Older blog entries for avriettea (starting at number 109)

19 Apr 2006 »

We all know those people who "escape" the business. The go on to some other profession where they don't make as much, but they claim to be happy.

That's fucking defeatist.

You know, this profession takes a lot out of people. And I'm not incredibly thrilled right now. What the fuck do you do? I'm not quitting, because I'm just not that weak. Where do you go from here?

I could go get some Unix admin or Perl hacker job, but you know, I've fucking done that. And I'm frankly fucking good at both. I don't want to step down from where I am and go do the same shit in perpetuity. What's the next thirty years going to be? More of the same old "fix the broken computer thingy", or "help the stupid customer understand their expensive hardware"?

If there were a word, like, ennui, combined with belligerent, that's about where I'd be. Fuck. You. World.

12 Apr 2006 »

After a year, more or less, of solid working, I am finally going on vacation. We'll be on Maui until the 25th, for our wedding anniversary (happy 1!).

Been borrowing a Nikon D70, and totally fell in love. I can't shoot Canon anymore. The Nikon just feels more real to me. So, I bought a D200, and in Maui we're going to be shooting both. Hopefully we'll have enough glass to get some nice pictures.

I love my wife.

23 Mar 2006 »

I'm not normally inclined to quote myself, but today is A Special Day.

Lastly, I'd like to mention it, because it's been bothering me some. Next time you're out in the car or otherwise in public, and you flip somebody off, cut them off, make a face at them, whatever.. Think for a second before doing so, "just how stable is this guy? I mean, is he going to flip out or anything?" Let me just say that you never know the guy you're tweaking there isn't psychotic. That is to say, he might be. And you might just be the little push he needs to go apeshit. And maybe, just maybe, he's got a trunk full of guns, and maybe he's had a long day. Certainly wouldn't be the first time it has happened, and you won't be the last, either. Just give it a thought.

I really wonder what compels us to treat eachother the way we do. It's not enough to tell somebody they're wrong. We frequently do so, and then remind them of it. And after we've reminded them of it, we remind them why we don't like it, and that we ourselves are superior -- we sure wouldn't have done that. Somehow, we need to re-visit the well of discontent and simultaneously claim the higher moral ground.

In addition to other events which I won't chronicle here, an example of this can be found at Wikipedia. It is hard to believe that a project with the openness of Wikipedia would have an institutionalized punishment body, and that users would repeatedly request that other users be punished. It's gone beyond dispute resolution in society. We don't just want to fix a problem or end a dispute, we want to bring people to justice. We want to have witness and victim testimony in the "penalty phase" of trials.

It all makes me sick. I think I've grown out of the phase where we bully people and make them eat dirt in the sandbox. Why hasn't anyone else?

25 Jan 2006 »

It seems strange to me that anyone who manages employees would not assume good faith in the actions of said employees. What reason has any employee to hurt the organization that pays them? Surely when that is the case, it is apparent to everyone around the employee. I would have thought this would be self-evident. That employees should not regularly have to defend their actions, but rather to explain what they've done. In situations where an outside party (be it internal to the company, or external) has issue with an employee, why would the "supervisor" favor the outside party over their employee? That can only breed (to borrow the USMC term) hate and discontent within the ranks. While the "slippery slope" is a logical fallacy, what other outcome could that have? Once the seeds of discontent are sown, how can anyone possibly undo that? And, having been done, what other outcome can there be but intractable productivity and morale problems?

In a similar vein, I probably shouldn't be surprised by this list of New Manager Mistakes. Nevertheless, I am surprised. Both because these "mistakes" all seem so glaringly obvious, and because I've gone through at least a few of these several times. What is most shocking about the About.com editorial is the statistic that "50% of new managers receive no training." How can a company do that to itself? The responsibility for productivity has to rest on the managers. Individual contributors, while contributing "work" and generating revenue, typically do not self-organize and self-manage. It takes management to help align employees with goals, and to offer guidance or assistance when the contributors need it. To poison that well, by creating managers who are unable to see the delineation, or their role in the process of getting work done, companies cause themselves to become less innovative, less productive, and frankly be doomed.

Somewhere along the way, "management" became a prize for being at a company for a long time, or making/having the right friends. In reality, isn't being a manager sort of similar to being handed a mop? "Here, go do some administrative stuff, and don't worry about contributing. Just make sure those who work for you contribute." Why on earth would that be something people would attack eachother over? The pay certainly isn't any better. The nature and description of the "manager" position has changed over time. The billet now attracts the exact wrong type of person for the position. Those who are power-hungry, abusive, or inept at doing their own job.

