Jun 30, 2007
Handy Dandy Railroad #9 (Porter 1941, ex-USN #9) steams around the Denton Airpark earlier today. This is the first locomotive I've seen in a week, so it was going to be the railroad photo of the day no matter whether I saw another locomotive or not.
—orc Sat Jun 30 20:54:07 2007
Jun 29, 2007
As part of the great Mastodon reanimation project of 2007, I've been going back and digging out some of the code that went into earlier versions of the distribution. A lot of this code was stored inside a proprietary source code control system that used to have a free version which was withdrawn when the company went into a screaming hissy fit about someone reverse engineering the wire protocol it used, so one of the prerequisited for the nouveau mastodonte was to find a new SCCS, move all of the old archives over to it, and port the SCCS itself back to SLS Linux.
After short consideration, I chose Linus Torvald's Git package (because it was written, more or less, in C, and because it supported decentralized repositories, which I grew fond of when I was using the hissy fit control system several years ago.) Git, as befitting a Linux program, isn't exactly what you'd call portable, unless by "portable" you mean Redhat, Debian, and MacOS X; I had to slightly tweak it to make it compile on FreeBSD 4.8 & 4.4, and I had to extensively tweak it to make it work on Mastodon and the SLS system that Pell still runs.
I did the first rounds of tweaks by hand, but only as a training run. My current version of git 1.5.2.2 (which I will send back to the maintainers, but I don't expect that it will be accepted) has had most of the autoconf/edit-10,000-variables-in-the-makefile "configuration" code ripped out and replaced by a configure.sh script, which is not fully tested, but now builds the code on MacOS 10 (where it passes all the tests) and Pell (where it doesn't, because the test suite wants to be able to use trap - exit, which the fairly old version of b*sh there doesn't understand (it appears to be able to handle pull/push/clone/commit, which covers most of the stuff I want it to be able to do for now; the tests not working properly is annoying, and if they bother me too much I'll tweak that part of the tests so they won't die with missing features.)
It's a patch, and you patch and make my slightly-more-portable git by doing:
cd git-1.5.2.2
patch -p1 < git-portability.patch
chmod +x configure.sh
./configure.sh
make
I don't know if any of the fancier transports work on older Unices, or if any of the code that uses poll() [implemented as a wrapper around select()] works, but I can push changes
over to pell, pull them back out, and clone packages into and off of the system. So it counts as New Code!™, at least up until the point where Git migrates out from under it.
—orc Fri Jun 29 21:40:53 2007
Jun 28, 2007
Yr. humble narrator, caught on the wrong side of a computer display.
—orc Thu Jun 28 07:30:03 2007
Jun 27, 2007
In the last episode of the saga of my descent into mac depravity, I mentioned that it might be nice to have MacOS 10 running on my Toshiba notebook, because the Toshiba has got a few features that my macbook doesn't have, such as
- multi-button mice
- a twiddle-stick mouse
- more USB ports
- a MMC/SD port for easy picture snarfing from my Pentax
- a less-sucky keyboard
So, since I've got a working wireless connection here at the beach, I decided to look online to see what the search engines could bring up. I found a few dozen pages that claim to describe how to install macOS on cheap PC hardware, which was no surprise, but before I ran into those pages I found some pages by dedicated macfanatics where the ridiculous thesis was put forth that Apple would never sell macOS separately because nobody can make any money selling operating systems without bundling them with hardware.
Okay, that makes sense. I've certainly never seen any company make money by selling operating system software by itself. It's not as if the richest family in the world got there on the back of such a company, either.
Hmm. Maybe not. But as far as I know Mr. Gates doesn't wear black turtlenecks, so he doesn't count.
—orc Wed Jun 27 09:37:49 2007
I still loathe loathe loathe the one-button mouse and wish the machines came with a proper twiddle-stick mouse, but there's something to be said for having a Unix machine with a functional user interface, even if it is a Unix box that sticks firmly to the delusional belief that the major problem with Linux is that it's got too much backwards compatability.
I wonder if I could put MacOS on my Toshiba notebook?
—orc Wed Jun 27 09:06:19 2007
Jun 26, 2007
(This is a variant of the earthy-crunchy biscotti recipe I developed several years ago. Why am I posting it, you might ask? Well, I forgot to bring my recipe book along for this holiday, and had to find my basic biscotti recipe by looking it up via g**gl*. So I'll post this one to the web as well, and that way it will live forever.)
Ocean Isle Beach biscotti
- Start heating the oven to 350°F.
- Mix
- Stir in
- 1.5 teaspoon orange extract
- ~1/2 teaspoon ground black pepper.
- 1 cup sugar
- Sift in
- 2 cups all-purpose white flour
- 1 teaspoon baking powder
- 1/4th teaspoon salt
- After mixing all of this into a sticky doughball, mix in
- Somewhere in the ballpark of 1/2 cup pecans (this is the South we're talking about, after all)
- Somewhere in the ballpark of 1/2 cup finely chopped candied ginger
- Somewhere in the ballpark of 1/3rd cup raisins
- Grease a cookie sheet
- Form the dough into a 3 inch diameter cylinder
- Put the dough on the cookie sheet, then flatten it down to about 1 inch high
- Cook it for ~35 minutes
- Let cool (on a cooling rack) for 15 minutes
- Cut into thin slices
- set the slices back on the cookie sheet, then place them back into the oven for 15-20 minutes
at 325°F.
- Set the finished biscotti on the cooling rack to cool, provided your friends and relations leave them alone long enough for you to get them over to the cooling rack.
—orc Tue Jun 26 20:11:03 2007
Jun 24, 2007
A little sea anemone that was clinging to a fragment of a plastic coke bottle before we captured it (we returned it to the briny deep after everyone had a chance to look at it.)
—orc Sun Jun 24 21:39:24 2007
Have I complained about the FSF ``configure'' program before? Of course I have, because it's a classic example of a program where the developers have completely and utterly forgotten what they were originally trying to solve. But it's like a clown car of stupidity, and the "what were they *thinking*??" moments just don't stop coming.
Today, I was trying to build a pdp11 emulator (as part of Mastodon, really; since I'm trying to keep the core of Mastodon nice and tiny, what better to use as a reference than a v7 disk image from tuhs?) and discovered, to my intense dismay, that it used the FSF configure program and didn't just come with a makefile. But if I was going to compile the program, I really didn't have any choice but to fire up configure and see what it would complain about, under the assumption that I could work around most of the "I don't like your libraries!" glitches.
Well, it didn't complain about libraries. No, it complained about the system type; apparently it isn't enough that MacOS 10 is freebsd, and that it will compile most any regular old Unix program out there (up to and including levee, which does not require any of the super-modern incompatability crap that's been stitched onto Unix like an ungainly frankenstein goiter.
And when configure couldn't find the system type, did it wade ahead and try to find out what things were actually provided by the system so you could go ahead and make something despite it not being a blessed by configure type of system?
No, it did not. Configure just complained and quit.
Reeeeeeeeaaaaallly useful for an autoconfigure program.
