Jun 30, 2010
A brace of jellyfish help me celebrate the festival of low fidelity that is the Olympus D510 CBC.
—orc Wed Jun 30 15:53:20 2010
Jun 26, 2010
Silas at the Legoland water park.
—orc Sat Jun 26 22:23:45 2010
Jun 25, 2010
Dust Mite and a friend relax in our elegant hotel room in Carlsbad.
—orc Fri Jun 25 21:59:44 2010
Discount has been pushed up to version 1.6.6 to correct one parsing error and
one defect that would make theme dump core if it tried to create an output filename
from a source filename that didn’t contain a dot.
The parsing error is that I match embedded []s inside []() constructs, so I can
use images as the link for a url (I do this all the time on tsfr), but I was handling
escaped ]s and not handling escaped [s. This meant that a link like
[this\[is\]a](test)
would not parse, because when I encountered the \[ inside the link, I’d treat it as just another [ that I needed to match. And it would match the trailing ], and then discount would run forward trying to find another ] that it could match against.
The theme defect is a bit more stupid than this, and it’s a result of what I wrote it for. I wrote theme as a demo to show how the discount library works, of course, but I did it so I could redo my web site into markdown (and eventually write an apache content handler that dynamically generates html from markdown, but I’ve not actually gotten around to it yet :-() And how I did that was a wholescale renaming of all of my .html files into .text, which I then went over and stripped out the html and replaced it with markdown. And, as a convenience, I wrote things so that all I needed to do was give it the name of a sourcefile and it would automatically generate the html filename by removing the .text and replacing it with .html.
“But what”, you may ask, “happens if there’s no .text on the source filename to replace? Do you just append the .html?”
That’s what you’d think, but, no, theme was doing a p = strrchr(sourcefile, '.') and then, without bothering to check for teeny details like if ( p != NULL ), would just charge ahead and do strcpy(p, '.html') with the expected hilarious consequences if p was NULL.
Siiiiiiiigh.
But it’s fixed now, and this New Code! is ready for your compiling enjoyment.
—orc Fri Jun 25 09:10:56 2010
Magicfilter (remember magicfilter? It’s a printer filter that lets you print anything from the command line) has been trivially updated to version 2.3.1 because I needed to rename getline() into getrule() to avoid namespace collisions with some stupid gnu extensions in gl*bc, and I needed to modify some functions in configure.inc to deal with some stupid gnuifications in MacOS 10.5.
There aren’t any other enhancements, alas. I’ve been meaning to write an xml to goodcode mapping function so I can use the foomatic database for the postscript and pcl output rulesets, but that’s fairly low on my list of priorities because I’d need to completely rewrite the MacOS printer subsystem to use something other than C*PS so I could debug the whole printer process.
But, aside from that, it’s New Code that’s ready for those of you who still actually use magicfilter.
—orc Fri Jun 25 01:05:57 2010
Jun 24, 2010
I dropped the Olympus when we were at Legoland yesterday, and, aside from the case being a little sprung it seemed to be okay. But, alas, it was not. It’s been slowly deteriorating ever since then; first it started to write images even slower than before , and then it went to irregularly locking up, and finally it has gone to only letting me take one picture, and then forcing a power cycle before it would take another.
It was free, so it’s not a total waste. But it’s still annoying. Particularly when I take a framing picture of a passing train only to realize that I’ll have to powercycle before I take the real picture thereafter.
—orc Thu Jun 24 19:16:26 2010
We drove out for an errand this morning, and as we were heading north along the Santa Fe mainline, I saw a southbound Coaster train pop out from under the distant bridge and scuttle towards the station we had just driven past.
The CBC, as usual, failed to respond to the moment, so I got this one washed out distant photo, but the train had zipped on by by the time it finished s-l-o-w-l-y writing the image out to the smartmedia™ film card.
—orc Thu Jun 24 13:30:34 2010
Jun 23, 2010
Russell disapproves of having people take pictures of him, but at least he smiles while he’s trying to ruin the photo.
—orc Wed Jun 23 23:31:34 2010
A snowy egret ignores the people with cameras at Legoland California.
