short news

June 30, 2005

1. HR told me that the probationary period is not a requirement and can be negotiated and that this job would not be a lateral transfer and therefore I can negotiate for salary. Therefore, I have started
to negotiate. 🙂

2. apparently I confuse eyedrops with phones.

3. icon has too much time on his hands.

So all in all, some good news, sorta 🙂

The last 7-8 months of my life have been complex. I’ve been involved in a long, somewhat arduous and definitely ridiculous process with duke about my transferring into another position at oit doing solely the linux@duke work that I’ve already been doing in addition to my current job in the physics department. Most of the time I’ve been on hold waiting for a decision to be made about whether or not they would be creating the position. Then, waiting for the position to be posted. Then, having an interview. Finally, waiting to find out if I would be offered the position created for a job I currently perform.

Well, today I found out. I was offered the position at oit. I was offered no additional salary despite the fact that I would lose the private office I currently have; the flexibility and latitude I have with regard to machine and service deployments and; other, minor benefits. Then, to top it all off I’ve been told that I’ll be on a required 90-day probation period – a time within which I can be fired without cause, warning or justification. Now I went through one of these periods when I first started working at duke almost 6 yrs ago. (sept 1999) but I didn’t figure I’d need to go through one again.

To be fair, I’ve been told that the probationary period is a requirement of HR at Duke and that there is nothing that can be done. Moreover, that for a lateral transfer I cannot be given a raise, nor a salary decrease. I’ve contacted HR to get confirmation on this and to hear about my options. I’m still waiting to hear back from them. I’ll call again tomorrow. If this is an HR rule I’m going to look into how rules can be bent. Then I’m going to ask why it is that a person not working for Duke has more rights of negotiation than someone who is already employed here. It seems odd to me that I could potentially negotiate for a pay increase if I weren’t working for Duke. Very odd.

The pay rate isn’t what irritates me, interestingly (though it could be what would appease me). It’s that I’ve worked here for 6 years; I do not think I’ve received anything lower than ‘Exceeds Expectations’ on my yearly performance evaluations; to top it all off I received a Meritorious service award from the University for all the stuff I managed to get done in the last 5 years; yet if I take this position I could still be fired, without cause, for the first three months I’m in the position. That bothers me.

The paranoid in me says, “this is just a scam, you take the job, you’ll get pink-slipped in the first few days and you’ll have no job at all and no legal redress.”

The part of me that doesn’t need any psychiatric drugs (:-D) says, “nah, it’s normal routine, everyone has to go through it, you’re not special. You just need more sleep, you’re being crazy.” The trouble is that I do not entirely believe either part of my own thinking.

Icon raised a good point today: I often say that one should not attribute to malice what can be adequately explained by ignorance or incompetence. This is true. But when you have past evidence of malice wouldn’t you be remiss to discount it? That’s the question. Is the evidence of malice great enough for my paranoia to win out? Maybe even more importantly do I want to work for an organization where I have to ask myself these questions before I take a position with them? The answer is, sadly, maybe.

Working in physics is not a bad job. Heck, by most standards it is an excellent job. It is flexible, a great number of the people I work with are quite smart and reasonable. I get to play with a number of new toys every few years. It is, however, a job I’ve already done. I’ve done it multiple times, ultimately. Hardware and software cycle many, many times. Is it worthwhile for me to continue doing it? Would it just become a place where I wait around for a year or so and look for something new? Where’s the challenge? Where’s the drive?

To answer my own question: This could be what open source contribution is for. Have a job that doesn’t challenge you but it does pay the bills? Then do your job and when you get home energize yourself by contributing positively to open source development groups/teams/communities. Alternatively, maybe I should work a job where I work a normal schedule and call it quits at the end of the day. Where I come home and do whatever it is I feel like doing rather than doing more work.

Then let’s look at what the motives are for changing jobs?

  • I get to work in a cubicle. Great!
  • I get to ask for permission before trying out new ideas. Great!
  • I potentially get hardware for the linux@duke servers that desperately need it. This would be nice, however, it isn’t a selling point for ME, it’s a selling point for linux@duke.
  • I, potentially, get to play and organize a bigger set of people for linux purposes on campus. That could be cool. Though I’m not sure how enticing that is given my work with Yum, Fedora and to an extent CentOS in much larger development and user communities than Duke could ever offer.
  • I could have fewer operational responsibilities which could mean fewer on-call times and less things weighing on my shoulders. This could mean more of a life and a chance to try out things I never have. The question here is, given that time, would I take it?
  • I could have more opportunities to bash some people in the skull and help fix the morass of problems that have snarled duke’s IT infrastructure for so long.
  • Potentially implement a massive vision to organize and coordinate universities in linux/open source development and make the world safe. Okay, so this is kinda a pipe dream but hell, why not, right?