I can expect that configure might not be happy with Mastodon, because Mastodon is a Linux, but it's a Linux that's not encrusted with all the FSF bloatware that all the commercial Linuxes have, but Darwin? Darwin that ships with EVERY SINGLE ia32 mac box, and can also be found on many many MANY PPC mac boxes? Sheesh. That's not an autoconfigure program, it's an iron maiden with bright paint and smiley face decals.
—orc Sun Jun 24 16:25:34 2007
Jun 23, 2007
Sample dialog from our trip from Raleigh down to Wilmington:
(me) Oh, look, a couple of engines!
(everyone else) That's nice.
(me) We should stop so I can get a picture of them!
(everyone else) [rude derisive laughter as we drive right past the railroad crossing, leaving me with this single blurry picture of a GE widehood nose.]
But we did end up stopping to get barbeque, so I guess it was a fair exchange.
—orc Sat Jun 23 20:31:55 2007
We loaded up all of the computing machinery we thought we might need (and then some!) before we headed out east for our holiday. But there was one thing we forgot to arrange, and that was net connectivity of any sort.
Fortunately wireless networks have become popular, and quite a few of the ones that are set up are deliberately left open so that nearby neighbors can piggyback onto the connection before they (tardily) call the local vacation isp and get a modem connection for a week.
(And, since this is North Carolina, they've got barbeque. The barbeque we can get in Portland is pretty good, but it's not even close to being in the same league as some of the little greasy spoons. BBQ plus considerate wireless networks is a very nice thing to be greeted with after a 2500+ mile airplane flight across the United States.)
—orc Sat Jun 23 20:18:16 2007
Aside from finding the occasional dead poisonous thing, the bears spent most of their day running around playing on the beach and wearing themselves out to the point we had to all but sit on them before we could force them into the arms of somnia.
Many pictures were taken, but you only get to have seven of them inflicted on you.
Today.
—orc Sat Jun 23 20:10:42 2007
Russell found a silver dollar sized black widow corpse between bouts of getting soaked to the skin in the Atlantic Ocean today.
—orc Sat Jun 23 17:41:41 2007
Jun 22, 2007
... because if I had, I would have felt obliged to use the nasty little white earbud headphones that came with it, and my attempts to be cool would have been rudely interrupted by my hurling the iPod to the floor, then attacking the remains with a 10 pound sledge.
In my listening tests of ogg vorbis vs. aac, aac-encoded music sounded a little bit too etched for my tastes; it had a somewhat better soundstage, but that soundstage is damaged by the razor sharp high-frequency ringing that drapes over every treble note. It's certainly not a fatal problem with aac; good speakers bring out enough of the rest of the music so that the only thing I notice is a bit of fatigue after listening to an album or two, and when I'm carrying the iPod on holiday, I'm not going to be doing much on the order of critical listening when I crank the Mekons as background music for an evening of Mastodon hacking at the beach. But if I put in the nasty earbud headphones, about all I get out of the music is vague muddled baselines, heavily colo(u)red midranges, and über-sharp treble spikes being hammered into my eardrums.
The iPod is a superbly designed little computer (all it's missing is a network connection for me to buy a bunch of them and use them as super-tiny super-quiet servers with groovy front panels,) but the earbuds sound like someone at Apple was shopping at Fast Eddy's back-of-the-van stereo mart (a more charitable guess is that every single marketing person at apple have their ears rolling off at 15000hz, and people like me who can still hear sounds up to 23000hz [the only sign of high frequency degradation is that I've developed a notch from ~21000hz to ~21750hz. I can still hear the scream of dying flyback transformers, but go much below that and there's this mysterious notch] are just shit out of luck in Appletown, but you'd think that some of their beta testers might have pointed out this misfeature.
—orc Fri Jun 22 23:33:37 2007
Since we're going to be spending the rest of the day in airplanes and airports, it's not very likely that we'll see any (resolvable) trolleys while we're doing it (and we're going through Houston, but, alas, the Houston airport != where the trolley line is.) So the trolley picture of the day is the only trolley picture I managed to get as we zipped towards PDX.
—orc Fri Jun 22 11:09:45 2007
The bears watch a flight pushing back from our gate at the Portland airport.
—orc Fri Jun 22 10:50:29 2007
Postoffice has been pushed up to version 1.4.2 with the correction of a couple of silly bugs that stopped it from functioning properly on an am64 version of ubuntu and stopped it from configuring on MacOS 10 (I did an overfancy method to detect whether BIND8COMPAT was needed, and as a result I could no longer find the resolver on macos. For 1.4.2, I prototyped the change on a MacOS 10 machine, so it's more likely to actually work now.)
I'd be more detailed about the changes, but it's 3am and I've got a plane to catch tomorrow at noon (which means awake at 8, to the airport at 9, then a two hour performance of American Security Theatre to entertain us before they let us onto the airplane. Be thankful this release announcement wasn't a mpeg of me pounding a beer bottle against a table while grunting Oook! Oook!
—orc Fri Jun 22 03:11:12 2007
When you're packing for a holiday, don't forget the Dust Mite!
—orc Fri Jun 22 01:58:49 2007
Jun 21, 2007
A Yellow Menace SD70 and an ex-SP tunnel motor scurry south at 17th and Powell.
I've discovered that I can focus my Lensbaby by simply wrapping my left hand around the body of the camera and using my index and middle fingers to adjust the focus. This leaves my right hand free to twiddle settings and press the shutter release, with the only problem being that the pressure of my fingers tends to roll the focus off towards my left. This can be a problem for portraits, but when a train is barrelling in from the left it works out nicely to have the holgavision™ be off on the right side of the print.
—orc Thu Jun 21 23:22:59 2007
Jun 20, 2007
It's not that I don't back up my home server. I do. But I back it up to another machine that sits right above it in the rack, so if my server catches fire the flames will zap the backup machine en passant. This has been pointed out as a somewhat suboptimal solution, so I've been working on a redesign of the home network (replace the two home servers with one machine that's got a couple of tiny [and quiet. And low-power -- the two mini-itx boxes I have right now pull a combined 95-97watts to drive the system disk, the two backup disks, and a couple of home disks. The single replacement machine consumes 19 watts] SATA drives and a flash disk for /, then move the old disk pack into the colo so that I've got full backups in two separate locations. But that's not going to happen for a while (the process of moving hardware into and out of the colo isn't difficult, but it's time consuming, particularly if I have to do an OS rev as part of the hardware shuffle) so as a short term solution I went out and bought a 160gb seagate USB disk as a temporary backup that I can drop into my purse and carry with me.
So I've got this USB disk, and I want to attach it to downbelow (running FreeBSD) to make sure it works. It starts making an annoying beep-beep-beep noise, the kernel complains about being unable to initialize the hub, and that's all that happens. Is the disk bad? No, of course not; if I plug it into my Windows laptop, it fires up and I can (after scrubbing off the point and drool backup manager they provided with it) format it and read and write to it without any problems. Ditto for the commercial Linux distribution I've got on my workstation at work. But just not the EPIA 5000 that FreeBSD is installed on.
Okay, so this means I don't have any *Unix* machines that will talk to this disk. So this would be a problem, no? Actually no; the Windows machine has a copy of Qemu on it, and I installed Mastodon on a Qemu partition earlier this week. So I plug the disk into the Windows machine, fire up Qemu, assign the disk to my Mastodon vm, fire it up, and Bob is my uncle.