—orc Wed Jun 23 22:45:47 2010
Jun 22, 2010
The plan was a nice loop up OR224 to Timothy Lake, then over Barlow Pass, back through Government Camp and (after a screaming descent down to about 10 miles east of Sandy) Sandy, then backroads across to Edgefield, followed by city streets to home. I’d ridden the first 50 miles of the loop a few times, so I knew that the start would verge on insanely hard, but then it would temper to a fairly moderate climb up to Ripplebrook, then a relentless stiff grade 2500 feet up the mountain to Timothy Lake, followed by roller(ish) riding over to Barlow Pass and (basically) all downwards from there. And, except for a handful of subdivisions around Oregon City and Gresham/Troutdale/east Portland, it would all be riding out in the country, even on the 30 miles of US26 from west of Barlow Pass to the middle of Sandy. And, judging from my performance (or lack thereof,) it would all be doable by everyone up to and including the slowest of the slow.
So I asked around for people to help me do a checkride on June 19th, and got nibbles from Ed Groth & Quang Vu Dang, who met me at the ung-dly hour of 5am and headed off for points east.
And, at least at the start of ride it was pretty much as I imagined; 17th & Milwaukie Ave are fairly urban, but pass by quickly, and then it’s a short wind along a bicycle path and back streets to the first (short but) steep climb on River Road, which then settles down to quietly rolling along all the way down to Oregon City, where we crept up the second steep climb to the top of the bluff, then headed south along Central Point Road until we fell off the Boring Lava at the edge of the urban growth boundary and headed down New Era Road on our way to the first control at Canby (interrupted only briefly by a stop when my cue sheet peeled off my handlebar bag and fell to the ground on the ramp down from the southern edge of Oregon city.) Past Canby, we headed east through a mix of woodlots & fields (there is suburban development around the edges of Canby, but not much, and by the time you cross the Oregon Pacific RR on Township you’re pretty much out of city until, um, you roll into Portland again.
I ran into a piece of broken glass somewhere between Canby and Estacada, and got a slow leak in my rear tire that didn’t become noticable until we were about ¾ths of the way up the insanely steep ramp that is Buckner Creek Road, so we burned 25 minutes of our time as I s-l-o-w-l-y replaced the tube. But we still reached the pacing control at Estacada a hour up on the clock, so we were doing okay, and at that point we’d finished the steepest climbs (and over a third of the total climbing) of the loop, so it all looked good from that point on.
Until Ed got a flat on OR224. Followed by me running over a stone and getting my second flat of the day. 35 minutes stopped while we repaired the flats, and all the lovely spare time that we’d banked went right down the tube, so we reached (more or less) the top of the ramp on NFD 5810 with about as much time up on the clock as I had when I went up to Timothy Lake on my Trek, but that was okay (and the climb up was also as expected; it started with a speedy run up the scenic-despite-the-weather OR224, a slowish run up the last mile or so into Ripplebrook (where the store was open, and this time I found the water faucet thanks to Ed showing me where the thing actually was,) and then the expect slow creep up the relentlessly graded ramps on NFD57 (punctuated by level(ish) parts where we rode along, dodging the occasional newt & lizard, through a tree-lined corridor with small bits of sunshine peeking through the clouds) and NFD 5810.
The funny thing about NFD 5810 was that last year, even though I had a much smaller alpine gear (~30" instead of the ~40" on the MLCM,) it still seemed like a harder climb than it did on Saturday. It’s still a relentless grade (pitching up to 7% in places) and it goes on forever, but Ed & I spent some of the climb joking about how easy this was compared to the fresh horror that is Buckner Road.
And then we reached NFD 5820, and everything proceeded to fall apart.
I remembered NFD 5820 as being paved. It is not. It was, once upon a time (we kept seeing patches of pavement as we rolled along it) but it has been resurfaced with gravel. And it goes up 300 feet to ~4000 feet, drops down to Anvil Creek (but it’s
not washed out here, unlike NFD 5810), climbs back up to ~4000 feet, then plunges,
all gravelled, about 700 feet down to the junction with the other end of NFD 5810.
