Open Source: What You Own

Posted in: MySQL, Technical Track

My parents instilled upon me many values that I keep with me today. My twin brother and I are the youngest of four children, coming from a lower-middle class background. We children had the inevitable fights over material possessions, screeching “Mine! Mine!”

My father’s response to this was to look at us and say “These toys are mine; I bought those toys with money I worked for. What’s yours is what you make with your bodies.” While the sentiment is arguably harsh, crude and bordering on vulgar, I cannot argue that he had a certain point.

If you do not truly own something, you will be left squabbling like a child when your perceived ownership is threatened. When you assumed you owned something and the truth comes to light, you will be massively insecure and have a sense of injustice.

A few points from OSCon are haunting me and getting me to think about what FLOSS means to me, and what I want from it.

— Open source is important even if you never read one line of source code or make one modification. The fact that anyone can read and write the source code is critical even if nobody besides the original engineer(s) ever does.

Many people love owning their own home; geeks are no different. But a geek owning their own home will likely make changes to it; in the late 1990’s many engineers wired up their homes with Ethernet cables to the bedrooms, bathrooms and kitchen. Whether installing solar panels, knocking down or putting up walls, a home is the ultimate open source tool. There are many manuals to read and a pretty harsh EULA. But the fact remains that even if you do not change the inner structure of your home, the fact that you own it and could if you wanted to is critical.

— Proprietary products from Apple, such as the iPhone and Mac OS X operating system, work well enough that open source aficionados are having internal ethical dilemmas about owning them. I am not a developer, however I have modified some open source code in my time. I even have something to show for it — in the left-hand menu of Nagios, the text box where you can enter in a host name, hit “Enter” and go immediately to that host’s page was my patch. I often found myself wanting to go to a specific host as fast as possible. So, I patched my own version, then submitted it onward, thinking it might do some good.

Part of me thinks, “if it works well enough, I never need to read or write code to make it better. Therefore, I don’t need open source.” The problem is that this idea is self-fulfilling. If a product is closed source and good enough, very few people even consider the need to modify what they “own”. If a product is closed source and not good enough, either the product’s engineers make the product better, or the product dies. With open source, anyone can look into the code and make it better, whether it’s a simple addition like a text box or something much more complex.

Open source gives out more power. Traditionally, bug fixes and features have to be critical enough to be worked on. Often “critical” does not just mean “severe” but rather, “someone wants it badly enough” to either call in a favor from people inside the company, or frankly, to pay the company until they fix it. You either have to know someone or have big bucks; with an open source product, you can modify your own copy and perhaps even share your modifications with others. (Sharing is not a requirement for open source, but as MySQL community developers know, being able to share modifications makes open source software able to grow.)

— I do not own services on the internet. I use a hosting provider for my website, web services for e-mail and my to-do list, and even I store some documents online. I cannot look into the guts of my hosting provider nor change the way hosting works. It would be dangerous for it to be completely open to me, because I do not own the service, I only pay for the privilege of using the service (same with e-mail systems). If I could look into the service and change how it works, in theory I could change it so that I can see anyone’s data, not just my own.

“But surely you own the data!” Yes and no . . . even if I accept a EULA that says “the end user owns all the data, the service provider owns none of it” (very few, if any, services offer this type of agreement) the fact that the data is stored on machines I do not own and goes through applications I do not own means that effectively, I do not own my data.

Like (other) Apple users, I am content with this trade-off, for some purposes. Using a closed source e-mail system is a convenience for me, as is running a proprietary operating system. I could set up an open source e-mail system, and indeed have had the responsibility of running one in a previous job, but I am lazy. Is that selling out?

“Selling out” is one way to put it. But it is not the whole story. I have run Ubuntu, Fedora, Windows, and Mac OS X on various laptops, and I actually find running an open source operating system too much work. I understand that folks differ in their experiences; this is mine, and it means that I am willing to trade open source ownership for convenience. The operating system is an extreme example of open vs. closed source, simply because the user experience is so different.

Blogging software is a very good example of open vs. closed source software. There is plenty of open source software out there for blogging software (see Ohloh’s search results for “blog”). Both Pythian, my employer, and I use open source software for blogging. Why would anyone use a closed source blogging tool when there are so many open source tools with similar levels of user experience? Preferences may differ, but most open source blogging software is easy to use, so why would anyone use closed source software for a blog?

I also use a blogging service that is based on open source software — Livejournal. I definitely do not have control over the data, but I can own a copy of the software . . . but as I cannot actually change the code running on the website, I do not actually own the software that I run.

So where does this leave cloud computing? Unless you are the service provider, you do not own the software running a cloud computing system. If the source is open you may own a copy, but that is not the same thing.

At OSCon, someone said “I do not want to own an iPhone because I cannot truly own it unless I can hack it.” I do not put such a requirement on everything I own, but I understand the mentality, and I do think we need more open source in our lives. Perhaps it is hypocritical of me to own an iPhone and yet want to push to make it run on open software. However, open source licenses in a Web 2.0 world are as irrelevant as Microsoft*, as Tim O’Reilly points out.

We need a standard of what an open source service truly means.

* With apologies to Brian Aker. Although from what I understand, the issue is that Microsoft is very obviously the dark side that very few open source proponents are willingly lured into. Apple lures open source proponents to the dark side of closed source with sexy interfaces and software that mostly works. Apple gets some of the most staunch and opinionated open source proponents to compromise their values and that is particularly dangerous.


Interested in working with Sheeri? Schedule a tech call.

1 Comment. Leave new

Cloud Computing « Oracle and other
October 1, 2008 2:33 pm

[…] mi ha ricordato anche un altro post di Sheeri Cabral che lessi tempo fa su pythian: “Open Source: What You Own“, l’argomento mi sembra pertinente. Mentre scrivevo questo post ho notato che sempre su […]


Leave a Reply

Your email address will not be published. Required fields are marked *