It's pretty sick, but it works. And this will give me my offsite backup, even though it's in a way that will make many system administrators get sick to their stomaches.
—orc Wed Jun 20 20:22:21 2007
North carries a load of commuters up the hill.
—orc Wed Jun 20 18:46:07 2007
Amtrak twinkies #116 & #117 pop out from behind a warehouse and head across the Powell Street bridge as they approach the 75,000 grade crossings in downtown east Portland.
—orc Wed Jun 20 18:41:03 2007
At lunchtime today, I detoured through downtown to stuff a hardware watchdog into gehenna, and when I hopped on a bus to return to work, it ended up detouring around Robin Hood's barn to avoid construction. Part of Robin Hood's barn was along the trolley line, and we ended up playing tag with Orange/Yellow for quite a few blocks. It's not particularly easy to take pictures of a streetcar from inside a bus, particularly when traffic keeps moving, so every picture I got was somewhat blurry. One of them turned out to look kind of oil painty (the tiny one is deliberately reblurred to defeat the normal sharpening you get from reducing a photo,) so I thought it was neat enough to keep.
—orc Wed Jun 20 18:32:25 2007
Jun 19, 2007
Last week, as part of my plan to make my public servers a little lonelier, I went out and bought a pair of Berkshire Products PC watchdogs to wedge into them. The watchdogs I bought were the USB versions, which seemed like they'd be the easiest to cram into my FreeBSD boxes after doing a bit of tweaking to the existing (Linux) device drivers.
Well, it turned out that the Linux device drivers weren't really suited (they were doing, in kernelspace, reads and writes to the watchdog) for moving to FreeBSD (and they'll certainly not work on Pell, because 2.0.28 doesn't have support for USB,) so I had to sit down and write my own driver program.
It's not actually documented yet (yeah, yeah, what else is new? My documentation always lags my software development by at least one release), but it's got a -h flag that spits out the following help screen:
usage: pcwatchdog [-f device] [-h]
pcwatchdog [-f device] command{,command}
If not specified, pcwatchdog uses the device /dev/uhid0
the commands are:
ping -- ping the watchdog card, resetting the alarm clock.
temp -- read the pcwatchdog thermometer.
status -- read the current status of the pcwatchdog.
switches -- show the current settings on the dip switch bank.
version -- show the firmware version#.
armtime -- show (=set) the startup arming delay.
alarm -- show the remaining alarm time (=set the default alarm time).
trigger -- show (=0; clear) how many times the watchdog was been pinged.
enable -- turn the watchdog on.
disable -- turn the watchdog off.
relay -- show (=set) external relay parameters.
buzzer -- show (=set) buzzer parameters.
count -- show (=0; clear) how many times the computer has been reset.
pulse -- show (=configure) how long a reset lasts.
reset -- press the Big Red Switch.
the command ``pcwatchdog command=?'' returns a description of
the valid arguments for the command
Fortunately, the FreeBSD usb stack lets you use read() and write() to HID devices, so writing the driver was merely a case of spending my lunch hour writing a little front-end program that translates english into pcwatchdog command. And then I simply put an entry into crontab to run it on 2 minute headways and (modulo FreeBSD 4.4 not supporting read() and write() to HID devices) *poof* I've got a watchdog.
A mean-hearted person might notice that this watchdog program doesn't actually attempt to autodetect a Berkshire PC watchdog, but just blindly attempts to talk to the device. Well, that's because I couldn't find any simple way to do an ioctl that would tell me what USB device is associated with an open file descriptor. There's one that you can use to do a device walk, but it doesn't actually appear to do any mapping from usb address to usb device name, so it's enough to convert this whole process into a more-than-one-lunch-hour procedure, so I'm leaving that off until after I've got a chance to grovel through the kernel source looking at USB ioctls.
But as it stands it's perfectly usable for a server that doesn't see much in the way of plugging and unplugging USB hardware.
—orc Tue Jun 19 23:40:09 2007
The mastodon wiki is now populated and running. Now to start scraping up bits of source code so I can start regenerating the build environment.
—orc Tue Jun 19 11:35:44 2007
Jun 18, 2007
One of the new Trios (Orange/Yellow) lines up alongside one of the old(er) Astras (Red/Orange) at the Gibbs St temporary terminus of the "downtown" trolley line. (Given that it goes up to 23rd NW and down to the south end of Homer's Folly, it's not much of a downtown line anymore.
One nice thing about this picture (for me) is that is shows one easy spotting difference between the Astras and the Trios; the Astra has a little clerestory on top while the Trio has the roofline bumped up to the level of the air conditioner.
—orc Mon Jun 18 23:59:03 2007
... if I mention that it took approximately 1/100th as long to install Mastodon INST0067 on the Microsoft Windows version of Qemu+Kqemu as it did to install BETA0066 onto Virtual PC? (And INST0067 is a network-only installer, which worked just fine in a qemu environment. I had to use BETA0066 for the Virtual PC install because the only network card that Virtual PC gave me was an Intel e1000, which is, not too surprisingly, not supported by the 2.0.28+logo kernel that Mastodon uses. Qemu emulates an NE2000, which, since it's been around since the beginning of time, *is* supported by 2.0.28.)
Other things of note were that Microsoft Virtual PC pegged one of the processors on my laptop (== many extra watt/hours and a very loudly whirring fan) while the only time Qemu came close to pegging a processor was when I compiled Levee, Postoffice, and one of the fast greps from comp.unix.sources.
And neither of those emulators supported DMA on their emulated ide drives, so it wasn't a matter of DMA vs pio. Sure, Qemu was written by the scarily talented Fabrice Bellard, but surely Microsoft has some people who can club a virtualizer into running at reasonable speeds, no?
—orc Mon Jun 18 11:46:14 2007
Jun 17, 2007
After quite a few years where it laid dormant despite the best efforts of various contributers to make me get off my butt and start working on the next release, I've finally decided to make another run at updating Mastodon to the next release.
Do I have the infrastructure up and running?
No, I do not; the old build machine is lying on the basement floor serving as a spider hotel even as we speak.
Do I have an architecture yet?
No.
Do I have a development plan?
Nope.
How about a version control system?
Uh, uh.
Have I made *any* preparations?
Well, yes, I've got a hat.
A hat?
Yes, a hat. A Linux distribution hat. A hat with "mastodon.biz" on it. I got it at GoDaddy. And it lights up saying "mastodon.biz" in great big neon letters, so you can maintain the distribution after dark when it's less stroppy.
I see, I see.
It's true. The domain name is the very first thing I've set up for the new development effort. The only other stuff I've got for it is various ideas
that are running around in my head:
- It's going to use Postoffice as the MTA, of course.
- I'll probably use Levee as the vi de jour (nvi is fine, but bloated. The real vi [Gunnar Ritter version] is, of course, perfect, but it's fairly large and I didn't write it. Vim has syntax highlighting, and is thus the work of the devil. So that makes Levee the perfect choice.)
- I'm going to continue my quest of deGNUifying the system programs (tcc claims to be able to compile modern Linux kernels, and the thought of tossing gcc out the window is very appealing.)
- The 2.6 kernel (the only way I can support modern devices.)
- Some bleeding-edge library. Either gl*bc or one of the dietlibcs.
