Shallow Thoughts : : Oct

Akkana's Musings on Open Source Computing and Technology, Science, and Nature.

Sun, 31 Oct 2004

Image Formats for the Web

Every time I see someone ask about image formats, I think "Someone really ought to write up a howto explaining the difference between GIF, JPG and all the other formats, and what they're good for." There probably are documents like this, but I've never seen one. So I wrote one. Image Formats for the Web and Elsewhere. (I'll probably give a Toastmasters talk on the subject as well.)

[ 15:25 Oct 31, 2004    More writing | permalink to this entry | ]

Thu, 28 Oct 2004

More ballot shenanigans

The Florida post office has somehow lost 58,000 absentee ballots in Broward County, FL.

They say they're mailing out new ballots, but that (not mentioned in the article) blithely assumes that everyone who voted absentee did so frivilously, not because they were, say, going to be out of town during the election.

By a staggering coincidence, Broward was the county which gave Gore the biggest margin in 2000.

Meanwhile, news from the EFF from last week when I was out of town: Santa Clara County poll workers are being trained not to offer voters a chance to use paper ballots instead of electronic voting machines.

I've been rather hoping that the EFF (or someone) would organize protests near polling places, trying to inform voters of their rights. But no such luck. Instead, they've set up a site with a big flash movie with monotonous music and no information that couldn't have been shown better in a simple fast-loading html page. If you want to watch the flash movie, it's at but there's really nothing else there besides the movie.

Spread the word anyway. Tell everyone you know in the affected counties (Santa Clara, Orange, Alameda, and Riverside Counties. Napa, San Bernardino, Merced, Plumas, Shasta, Tehama, and Riverside counties) that they can request a paper ballot, and that way leave a paper trail that can be verified in case of a recount.

Tags: , , ,
[ 23:43 Oct 28, 2004    More politics/election04 | permalink to this entry | ]

Wed, 27 Oct 2004

Pictures from the trip

Photos from the trip are up (except for panoramas which still need to be stitched).

Tags: ,
[ 11:18 Oct 27, 2004    More travel/anasazi | permalink to this entry | ]

Mon, 25 Oct 2004

Red Stone and Redstone

Yesterday and today were travel days -- supposedly nothing much to report. But it turned out otherwise.

Nothing much yesterday except the herd of bighorn sheep grazing by the side of the road as we left Moab. (We had planned to stay in Moab for a few days, but the weather turned sour.) The drive through the San Rafael Swell is always impressive, but I've written about that already.

Today, first, a quick stop by Kolob Canyons, a small branch of Zion National Park accessed right off I-15. It's marvelous: a very short road loop with stunning views, and three hikes of varying lengths. We didn't do any hikes due to weather and health issues, but we'll be back!

After leaving St George and Utah and before entering Nevada, I-15 briefly passes through Arizona in the impressive Virgin River Gorge. Arizona doesn't bother with trivialities like nice roadside view areas like Utah and Colorado do.

But there's a BLM area flaking the north side of the gorge, with a dirt road: the Beaver Dam Mountains Wilderness Area. We went a little way up the road; we didn't find views of the gorge from there, either (perhaps farther up?) but the rocks were quite interesting, evidently a mixture of rhyolite and basalt with some bits of tuff and river cobbles (did the Virgin make it up this high before the area was uplifted, or are the cobblers from streams which used to run from higher still?) We'll be back to explore further (with a BLM map, I hope).

Returning to I-15 and crossing into Nevada, we chose a detour: instead of following the interstate through the rush-hour traffic of Las Vegas, we swung left onto a little highway that cuts down by Lake Mead, marked as "scenic" on the map?

Getting through the tiny town of Overton took longer than we expected; its "so ridiculously excessively low as to be obviously a speed trap" speed limit zone went on forever. But we finally emerged out the other side, passing the Lost City Museum (curiously, just last week we'd read an article in the LA Times about an old town near there which had been buried for most of last century by Lake Mead, but which had re-emerged in the last few weeks due to record low water levels, creating great interest among historians). The scenery began to get interesting right away. It offers very little in the way of views of the lake (unless you drive down the side roads leading to the lake itself), but the area is "painted desert" of bentonite or a similar ash, punctuated by jagged peaks of volcanic rock. Most of the land is part of the Lake Mead National Recreation Area. Numerous parking areas are located at small oases named This-or-that Spring. Some of the springs are visible from some distance as a grove of palm trees. Are any palm trees actually native to the American southwest, or were they all introduced by settlers?

Update: Apparently the origin of these palms is a point of dispute, but there's quite a bit of evidence arguing for their being native to the area. William Spencer sent me a link to a page discussing the issue and the fight to save the palms.

This goes on for miles, and then gradually bits of brighter color begin to appear, in the shape of red sandstone. We stopped at a parking area on the left, and found a true jewel: Redstone, a little rest stop with a trail of maybe a mile which goes out around the vividly red rocks, with occasional interpretive signs which are interesting and not patronizing. The rock is Aztec Sandstone, formed from dunes which covered the area some 130 million years ago, with wonderful cross-bedding and weathered textures, and nearby mountains of black basalt to provide contrasting color.

After taking the Redstone hike, we continued on the highway, stopping at some of the pullouts, including one which included an interpretive sign describing the "bowl of fire", resulting from a layer of Aztec sandstone which swelled into a domed shape, then eroded from the top, leaving an outer ring. The fiery red ring is easy to see among the darker layers surrounding it.

Presumably the nearby Valley of Fire state park is also Aztec sandstone sculptures; it looks like it from a distance. We wished we'd taken that route, and will next time.

The scenic highway ends in Henderson, leaving us to fight our way through yet more heavy traffic (no matter which way you approach Las Vegas, or at what hour, or how hard you try to bypass the center of town, somehow you always end up in a traffic jam!) to return to I-15 and head down to our destination of Primm, musing on the long, gradual talus slopes so typical of the Mojave desert, and how superficially similar they look to a shield volcano like Mauna Koa. I wonder how the angles of repose compare? (Alas, there's no internet in Primm, so that's a question for a later time.)

