2010-05-14

Target audiences.

A question in IRC that I missed due to eye problems was basically this:

Who did Red Hat Linux have as a target audience, because it didn't seem like they did and they were fairly successful.

Actually Red Hat Linux had several audiences:

  1. The developers had to get stuff done. If Alan Cox needed joe to edit, then joe was always in a release. If rpm development needed xyz-lib then it was there. For the most part, it just had the things people felt they needed to make the distro what they wanted. [Name for audience: OS Developers.]
  2. The staff of Red Hat. It was a requirement for everyone to use Red Hat Linux as their desktop. The only people who got away for a short time with it would be new transplants or for short term jobs (filing reports to the taxman that required a Windows 3.11 utility). If something broke marketing's desktops it usually got fixed pretty quick. [Name for audience: Small business staff.]
  3. Spouses and parents. This was the big target and the most demanding in some ways. Well especially for 2 people: Mr Troan's wife and Mr Szulik's dad. However, this target was usually the next release target.[Name for audience: casual users.]
  4. People who wanted things. Be it customers or the larger set of users downloading from the internet, the requests of things from users would be weighed against "Can we support this?", "Will it fit on the Cdrom?", or "What has to go to put this on?". Since we originally were trying to keep things down to 1-2 cdrom's.. most requests for things would end up on the "goes to powertools.", "will the emacs or the vi users raise a bigger stink?", "we will try for it next release." In most cases we can call this audience: home-DYI user.

After RHL-6 came out.. target audiences changed. Where the money was coming in wasn't the home-user crowd but the server organizations. The needs of those organizations pushed RHL more into moving their needs up the ladder, and the DYI user further down the stack. Eventually many of those people went to other distros (Gentoo was really really big in 2001.) or putting stuff into fedora.us (and then either going onto Fedora Project or doing their own thing.)

So what does this say about the target audience(s) of Fedora? I think that in the end #1 is the same: OS developers. #2 is where things get wishy washy and people either think they are it or they will never be it. I would prefer for us to look at #2 being the same as it was for RHL: 'the staff'. There are a lot of cursory things that need to be done (setup fudcons, review packages, make nice icons, or make webservers go) to get a release done, and without that work audiences 3,4,5, etc aren't going to be happy either.

So what is #3? Well that is the 99 million dollar question, that we successfully avoided for nearly 6 years until last October:

Someone who
  1. is voluntarily switching to Linux,
  2. is familiar with computers, but is not necessarily a hacker or developer,
  3. is likely to collaborate in some fashion when something is wrong with Fedora,
  4. wants to use Fedora for general productivity, either using desktop applications or a Web browser.
Ok so we aren't looking at Mr Szulik's father nor (to update names) Mr McGrath's wife any more. Now while it doesn't define who our target should be, we can see from data that is collected in smolt and via maps of mirror requests who currently uses Fedora.
  1. a sizeable chuck (40%) of the systems are running on less than 2 GB of ram and 2GB of disk-space.
  2. the majority of systems (70%) are running i686 versions of Fedora versus other types.
  3. the CPU of many systems are not the latest and greatest.
  4. updates are looked at from both high-speed places and slow-speed places (versus just one).
While I can't see other hardware, I would expect that monitors and such would be mainly lower end ones. That gives a pretty good idea of who will be using the general release. To become more specific, I think we would need to target for spin(s): The sysadmin who wants to work from the coffee shop on his less than stellar spare laptop that the guy in marketing dropped off on his desk last week. Or something like that.

1 comment:

Moses said...

Interesting Post. I struggle with Fedora a lot. I think it's a great distro and it was the distro I switched to when decided to use Linux as my primary OS -- Fedora 6 time frame. I've since moved to Debian after a brief stop in Ubuntu land. Ubuntu is a disaster for anything but the most vanilla setups. It's one of the least stable mainstream distro's I have ever used (I've used 8.04, 8.10, 9.04 and 9.10) I'm using Debian Lenny right now and I am anxiously awaiting Debian Squeeze. I'm not really that happy with Debian Lenny because of the age of the packages and the lack of any commitment for a release date, however the stability of the system is ideal. I had considered moving to F12, which I do run on my laptop (4GB Lenovo T400s) and my netbook, but I have 3 monitors using to Nvidia cards on my desktop and there was bug that broke support for multiple monitors with the closed source nvidia drivers that made F12 no work on my desktop. There has since been a patch but incidents like that and moving to Thunderbird 3 before it was a stable release with no option to downgrade to Thunderbird 2 without do my own install/config/setup. Made me seriously reconsider Fedora for my workstation. I understand the need to push the envelop with the latest tech in Fedora, but it hurts Fedora from becoming a techies workstation. At the end of the day I need to get work done and not fix problems with my OS.

It would be great if every other release of Fedora was a workstation release which addressed the issues of the previous version and was supported with critical bug / updates for 18 - 24 months. That's the distro I would happily use on all of my computers including my 2 servers (both of which run Debian Lenny).