- I may have to give up a.out support, because having to write a new
version of the 4.8 series libc that supports all of the ansiisms and
other evil crud that's worked its way into the modern so-called "standards"
will easily evaporate all of my spare time for the rest of my life.
- Firefox or one of the other cycle-eating web browsers that Linux is saddled with.
- And I guess I'll have to have a X11 that's more recent that version 3.3.6.
I'll probably stuff a wiki into mastodon.biz so people can pester me for stuff that makes the code useful to people who don't live in Chateau Chaos. But as for right now I've got a hat, and that should be enough.
—orc Sun Jun 17 20:43:24 2007
Configure has been rolled up to version 0.59 by the addition of a handful of new functions and a little more work on trying to get vpaths working.
The new functions are:
- AC_SCALAR_TYPES
- Defines the DWORD(32 bit), WORD(16 bit), and BYTE(8 bit) unsigned scalar types.
- AC_QUIET
- A prefix that makes a detect function run without any output.
- AC_CHECK_STRUCT
- Check to see if a structure exists
- AC_PROG_AWK
- Looks for awk (or one of its less reputable clones.)
- AC_CHECK_RESOLVER
- Check to see if the bind4 resolver library (or one of its less reputable successors) is present on the system.
configure.sh also creates a config.md script which attempts to encapsulate the process of creating directory trees during make install.
—orc Sun Jun 17 14:04:22 2007
Postoffice has been pushed up to version 1.4.1. 1.4.1 is primarily a cleanup release that finishes incomplete parts of 1.4.0, but it includes a few bugfixes and portability enhancements.
Postoffice 1.4.1 may compile on 64 bit machines running recent versions of Ubuntu linux, too. After I released 1.4.0, I got email from Wink Saville with a bug report about configure.sh not being able to detect the resolver library. After I fixed that (I was looking for the function res_query(), which doesn't seem to consistantly exist. The name res_query exists, but in many versions of resolv.h it's simply a #define to _resquery. I already had one problem with detecting the resolver (Darwin, as befitting a FreeBSD derivative, takes even more liberties with published interfaces than any Linux distribution would ever dream of. In this case you need to #define BIND_8_COMPAT before the resolver HEADER structure become visible [this bug was reported by Andres Saloman a few releases ago]) so I took this new bug as an excuse to put an AC_CHECK_RESOLVER function into configure.sh which attempts to both detect the existance of a resolver library and the Darwin BIND_8_COMPAT brokenness) he reported that the code was dumping core in listq() when it attempted to call strftime() without first #includeing time.h, so I went through the code and added all the prerequisite #include's that were needed to properly prototype various library functions.
It's not all bugfixes, though. I did manage to write the manpage for usermap(7), so some of the official new 1.4 features are actually documented these days.
—orc Sun Jun 17 13:35:45 2007
Jun 16, 2007
Levee has been pushed up to version 3.4p with the addition of autoconfiguration via configure.sh, better support for randomly sized terminals (Linux consoles, terminal windows on GUIs), and another round of code cleanup to make gcc less whiny about my (K&R) C code.
This code was released because I got a letter from Felipe Augusto van de Wiel, from the Debian Project, asking if I'd accept some patches and push the mainstream code up to 3.4p so that Debian could use a new version of the code instead of carrying a decade's worth of patches to the last published version. I've not gotten his patches yet (that's going to be version 3.4q), but after sending a "sure, send me the patches" reply, I decided on a whim to see if the existing code would compile on one of my FreeBSD boxes.
It would not, so I decided to correct that problem by cutting out the old edit 50,000 things in levee.h configuration scheme and replacing it with configure.sh. And while I was at it I decided I'd clean up most of the cases where new versions of gcc whined about perfectly valid K&R C code (register i? Baaaad, at least according to &^$%!!? gcc) and change termcap.i to ask the tty what size it is instead of depending on the system termcap library to return the correct values.
And now, after seven hours of programming, here's 3.4p. It builds on FreeBSD 4.8 and SLS linux (both a.out and elf), and it might even compile elsewhere. But, in any case, it's new code!
—orc Sat Jun 16 11:22:28 2007
Jun 15, 2007
Dust Mite is a considerate driver and always waits when people are using the crosswalk.
—orc Fri Jun 15 23:45:36 2007
I didn't really expect to see anything today. I had a picnic to go to right after work, and was supposed to get from work down to Sellwood Park by 5:30, and even with that I didn't think I'd see train #14 because it had started out 2hr late and had worked its way back to 2 and a half hours late when I last checked before leaving work. But I wanted to ride the #70 bus so I could get a few blocks closer to Sellwood Park than the #19 could get me, so when I left work I hopped on the first bus to come by (at, um, 5:09. Not good for a planned 5:30 arrival) and rode it to 21st and Powell, because it was about as fast as a connection as Milwaukie and Powell, and it did put me alongside the railroad for about two blocks, so if something interesting came by (perhaps a Yellow Menace freight, or a P&W transfer, or even a very late BN/SF transfer freight) I'd have a chance to see it.
And because I was hoping for something interesting, I had my Pentax in hand and I wasn't walking quite a fast as I otherwise would have. So after I came out from the underpass, had crossed over to the west side of 17th, and started to hear the rumble of a fast approaching train (from the south; if it had been approaching from the north I would have heard a chorus of toots as it went over the 75,000 grade crossings in East Portland,) I had enough time to fire up the camera and take a picture when the not-so-late-as-I-thought Coast Starlight blew past on the final leg of the trip to Union Station.
The combination of the bright afternoon sun and the threatening clouds covering the sky? Oh, that's just typical Portland weather when it's trying to decide whether it's the rainy season or not.
—orc Fri Jun 15 22:48:04 2007
Jun 14, 2007
Most of the downtown bridges frame the Ross Island tug as it pushes a load of Ross Island downstream.
—orc Thu Jun 14 23:33:23 2007
Every time I've taken the #70 bus to work in the past week or so, I've seen this piece of graffiti as the bus went around the curve between 17th and Powell. Today, I decided that I would go and try to get a picture of it before it was tagged over by someone else. It's a pretty spectacular piece of graffiti, but the big surprise was on the other wall, where the taggers have been very busy little vandals indeed:
It's not as tidy as GPK IOK, but it's impressive the way the big tags fight for your attention against the background of hundreds of tiny tags.
—orc Thu Jun 14 18:47:21 2007
A proposed constitutional ban on same-sex marriage was defeated today by a joint session of the [Massachusetts] Legislature by a vote of 45 to 151, eliminating any chance of getting it on the ballot in November 2008. At least 50 votes were needed to advance the measure.
.
.
.
Because fewer than 50 of the state's 200 lawmakers supported the amendment, it will not appear on the 2008 ballot, giving gay marriage advocates a major victory in their battle with social conservatives to keep same-sex marriage legal in Massachusetts.
(--the Boston Globe, via Pandagon)
And the bigots won't be able to make a run at stripping basic rights from people in Massachusetts until 2012, either. It's just unfair that they can't strip civil liberties from a class of people they personally disapprove of (or are trying frantically to pretend they are not part of. The intersection between the sets of right-wing fundamentalist preachers and pathetic closet cases is impressively large) for another five years.