Photos of Kolob and Redstone.

Tomorrow: home!

Tags: ,
[ 21:23 Oct 25, 2004    More travel/anasazi | permalink to this entry | ]

Sat, 23 Oct 2004

The Confluence

I've wanted for years to see the confluence of the Green and Colorado rivers: the place where the west's two biggest rivers meet, mingling their different colored waters into the larger river which is the lower Colorado, flowing down to become Cataract Canyon.

The Confluence is hard to get to, though. The only viewpoint above river level is located in the Needles district of Canyonlands National Park. Sounds easy enough; but the only road that goes near it is a technical jeep trail called "Elephant Hill", involving tricks like five-foot rock drop-offs. A bit beyond our skills or vehicle. So instead, we drove to the beginning of Elephant Hill, then mountain biked from there. It's about 9 miles to the confluence overlook (then a half-mile hike from there), and about 6 miles back (it's a loop trail with one-way sections).

First we had to get to Canyonlands. We took the scenic route from Monticello over the Abajo mountains, offering great views of the lacolith triangle: the Abajo, Henry, and La Sal mountain ranges are all rock which has been warped upward by subterranean magma, without actually being made of volcanic rock themselves.

On the Saturday of Utah's week-long deer hunting season, the Abajo route was crawling with trucks filled with blaze orange clad passengers, pulling trailers laden with ATVs. Every pullout, every campground, was full of hunters. Ironically, twice during the day we had to slow down (and once, stop) for large groups of does wandering near or across the road. We never saw any bucks, but I guess the number of does on the road suggests that the deer population isn't in any serious threat from the hunters. But we nevertheless were glad we were going to be doing our riding in a national park today.

Elephant Hill is as technical as we remembered it from our last visit to Needles. We tried to ride up the hill, but gave up fairly early and walked the steep sections. The trail alternates between short, impossibly steep and technical rock sections (which we walked), moderately steep and technical rock sections (which we mostly rode, and enjoyed immensely) and long near-level stretches of deep fine red sand (fun if you don't mind sliding sideways).

Dave rode more of the rocky uphills than I did, and I rode more of the rocky downhills. I biffed on one downhill, coming off a rock ledge into deep sand and landing hard on one hand. No permanent damage.

No bikes are allowed on the half-mile section of trail from the end of the road to the overlook, so we had to stash our bikes in the bushes and continue on foot.

The confluence overlook is fabulous! It's just like the pictures: you can see the boundary where the two differently colored rivers mix to form one larger river. Apparently the colors vary depending on what's been going on upstream; every picture is a little different. Today, both rivers were muddy green, but different shades, with the Colorado being darker and clearer than the Green. On the horizon, you can see the three districts of Canyonlands: Island in the Sky (between the two upper rivers), the Maze (along the west bank of the Green) and Needles (where we stood, on the east bank of the Colorado).

The ride back was surprisingly easy, though going uphill through the sandy stretches was a workout. We got back to Elephant Hill just as a couple in a rented jeep began the first descent, so we had a chance to see how it was done. The Jeep handled the tough descent easily. I bet it didn't seem as easy from the driver's seat as it looked from the outside.


Tags: ,
[ 23:20 Oct 23, 2004    More travel/anasazi | permalink to this entry | ]

Fri, 22 Oct 2004

Goosenecks of the Gods

Our scenic loop to the Valley of the Gods began with cold, windy, overcast, drizzly skies. But as if to make up for the weather, we were greeted with a rainbow almost as soon as we hit highway 95. Or maybe that was to make up for the snow flurries we encountered a few miles later. Whatever.

We headed down highway 261, eschewing Natural Bridges National Monument. Been there, done that. We were headed for Muley Point, which turned out to be an unmarked dirt road turnoff just before the Moqui Dugway. Four miles of relatively good dirt road led us to two stunning viewpoints overlooking the sinuous San Juan river and points beyond, such as Monument Valley, Alhambra Peak, and Valley of the Gods. The wind was icy, but the view was worth it.

Returning to the highway, we headed down the Moqui Dugway (variously spelled Moki or Mokee, depending on which map you use; everyone seems to spell Dugway the same). This is a steep (11%) grade, gravel except on a few turns where pavement returns, winding 1100' down the side of Cedar Mesa to the bottom. Why it's gravel when the rest of the highway is paved isn't clear. But it's fun.

At the bottom of the Dugway, a BLM dirt road goes left into Valley of the Gods. But we decided to see the Goosenecks of the San Juan first.

At Goosenecks, the San Juan river travels over six river miles in the space of only a mile and a half. It's held up as one of the best examples anywhere of an entrenched meander, where a lazily meandering river on nearly-level terrain cuts a shallow channel, then rapid uplift of the area (in this case, the Colorado Plateau) causes the river to cut a deep canyon.

There are entrenched meanders all over the area -- such as Bowtie Bend and Dead Horse Point -- but nowhere are there so many, in such a short space. It's very impressive.

And that's all there is to Goosenecks of the San Juan State Park -- one amazing overlook. There's a trail somewhere (the Honaker Trail, namesake for the rocks comprising the upper two-thirds of the San Juan's canyon; the bottom third is the Paradox Formation, both Pennsylvanian layers of limestone and shale) but it's accessed from outside the park, and there's no information about it at the park.

We backtracked to the west end of Valley of the Gods Road and began our divine journey, following a guide we'd picked up at the visitor's center in Blanding. The first rock on the list was Balanced Rock -- I pointed it out. "No," said Dave, "that's got to be Lady in a Tub. That's exactly what it looks like." "Um, I don't see that on the list here." It turned out that this was an alternate name for the same rock, listed on the map but not in the guide. And indeed, it was a good name -- except that as we proceeded down the road, it became a Man in a Tub.