And while we crept along NFD 5820, we were having flats. About one flat per mile on
this over 5 and a half miles of gravelled roads. After the fourth flat, I looked at our time (we were now deeply behind the clock to the tune of a hour and change into the Clackamas Lake Ranger Station) and suggested that we abandon and cut the loop short on US 26. No, no, the consensus was that we should continue up to Barlow Pass. And then we had a fifth flat, and as we sat there by the side of the road doing tube repair (Ed’s bike had the flat, but I was attempting to patch the two tires that I’d punctured earlier so that when we had the next flat, we’d have a tube at the ready) we looked at the time, looked at the road, said “f*ck it!”, and abandoned the brevet on the spot.
And then all we had to do was ride another ~80 miles to get back to Portland, and it was about 4pm.
NFD 57 & 42 were, as expected, rollery all the way up to US 26 (ascending ~400 feet in 22 miles) but US 26 had some unexpected surprises in the form of a collection of really nasty false flats (the summit of US 26 between the junction with NFD 42 and the junction with OR 35 is Blue Box Pass, at 4000+ feet, and we didn’t realize just how much climbing we had done before we got there until the summit signpost crept into view. And then after dropping off that point of land we dropped into a river valley and false-flatted our way up to the OR35 junction.
The 350 feet of real climbing to get up above Government Camp was a relief compared to this because it was steep enough that we knew we were actually climbing.
And after a 45 minute hot food + coffee break in Government Camp (where the temperature dropped about 10 degrees while we were in the Huckleberry Inn eating) we threw on every layer of clothing we had and threw ourselves off the edge of the cliff for the screaming descent down US 26 to the river bottoms just east of Sandy.
I’m not sure just how fast I was going on the initial descent (because my GPS had run out of power a hour earlier, and I’d forgotten to bring a booster battery) but it was fast enough that the MLCM started to develop a nasty front wheel shimmy about halfway down.
(Or maybe earlier, but I was shivering too hard then to notice.) But we were moving; it took us about a hour and 15 minutes to ride the 26-ish miles between the lower end of Government Camp Loop and the eastern edge of Sandy, and 3-4 miles of that were the climb up (in the DARK and in the RAIN, I may add) out of the Sandy River gorge at the end of the descent.
Needless to say, we did not loop over to Edgefield, but just followed 26 (in the DARK, but, thank goodness, the rain had stopped by then) all the way into Gresham, punctuated by one final flat, then hopped onto the Springwater Trail for the last dozen miles home (arriving at the stroke of midnight,) where I sat down, ate three meals, dumped my photos onto the computer, and slept the sleep of the dead for about 7 hours before I had to wake up and help pack and clean the house so the family could fly down to That Paradise That Is California™ for a vacation trip to Legoland. And when I finally got out on the line – for a short trip to get takeaway food for dinner – I found to my lack of surprise that my legs just didn’t have enough power to push me up hills at even my normal slothlike pace.
So I’m going to have to revise this 300 before I submit it as a permanent. The fast boys (particularly the fat tire bunch) would probably love it (or consider it a worthy challenge) as it stands, but the one flat per mile section of the ride (that’s at least 75 minutes burned up right there) makes it much less doable for the 17+ hour crowd. I know that I’d be fairly fretful about the whole idea of riding over 5.5 miles of tire-eating gravel, even if the Clackamas Lake Ranger Station is only a couple of miles of walking past that. I suppose that I could just change the loop to take NFD 58 up to High Rock, then NFD 58 and NFD 5890 back over to the point where NFD 5820 becomes paved – that only adds about 10 miles to that part of the loop, but on the other hand it adds another 700 or so feet of climbing on NFD 58, which apparently gets fairly steep(er) as it gets closer to High Rock. But if I do that, I’ll probably have to drop the Buckner Creek section of the Canby-Estacada leg, because having two sections of mind-destroying grades is not really compatible with my wanting to build a permanent that slower riders can do in finite time.