Tch tch tch. How will the state of Massachusetts live with itself? Losing the respect of a bunch of hate-filled bigots must make everyone in the state so sad that they can hardly keep from falling down laughing in the middle of the street.
—orc Thu Jun 14 14:06:13 2007
Jun 13, 2007
Anti-war Senate Democrats Tuesday plotted a new showdown with US
President George W. Bush over
Iraq, but admitted they had erred by making supporters think they could end the war.
.
.
.
"We set the bar too high," he said, noting that under Senate rules, Democrats needed 60 votes in the 100 seat chamber to thwart Republican blocking tactics.
(--AFP article via Yahoo via Marisacat)
You know, this might have been a spiffy argument when the Democrats were in the minority and didn't have their grubby little fingers on the budget, but one of the spiffy things about being in the majority is that if you don't want the bill to see the floor the bill won't see the floor, and any blocking tactics are only as effective as the PR campaign behind them.
If the Democrats wanted to end the war, they'd defund it. The Pentagon has 3 months of money already budgetted, and that's ample time to fire all the parasitic mercenaries and withdraw from the country. The Republicans can do all the "blocking tactics" they might want, but if every bill that comes to the floor has had war funding surgically removed, none of those blocking tactics will put one penny of it back in.
And, furthermore, if there are any quisling Democrats in the Senate caucus, I'd think it would be much more productive to force them to show their true colors, so they could be tossed out on their ears in the 2008 primaries (this is assuming, of course, that they don't turn and run on the CfL ticket with the explicit approval of the Democratic leadership.)
But I guess that when it comes to the choice between acting as the majority party in the House and Senate, or grasping for that elusive 23% approval rating, the sensible thing (at least according to the pack of Evil Party toadies who claim to be the Washington press) is to lunge for the approval rating, because, hey, that keeps B*sh's approval rating above that of the rest of the government.
I'd like to meet the DLC consultant who managed to sell the Democratic caucus on the idea that repeated craven surrender (and whining about it) was the path to a lasting electoral majority. I'd have to shake their hand and complement them on their spectacular job of neutering the Democratic party, because there comes a point where you can only gaze in awe at how efficiently someone can teach a collection of (allegedly) grown-up people to stab themselves in the back so that their enemies won't tire their knife hands.
—orc Wed Jun 13 23:01:40 2007
A Yellow Menace GE helps me play around with my new lensbaby.
—orc Wed Jun 13 18:41:08 2007
baby_t.jpg "Silas and a selective-focus lens")
I bought a new lens today, and Silas was the first of the bears to come within range of the thing. As you can guess from the all-over soft focus, it's a manual focus lens and the battle between the lens, the camera, and my terrible vision results in this sort of artistic blur no matter what my preferences might be.
—orc Wed Jun 13 12:34:55 2007
Jun 12, 2007
Nothing like a 10 hour backbone outage to cut way back on the number of times the spam zombies batter themselves against my antispam barriers. Thanks, AT&T! Now if you could only do this without, um, losing your routers (normally a cold shudder of terror would ripple down my back when I see a traceroute going through sprintlink, but when the alternative is things coming to a grinding stop at the point where everything connected through to at&t, the sprint hopshuffle doesn't seem to bad anymore) and taking everything at pell off the net for long enough to have the big ISPs start either (a) bouncing or (b) sending "helpful" we've-not-sent-your-mail-yet-but-we'll-keep-trying! messages [I'm looking at you, google -- I know that sendmail isn't written in p*th*n, but someone in your PhD-laden Redmond-clone HQ should have already figured out that when a user sees the "helpful" warning message, all they can see is that their mail didn't get delivered so SOMETHING IS HORRIBLY WRONG!!!?!?!!™ (and do they send the complaints about this to the Googleborg? No, they do not. They send the complaints to me, and since they're my users I can't very well snap "if I got this mail, then you can safely assume that the network is functional again" at them)] I'd be, if not happy, at least not quite so homicidally disgruntled about the whole thing.
Sigh.
—orc Tue Jun 12 11:46:38 2007
Jun 11, 2007
The Coast Starlight will not operate on the Sacramento to Portland section of its route from June 9 through June 12. Union Pacific will be working on the stateline tunnel north of Dorris, CA, south of Klamath Falls. The entire railroad will be shut down, allowing UP to work on other projects on the closed segment simultaneously.
Train #11 (#1011 officially) will operate Sacramento to Los Angeles on June 10-12. Train #14 will operate Los Angeles to Sacramento on June 9-11. Coach equipment will operate Seattle-Portland to cover that portion of the Starlight schedule.
There will be no alternate transportation provided Portland to Sacramento.
(-- www.on-track-on-line.com)
Well, that might explain both the mysterious disappearance of yesterday's Coast Starlight and why the Yellow Menace had MOW trains as part of their afternoon traffic jam south of Brooklyn Yard on the 3rd. It's certainly more of an explanation than I've found anywhere else.
—orc Mon Jun 11 11:10:32 2007
Jun 10, 2007
At around 3:20 this afternoon, I decided that I'd take a little walk down to the Bybee overpass so I could see the 13 minutes late (according to the Amtrak website) northbound Coast Starlight zip by on the final leg of its trip into Portland. The sun had been popping into and out of the clouds all day, and when I reached the bridge it was out and shining brightly, so I thought that this might be a good day for taking more pictures of blue and silver engines.
It was around 3:30 and a couple of southbound trains were parked on the mainline leaving enough room for a northbound train to thread the needle between them, so I figured that I'd have about 10 minutes before the train came hurling into sight. I settled down and waited for it to appear.
And waited.
And waited.
Eventually, the Brooklyn Yard switchers (Y1309 [yellow] and Y1458 [espee grey]) came down past the bridge, turned around, then headed back into the yard.
But there was still no Coast Starlight. I dithered for a second, and saw the switchers coming back with a long line of empty container flats, so I crossed over to the other side of the bridge and took a few more pictures of them (the sun picked this moment to pop briefly out from behind the clouds, which is why the next picture looks a little less gloomy than the first one.)
The switchers parked on the yard lead for about 5 minutes, then pushed the flatcars back into the yard. It was around 4:15 now, and I had a batch of bread rising at home which needed to be punched down no later than 4:30, so I had to leave pretty much immediately. I went home, punched down the bread, wedged it into breadpans for the second rise, did a couple of errands, then, out of curiosity, decided to look at the online arrivals page to see just how late #14 was when it reached portland, and saw:
The scheduled arrival time is 3:40 pm. Presently, no further information is available. Please check back later for updated information.
Well, that explains why the train never showed up. I wonder if one of the twinkies lost a traction motor between Oregon City and Brooklyn Yard?
—orc Sun Jun 10 18:17:58 2007
Most of the driving we do is short-haul driving in town, mainly trips of 4 miles or less. This is very bad for our gas milage when compared to the EPA "52 mpg"; we average a little less than 40mph, because the Prius wastes a lot of gas heating up the engine block when we start out, and by the time the block has heated up we're done with the trip.
If not for car insurance rates, it would almost be worthwhile to buy or build an EV for our in-town trips and leave the prius in the hangar except for trips to the coast.
