts, documentation
for the professional :
better documentation

For my projects... no.

[jhouserizer]

I'm the "lead" of a couple open source projects that will never be mainstream, for two reasons: (1) The products target application developers (not lay-men) and (2) I don't have time to donate for the sole purpose of helping "stupid" users.

While reason (1) kind of makes my posting a little off-topic, reason (2) I think is true of a lot of open source projects - including those for products that do not specifically target the tech-savvy.

The reason is that open source is nearly always built from "donated" time, and most of us coders just don't have enough time to spend on such low-priority (as we see it) things as making the product easy for "dummies" to use. Sometimes I struggle to even respond to mailing-list questions that are obviously written by "dumb people" - I just think "it's not worth my time"!

This attitude probably even affects open source projects that are actively trying to target the mainstream. I'd imagine for most developers it's a constant battle between their personal attitude/desires and the project goals.

I'd say Mozilla and Evolution are the two best examples of success in making open source software that is usable by the main-stream. Kudos to those developers!

It can be

[Apreche]

Just most isn't. A good example is something like CDex. It's a small open source free software project that is relatively mainstream. The reason it is so successful is because it serves a useful function, is for windows, is easy to use and easy to install. It is also one of, if not the, best CD audio ripping program there is.

The reason that OSS isn't mainstream is because most of it is for linux, most of it is hard to use, and most of it is hard to install. Most of these have to do with the nature of being for linux.

Stuff like Mozilla, gAIM, CDex, etc. can become mainstream. But Open Source programmers make things for themselves, and generally don't have the public in mind. Companies that make commercial software have a primary concern of profit. They will only profit if their software can actually be used by lots of people. OSS programmers don't have this as their primary concern. When they do their software will become mainstream.

LIstening to the smartest customers

[Crag]

While it's true that Libre Software developers work more closely with the users who contribute back to the project in some way, and those users tend to be the smarter, geekier users, the biggest difference between free and proprietary software is that free software encourages users to become smarter.

The value in learning the nitty-gritty details of a proprietary product are lost when the vendor makes incompatible changes to scare off potential competition. The proprietary vendor wants no help from the users. He wants his users to send him money on a regular basis and not ask questions unless they will pay for answers.

The Libre developer doesn't give a rats ass what the user does with her software. That's what makes the software free. The developer prefers to get something back for her effort, so she has a motive to make her software approachable and to provide her users with means to contribute back to the project, and often that means encouraging the user to get smarter, directly or indirectly.

This is a gross over-generalization, but /. is not the forum for full-blown research papers as comments, so I won't defend my thesis further.

Re:It IS mainstream already

[swordboy]

Most of the net and probably most corporate and military servers runs apache and sendmail

Linux on the server is mainstream but...

I think what he is trying to ask is:

Will Joe User ever be able to sit down at a given open source workstation (i.e. - Linux on the desktop) and find enough consistency with every other open source workstation such that he/she can get something done without spending countless hours reading HOWTOs, message boards, distribution-specific documentation and performing mind-numbing tweaking at the four corners of the operating system?

I had the day off today so I installed Redhat 8.0 (SURPRISE!) and tried to get Mozilla 1.2 up and running with anti-aliased fonts. I wasted the whole day and I am glad to be back on Win2K (call me stupid or whatever... half the font stuff made me feel like a criminal - why isn't it *on* by default? I'll pay big bucks for that...). Linux is shooting itself in the foot with that respect. Everybody hears so much about Linux so they install it only to be disappointed to such an extreme that they'd never want to bother again (I know that I do not).

WTF?

I would be GLAD to give several hundred dollars to any company that can make a consistent, user-friendly, non-MS OS for my x86 hardware (all of it, not just some). Is this possible? Apple - where are you?

Linux will be ready for the desktop when Gnome or KDE drop dead (I can't wait) and some consistency settles in. Until then, I'll run BSD on my servers (the documentation is much better as a result of the consistency) and Windows on the desktop.

Cheetos!

Why wouldn't they?

[Kjella]

Certainly, if all the people who were pirating actually had to pay for Photoshop, they'd probably consider Gimp instead. Some might still have found they need it, but most would settle for something free, or something cheaper. Unless you have ethical or juridical (think:companies) concerns with not having a legal licence, Photoshop is, and presumably will be superior to Gimp for a long time.

