${me:-whatever}

A lot like real excrement, it’s a lot harder to give shit away than it is to make it!

It’s hard to believe just how much work we put into being ready to give this software away. You see, when you write something for yourself –and remember that YaketyStats has been in use for 3 years or more in one form or another by my work– you can make all kinds of assumptions about the environment and you don’t have to document anything because if you have a question, you just ask a co-worker.

However when you give it away, first you have to yank out all of those assumptions. That takes a very long time because it’s a hard thing to think about. Confirmation bias ain’t just about new information, bubsie. While you’re tearing out assumptions, you’re making new bugs. You’re fixing old bugs that didn’t really matter to you, but would probably matter to others. Your bug fixes make new bugs. You start to have confidence issues about anyone wanting your crappy code anyway.

You have to make a website, and like all of the rest of this, it’s way harder than you’d think. You can go with the siloed antiques you know (Mediawiki and Bugzilla) or you can try something new on a platform you’re uneasy with (Jira/Redmine) and then there’s all of the time you spend vacillating in between. Don’t forget the time it takes to set it all up and tear it all down, to learn about the new platforms… you get the idea.

Once you pick your website stack, it’s time to document your every little thing about your project. You get to document how to install it, how to use it, why it does all the crazy crap that makes sense to your environment and possibly no one else’s. You word it, you re-word it. You wonder if you’re still making assumptions. You are making assumptions. You re-write it again. You re-organize it. The new organization is better, but you need to re-write some if it to fit the new organization.

Oh, and the whole time you’re doing all of this you’re wondering to yourself if you’re throwing a party which no one will attend.

Then the release date you thought for sure you’d make comes and goes and you make a ton of last minute changes. You drop your “announcements only” WordPress install in favor of using the semi-crappy News feature of Redmine. You re-organize all of the documentation again. Then you try to write a “press release” and ZOMG BBQ lemme tell ya — writing press releases is something that I am in no way equipped to do. So then you enlist the help of others. These others are much more tuned to writing this sort of copy, but they don’t really know your software or why it’s cool and different. You’re grateful in any case. This is the best that the blind leading the people who don’t know where we’re going can come up with:

Athens, GA (Oct. 14, 2008)

We’re happy to announce the WORLD-WIDE release of YaketyStats, a new tool for gathering and graphing statistics about your UNIX/Linux systems.

It has an intuitive, AJAX-powered web interface for graphing data. Graphs are built on the fly, can contain stats from multiple systems and include “Google Maps”-like dragging.

YaketyStats is easy to install, maintain, and extend and is free Open Source software licensed under the GPLv2. YaketyStats supports Firefox and is built on Perl, PHP, and RRDtool.

If you’re tired of having to micro-manage your existing stats system or you don’t have one, you should check out the YaketyStats website and demo video at yaketystats.org

I’ll pretend it’s tomorrow and publish this now.

Tags: linux, YaketyStats

First and foremost, I have no idea what scrollkeeper is or does, but whatever it’s doing: it’s doing it wrong.

I installed a couple of boxes yesterday that had modest hardware and they each spent several minutes running scrollkeeper-update at the end of the install and then they both got to run it again when I installed updates. And in case you’re unfamiliar, this is a process that consumes all of the CPU it can get. How can anyone find this acceptable?

Actually, I lied. I do know what it’s for and that’s the saddest part. It’s for indexing help documents. I guess since most people don’t like man pages, someone felt the need to write this system-crushing utility to index some help documents I’ll never ever read.

Wait, I’m wrong, that’s not the saddest part. The saddest part is that every distributions packaging of GNOME (I guess, based soley on the things that threaten to be removed if I try to uninstall scrollkeeper) make binary packages depend on this awful piece of software. If you want to use gnome-terminal (or whatever,) you’re stuck with scrollkeeper. What’s wrong with having a gnome-doc package or something?

I could guess about why it’s slow… I see those horrible letters “XML” some of the dependencies, so I could easily point at that, but of course it would only be speculation. While I’m speculating, I’ll go ahead and offer that I can only imagine that were I to use these help documents sans the helpful indexing of scrollkeeper, I’d actually spend less time waiting on my computer than I do with the help of scrollkeeper.

To establish some “cred” before my next amazing feat… er complaint, I’ll inform you that I’ve been using Linux since 1994. I’ve been a professional admin for 8 year or so. To top it off, I think I have a fairly good sense of humor. In fact, on more than one occasion I’ve had friends and even total strangers describe me as “hilarious.”

With that said, I feel fully qualified to say that User Friendly is about as funny as Sinbad which is to say: not. I’ve never laughed. I’ve never smiled. In fact the most positive response I’ve ever had was mild annoyance. I can only imagine that this comic has a following because people feel that they need to laugh to be part of some community that only exists in their own minds.

Linux magazines, hear my prayer. Stop syndicating this crap.

Tags: linux