Greetings, live from LinuxCon in Sandy Eggo.
Two things I want to talk about:
First up: Fedora Board Meetings. We do a public IRC meeting every other week, on Wednesdays, at 18:30 UTC (11:30 pacific, 2:30pm eastern, or use a time converter that I’ve conveniently already preset with the time in this link.
These meetings are open to everyone. We set aside time at the beginning of every Board IRC meeting to take questions/concerns/comments/otherwise from folks who wish to join the meeting (we used to do this at the end, but it always seemed to fall into the “we ran out of time” situation). So consider this a friendly reminder, or an announcement for those who didn’t previously know, that you are welcome to join and observe, participate, etc. – sometimes we have no questions, and sometimes there is lively discussion.
If you’re not familiar with IRC yet, this page is a good place to start. We meet in the #fedora-meeting channel on irc.freenode.net. There is also a wiki page with some light information about meeting structure and protocol for Board meetings, which is useful to read as well.
Second: I wanted to type a bit about user surveys. It’s an old board ticket, but has particular interest to me, and I’ll elaborate on why that is. 
So way back in Ye Olden Days, I was a new person to the Fedora Marketing team. One of the first things I was very interested in was the idea of market research – I’ll get to my interests there in a moment – and making a page about moving forward with some various aspects of research was, literally, the third thing I ever put on the Fedora wiki. The first two things were release name suggestions. That was September of 2009. We embarked on the epic journey of Lime Survey packaging, and, well, eventually I got sidetracked by other things (FUDCon, becoming employed, etc.)
But the idea is still near and dear to my heart. Before Red Hat, before the motherhood period of being a stay at home mom, before Intel, I worked at a industry analyst firm, cranking out reports on server, PC, PDA (yes, I just dated myself) usage, sales, dissection of usage by vertical markets and size of business, etc. I find data fascinating. And part of that job was surveying people for various things.
Why I think this type of thing is useful? A few reasons. From strictly a “how many” perspective, which was the bulk of my reporting at that point, it’s incredibly useful data to a variety of information-consumers; if i manufacture parts for a PC, it’s helpful to know that if I make, say, memory for laptops, that it is unlikely that I will sell 4 billion pieces of memory, if we generally assume 1 or 2 sticks per laptop, and worldwide sales of laptops are 150 million per year. You laugh, yes, but I have seen forecasts in my time that wound up equating to “45 set top boxes per man, woman, and child, sold in one year.” At Intel, I was on the data-consumer side of this, looking at new opportunities for specific chips, so looking at this type of data could help me establish how big a market was, vs. how much we were already selling into it, etc. And finding new markets altogether was always awesome.
From a more general usage survey perspective – which was more in line with size-of-business segmentation that I did – that type of information is useful to vendors for tailoring their needs to different types of markets, or identifying which ones they can serve the best. For illustration: If businesses between 1 and 5 employees typically use cell phones to conduct business operations, because they don’t want to screw around with maintaining phone infrastructure, and businesses between 1000 and 5000 employees use some sort of PBX or VOIP stuff, and I am a vendor that is selling a magical pink unicorn that makes VOIP dead simple and lowers costs, I could tailor the targeting/marketing of the small businesses, because they don’t have existing solutions and because they struggle with barriers to implement, and target the enterprises differently (cost savings, etc).
Anyway. My point is this: I find it interesting, I find it gives useful information to people.
We’ve talked a lot in Fedora over the years about where we are going, what we are going to do, etc. It’s always controversial. I think one of the key sticking points is this, and again, I love metaphorical illustrations, so: If you have a group of friends, and you want to go to dinner, you have to pick a place that works for everyone’s diet, you have to pick something within budget, etc. Lots of considerations. You never, ever say, “Let’s go to dinner in Paris,” and assume that works. Particularly if you are not close to Paris, if you don’t have a plane, and you only have a boat. If you live in Paris, then that’s totally attainable. When you go to dinner, you consider where you are starting from.
But we often lack any consensus, at least, in my humble, often-wrong opinion, about Where Fedora Is Today. And in many discussions, I see a wide variety of assumptions about usage, and they have vast differences, like, oceans-apart, totally conflicting differences. Coming to agreement on how to solve a problem when there’s no common understanding of the underlying assumptions… well, that’s kind of like, making a map to dinner in Paris that only covers the last mile of walking, and doesn’t cover the “where we started from, and do we have a plane” type of stuff. I’ve probably mentioned about 40 times that I’m sort of into planning things, so I think this is a good first piece of that type of thing.
So. Still with me? Haha.
A few good things to note about user surveys:
- Doing them consistently (ie: with the same questions or only slightly changing what you are assessing), on a yearly basis, can give you a good way to measure “things.” “Things” being – if you focus on addressing a certain area, for example – you can see if the work done made a measured difference in following years.
- Much of it is about writing good, clear questions. Unbiased questions, without a particular slant to them.
- Be clear when working with folks to develop the survey about if you’re looking for opinions or actual data. ”What is your opinion of __________” is different from “what is your primary use of _______.”
I’m really thinking of the usage data points for this survey – how do you use Fedora, what applications do you typically use, what type of hardware (desktop/laptop), that type of thing. But I’m still rattling ideas around in my head – we’ll probably tackle this more fully in marketing-land, though it has been, as I mentioned, a board ticket for some time. There’s also other ideas along the lines of doing strictly community-people surveying, but we shall see. And of course things like – tooling – figuring out a process for translations – figuring out how to get the word out to a lot of places that we’re doing surveying – etc.
Anyway: I think it’s an important thing to do. It helps to plan, prioritize, and give people new ideas about ways to contribute or places to improve things, and a way to measure improvements or progress (aka: mustard). I’ll write more about what kinds of specific questions I’m thinking about and how people can get involved over the coming days.
And with that: I am off to keynotes and booth duty.