And I discovered one really annoying thing about US 26. It’s got rumble strips. And they’re placed on the shoulder side of the white line. And they aren’t visible when it’s wet and dark. And they aren’t continuous. And the shoulder isn’t reliably wide. So when it’s dark and raining, and you ride close to the white line so you don’t end up pitching down an embankment or into a drain, you will ride into one of them, and they’ll suck your wheels into them.
The first few times (of maybe 6 or 7 times) I hit the rumble strip I ended up being squirted out into the travel lane as I fought to get out of the velovibrotron without flipping the bicycle over in the process. If US 26 had been busy (instead of it being ~9:45pm on a Saturday night) this could have been unpleasant. Admittedly, this is not likely to be a problem for riders on a permanent (because I’m not going to make this one available in the short-dayed months, so the section on US 26 will happen during more daylightish hours if you’ve not already DNF or DQed out) but it makes the idea of routing along Marmont Road seem awfully appealing, despite there being a couple of extra-annoying climbs there.
In the category of things that worked vs. things that didn’t work, well, I’m not sure. NFD 5820 didn’t work in a fairly serious way, but it did provide a nice test road for the magnetic latch I was field testing. The 40" low gear was pretty painful when I was climbing Buckner Creek Road (but some of that was because the chain in the MLCM is worn to the point where it’s skipping), but I didn’t have to walk any of the ramps except when I walked up to a driveway to fix my first flat. I am fretful about the high speed shimmy that developed on the 2500' descent out of Government Camp, but all I need is $3-400 and a call to a framebuilder to get something with enough trail to fix that wagon (or I could just not descend long ramps quite so enthusiastically in the future.)
But, aside from that, it was a very nice loop, and all I need to do is correct the tire-eating part and it will make a nice 300km permanent that fulfills my desire of not having to ride 45km to get to the starting point. I’ll fix the routing and do it again later this summer, and maybe then I’ll be able to place the info controls before the sun goes down.
And I borrowed a camera, so there are pictures documenting the insanity.
—orc Tue Jun 22 23:23:54 2010
Jun 20, 2010
Yesterday, when I rode up to Mount Hood and back, I took along, as part of my food collection, a package of Clif “shot bloks”, which I ate about half of, then dropped the rest into my pocket for later consumption (that never happened, despite me not getting home until midnight.)
Today, I emptied out my pockets and dropped my amazingly sweat-soaked plus 4s into the wash, but I managed to miss the pocket containing the shot bloks.
I didn’t even realize that they were there until after the plus 4s came out of the dryer, and I was folding my clothing so I could put it away.
The horrifying discovery? No, it was not that the lining of my zipper pocket had become a mass of melted and congealed gelatine. No, it was that the shot bloks looked just like new, without any signs of being dissolved or otherwise mangled by the washer or dryer.
And I’ve been eating these things?
Eeeeeuuuuuwwwww!
Excuse me; I need to go and scrub out my digestive tract now.
—orc Sun Jun 20 23:01:09 2010
Jun 18, 2010
The first two lid latching mechanisms (velcro, plastic snap) that I experimented with on my handlebar bag prototypes turned out to be less than ideal, so I’m trying out a third – my prototype 2 bag has been retrofitted with a magnetic purse snap, which will be tested on a R300 tomorrow morning.
The wonderful photo quality is courtesy of the Olympus D510 CBC, which is a terrible substitute for a real camera, but it’s all I’ve got :-(
—orc Fri Jun 18 21:41:03 2010
Dust Mite hangs around with a few friends.
—orc Fri Jun 18 21:07:55 2010
A panoramic view looking north from Bohna Park Road; I was out on a short loop to Boring, but had to stop to get this picture (actually “these 22 pictures, which I then photostitched together”) once I realized how pretty it was.
—orc Fri Jun 18 21:01:07 2010
Jun 17, 2010
Perhaps I’m too used to using a SLR (aka WYSIAYG) camera, but when I was composing this picture (taken through the screened side of the Springwater Trail bridge over the Yellow Menace) I forgot that when I was carefully composing it to make sure that none of the screen was visible, I was not taking into consideration where the actual lens was pointing.
Whoops.
—orc Thu Jun 17 21:29:33 2010
I took this picture at ~18:00, so the sun, even though it was hidden by the clouds, still a long ways above the horizon. So I’m not really sure why the reflective tape on the side of these Yellow Menace locomotives was acting quite so reflective as it did here.