Sadly, some of it probably comes down to money. I might be willing to suck it up for a lot of the problems I’ve explained above if it wasn’t a zero-sum-game with regard to salary and benefits. I’ve got obligations and responsibilities to folks just like everyone else. I’ve found that some of those become easier with more money. Depressing, isn’t it? Though I’m reasonably certain there are jobs I’d be willing to take for much less money. Though, at the moment, I don’t know what that would look like.

Right now, I think I’ll wait and talk to HR and find out what options I have.

To say the least I’m feeling kinda burnt-out and I am wondering if my best interest doesn’t lie elsewhere. That maybe I should walk away while I still have some whispering shreds of an un-embittered personality left. I remember when I first got here. I had a lot more hope and I was a much nicer guy. I was also 6 years younger. It might just be I’m getting older and grumpier. I don’t want to believe that, yet.

I’ve prattled on for a long time and I think writing this has helped me see things. I’m in a lot better position than a lot of people. I get to pick from two jobs. In normal circumstances both are reasonably well-paying and stable. That’s a lot more than a lot of folks can say these days and I’m not ignorant of that distinction. But as in most situations you’ll get the most or least out of life from the choices you make. I’m just trying to work out which choices are best for me and this is one good way to banter back and forth with myself.

I welcome advice (within reason :)) if you cannot find my email address via google you’re not trying very hard.

Fedora retrospective

June 16, 2005

Back in October of 2004 I wrote an entry about FC4 and what needed to happen to make fedora a better place to be involved with. By a freak click of the mouse I read through that entry today and was both pleased and unhappy with the results so far.

Let’s look at a few of the items and see how we’ve done:

  • communication: the steering committees for a variety of fedora projects/subprojects meet and talk and from what I can see actual results are coming out as a result. That’s a pretty good improvement from 8 months ago.
  • leadership: We’re not quite there yet but, as always happens, a number of defacto leaders have emerged from the fray. Greg has been providing the most cohesiveness and consistency for fedora overall and Karsten has given much-needed focus to the docs project. However, we still need to move further and have more community leaders. We may find that some of that emerges as the details of the fedora foundation become more clearly defined.
  • update posting and notification: Sadly we’ve not gotten to this one. Happily there is an intern at red hat who has been tasked with this and is actively working on the bits to make this happen this summer. I’ve been discussing storage of this metadata in the repodata with Paul and Peter and I think if I can get some hours to focus on it we can come up with a workable solution.
  • Fedora Extras: Damn it that didn’t occur. We went from manual builds and a forced march to functionality to almost real viability and functionality. We’re no longer relying on manual builds, we’ve got tons of checkins happening every day and we’re acquiring new packagers at a reasonably good pace. We’ve got a lot of bugs to shake out of the system, yet, but I think that will come in time. There are people inside and outside of red hat who are noticing fedora extras and realizing: “Hey, this doesn’t entirely suck.” We still have a number of detractors and we need to deal with some of their concerns to make it work but at this point the limitations seem to be almost exclusively technical save a few organizational issues. There’s very little hold up for other reasons.
  • Mirroring: Not much changed – there MIGHT be some things on the horizon – we’ll see
  • Package metadata: Well we got rid of 2 of them: rpmdb-fedora and the yum .hdr files. Yay us. We still have comps.rpm to go (DIE DIE DIE) and the hdlists. That’s a big improvement as opposed to what used to be shipped around. We also now have set of tools to do more with the metadata we have. Repoquery in the yum-utils package (now in fedora extras) allows some pretty arbitrary queries of the data.
  • Technical priorities: Not quite there yet but I think the next few weeks of FC5 pre-devel will be valuable for shaking out what the goals really are. On the plus side Josh has compiled this list and that provides some more focus.

So it’s kinda cool – we made some movement on some of the issues and in other cases we’ve made progress on items that will let us move on the issues. In almost the almost 8 months since FC3 came out we’ve come a long way. Fedora is not as much scoffed at by folks when someone claims it to be a community-involved project. That, to me, is kinda a big deal.

Maybe I need to think about what FC5 holds and work on another entry like the last one. Maybe someone else should, too.

Test post from Drivel

June 14, 2005

It seems to work – but balks about an unknown method using the MetaweblogAPI in pyblosxom. I think at some point I need to check the set of methods available in that code and figure out which one its missing.