It's a while before the next Named Rock on the guide, but that's okay; there are fascinating rock formations everywhere. The light was difficult for photography, since it was still mostly overcast, but that made for dramatic light when the sun did come out.

And a few miles in, I spotted an even more interesting formation: a tarantula making its way across the road. We go tarantula spotting every year, but the season when the males go wandering aboveground in search of females is so short that we often miss it. This year we were sure we'd missed the season at home; so finding one here was serendipitous. This one appeared to have no inclination to get off the road, so we had plenty of time to shoot photos (including "tarantula walks over the camera" and "real tarantula completely ignores our rubber tarantula") while we gently tried to persuade him to walk by the side of the road and not in the middle.

We invented names for unnamed rock formations, like "Mohawk with Squirrel on Head" and the nearby "Organ Grinder's Monkey, with Drum". Rooster Butte should have been Senorita Butte -- a Spanish dancer with full flowing skirts. Occasionally the road became mildly technical, with rocks or gully crossings. "Chacoan speed bumps!" exclaimed Dave. Two painters had set up camp right in the middle of a wash, with their easels right by the road -- maybe dust is part of the art, and a flash flood just gives an artist more inspiration. Setting Hen Butte (its official name) has giant sandstone eggs all around it.

Too soon, we found ourselves at the other end of the road, and the highway. But before heading back to Blanding, we took a detour to Sand Island, near Bluff, to see what was there. What was there was petroglyphs -- a whole wall of them, comparable to the much more famous Newspaper Rock to the north near Monticello. Excellent rams and elk, snakes, and other figures. But what interested me most was all the Kokopelli-like figures. Kokopelli (the dancing flute-playing trickster) shows up in nearly every gift shop in the southwest. He's so prevalent that a mapmaker in Moab (Cheap is Real) comments on the back of each map that it is a "100% Kokopelli-free product"). Yet in the rock art I've seen, I have yet to see an actual Kokopelli -- until Sand Island. Sand Island is definitely not a Kokopelli-free zone. But it's a great set of petroglyphs.

Photos of Goosenecks, Valley of the Gods, the tarantula, and Sand Island Petroglyphs.

Tags: ,
[ 22:48 Oct 22, 2004    More travel/anasazi | permalink to this entry | ]

Thu, 21 Oct 2004

Aztec Ruins

Driving around Farmington, NM is a little different from driving around California.

Heading out of town, we passed the Permian Power Tong building. I guess you'd better be careful when complaining about your electricity bill in Farmington! Especially if you want to assert that it comes from the Mesozoic, or something. Not long after that, we passed Jimmy's Swabbing Service. I don't think I want to know too many details about that, nor about the Four Corners Bull Test Station we saw later.

Update 11/8/2006: Someone from the Permian Power Tong wrote to let me know that they're an oilfield service company, not an electric company.

We stopped at the Aztec Ruins, so misnamed because early white settlers apparently thought these Anasazi ruins were left by the Aztecs (?). It's a small park, with one trail, but the ruins are excellent and the guide is full of information about the architecture. The structures were originally built by Chacoans and most of the lower masonry is similar to what we saw in Chaco Canyon, but was later modified (for repairs and additions) in a style more similar to Mesa Verde. Then, much later, some of the masonry was re-done by the park service in a well meaning but misguided attempt to stabilize the fragile structures, with the result that there's a lot of modern concrete, metal drains, and other anachronisms and apparently it's sometimes hard for modern researchers to be sure what came from which era.

The Chacoan work is the most beautiful. They liked to alternate layers of large bricks with small, or red with other colors, whereas the Mesa Verdeans used fairly uniform large bricks everywhere. Someone who came along later (perhaps the Mesa Verde group, perhaps a later tribe) added rounded river rocks in places, from the nearby Animas river. The Animas may also have been used to float the hundreds or thousands of logs needed for the roofs of the structures; the wood apparently came from the mountains, near Durango, since it's wood which wasn't available locally.

Although the park service tries to be much more careful now, we saw some modern repairs on the structure while we took the self-guided tour: Navajo bricklayers pounded sandstone with a hammer, chipping flakes off to make it the right shape to fit into the spot being repaired.

Outside of the park, we explored the town of Aztec, which has a nice little suburban downtown area surrounded by miles of scrubland with residential trailers. We noticed that the downtown area had a predominance of Kerry signs, unlike Farmington and the rural areas outside Aztec where Bush signs prevailed.

We took back roads from Aztec, eventually passing through Mancos (the Mancos Motocross, Now Serving Elk Burgers -- what more could you want? -- and the Reptile Reserve of Southwest Colorado) and the poshest highway rest stop we've seen anywhere, at Sleeping Ute Mountain, which offered its own hiking and pet exercise trails.

Our plan was to stay tonight in Monticello, UT, which is close to Canyonlands' Needles district and lots of other interesting places. The first hotel we tried should have given us a clue as to what was coming: the sign proclaimed "Big Buck Display!" A big dollar bill? wondered Dave. But it turned out this is the beginning of Utah's week-long deer hunting season, and that Monticello is the deer hunting capital of southeastern Utah (for some reason). We pushed on to Blanding instead.

Blanding looked like a bigger town in the AAA guide (more hotels) but isn't really. Fortunately, the Best Western has wi-fi (the only place in town, unlike Monticello which has two hotels and a cafe). The router gives the wrong address for the DNS server, but we guessed at the right address and edited /etc/resolv.conf, and things work okay as long as you remember to do that before making any net connections (otherwise the wrong DNS info gets cached by some proxy server somewhere).

Dave went to the office to see if anyone knew about this. He was told: "They just fired up the system two weeks ago, and it has been slow," but no one knew any more detail than that.