It isn't to say that all management is like this. It really tends to be upper management and the lowest tiers of management. How those "low tier" managers attain positions of "VP" or "Executive" is really beyond me. Further, how the very good managers don't move up in companies is also beyond me. I can only guess that the traits that make a good manager (primarily empathy and a healthy work ethic) are anathema to some of those higher-up positions.

Quite a few friends have told me recently that while startups don't pay especially well (and indeed sometimes do not pay at all), that the environment is much healthier, and individual contributors have more freedom than they do in large juggernautish organizations (Lockheed, GE, various DOD organizations, etc). One of these friends told me that the amount of freedom in a position is inversely proportional to the amount of compensation offered. I think the jury is still out on that, but anecdotal evidence seems to support her claim.

21 Dec 2005 »

Thank you, Wikimedia, for the Christmas Card. It's a nice touch. I find myself wondering, at the end of the calendar year, whether my donations are tax deductible. And, since I still believe in the cause, I added articles and did some copyediting and stuff, rather than just pour money into Florida. Or wherever the wikimedia foundation lives these days.

Taking some time off work this year, interesting phenomenon of having to "burn" vacation or "lose" it because I can't accrue more than six weeks. This is problematic because my wife only has two weeks of vacation, while I will have six for the foreseeable future (including the "burning" of vacation days). Perhaps I need to take vacations by myself? That seems weird.

11 Dec 2005 »

I have not posted in some time. Partially because I'm not sure what to post. I have been doing a lot of work in lots of areas that are corporate-confidential (as opposed to national-security-confidential), and can't much talk about them.

I have posted all the pictures from recent events (I/ITSEC in Orlando, Supercomputing 05 in Seattle, the ARM Developers' Conference) on flickr.

I also have some interesting ideas to bounce off the Linux community involving the ARM11 MPCore cpu, but haven't really come up with a good way to do it.

Primarily, I'm suffering a sinus infection presently, and doing a lot of reading.

5 Oct 2005 »

From the "hidden meanings" department:

I hesitated, a terrible feeling growing in me that whatever I did next it would be the wrong thing, the wrong thing entirely, and I would suffer for it forever more and worse still so might she, but eventually, while she rocked back and forth and moaned piteously to herself, I put my goblet down at my feet and got out of my seat and went to squat by her. I reached out one hand and placed it gently on her shoulder. She did not react. I let my hand go back and forth with her rocking, then slid my hand round her shoulders. Somehow, touching her like that, she suddenly seemed smaller than I had always thought her.

She did not seem to think I had committed any terrible transgression by touching her so, and, finding my courage and taking it by the scruff of its neck, I moved closer to her and put both my arms around her, holding her, slowly stopping her rocking, feeling the warmth of her body, and tasting the sweet air of her breath. She let me hold her.

25 Sep 2005 »

$17

Are you kidding?

hands me the drugs

Colin

Work is becoming something like 70 hours a week. Only I just figured out, I'm not counting time away from my wife, my home, and all things familiar. I'm in some fucking town I don't want to be, without an office, spending money I'm expected to meter, etc.

Oh, and I'm doing 3/4 time school. Whatever that means to you, it means 20 hours a week to me. So some 90-95 hours a week of stuff to do, frequently in some strange hotel room, in some strange city, and what is it worth?

It's worth nicotine, again, apparantly.

I'm going to see my fucking mountain tomorrow, and hopefully I won't wind up in the goddamn ER again.

"hate" is too unidirectional.

16 Sep 2005 »

So, what gives with all this embedded and realtime stuff? I've been at the Embedded Systems Conference in Boston this week, and I am really quite shocked at the fact that everybody I talked to about it -- outside the conference -- was also involved in, or at least watching the RTOS and embedded fields. In fact, I spoke with an old friend of mine who lives in the area and happens to work on a project involving airborne lasers (before you ask, apparently they are quite tired of the "frickin lazerbeams" jokes). It turns out that they're very interested in RTOS applications, and have been working with Linux on a 200mhz-ish Geode. I won't comment on the performance, other than to say it wasn't great. However, I am concerned that while the performance of CE is in fact quite good (7-10 microseconds), I have heard from people at the ESC that the code is something of a nightmare (the plan is to go into said code in detail when I get back to the office). It seems that many of the new devices we're basing technological progress on are having ever more complex "processing" requirements. One of the instructors at the conference said that a new BMW will have as many as 75 processors in it, and in fact has a data bus (that part I knew). Many of these devices are either safety-critical, or space/weight/heat constrained. So we start seeing operating systems everywhere. It's really quite fascinating, and when I took this job, I never expected that the embedded sector would be so active. However, it appears that I have a pretty interesting decade or so ahead of me.