Yum Upgrade Faq

June 13, 2005

I added a Yum Upgrade Faq to the fedora wiki. It just helps document how to go about upgrading from release to release via yum and the common problems that occur. If you find items that you think should be added there either add them yourself if you have Edit access or send them to me.

I’ve done 3 machines from FC3 to FC4 now and they seem to work reasonably well.

mock 0.3

June 12, 2005

Well mock 0.3 came out in a hurry. Toshio Kuratomi was kind enough to point out that any member of the mock group could run whatever they wanted on the local system, as root. This was due to mock-helper having an ‘env’ command defined for some odd reason. It was a leftover from mach-helper and not needed anymore for mock. So, I removed it and made a new release post-haste.

Many apologies for not noticing it to begin with and thanks to Toshio for being so kind as to point it out.

mock and yum-utils should be hitting fedora-extras tonight sometime.


June 10, 2005

Do you know what sucks? Vertigo sucks. My mom gets it during stress and I have it too. What this means is that your inner ear, which determines balance is off. It relays information that says: “Yes, we’re sitting firmly on the wall while the rest of the room spins”. Last night it got pretty bad. I laid down, enjoyed the room rotating in a loving manner then a couple of loops while I put one foot down to try to keep from throwing up. Sometimes it’s not all that bad. It’s a good indicator when I’m stressed out and need to relax some but sometimes there’s nothing I can do about the stress so it just gets in the way. Last night it made me go to bed which helped. We’ll see how well I fare today.

round 2

June 9, 2005

Today was markedly better than before. The meetings in the morning didn’t make me think about creative ways to kill myself or others and that’s always a plus. I was greeted by what could only be consider extremely positive news about the power outages scheduled for this weekend and then the interviews in the afternoon didn’t cause obvious, excruciating pain. Still not sure what I’ll do if I’m offered that job. There’s neat things that could happen working there and there are some extremely crappy things that could happen, too. Gotta love it. Either way duke’s going to be looking for a senior linux administrator/programmer. It’s just a matter of where they end up sitting.

Bought a new moleskine today while out to dinner. I like those things a lot. This is one of the bigger ones, gonna try to keep it on my desk at work for making longer lists and whatnot. We’ll see how that works out. I’m flipping into the second pocket-sized moleskine but this one has graph-paper pages rather than normal ruled pages. They’re a touch smaller lines so I’m curious what impact it will have on my handwriting. Can’t make it much worse, I hope.

Bad Bad Bad

June 7, 2005

Today I had my first job interview for the linux@duke position that was created at oit. I didn’t feel that the interview went well. There was no technical discussion. It was all a discussion of whether or not I’d be capable of dealing with the folks at oit and some of their work perspectives. Despite that my resume and these people’s personal knowledge of me should have shown them that I’ve been able to get a diversity of people to do what I think is a reasonably impressive set of results.

It’s just weird. I always thought that one of the points of achieving things and getting results was that it could be used as evidence on your behalf.

Very saddening meeting.

I guess we’ll see how tomorrow fares.


June 4, 2005

So we made the first release of yum-utils today. It’s 0.2 and it’s available in yum’s download dir. It’s got a great set of tools in it and I think those people who have ever complained to me about some crack^Wadvanced feature yum needed that I balked at will find these extremely handy. From the readme:

— Introduction —

Yum-utils is a collection of utilities, plugins and examples related to the yum package manager.

If you encounter any problems using these utilities feel free to send an e-mail to the yum mailing list:

If you have created an utility related to yum and you’d like to look into getting it included into this set please e-mail the yum-devel mailing list:

— Description of the utilities —


Check for unneeded packages and dependency problems in the system. (also cleans up old kernels)


Check for dependency problems in repositories.


Look up oldest or newest packages in a directory. Can be used for cleaning up repositories for example.


Query packages and groups in repositories similarly to rpmquery.


Generate RSS feed from repositories.


Install build dependencies of source RPMS.


Download packages (and optionally their dependencies) to arbitrary directories without installing them.

— Authors —

Various people have made a contribution to this collection:

– Gijs Hollestelle

Maintainer of yum-utils, author of yumdownloader and package-cleanup

– Seth Vidal

Maintainer of yum, author of repoclosure, repomanage and repo-rss

– Panu Matilainen

Author of repoquery, yum-builddep

– Sean Dilda

Author of the update on boot init scripts

So I’m clearly kinda happy about this and I hope yum-utils will show up in fedora extras real soon now.