Tags: ,
[ 21:08 Oct 21, 2004    More travel/anasazi | permalink to this entry | ]

Wed, 20 Oct 2004

Chaco Canyon

I've been curious about Chaco Canyon ever since as a kid I read an article in Sky & Telescope about the Anasazi Sun Dagger, a rock structure whereby at the solstices and equinoxes the sun creates a narrow sliver of light projected onto a spiral petroglyph.

Unfortunately, it turns out that the Sun Dagger is not open to visitation (by the public or even by most researchers). In the 1980's it was deemed too fragile for visitors, and the site was closed down. There are some other astronomically oriented petroglyphs, but no one seems to know exactly where, or to have a complete list.

Getting information on Chaco is a bit difficult. There's not much useful information on the web, the park doesn't have specialized handouts like a lot of other parks (many parks have one-page handouts available for the asking on subjects such as geology, astronomy, petroglyphs, etc) except for one giving a brief listing of the available hiking trails. The ranger at the visitor's station was somewhat reticent: he recommended a couple of hiking trails, and told us that the Sun Dagger was located high on Fajada Butte, but not much more. I noticed a picture of some petroglyphs thought to depict a supernova, and asked where they were, but he apologized "Sorry, I don't go out that trail much".

Nothing to do but go try some stuff and see what's cool. We visited all the ruins along the park road, then headed up the steep trail to Pueblo Alto and the Pueblo Bonito overlook, which begins by a scramble through switchbacks over broken rocks, followed by a steep ascent through a narrow gap in the rock wall. Fun! And daunting: but it turns out that once you squeeze through the gap, you're up on top of the mesa and mostly finished with climbing.

The mesa top is interesting rock: white, layered mudstone, full of interesting embedded objects (presumably plant fossils, though some of them actually look like bone). The Fajada Butte interpretive sign, the only mention we found of park geology, says of the butte: Cliff House Sandstone forms the upper layer with deposits of fossil shells, clams, shark teeth, and marine sand. None of these fossils seem to correspond with what we saw embedded in the rock along the Pueblo Alto trail. More research is required.

The view of Pueblo Bonito from above is marvelous and well worth the short and interesting hike. The semicircular shape of the great house, not obvious from below, is striking when viewed from above.

The hike up to Pueblo Alto was pretty, and enjoyable as a hike, but Pueblo Alto itself is much less interesting than the ruins down in the canyon. We wished we'd gone the other way on the loop trail for more birds-eye views of the canyon houses.

Another interesting aspect of Chaco: their astronomy program. They have a fixed observatory (a dome housing a truss-tube dobsonian of about 18") and something outside on a tripod (probably a big Schmidt-Cassegrain). The visitor's center was full of photos of astronomical objects, as well as some information about light pollution. It's nice to see a park so interested in astronomy, especially with the sort of skies they must get at Chaco. Alas, we weren't able to stay the night.

But Chaco's big mystery is the "roads". The park literature talks about the amazing roads the Chacoans built, stretching for hundreds of miles between Chaco and neighboring settlements in many directions, used for trade between tribes. On the Pueblo Alto hike, a short segment of one such "road" is roped off and signed: a wide rectangle of more or less bare rock, perhaps ten or fifteen feet on a side, lined generously with rocks on two sides. With a lot of imagination, you could imagine a boulevard continuing in this fashion, rocks lining the left and right sides of the "road" like a huge version of some national park trails.

Dave smelled a rat, and dug further. These "roads", apparently, were originally detected as unexplained straight lines appearing in infra-red images, using NASA's TIMS system. Archaeologists subsequently searched the ground and found some short segments which looked vaguely road-like, and drew maps connecting the segments. Here's one such map of the Chaco road system. Notice anything unusual? Like the fact that the ground map doesn't actually match the lines in the IR image? Note also how straight the "roads" are in both theories.

It gets even weirder. One of the park's roadside pullouts points to a "Chacoan stairway" high on a mesa, and comments that the stairway was part of one of the roadways. The stairway is there, and it's neat. There are other stairways elsewhere in the park -- we saw photos (though the one section we saw up close, on the Pueblo Alto hike, was a bit too subtle for either of us to find the "stairway" on the indicated rock).

Why would the Chacoans build roads like this? It makes no sense. Why would a prehistoric people with no wagons or pack animals need rock-lined ten foot wide "roads", arrow straight and made without respect to the local topography?

Let's look at this practically. You're a Chacoan heading out to trade with someone in a pueblo to the south, or a southern resident travelling to Chaco. You have a choice between following a straight road, which requires you to climb up onto an 800 foot mesa, then down a precipitous set of rock stairs which lead to a steep scramble back down to the canyon bottom; or you can walk a quarter mile west and stroll through the huge gap between two mesas, without having to climb or descend at all. You're travelling on foot, carrying your pottery or baskets or whatever it is you're bringing to trade. Perhaps you have your family and kids along. Which route would you choose?

The stairways are there; and the "road" segments are there, too. But that doesn't mean that they connected to form hundred-mile long roads between communities. The stairways are useful for locals who want access to the mesa tops -- perhaps for defense, or religious purposes, or just for sightseeing. The short "road" segments on the ground -- who knows? Perhaps parade grounds. Or maybe they were malls, where vendors lined up to spread their wares out for customers to view. There are lots of possible explanations!


Tags: ,
[ 22:28 Oct 20, 2004    More travel/anasazi | permalink to this entry | ]

Tue, 19 Oct 2004

White House to Ship Rock

The weather wasn't really much better this morning, but we decided to hike the White House trail down into the canyon anyway.

Good move! It's a beautiful trail which definitely belongs on a top-ten list of park trails (along with trails such as Hummocks at Mount St. Helens). And as a bonus, it's not even particularly strenuous -- the canyon is only 600 feet deep at that point, and the trail is fairly gradual. It descends from the cross-bedded riverine rock of the Shinarump member of the Chinle formation, down into the thick de Chelly sandstone, where it winds through little tunnels and around switchbacks, past shrieking squirrels and soaring ravens, giving ever-changing views of the canyon floor.