—orc Thu Jun 17 20:07:03 2010
Jun 15, 2010
The lid of my handlebar bag is not a very tight fit, so when it rains like the dickens, as it did for about 6 miles of my loop through east portland today, water actually leaks in and collects in the bottoms of the side pockets.
Fortunately my Olympus D510 CBC recovered from the shock of a cold shower after I popped open the battery bay and let it dry out for a hour or so.
But the next bag (I’ve a yard of a more brick red cordora, as well as a yard of a lovely orangey cordora, and Mill End is going to be getting some waxed canvas soon, so I can make up some bicycle luggage for the more hardcore randonneurs) will have the lid be just a little bit wider than the body so that the rain will have to work harder to sneak on in.
Oddly, my saddlebag, which tends to gape when it’s loaded (and which is kept pushed back by my saddlebag support), didn’t get any leakage, and the double layer of shell + lining kept the rain out without any drama. But I don’t tend to open and close the saddlebag nearly as much as I do the handlebar bag.
—orc Tue Jun 15 22:54:15 2010
EPT 187 sits at the end of a siding just west of what used to be Golf Junction when Portland Traction still had a line running south to Oregon City (the ROW is now going to be the “trolley trail”, except for about ¾ths mile going up the ramp from downtown Milwaukie to just north of Oak Grove; that ¾ths mile is going to be part of the new Tri-Met Milwaukie branch, which, other than that, will be built along roads and the SPYellow Menace / Portland & Western ROWs.)
—orc Tue Jun 15 16:29:28 2010
Jun 13, 2010
Last week my Pentax bit the dust by falling out of my handlebar bag, which was bad, because it will cost several hundred dollars to replace if I’m lucky and can find a cheap auction on ebay. So I decided that I’d just sell off my macbook air and go back to the old macbook instead, but when I started moving all of my software over to it I discovered, in quick succession, that (a) the cd-rom drive in the macbook was dead, then (b) the macbook itself would not reboot when I powered it off to do a force eject on the macos install cd that was going whiirt-whiirt-whiirt in the drive. It’s been sitting there with a little spinny wheel under the macos logo for, oh, about 15 minute now, and in the grand tradition of macos it’s not actually telling me what’s going on there that the New! Improved! version of MacOS that Apple’s software update program installed for me broke CD-rom support (to the point where I had to reset the bios to get it to recognise (via an <option> boot) the cd-rom again,) aaaand the OS update script was gumming up a command line (I only noticed this by an <open-apple>v boot where I could actually see the Unix kernel boot messages scrolling by, then abruptly and ominously stopping with a string of “command not found"s for every argument on a commandline which moved in a new version of sudo) which stopped the boot process dead in its tracks, thus converting the Macbook into a ominously humming boatanchor.
Stupid stupid computers.
And it takes effing forever to install MacOS 10.4, then upgrade it to MacOS 10.5 (which, now that 10.6 is out, will probably suffer the same sort of “oh, hey, that’s an old machine and OS – here’s an update that will totally gum it up for you!” failure sooner or later.)
But, after a brief 24 hours of rsync'ing my iPhoto library over (a 2mbit wireless link isn’t that slow for day-to-day activity, but it’s really slow when running rsync on a large machine) the Macbook Air is now out of service and ready to be sold.
*sigh*
—orc Sun Jun 13 14:34:52 2010
Jun 12, 2010
This old Kodak point&shoot used to belong to Russell before it had an unfortunate run-in with a tree branch (that ended up crimping the lens and shattering the quality plastic gears that operated the zoom/focus mechanism.) Russell now has a bigger and better Nikon point&shoot, and this one rots away in the project box.
But it still works, and I was able to do an experiment with the shattered lens where I removed the optics and most of the barrel, leaving just enough for the camera to spin when it powers up (if it can’t actuate the lens, it gives me a lens error and shuts down), then assembled a q&d lightbox and held up one of the lenses from my Auto 110, I could actually get properly focused (if somewhat hazed due to light leaking in around the corners of the box) images.
