I’ve been reading several responses about the news story of a girl who “dropped out of school” because of Ubuntu. I’ve read several responses, including Aaron Toponce’s response, and it got me thinking about a topic that has been on my mind for awhile.
You can read the story yourself and the follow-up backlash from the Ubuntu Community. I was extremely sad, but not at all surprised, of the backlash that this woman got. I’m sure the negative response is from a relatively small percent of the community that is just extremely loud. I also think the title of the story is a little misleading and that it can be taken as bias. However, the root of the problem lies in this: a woman thought she was buying a consumer product, when in fact, it wasn’t.
Difference Between Consumers & You
I don’t care who you are, but if you are reading this, you already have technical edge on 95% of all computer users. You are automatically disqualified from being able to say “Ubuntu is easy to use.” “Installing software is easy on Linux.” I’m sorry, but that is just the case. The definition of “easy” for you and I is completely different. To understand the average consumer, what you really need to do is observe the consumer trying to use your product.
A consumer expects to be able to stick a CD in have it work. A consumer expects to be able to download a file and double-click. If they can’t do that, to them it is broken, and in a consumer context, they are right. Easy-of-use when starting at a zero knowledge point is crucial for consumers. They don’t want to spend time learning new things. The learning curve an average consumer is willing to endure is extremely low. I’m sorry, but I don’t know of a single Linux distribution that is ready to be the main computer for the majority of consumers.
Right now I think Linux can make a great secondary computer for a normal consumer. It is easy enough to do Web, Email, and IM. School work can be done with OpenOffice. But for everything else that is out there for Windows and OS X machines, I really do think the learning curve is too much for normal users. (Heck, normal users struggle with Windows, something they have been taught for ever!)
Consumers Don’t Care About Freedom
Free as in freedom is useless to a consumer for the most part. My wife isn’t going to custom compile her applications. My mother doesn’t care if she has the option of modifying her programs. Freedom holds very little value in the sense of being able to change. Consumers, however, love freedom when it comes to installing on any and all machines, and freedom as in price. Free as in beer is something almost every consumer will love. However, it is not out of principle, but usually that their wallets stay full.
Education Is Expensive
It takes time to learn something new, and consumers have a completely different tolerance to education than technical users. Most of the time, technical users enjoy learning new things about technical topics. When something isn’t working, googling the problem isn’t that big of a deal. The average consumer, however, hates it. Their tolerance to “figuring stuff out” is extremely low. If you don’t enjoy technical stuff, learning how to do things is a pain.
Educating a user is expensive in the sense that it takes a lot of time. Both for the person putting together the education, and the time it takes for a consumer to learn. A consumer, like the women in the news article, doesn’t want to learn a new operating system. She just wants to write papers and log onto her website. Ubuntu is not a good solution for her if she doesn’t want to learn something new, because the onus is on her to figure stuff out. Most consumers have never used an online forum, and they have no clue about IRC. For her to get help, not only does she have to find a good place to ask, and ask in a correct way, she has to learn how to log onto IRC and how to use Forums. The education requirement is very high currently for people starting at zero knowledge.
I’ve seen people in the open source community throw out the idea of “education” like it is this free commodity. However, in all honesty, educating people takes time, and time is a finite resource that is very valuable for most people. Asking someone to learn something completely new, especially when the onus is on the user to take care of themselves, is a lot to ask for.
Power & Polish
In my mind, the way I visualize computing and usability is by using the two terms power & polish. Power is the raw ability to do something. Polish is the ease doing that something. Lets look at an example:
Linux is very powerful. If you know your way around the command prompt, you can accomplish just about anything. I hate administering Windows Servers because I lack the power that I have when administering a Linux Server. However, I wouldn’t say Linux is very polished in comparison to OS X. Apple has placed a huge amount of time, effort, and work into polishing the interface and usability of their operating system. The reason I like the term “polish” is because polishing something doesn’t give you anything new, but just makes something already in existence better. It also requires a great deal of work.
Typically, I highly dislike “polishing” my applications as a developer. I already know how to use it, and use it well, because I made it. However, making it easy for everyone, from the intelligent to the confused, is extremely time consuming. Its amazing just how clueless I think users can be, however that polish is extremely important if I expect users to use and enjoy my application. Lets look at rsync for Linux. It is extremely powerful, and I love using it every day. However, you have to know what your doing. To polish rsync to a point it would be at a consumer level, we would have to make a GUI. That GUI would have to built in tutorials and help functions. We would have wizards and setup helpers to make automated backups. It would have to explain every little detail in their help functions, as well as handle every possible error, user and application. I would have to make it easy to open any where and use it anywhere. The time it would take to that version of rsync would be a lot. What open source developer wants to do that, especially since all the people currently user wouldn’t use that GUI at all? Technical people tend to favor power over polish. Consumers, however, love power, but are lost without polish.