At the bottom, the trail skirts a Navajo ranch (no photography please) then follows the stream bed, lined with cottonwoods in glorious fall foliage, to the eponymous ruin, surrounded by fences to keep out vandals and well-meaning but overly enthusiastic tourists. Nearby, an unattended horse grazed, and a local rancher followed his sheep herd as they browsed along the riverbed.

Impressive ruins. Lovely trail. Go see it.

After climbing back up to the trailhead, we went off to explore the north rim (which is technically a different canyon, del Muerto rather than de Chelly). The north rim viewpoints are sparse, but well chosen; they show more ruins, from shorter distances, than the south rim viewpoints.

After leaving the park, we debated whether to go south to Gallup, or north to Shiprock and Farmington. Shiprock won. But after turning onto highway 13 to cross the Chuska mountains, we questioned the choice. Large signs warned of upcoming highway construction, road closure, and seasonal (winter) road closures over Buffalo Pass. This not being winter yet, we proceeded with trepidation. Our fears (and the warning signs) were unfounded: although the road is narrow and twisty, the pavement is excellent and the views outstanding.

Just past the summit, we got our first view of the immensity of northwestern New Mexico spread out before us -- and immediately realized that Shiprock was not what we had seen yesterday from Spider Rock overlook. Shiprock is unmistakable and striking. It sails on an immense flat plain, tossed on waves of sage, trailing a wake of basalt behind it. It dominates the landscape for many miles in any direction.

Shiprock is a giant volcanic neck: lava which sat in the neck of a volcano, and hardened there. Later, the volcano and its surroundings eroded away, leaving only the neck. But there's more: in addition to the neck, Shiprock's lava also squeezed through a dike, a vertical seam stretching for many miles on either side of the volcano. After the surroundings eroded, what was left was an immense wall of lava, only a few feet thick but some fifty feet high and miles long.

The triple-A map showed a dirt road just east of where the highway crosses the dike, leading up alongside the rock. Sure enough, the promised road appeared just where the map said it would. Woohoo! It turned out to be an unmaintained jeep trail, a nice challenge for our little RAV4 (which had no trouble with it). The road parallels the dike up to the neck itself, giving wonderful views from any angle. Unfortunately the area right next to the neck is spoiled by grafiti, but the rest of the area is fabulous.

We pulled into Farmington later than expected, after stopping to help a Navajo family whose truck had broken down. Unfortunately we didn't have any mechanical insights they hadn't already tried, but we gave one to the nearest store to call for backup. I hope everything worked out all right.

Farmington is the Big Gorilla of the four corners area, by far the biggest town around. Happily for us, it's also fairly well wired, and nearly every motel sports wi-fi that actually works (the only catch being that they fill up surprisingly early on weeknights; we're still not sure why). It's a deceptively large town, with a small college and the usual assortment of restaurants and businesses, several rivers, and plenty of farmland on the outskirts, befitting its name.

Photos of de Chelly and Shiprock.

Tags: ,
[ 23:29 Oct 19, 2004    More travel/anasazi | permalink to this entry | ]

Mon, 18 Oct 2004

The Dogs of Tseyi

I managed to wheedle Dave into taking the back roads from Winslow to Chinle, crossing more of the Navajo nation rather than staying on the interstate we've seen before.

Good move! The roads are fine (if a bit slower than an interstate highway) and the scenery is terrific. Dave reciprocated by making an impulse turn into the Little Painted Desert county park's overlook -- empty except for us, the vista across striped layers of bentonite (is that Moenkopi, or Morrison?) rivals its namesake to the southeast in everything but size.

Near Castle Butte, a striking wall of basalt curves gracefully across a plain, an obvious remnant of a vertical dike from which the surrounding, softer rock has long since worn away. This is what created Shiprock, a larger and more famous formation of the same type which I'm hoping to get a chance to see later on this trip; but the thin, curving walls near Castle Butte, with their spiky towers, are marvellous examples.

The roads through this part of the Navajo reservation (perhaps it's true everywhere) are open range. Cattle grazed near the road, and at one point I had to stop suddenly when a horse decided to trot across the road in front of us.

Canyon de Chelly sits right on the edge of Chinle, closer than we'd realized from the map. In fact, Chinle, "where the water flows out", is located right at the mouth of the canyon, where the surrounding mesas drop to the level of the river at the canyon's bottom.

De Chelly itself is really Tseyi, meaning "in the rock" in the language of the Diné (i.e the Navajo). The Spaniards had difficulty pronouncing this (sometimes spelling it "Chegui"), and when early American settlers moved in, they mis-heard it again and assumed they were hearing "cañon de chelly", Spanish for "canyon of rock", pronounced, more or less, "dee shay". But the Tseyi name is still prominent in town and in park literature, this still being Navajo land. The park literature says it's pronounced "say-yee", but a Diné woman in town pronounced it for us more like "tsay-yeh".

The park literature mentions that there may be some stray dogs wandering in the park, and warns not to feed them. The town of Chinle has a problem with too many stray dogs; feeding them "only makes the problem worse." It doesn't mention stray horses, though quite a few wander the mesas above de Chelly and occasionally cross the roads.

We followed Dave's Rule of Parks: go to the end of the road first, because that's where the really good stuff is. The end of the road for Canyon de Chelly is the end of the south rim road, or Spider Rock Overlook. Spider Rock itself is an impressive spire of sandstone (de Chelly sandstone, in fact: a thick desert dune deposit like Navajo sandstone, only much older, at 230-260 million years, and also much redder) standing in a wide, flat canyon of green and autumn gold.

On the horizon far beyond Spider Rock stood a striking dark butte. Our first view of Shiprock? (No, as it turned out.)