—orc Sun Jun 10 00:01:51 2007
Jun 09, 2007
Blue/Green sits at the Gibbs St. station on friday afternoon. There's nothing about the trolley that you've not seen before (I've only taken the "trolley from the Ross Island bridge" picture about 70,000 times so far,) but the picture clearly shows that the loop -- and the overhead wire --through condoworld south is connected to the rest of the line, and that the only thing separating the western track from the loop is a carefully placed porta-potty.
Now, I don't know about you, but if I was looking for a bathroom and the only one I could find was one that was being used as a stop block for modern streetcars, I'd suddenly find that holding it in for a while was not such a heroic effort.
—orc Sat Jun 9 23:31:10 2007
Postoffice has been pushed up to version 1.4.0 with the addition of usermap (personal alias files) support.
A usermap is simply a personal
alias file (formatted like the aliases(5) file) which is placed
in a home directory, and which is referred to by a usermap option
in postoffice.cf(5).
The usermap option is formatted as
pattern:target{,target};
the pattern is a shell-style wildcard, with the addition of using the
~ to match any valid user in the domain. A match is either an
alias or the special token ~/filename, which is the address of
a personal alias file. When a usermap is called,
postoffice will try each target, stopping
when it matches one.
For example, the usermap *-~:~/.alias,bounce will
match any mail address of the format something-user. If
it matches an address, postoffice will first
see if the address is in ~user/.alias, and then if it doesn't
find a match there it will map to the fixed address bounce.
There are also a couple of trivial bugfixes in 1.4.0
Postoffice wraps smtp sessions in several
layers of timeout. There were some cases where an alarm would fire when
the default signal catcher was running, so you'd end up with a core dump
and a crash warning for a completely normal timeout.
A second, and more annoying, bug was discovered when testing the
new usermap code. Postoffice is fairly relaxed
about the order of MAIL FROM and RCPT TO commands and
will accept them in any order as long as both of them have been
issued when a DATA command arrives. But some of the sendmail
filters (milters) I use are not so forgiving, and if they get a
RCPT TO prior to a MAIL FROM, they will freeze and lock
the mail session. This is bad programming, to say the least, but
it's broken in a sendmail-compatable fashion so it's not likely
that it will ever change. So I've crippled postoffice (if
built with --with-milter) to whine bitterly about RCPT TO
without a prior MAIL FROM.
A super-trivial bugfix is a tweak to configure.sh to test for
the existance of malloc.h so I can only include it on machines
that actually have it. This is a generalization of the OS_DARWIN support
that Andras Salamon contributed for 1.3.8c, and should make the code a
little bit more portable to other machines that use C compilers that blindly
follow the whims of the (break-all-of-the-)standards committees.
WARNING: This code is still very new, and may still need more work. It's running on Pell and Downbelow right now, but nowhere else, so use caution when playing with it. Oh, and it's not documented yet, except in this release note.
—orc Sat Jun 9 15:57:45 2007
Jun 08, 2007
The United States Navy isn't the only organization that has brought their ships to Portland for the Rose Festival; this tiny square-rigger and a mate have been sailing around the river for the past week, and doing it conveniently close to the Ross Island bridge and the #19 bus.
—orc Fri Jun 8 21:50:42 2007
Is a Dust Mite a collectable figure?
—orc Fri Jun 8 21:42:07 2007
After I stopped for a twinkie or two, I walked over to the Seven Corner New Seasons to pick up some peaches, plucots, and bananas. When I left the store, I heard the frantic tooting of a train heading northwards along the SPYellow Menace mainline through downtown, so I though I'd missed the last train for a while. But, no, when I reached Powell Street, I saw this train parked on the mainline, and by the time I'd crossed under the Powell St bridge, the train had started to move. So I had to take a picture, because otherwise how would I have my daily ration of yellow engines?
—orc Fri Jun 8 20:53:34 2007
_t.jpg "Amtrak #14 zips towards the 12th/Clinton/11th street crossings at ~5:15pm")
The northbound Coast Starlight, running about 90 minutes late, approaches the 12th/Clinton/11th street crossings tonight. It was running late enough so I could leave work and actually scout out a different place to take pictures from before the twinkies popped around the curve leading from Brooklyn Yard.
—orc Fri Jun 8 18:36:55 2007
Jun 05, 2007
When I change busses at Milwaukie & Powell, I walk north to the bus stop closest to the SPYellow Menace mainline so that if something interesting comes by I'll be in a good place to see it (and maybe take pictures of it.) Today, the interesting was a pair of AC45C's (this one and 5299), which were running light down to Brooklyn Yard. I'd just received a used Sigma 28-135 lens, so I was, for the first time in months, running around with a zoomable lens, so I took full advantage of it as the train approached and went by. Most of the pictures weren't anything special, but the sun peeked out from a cloud just as the Eng!s cleared 12th, so I could zoom in and get this picture of 5533 fractionally in the shade.
—orc Tue Jun 5 19:41:20 2007
Around 5:30 am yesterday, Gehenna (the machine that contains this weblog, as well as a half dozen other services) fell over dead. Unfortunately, yesterday morning was a dentist morning, so by the time I woke up it was too late to do anything about it until I finished with the dentist's appointment (at 11:30) and could go into the co-lo to see what happened. When I got into the co-lo, gehenna was well and thoroughly dead, to the point where all I saw on the console was a little bit of "hi! I'm a dead vga!" scramble at the top of the screen.
I hit the Big Red Switch (on this machine, the Big Red Switch is the power cord) and gehenna chirped happily back to life, with no indication that anything had gone wrong except for, um, having to have me reboot it.
I spent part of yesterday afternoon researching replacement hardware and pricing PC deadmans switches, because if the machine was going to fall over and leave
nothing but dead vga scramble, that meant that something was seriously wrong with the hardware. The deadmans switches were extra paranoia; as the hardware in Pell and Gehenna gets older, there are more and more (at convenient 100 day intervals, so the people at the co-lo won't completely forget who I am) mysterious failures that are "fixed" by a therapeutic whack at the reset switch, and I'd rather have this whack happen without having to go downtown and back.
Last night, at approximately midnight, Gehenna fell over dead again. I was preparing to go to sleep, so there was no way I'd throw on my clothes and scuttle down to the co-lo. So I left it until this morning, when I packed up a new motherboard (scavenged from the Glass PC) and a new power supply and shlepped off to the co-lo before going into work.
My plan was to just replace the motherboard and scuttle back to work. Shouldn't take more than 10 minutes, right?
Well, no. It took about 10 minutes to take the old parts out and put the new parts in, but then it took another 40 minutes of fighting with the kvm switch before I realized that it was dead (but dead in that sort of charming way that kvm switches die; it was relaying vga signals to the monitor, but other than that it was merely a quietly humming doorstop) and that I'd need to remove it before I could actually talk to Gehenna and get it to boot properly.
So it's dimly possible that Gehenna was "fine", but was just being killed by debris from a disintegrating kvm switch (it's happened before; I've got some systems at my day job which will not boot if their keyboards are connected to a kvm switch) and all I needed to do was to pull the horrible thing out and everything would work perfectly. But by the point I'd realized this I'd already swapped out everything except the disks, and I didn't want to unswap anything for fear it would die AGAIN and I'd have to do another run for the co-lo.