So I noticed another thing at the event. I generally keep a pretty low profile, wear my badge out of the way, don't wear the company logo-wear, etc. It's not because I'm necessarily embarassed to work for Microsoft, it's just that it gets in the way of reasonable discourse for it to come up. I generally prefer to talk to somebody for a while, and only mention it if somebody actually asks me. People are generally surprised at it, and the question of "Well, so what brought a Unix guy to work at Microsoft?" Anyways, I was actually at a competitor's display at ESC, talking to them about the system they were demo'ing. They had a low-power PPC chip using a TOE to spit about 850mbit, using 40% CPU (that seems pretty crummy to me, but I don't know enough about the CPU they're using to really assess that). We got to talking about their (Linux) product, and some of the things they were doing, what they thought their roadmap was, etc. The guy makes a pretty obvious glance down at my badge (which I was wearing off my belt, rather than about my neck like most of the attendees). Mood immediately changed. What had been a lively conversation became sharp disinterest on his part. The guy became downright rude. Now, correct me if I'm wrong here, but isn't everybody at this event a competitor? I mean AMD is there, Intel is there, and they're not fighting with eachother. This sort of scenario was actually repeated a few times. It was pretty surprising for me. I can't understand why a Microsoft employee showing up at a Linux event would be a bad thing. People screech at us to not be such myopic, navel-gazing tyrants, and when somebody does actually come out and see what's going on in the community, this is a bad thing? Sigh. I suppose assholes are everywhere.

Good ESC speakers: David Kalinsky, Bruce Douglass from iLogix, and Celia Santander.

Kashmir indian restaurant on Newbury street in Boston is quite good. Tealuxe, also in the same neighborhood, has a great Yunnan tea.

Redmond next week (not going to make it to Heidelberg), followed by ARM conference in norcal, followed by another hike up to Redmond, and then some much needed vacation.

30 Aug 2005 »

I'm not generally one to belch forth two entries in a given day, but I suppose it merited mentioning, and the last entry was just so goddamn long, so this ought to get it off recentlog as well.

subversion

So after looking at the output from the init script, from apachectl, httpd, and everything else, I could not figure out what in the hell was wrong with the thing. It came up, ran successfully, and then completely barfed. Turns out (after some stracing, which I hadn't had time to do before), it was bitching about mod_rewrite. This is irksome, since it isn't mentioned in the httpd.conf, and it wasn't immediately clear to me why rewrite was being used. The error it was bitching about was not being to come up with a lock related to mutexes. "function not implemented". None of the mailing list archives on the web were particularly useful in debugging. A bunch of mandrake weenies were complaining that their kernel had broken apache which is, well, absurd. Somebody did suggest that apache had to be rebuilt. Since I still had the source sitting around, I checked to see whether my installed binary was the same one I had sitting in .libs. It was. Recompiled, and they were different. Interesting. So, install it, then rebuild and reinstall with DAV enabled, and voila, it works. I am guessing that within the year of uptime I had updated my libc or something which inadvertently broke mutexes for mod_rewrite, and a recompile fixed it. At any rate, the kernel itself didn't change.

development

What I really find interesting is that all the naming of individual components was influenced by Charles Stross books. It is strange to me that the evidence suggests that Mr. Stross' books motivate me to sit down and write something novel, like a new protocol. It would seem that upon finishing the book, I move on, and abandon the project. Some of the stuff is pretty cool (for example, using YAML on the wire to abstract other protocols, which is a frequent goal of mine), so I'm not real clear on why I abandoned it to begin with, although I do recall being short on time.

So I've been thinking of bringing some of them back up again, and actually have the yaml project in particular working again. Unfortunately, most of my documentation is based upon the specifics of what it was doing, not what I was planning on doing with it. So I kind of have to figure out a use for software I've written.

Kind of exciting, though. Almost like I had left some presents behind for myself.

100 older entries...