The other attraction of Canyon de Chelly is the Indian ruins. Anasazi cliff dwellings pepper the cracks in the canyon walls, and are visible across the canyon from many of the overlooks. Bring binoculars (and a good zoom lens, if photographing).

The star ruin of the park is called White House, and it's accessible via a trail which climbs down from the south rim and crosses the canyon. It was beginning to rain as we arrived there, as well as nearing twilight; we hope for good weather tomorrow morning. We had to drive around a tired looking black dog lying on the (presumably warmer) roadway, seeming unperturbed by the cars going by and disinclined to move. Another dog followed tourists around with a hopeful expression.

And Dave's Rule of Parks? It doesn't work as well at Canyon de Chelly as at most parks. White House is far better than any of the ruins visible from the farther overlooks; and in fact, the very first overlook (last for us, since we were visiting them in reverse order), called Tunnel Canyon, gave a lovely view down a narrow canyon to the riparian zone below. Maybe we were just lucky with the light, arriving at Tunnel as the setting sun pierced through a hole in the otherwise unbroken cloud layer. There's a trail going down from Tunnel, too, but it's only open for guided tours. (Access into Canyon de Chelly requires a guide, except for White House trail, because some 40 Navajo families still live and farm inside the canyon.) After appreciating the lovely light, we chatted with a Diné woman selling jewelry, and watched a couple of puppies trot in, search for food, and then run off toward home.

The town of Chinle is neither depressing, like Tuba City or the area around Monument Valley, nor modernized, like Kayenta. It's small and sparse, with only two hotels (plus the one inside Canyon de Chelly) and few restaurants besides the two associated with the hotels -- a few fast food eateries and a pizza parlor. Yet at night, lights (mostly low-pressure sodium, I was happy to see) twinkle from a wide area, hinting that there's quite a bit more to the town. We tried to explore, but couldn't find our way to the pockets of light we could see from the main part of town. So we reluctantly settled for a dinner at the Holiday Inn's restaurant, which was surprisingly good. Native American towns don't seem to succumb to chain-hotel-itis quite so much as other towns do.

And the dogs! Everywhere you go in Chinle, a few dogs appear out of nowhere to follow you. Dogs fade in and out of the plants along the roadside, and haunt every park overlook and restaurant parking lot. Most of them look quite young -- which may bespeak a short lifespan -- though most of them also look fairly healthy and friendly. They wag, and play, and appreciate a head scratch, and otherwise behave pretty much like pet dogs everywhere.


Tags: ,
[ 22:32 Oct 18, 2004    More travel/anasazi | permalink to this entry | ]

Sun, 17 Oct 2004

Vignettes from a road trip

I-15 follows the San Andreas fault as it cuts through the San Bernardino mountains. The spectacular exposed hogbacks, reminiscent of the Devil's Punchbowl a few miles north along the same fault, or perhaps of Colorado Springs' Garden of the Gods, leave no doubt that massive geologic forces are at work.

Around Victorville, the power towers stand like four-footed animals with huge wings outspread -- power pegasi. But beyond Newberry Springs, at the western edge of the Pisgah Crater lava fields, they change to the broad-shouldered power kachinas seen in parts of Utah. Nearby, a raven practices no-flap take-offs, presenting outspread wings to the constant gale, lifting smoothly a few feet off the ground, then floating gently back to earth to try again.

A commercial on the hotel TV advertised a laser level using "refractive lens technology". Wow! What a breakthrough!

Tags: ,
[ 23:33 Oct 17, 2004    More travel/anasazi | permalink to this entry | ]

Wed, 13 Oct 2004

SuSE 9.1 vs. Ubuntu "Warty"

I took a break from housepainting yesterday to try out a couple of new linux distros on my spare machine, "imbrium", which is mostly used as a print server since Debian's CUPS can't talk to an Epson Photo 700 any more.

The machine is currently running the venerable Redhat 7.3 -- ancient but very solid. But I wanted a more modern distro, something capable of running graphics apps like GIMP 2 and gLabels 2. I considered Fedora, but FC2 is getting old by now and I would rather wait for FC3.

First I tried SuSE 9.1. It was very impressive. The installer whizzed through without a hitch, giving me lots of warning before doing anything destructive. It auto-configured just about everything: video card, ethernet, sound card, and even the printer. It missed my LCD monitor; X worked fine and it got the resolution right, but when I went in to YaST to enable 3D support (which was off by default) it kept whining about the monitor until I configured it by hand (which was easy). It defaulted networking to DHCP, but made it clear that it had done so, which made it easy to change it to my normal configuration.

SuSE still uses kde by default, which is fine. The default desktop is pretty and functional, and not too slow. I'll be switching to something lighter weight, like icewm or openbox, but SuSE's default looks fine for a first-time user.

I hit a small hitch in specifying a password: it has a limited set of characters it will accept, so several of the passwords I wanted to use were not acceptable. Finally I gave up and used a simple string, figuring I'd change it later, and then it whined about it being all lower case. Why not just accept the full character set, then? (At least full printable ascii.)

Another minor hitch involved the default mirror (LA) being down when it got to the update stage. Another SuSE user told me that mirror is always down. Choosing another mirror solved that problem.

Oh, and the printer? Flawless. The installer auto-detected it and configured it to use gimp-print drivers. (from gimp-print) worked fine in a full "test page with photo", and subsequent prints (via kprinter) from Open Office also worked. Good job, SuSE!

The experience with Ubuntu Warty wasn't quite so positive. The installer is a near-standard Debian installer, with the usual awkward curses UI (I have nothing against curses UIs; it's the debian installer UI specifically which I find hard to use, since it does none of the "move the focus to where you need to be next" that modern UI design calls for, and there's a lot of "Arrow down over empty space that couldn't possibly be selectable" or "Arrow down to somewhere where you can hit tab to change the button so you can hit return". It's reminiscent of DOS text editors from the early eighties. But okay, that's not Ubuntu's fault -- they got that from Debian.