And the ME-6000 motherboard from the Glass PC has a bios option to restart after a power failure, so I can ask the people at the co-lo if they could pull the power cord before running down the next time (oh yes, there will be a next time. Even if I put in a peer to peer failover mechanism, there will still come a time when that code will fail and the bottom will fall out) to replace hardware or kernel.
<obsigh>And did I mention that today is the day I'm switching network providers to Qwest, and they started the day by, um, disconnecting the network without bothering to mail me any instructions about how to connect to the Qwest routers? *sigh* </obsigh>
—orc Tue Jun 5 10:23:15 2007
Jun 04, 2007
I went into the dentist's office this morning to successfully get three teeth repaired, and as a result my mouth has been aching all day. This means I've
been eating primarily soft foods, and given that the peaches are starting to come in, this means I've been eating lots of peaches. The last batch of peaches vanished down my maw at lunchtime, so I went past the big big-big-store after I left work so I could refill the peach bin at home. The busses were very crowded this afternoon, so I couldn't even get onto the #9 bus that I would normally take out to 21st, but had to cram myself onto a #19 and walk from Milwaukie and Powell.
I had my camera in hand, so when a fast-moving Yellow Menace freight blew into the 11th Ave crossing it didn't catch me completely unaware. (unlike the Coast Starlight [not pictured] which rudely zipped past when I was two blocks away from the 15th Ave footbridge -- I was crossing 16th when a flicker of movement caught my eye and the train flashed by in a streak of silver paint and double-decker windows. I actually pulled out the camera and made a couple of running steps towards 15th before I realized that at 40+mph there was no way I'd make the 200 feet to the next street before the train reached the 12th/11th street grade crossing.)
—orc Mon Jun 4 22:35:48 2007
Jun 03, 2007
As part of a long complex scheme I'm trying to work up to crack my spectacularly impressive needle phobia (and let me tell you it makes it *really* fun to go to the dentist; I'm pretty indifferent towards dental work, to the point of not even noticing root canal work [I had a molar detonate a few years ago, and once the dentist managed to get the needle into my mouth he was shocked to have the rest of the procedure, including pouring molten structural plastic into the tooth, go past like assembling furniture. "So, when am I going to have the root canal? Next session?" "Um, we just *did* it?" "Oh. Sorry I didn't panic over it."] but the initial application of anesthetic is spectacularly stressful) I decided that I would try to use tranquilizers and trance music to duplicate the single time I managed to get a shot without freaking out during my adult life (I was having a brain MRI to try to figure out a batch of numbness on my right side. It wasn't successful at finding anything, but after 30 minutes of lying in the MRI machine listening to it go BANG! BANG! BANG! KLUNK! KLUNK! WHAM! WHAM! WHAM! I was so relaxed that when they proposed to inject me with a bunch of indicator I was perfectly willing to let them poke holes in me and pour in all the indicator they wanted.)
So I decided I'd go out, buy an Apple iPod, pour all of my digitised music onto it, then listen to a carefully mixed batch of trance, industrial, and whalesong before and during the shot part of my next dentist's visit.
Simple enough, right? Well, not quite; when I digitized our ridiculously huge music collection a few years back, I did a lot of listening to various lossy music formats, and decided that Xiph.org's Ogg Vorbis format was the least horrible of the lossy formats (it was much better than the popular, and encumbered, .mp3 format. Since this was the dark days before the expiration of the .gif patents, the detail that it wasn't patent-tangled was a big plus, and at that time I expected that I'd be playing it on home-built stereo gear like the Ogg-O-Matic, so the format it used didn't matter that much.) This may have been a good idea at the time, but it meant that I had 20+ gigabytes of music that was in a format that most commercial music players (including, of course, the iPod) couldn't read.
So when I got a music player, I'd either have to find one that supported ogg vorbis (few and very far between. Ogg Vorbis is the beta format of the 2000s) or one that could be loaded with a different OS that could read the format. The iPod was fortunate enough to have two different alternative OSes available; Linux or the Rockbox player (which claimed to support Ogg Vorbis format). I've been making my living writing and maintaining open source software for about 13 years now, so I thought that with these two alternatives around I'd be able to find one that worked.
Well, iPodLinux was a bust. The 5th gen video iPod isn't supported (usb isn't supported? How the devil do you use the damned this if you can't get music to it?) But Rockbox claimed that it supported the 5th gen video iPod (the only thing I could find when we went out shopping today), so I thought I was set.
I pulled a copy of rockbox (actually two copies after the first one loaded and failed to work), used the handy installer to write it to the iPod, recoiled in horror at the thing they call a "user interface", found a couple of skins (not via the rockbox web site, which is not what I'd call intuitively designed), stuffed them onto the box, poked around enough to be comfortable with them, then dumped (s-l-o-w-l-y because I was going over the house wireless network) all of my .oggs over to the box.
After the oggs were dumped over there (I couldn't actually do anything while they were transferring, because the iPod was in "I'm a disk! I'm a disk! And you can't do anything with disks!" mode during this transfer) I poked around trying to build a playlist (with no success. The playlist maker said it was loading all of my files into a database, but after running a counter up to the number of files in the machine, it silently returned to the "create a playlist" menu item and did the same thing again when I selected it.
Okay, so I'd have to do some research on how to make that work. But I'd already been cooling my heels for quite a few hours waiting for it to load, so I was impatient to actually hear some music. So I used the file browser to pick one of my Alison Krauss album, selected one song, and sat back to listen to the music.
For about 10 seconds. And then there was a horrible SSSKKKRRRAANNKKKSSSRRRGGG sound and the iPod locked up like a drum, requiring a [menu][select] reboot to shock it back to life.
Ooooo-kay. That was bad. But I didn't know just what kind of bad it was, because rockbox had blanked the screen five seconds into play and I was doing [menu][select] on an iPod that looked as if it was a doornail.
After it restarted, I went and picked a different song, selected it, and 10 seconds later (after the screen had blanked) the exact same crashing sound happened and I was left with the corpse of my iPod.
So I looked at the rockbox bug tracking system, and found reports of .oggs crashing the system. But they were old .oggs, from a pre 1.0 version of libvorbis, and I'd encoded my files with libvorbis 1.0 (these encodings are perfectly happy to work with videolan and foobar2000, btw,) so that shouldn't be a problem, right?
So. I [menu][select]'ed the thing back to life, went into the file selector, picked a third song, and started twiddling with the controls to keep the screen lit so I could see wtf was going on. 10 seconds passed, the sky fell in, and a little "ogg decode error (something)" popped up on the screen.
Lovely.
Now, Ogg Vorbis might be the Beta format of the 2000s, but it's a darling of the Open Source®™© crowd, so you'd expect that one of the things that would be important to an open source®™© music player would be the ability to play that format? Well, um, no, apparently not. A second browse of the bugtrack database left me with the distinct impression that nobody involved with rockbox actually has any music encoded in the Ogg Vorbis format, and instead went to all this trouble to build an open source™®© player that poorly emulates the bog-standard iPod OS.
And it is a poor emulation. Where the Apple OS for the ipod has smoothly scrolling pages that roll left and right, this one has a series of text menus
that would be perfectly at home on a V7 Unix some 27 years ago. And while the iPod uses nicely antialiased fonts, the free alternative uses blocky x-windows style fonts that would be right at home on an Atari ST some 24 years ago.