It would still be a point-and-shoot, I’d still have to fabricate a viewfinder (a rangefinder is well past my mechanical abilities), and the sensor chip is about the size of a gnat’s elbow, but it would still be a good cheap placeholder until I can sell enough stuff to buy something evil and a m42 to µ4/3rds adapter (Pentax *istDSes seem to be fairly sought after on the used camera market, so I’d have to spend >$200 to get one, and that’s getting up into the evil camera department right there.)
—orc Sat Jun 12 23:56:26 2010
Dorrie walks over to be scritched.
—orc Sat Jun 12 17:53:02 2010
Furnish him with your details your files would be updated and he will be releas
ed to you by the Federal Government including taxes, custom paper and clearance
duty so all you will ever need to pay is 385.00 only.
The undisclosed organization.
The office of the Central Bank Of Nigeria, Mr. Patrick Aziza, Frank, Anderson,
none officials of Oceanic Bank, Zenith Banks, Kelvin Young of HSBC, Ben of FedEx
, Ibrahim Sule, Dr. Usman Shamsuddeen and some imposters claiming to be The Federal Bureau of Investigation FBI, where deployed to investigate on issues of deb
t funds as a result of the mandate from US Government to make sure that any huge
amount of money like yours that enters United States Dollars to be 100 risk free and free from any hitches as it’s our duty to monitor everything in regards
to your designated home address without any further delay, extra fee or any authority raising eyebrow.
Scam via the above fax number so we could act upon it immediately. Help stop cyber crime.
Card will be effected to your transaction, hence the
chairman is presently under investigation for conspiracy with your
representative with intension to divert your funds.
Yours Faithfully,
You are advised to desist from contacting anyone including your agent or
representative in these regard, International Monetary Fund I.M.F ,
- YOUR BANKING DETAILS
- YOUR ADDRESS
YOUR OCCUPATION
YOUR BANKING DETAILS
- YOUR ADDRESS
- YOUR BANKING DETAILS
- YOUR ADDRESS
- YOUR NAME
- YOUR TELE FAX NUMBER
- YOUR TELE FAX NUMBER
- YOUR OCCUPATION
It has come because Change can happen .
Not illegal.
Valued US 15.8 Million United States Department of the individuals and organizations who are yet to receive your payment because it’s more guaranteed, since
over 5 billion was lost on fake cheque last year 2008.
- of Birth mm dd yyyy
- Driver’s License Passport Copy
- Marital Status
- Current Employer Name
- Middle Name
- Home Phone Number
- Cell Phone
- Occupation
- Age div
By Alh. Ahmadu Giade of the Senate president in Nigeria, because that is the contact information of the Senate president in Nigeria, because that is perpetrated by those impostors.
Affairs grows to be 100 risk free and free from any impostors or offices claiming to be The Federal Bureau of Investigation.
Receiving your payment totaling 6,100,000.00 will be used to conduct lawful underlying principles of verification on your name appeared as owner of funds value
d US 15.8Million United States Dollars to be 100 risk free and free from any impostors or offices claiming to be The Federa l Bureau of Investigation FBI, where
deployed to investigate on issues of debt funds as a matter of urgency to contact the ATM Card Center via his contact details above and furnish him with your details your files would be updated and he will be released to you.
We noticed that over the past you have been impersonating the likes of Prof. Soludo of the event was a solution to your payment because it’s more guarantee
d, since over 5 billion was lost on fake cheque last year 2008.
- Home Address
- Home Cell Phone
- Occupation
- Age div
New are hereby advice only to be in possession of your funds but to no
avail. Our investigation continued to the release of your ATM card will be used
to conduct lawful underlying principles of verification on your behalf which will cost you 385.00 only nothing more and no hidden fees as everything else has been taken cared of by the Federal Ministry of Finance that your name appeared as
owner of funds valued US 15.8Million United States of America. This is as a result of this we hereby advise you to reconfirm your personal fund.