Polish puts the onus on developers instead of end-users to make things ease and “user-proof.” I think Linux has made a lot of headway in the sense of Polish. The technical knowledge required to use Linux has dramatically dropped from ten years ago. I remember going to a store, buying a copy of Mandrake Linux, installing it on an old machine, and thinking “what the heck is this?” I was so confused and lost. Now I think I could get going a lot better if I tried it again at that skill level.
Death & Taxes
I think this is the best example of helping developers understand how consumers feel about using computers. My father-in-law, an owner of an accounting firm, has a saying he loves from Benjamin Franklin: “Certainty? In this world nothing is certain but death and taxes.” Taxes is something that we all will deal with and pain. They are also extremely complicated. The more money you make, and the more stuff you buy (house, cars) and things you spend it on (kids, education, medical, bills) the more complicated they get.
It is 100% possible for an individual to do all of their taxes for their entire life. All of the resources are available “in theory.” You can look up and read all the tax laws. You can contact your sate or federal IRS and ask very specific questions and go through hell trying to get answers. If the IRS performs an audit on them, they can read up again on all the laws, show them all of their records, and fight to prove they have been honest. They also can pay the consequences (ranging from fines to jail time) if they mess up and the IRS penalizes them.
But who would want to do that? That is why we have accountants and programs to do it for us. A few people take a lot of time to gain a full understanding, and then help normal people get through filing and paying their taxes. There are people, including my wife, who really enjoy learning about all the complexities and become an accountant, but it isn’t for everyone! So why do open source developers expect all people to be “accountants?” What would you do if your accountant told you to “RTFM?”
Yes, the resources are out there for consumers to find answers to their problems, but it in not always easy, and most of the time very difficult and over their heads. Consumers and end-users frustrate us developers, I know this, but we have to be understanding.
Consumer Version of Linux
I don’t think I could call any version of Linux right now available “a consumer version.” However, I do have my predictions as to what needs to happen for a consumer version of Linux to become available.
- One Consumer Distro – I think we’re going to need to have one major consumer version of Linux. Not ten, five, or even two. There needs to be one version that is considered the “starting point” for consumers being introduced into Linux. There can be “others” out there, but there needs to be one great version that all OEMs and vendors carry and support.
- More Polish Than Power – I think we’re going to need that distro to have a lot more polish than power. Consumers don’t need or want unwieldy power, they need easy control.
- Easier Than Windows / OS X – If Linux wants to become mainstream, it needs to be easier and better than windows. The goal isn’t being as easy, but totally surpassing it.
- One-Click Install / Download & Install – It needs to be easy for users to download and install applications. They need an easy way to find those installed applications. It also has to be relatively easy for developers to create these distributions.
- Major Company Backing – I think one of the reason Ubuntu has been so successful is that it has had a great mix of company backing and a great community. I think Red Hat as well has found a nice niche in the market. The question is whether or not
- Inexpensive – I don’t think this version could be 100% free (as in cost). The problem is the total amount of polish required to put into this project, I don’t think you could get any team of open source developers to spend such a large amount of time on usability with out an open revolt. Preventing all levels of stupidity is just extremely monotonous and time-consuming. A low cost OS (like $39) would be great. I imagine an Ubuntu Plus project, and it would have a relationship like Fedora, CentOS, and Red Hat has.
- Quality Customer Support – There will need to be several options of customer support. Right now there really aren’t any options for people to get paid assistance. But pointing people to forums, mailing lists, and IRC is not an option as being the first line of defense for getting support. Consumers pay for support so they can ask their really stupid questions. This support should come with the purchase of a new machine.
I honestly don’t know. I would love to see this soon, but all of the current distros just are not ready to handle a full on-pslaught of stupidity that regular consumers bring. Ask anyone in technical support. I believe technical support is one of the most loathed jobs in the industry, and there is no way the current Linux community is ready to be subjected to it. I’m kind of glad that Linux, in its current state, isn’t 100% mainstream yet. Knowing Linux, and dealing with Linux users can be a haven of refugee from the common issues of regular users. I honestly think in order to have a consumer version of Linux take a strong hold in the market, there is going to have to be a very strong business behind it that knows how to rally a great community around their product.