And it crashes on the file format I'm using it for.
It's not the iPod's fault that Steve Jobs is a jerk, or that it's a proprietary OS that doesn't let people wedge new codecs onto it, or that the server-side part of the file transfer protocol is the hideously bloated monstrosity that is iTunes. But it's not a particularly useful tool for me to use if I can't use the stack of digital music that took me a week (and a 14 drive cd tower) to slurp up and write to my jukebox directory.
And the open source alternatives (that aren't actually alternatives unless you've got a pathological dislike for properly designed user interfaces) are nothing but a waste of my time, energy, and stress levels. (it would be useful to have a version of iPodLinux on the shuffle, since no screen := a teeny tiny web server.)
So, instead of putting an "open source alternative" on an iPod and having a teeny tiny music player for my next dentist's appointment, I'll be loading all of my music on my Windows laptop and and using foobar2000 (it would be an inferior UI, but it plays Ogg Vorbis format music without crashing) to play trance music for me.
And the iPod? Well, the best has been wanting an iPod for a while now, so it will hopefully be seeing some use before it sneaks off to the secret clubhouse where so many of our electronic gear has hidden themselves in.
Maybe I'll take one of the big apple stickers that came with the iPod and slap it onto my Windows notebook, then tell people it's an iPod with a hyperactive pituitary gland.
—orc Sun Jun 3 23:13:39 2007
Jun 02, 2007
This afternoon, while I was doing a bunch of clean-up chores around the house, I heard an unfamiliar locomotive whistle blowing repeatedly, as if it was attached to a train either coming south or going north on the ex-SP line up through east portland. After listening for a little while, I decided that it was coming south, and if it was an unfamiliar whistle, it might be an unfamiliar (and by unfamiliar, I mean "not Yellow Menace") locomotive and I could go out and take some pictures without being ridiculed when I returned to Chateau Chaos. So I grabbed my camera and walked down to the Bybee bridge to see if I could get down there before anything interesting came by (the last time I tried to do this was when I heard a steam whistle from the Brooklyn roundhouse. I didn't make it down to the tracks that time, but I suspected it was a steam locomotive because there was a huge blast of black smoke from under the Bybee bridge just as I reached the edge of Westmoreland Park.)
When I got down there, I realised that even if it was an unfamilar engine, it wasn't likely to have passed by, because there was a traditional Union Pacific style traffic jam at the south end of Brooklyn Yard; the east main had two northbound trains stacked up facing a southbound train, while the west main had (blocking the south yard crossovers, so the two northbound trains on the east main couldn't even dodge around the southbound train) a loooong mixed freight standing there, and way down at the end of the new (put in over the screaming objections of {East|West}moreland residents who were tired of the sounds of late-night switching) yard lead was a fifth train sitting mysteriously dead on the line.
As I sat there and wondered what the dispatcher was going to do, I saw one of the Yellow Menace crew SUVs driving slowly down towards the head end of the fifth train. It stopped up by that train, then turned and slowly proceeded back towards the yard, followed eventually by the train, which turned out to be a fairly long container freight that disappeared off into the container yard after it passed by.
_20070602_t.jpg "A SD70m leads the first train north")
That train cleared the west main, so that waiting train was cleared to go, and after a mysterious 10 minute wait (why? It's not as if the line was blocked, given that a northbound train had just cleared) it started up and clattered south. And when it finally departed, the southbound train on the east main crossed over and follow it southward.
_20070602_t.jpg "A mess of SD70ms and AC4400CWs")
With the eastbound main cleared, there was nothing to stop the two remaining trains from heading north, and, after a long wait, the first train proceeded northwards through the yard (this was a train of MOW hoppers, all of which had little solar cell grids mounted on them, with wiring leading down to a mysterious grey box on the end platform.) By this time, I was beginning to suspect that the day was going to be all Yellow Menace all of the time, but since I'd waited so long I figured I might as well wait for the last train to go past.
_20070602_t.jpg "A single SD70m pulling a train of MOW equipment")
As expected, it was a Yellow Menace unit. A single SD70m, pulling a second MOW train, and this one was all of the ballast tending equipment that would follow along after the first train, cleaning, sweeping, and washing the ballast until it shone. The MOW train was a pleasant surprise, but it wasn't the only one; normally (as in "every train I've seen on the Union Pacific in the Portland area") engines come in bunches of two or three, with occasional clusters of 20 or so, but #4881 was pulling this train all by itself.
_20070602_t.jpg "A Plesser & Theurer track cleaning machine")
As far as I know, I'd never previously seen a Plesser & Theurer ballast cleaner, but there one was, tucked in at the back of the train. So it was actually three pleasant surprises on one train.
After this train vanished to the north, I looked to see if any other trains were within sight. Nope. But I did hear another one of those unusual toots, so I suspect it's just that one of the switch engines had an old SP horn that has managed to avoid being replaced with a standard UP blat-blat horn.
And when I got home, it was, yet again, "Oh great, more yellow engines." Sigh. I wonder if they'll have the same reaction when the working engine and the 700 run their excursion trips on the Portland Traction line this July 12-14th?
—orc Sat Jun 2 22:21:09 2007
Jun 01, 2007
I was looking over some of the spoiled pictures from last week (spoiled because of the gunk on the windshield) and found one that isn't that bad, even with the large blurry glob in the middle of the frame. This is another picture of the pumpkin orange units from last monday, and here it almost looks like Mount Hood is resting on top of that overpass behind the engines.
—orc Fri Jun 1 22:48:34 2007
![[apples,peaches,Dust Mite]](/~orc/fdmb/Still_life_with_apples_and_peaches_t.jpg)
Still life with apples, peaches, and Dust Mite.
—orc Fri Jun 1 22:41:16 2007
Two tugboats proceed downriver at around 1pm this afternoon. The barges don't look like Ross Island Sand and Gravel barges, but if they aren't I wonder (a) what they're doing on this part of the river, and (b) why is the Ross Island tugboat acting as the wingboat?
—orc Fri Jun 1 22:34:30 2007
When I was taking the bus home for lunch, I spotted the yellow side of OrangeYellow at Gibbs St. I was too busy gaping at the rather bright color scheme (the city doesn't believe in dull paint schemes on the Skodas (et al), so even on the more sedate cars it's an awful lot like the psychedelic trolley reborn, but this one makes the older cars look muted.
I didn't even remember that I had my camera in my hands until the bus was past the trolley line.
—orc Fri Jun 1 22:17:25 2007
—30—
baby.jpg "Silas and a selective-focus lens")
.jpg "Amtrak #14 zips towards the 12th/Clinton/11th street crossings at ~5:15pm")
_20070602.jpg "A SD70m leads the first train north")
_20070602.jpg "A mess of SD70ms and AC4400CWs")
_20070602.jpg "A single SD70m pulling a train of MOW equipment")
_20070602.jpg "A Plesser & Theurer track cleaning machine")
![[apples,peaches,Dust Mite]](/~orc/fdmb/Still_life_with_apples_and_peaches.jpg)
![[rss feed]](/~orc/feedbutton.png "Atom + RSS feeds available here"){kind=small}