"Free" copies of Windows, MS Office etc. is what is keeping free software from the markedplace. And I think Microsoft knows this. Noone is going to feel that they've "hurt" Microsoft by not adding another 0,000000001% to their bank account. I don't think there's much software that a majority needs and would be willing to pay for. 50%+ don't need Photoshop. But if they can have it anyway, why not. It's like having an off-roader without ever going off-road. It's not that you actually do it, but that you could do so.

I know. At a work place I had to make do with Paint to make some simple figures, because there was no budget to get me anything better. Ok it was simple lineart, but still... I'd want nothing more than to install Photoshop/PSP/whatever, but I couldn't.

Kjella

define mainstream

[timothy]

If you could buy something in a Texas Wal-Mart a couple of years ago, would you say it deserves the name "mainstream"?

If yes, then it's too late for Linux to escape, because I've done that.

Nicely boxed, manual-included Linux distros have been around for years (in national chain stores), and "open source" covers things a lot less radical, like say the Phoenix browser. Lots of Windows users don't think of themselves as too far from the mainstream justs because they're using a better browser than IE :)

tiothy

Software "for sophisticated users"

[ChaoticCoyote]

What's wrong with software for "sophisticated users?"

Nothing.

Somewhere, pundits have declared that Open Source and Free Software must appeal to the masses in order to be a "success."

If anything, the desire to attract the masses is a primary reason why commercial software stinks. It's bloated, complex, and wasteful -- because it tries to be everything to everyone.

Open/Free Software, on the other hand, lacks the financial incentive that dilutes creativity and effectiveness in commercial products. "Free" has many connotations, including the freedom to be original and precise.

Open/Free software can not be treated as a monolithic block; "popularity" means very different things to developers of various projects. Where KDE and Gnome care deeply about being popular, many (many) other projects do not.

Freedom is about choice -- some projects chose to chase popularity, while others focus on being the best available tool for a discerning audience. Trying to declare a goal of "popularity" for all Open/Free software is myopic at best and counter-productive at worst.

Re:It IS mainstream already

[Bilestoad]

And this is the question he should be asking -

"Will Free/Open Source authors/users ever admit that they have to listen to end users, or will they continue their elite cries of "RTFM!" whenever a "luser" can't do something that is perfectly fucking obvious?"

Sadly the answer seems to be NO as you gentlemen have shown. It's not whether Open Source/Free software is capable of beating Microsoft at their own game; I truly believe that MS would die the death of a thousand cuts if all the hackers out there just got their shit together and tried to produce a real windows replacement. However that isn't what it's all about. It's about producing cool stuff to be used by those in the know, and it's never about producing hand-holding software like Apple's iPhoto. Sure there might be something like iPhoto in some ways but was the primary focus to write an app that would act, as Gassee used to say, as a "tractor" app? Didn't think so.

Re:KDE and Gnome are paying attention

[Khalid]

Thanks to SUN which initiated a Gnome usability study; there are now explicit usability projects http://developer.gnome.org/projects/gup/ for Gnome and http://usability.kde.org for KDE. I feel that Gnome and KDE developpers have began paying attention to what heir usability contributors are saying, and there have been some (albeit) small steps in the right direction. But things will sure need some time to happen as is the always the case with open source. Open source need time. I am using Mozilla 1.2 right now, and it realy rocks ! IE 6.0 has been really left far behind ! in my opinion

Re:It IS mainstream already

[t0qer]

Well the solution they ended up going with (the name escapes me at the moment) was a $200k solution (not a typo)

The support contract specifically states that the software vendor has to come out on site and install it, train users, fix it remotely anytime and a 24hr turnaround for anything that requires a live person.

Perhaps if the makers of nuke, slash, ect all got off their collective asses, actually marketing their software in addition to keeping the open source strategy they've used and provided "services" as well as software perhaps this would have turned out differently. The fact that they rely "solely" on open source to market their code is just shooting themselves in the foot.

Hey Malda, just a question, is a 200k sale worth a dog and pony show to you?

Missing something

[wfrp01]

Every time this discussion comes up, the presumption appears to be that free software lacks mainstream appeal because of interface issues. While such considerations play a role, de-facto standard proprietary data formats and communications protocols play a far greater role in establishing the entrenched 'mainstream' computer interface with which people are familiar. Unless and until people wean themselves from their dependance on .doc, .xls, SMB, .NET, .mov, .wma, etc. they will find themselves locked into the familiar "mainstream" operating systems and applications. That is the crux of the matter, not pretty buttons and widget layout. With the MS anti-trust farce behind them, and palladium ahead of them, expect no mercy on the proprietary format front. Free software has a very tough row to hoe. Which is why free software's ultimate victory will be so much the sweeter...