The first step in the install, of course, is partitioning. My disk was already partitioned, so I just needed to select / to be formatted, /boot to be re-used (since it's being shared with the other distros on this machine), and swap. Seemed easy, it accepted my choices, made a reiserfs filesystem on my chosen root partition -- then spit out a parted error screen telling me that due to an inconsistent ext2 filesystem, it was unable to resize the /boot partition.

Attempting to resize an existing partition without confirming it is not cool. Fortunately, parted, for whatever reason, decided it couldn't resize, and after a few confirmation screens I persuaded it to continue with the install without changing /boot.

The rest of the install went smoothly, including software update from the net, and I found myself in a nice looking gnome screen (with, unfortunately the usual proliferation of taskbars gnome uses as a default).

Of course, the first thing I wanted to try was the printer. I poked through various menus (several semi-redundant sets) and eventually found one for printer configuration. Auto-detect didn't detect my printer (apparently it can't detect over the parallel port like SuSE can) so I specified Parallel Port 1 (via an option menu that still has the gtk bug where the top half of the menu is just blank space), selected epson, and looked ... and discovered that they don't have any driver at all for the Photo 700. I tried the Photo 720 driver, which printed a mangled test page, and the generic Epson Photo driver, which printed nothing at all. So I checked Ubuntu's Bugzilla, where I found a bug filed requesting a driver for the Epson C80 (one of the most popular printers in the linux community, as far as I can tell). Looks like Ubuntu just doesn't include any of the gimp-print drivers right now; I signed up for a bugzilla account and added a comment about the Photo 700, and filed one about the partitioning error while I was there, which was quickly duped to a more general bug about parted and ext2 partitions.

I don't mean to sound down on Ubuntu. It's a nice looking distro, it's still in beta and hasn't yet had an official release, and my printer is rather old (though quite well supported by most non-debian distros). I'm looking forward to seeing more. But for the time being, imbrium's going to be a SuSE machine.

Tags: , ,
[ 19:15 Oct 13, 2004    More linux | permalink to this entry | ]

Mon, 11 Oct 2004

Migratory birds singing in autumn

For the past week, the mockingbird and the hummingbirds have suddenly begun singing again -- the mocker only in the morning, the hummer sporadically all day. October seems like a strange time to be singing. I wonder if it's related to the decision whether to migrate? Both Anna's hummers and mockingbirds are inconsistent about whether to winter here or migrate south: some years they stay, some years they go.

Tags: , ,
[ 14:23 Oct 11, 2004    More nature/birds | permalink to this entry | ]

Sun, 10 Oct 2004

Pho 0.9.5-pre3: evil Metacity window sizing

It turns out that the problem with pho windows not resizing in metacity is this: when metacity sees a window that's slightly larger (in either dimension) than the screen size, it unilaterally makes that window maximized, and thereafter refuses any request from the app to resize the window smaller.

Mandatory maximize might actually be useful in some circumstances (anyone who's ever tried to run on an 800x600 laptop has doubtless seen dialogs which don't fit on the screen) but the subsequent refusal to resize makes little sense, and causes bustage in programs which work fine under other window managers.

A workaround is for pho to unmaximize before any window resize. This would be a bummer with an app where the user might click the maximize button manually; with pho, that's unlikely (I hope) because anyone who wants to run maximized is better off running in fullscreen/presentation mode (which now finally sets its background to black, hooray).

Get'cher Pho 0.9.5-pre3 here.

[ 19:21 Oct 10, 2004    More programming | permalink to this entry | ]

Fri, 08 Oct 2004

More on CA paper ballots

Unable to find any law stating the paper ballot requirement, I called the Sec. of State's office back, this time being forwarded to someone named Michael.

He told me that the requirement specified in the decertification action was a "directive by the secretary of state", not a legislative action" and so was not reflected in the election code.

However, the requirement is stated in the Voter Information Guide. I do not seem to have received my VIG, but it's available in PDF form (168 pages) on the Voter Information Guide page off the Sec. of State site. It's on page 167: "Counties using touchscreen/DRE systems are required to have paper ballots available upon request."

So there it finally is, in writing. Whew!

I strongly advise all California voters to ask for this option at their polling place on November 2.

Tags: , , ,
[ 14:41 Oct 08, 2004    More politics/election04 | permalink to this entry | ]

Verifying CA's supposed paper trail elections law

Our story so far: the nice lady at the Secretary of State's office pointed me to the PDF for Shelley's Diebold decertification as the proof that the upcoming election will allow voters to request a paper ballot. That PDF says that it modifies Division 19, Chapter 1 (commencing with Section 19001) of the Elections Code and Government Code section 12172.5. My goal is to make sure that this hasn't been superceded by subsequent recertification or other lobbying.

First I tried Leginfo, searching the Government and Elections codes for various combinations of the words paper ballot option election machine That gives lots of links (which I need to explore) which don't include 12172.5.

Searches on leginfo, I notice, always return exactly 20 results (two pages of ten), no matter what you search for. Somehow this doesn't give me a feeling of confidence.

To get directly to a numbered law, leave the search field blank to go to the table of contents for Government or Elections. Then wait a while.

It turns out that Government Section 12159-12179.1has nothing to do with voting procedures or technology, and doesn't have a .5. Hmm.

Well, let's try 19001 and see if it's related. Oops, the table of contents skips from 18993 to 19050 (which is something to do with making General Appointments, anyway).

The Election code, on the other hand, skips from 12113 to 12200, missing 12172.

The 19000s of the election code do, finally, seem to relate to the issue of technology used in polling. But nowhere in the 19000s can I find any mention of paper ballots.

A google search of "paper ballot" option on site leginfo-ca-gov returns no hits.

Is leginfo behind? Or was the lady at Shelley's office wrong about that provision still being current?

Tags: , , ,
[ 13:09 Oct 08, 2004    More politics/election04 | permalink to this entry | ]

CA will (apparently) honor requests for paper ballots

I've been waiting for months for the papers, or Wired, or someone, to give us the definitive word on California's proposed paper ballot option.

Back when Secretary of State Kevin Shelley moved to decertify some of the Diebold voting machines, he included a provision that voters who wish a paper trail may request a paper ballot in counties which use touchscreen voting machines.

But since then, many things have changed, many of the decertified machines have been recertified, and none of the news articles ever mentions the paper ballot option. I've been keeping an eye on the CA Elections and Voter Info site for some time, looking for help or information, but time is getting short to request an absentee ballot, so I mounted a search.

The Elections and Voter Info site has a FAQ -- they only link to the FAQ about voter registration, but that same page has answers about other topics as well, including voting systems. But no mention whatsoever about paper ballots.

The elections page also links to another site run by the Sec. of State,, which has lots of interesting information on things like Diebold decertification and recertification, but still no info on the paper ballot rule (or lack thereof).

Going back to the elections page, I called toll-free phone number for voter info, and spent a few minutes navigating a phone tree, which didn't include any options which seemed relevant; determinedly pressing the numbers for "other requests" eventually ended up in something that wanted to request info from me (for what? I wasn't clear) rather than let me ask questions of a human.

I hung up, and tried the Sample Ballot I received in the mail a few days ago. It has instructions for voting both on touchscreen and on paper, but no assurance that the paper ballot is actually an option for anyone receiving the sample ballot. The only phone number I could find anywhere in the sample ballot was one for requesting ballots in other languages.

Going back to the Secretary of State's web site, I found the phone number for the Sec. of State's office in Sacramento, and called long-distance. Navigating another phone tree (oddly, "Elections Division" is not in the first list of options; you have to choose "Other" which takes you to a menu which includes elections) and ended up speaking with a friendly and helpful woman there.

She assured me that yes, all voters in California would have the option of requesting a paper ballot at the polling place, and she offered to find it on the web site for me.

Several minutes of searching ensued. She initially thought it would be on the Voter's Bill of Rights linked off This turns out to be a PDF of a big-type poster, which, alas, says nothing about paper ballots.

She put me on hold briefly while she went searching, came back, and tried to remember the click-through route she'd taken so I could find it too. We followed several false leads, but finally got there: start at the Elections & Voter Information page, scroll way down to Voting Systems (under "General Information"), then click on Decertification and Conditional Certification for certain DREs to get the 9-page PDF of Shelley's original decertification of the Diebold machines, which, on page 4 item 4.b.1, specifies that every polling place must either (a) have a voting machine offering a "fully tested, federally qualified and state certified accessible, voter verified paper, audit trail" or (b) (1) Permit every voter to have the option at his or her polling place of casting a ballot on a paper ballot which may be satisfied by providing an adequate number of paper ballots to each polling place based on each County's assessment of the number of persons who may request them. The cost of additional paper ballots specified in this paragraph shall be borne by the vendor of the voting sytem that sought its certification or approval for use in California, or the vendor's successor in interest".

(Incidentally, this PDF is simply a scan of the successive pages of the document; there's no searchable text here, so google wouldn't help unless it had OCR capability.)

The woman at the Sec. of State's election division assured me that this was still in effect and had not been outdated by the more recent recertifications, and that it applied to every voting district (presumably there's no currently certified voting machine which meets clause 4.a?)

The status of this document (see page 3) is that it amends Division 19, Chapter 1 (commencing with Section 19001) of the Elections Code and Government Code section 12172.5. So that's the place to go to make sure this is still current. More on that later.

At the end of our conversation, I mentioned that this info was a bit difficult to get to, and maybe a clear FAQ entry, somewhere in the html of the site, might be in order. She agreed. Perhaps someone will update the web site before the election.

Tags: , , ,
[ 12:43 Oct 08, 2004    More politics/election04 | permalink to this entry | ]

Thu, 07 Oct 2004

Good article on laptop Linux

Linux Magazine has a good article by Jonathan A. Zdziarski on Linux on the laptop: Ten power tools for the mobile Linux user. He gives hints such as what services to turn off for better power management, and how to configure apmd to turn off those services automatically; finding modules to drive various types of wireless internet cards; various ways of minimizing disk activity; and even making data calls with a mobile phone.

Lots of good information in there!

Tags: ,
[ 20:51 Oct 07, 2004    More linux/laptop | permalink to this entry | ]

Fri, 01 Oct 2004

New showpix script now uses PHP

I'd been meaning for ages to write a PHP version of my showpix.cgi Perl script, to show images without needing a separate .html file generated for each image. I finally did it this morning, and it was much easier than I expected, and seems to run a lot faster than the perl CGI (not surprising, since PHP is cached in our web server and perl isn't; so this should be more scaleable and less load on the server).

The hardest part was writing the Python script to generate a new showpix.php for a directory of images, and that only because of all the escaping of quotes that needed to be done when telling python to print a line that tells php to print a line to serve up over http ...

Anyway, I've converted the Flume Trail images to use the new PHP stuff, and I've updated the page for the Imagebatch scripts to include PHP ability.

[ 15:40 Oct 01, 2004    More programming | permalink to this entry | ]

Democrat site completely fails with large fonts

Latest accessibility gaffe:

The web site for the Democratic Party is totally screwed up if your fonts are even one step larger than the default. Try it! (ctrl-+ in mozilla.) I wonder if it ever occurred to them that many of those precious Florida voters are older, and need to use large fonts?

I sent them a note, with screenshots of the top of the site and some unreadable text farther down.

I wonder if the Republican site is any better? I'm not sure where it is, and oddly, googling for "republican party" doesn't get anything that looks like an official nationwide site on the first page.

[ 11:38 Oct 01, 2004    More politics | permalink to this entry | ]