—orc Sat Jun 12 16:21:04 2010
Jun 11, 2010
Now that the Pentax is dead, I’m using the Olympus D510 as my emergency camera (I could use my old Polaroid PDC3000, but it writes images in the proprietary “PDN” format and I do longer have a Windows machine to run the (slow and awful) PDN->Tiff converter. And I don’t have a Compact Flash reader either, so I can’t even read the images off the film. Sigh. I guess I should sell the PDC3000 to someone who can make better use of it) and, unlike the *istDS, which can/could stretch a set of batteries out to about 1000 pictures, the D510 goes through a set of batteries in about 4 days. Which meant that when I got home tonight and tried to fire it up to get a picture of Dust Mite, it just beeped the “I’m out of power” song and shut back down.
Which left me with no camera except for the VGA camera that’s in the macbook Air.
Thus this post, a celebration of the crappiest possible FDMB photo ever made.
—orc Fri Jun 11 22:57:26 2010
Buckley dozes on the sofa after a long day of feline carnage.
—orc Fri Jun 11 21:14:27 2010
To round out the week, Postoffice has been rolled up to version 1.5.3 with the correction of one annoying defect in the mfdata() function in the milter library.
This defect wasn’t one that would make things explode, it just was a case of my mishandling the little scanner inside mfdata() that split the mail message up into headers and body, so as a result the entire message, except for the first line, was scanned as a body. Spamassassin would then not parse things quite as I wanted, and was rejecting good mail because it thought that the headers matched the bayesian patterns of known bad mail. (This was only happening for me with the unstoppable daily status messages from ebay, but some other people were seeing good mail misfiled into their SPAM folder.)
I believe I have fixed that wagon with this (tiny) release.
—orc Fri Jun 11 21:10:55 2010
Jun 10, 2010
The Bears huddle under the umbrella as they walk through the pouring rain.
—orc Thu Jun 10 22:13:40 2010
Discount has been pushed (after a long delay because I’ve been flakey this year) up to version 1.6.5 with the addition of a couple of features and the deletion of a few more bugs/defects.
The now-elided bugs are:
- properly handle html comments w/o spaces
- fix nested list handling with unindented
list items.
- Add config.h to the dependency lists for
cols & echo, so those functions will properly
rebuild after a reconfig.
- More plan9 fixes from Josh Wood; this one
adds the flag
_C99_SNPRINTF_EXTENSION to tell
the APE cc that, yes, I do really want to be
using snprintf() and I know that C99 has it
return the size it wants to populate.
- don’t describe
<style> blocks as mystery node during markdown -d
And the new features are:
- add a VALID_DOCUMENT magic flag to the MMIOT
structure, so that if someone attempts to double-free
it, the second attempt will just not work instead of
causing a core dump.
- restructure the blocktag[] array into a STRING(), so
users can add additional block tags as they desire.
- in relation to this, add the library function
mkd_add_html5_tags(),
which adds (globally, and non-removably) a handful of
new tags for html5 support.
As an artifact of me being flakey, these features & bugfixes have not been tested nearly as much as I would want, but I have written testcases for them, which they pass, and they don’t appear to make the markdown test suites fail for me. So they’re just the New Code! that you want to install, so you can test them for me.
I’ll be in the bomb shelter waiting for bug reports.
—orc Thu Jun 10 19:03:57 2010
Jun 09, 2010
Who would have thought that saving approximately 35,000 jpeg images would pretty much fill up an 80gb disk (as of now, after a couple of rounds of cleaning up everything but
iPhoto, there’s about 700mb left free on crepe.)
And because it’s a macbook air, it’s (a) not particularly easy to dismantle it to remove the hard disk, nor is it (b) easy to find an inexpensive larger disk for the thing (and, really, by “inexpensive” I mean “free” – I’m using an ancient free Olympus D510 as my temporary camera now that the *istDS is dead and in pieces, but whatever money I’ve got from selling off my more meaningless possessions (a chain of LED deck lights? WHY???) is going to have to go to get a replacement camera body, preferably one that has replaceable lenses, but definitely one that has a < 1 sec startup time and > 2 images/sec cycle time. And I don’t expect that my lottery ship will be coming in to solve that problem) so I guess I could just move everything over to the old macbook poison (that had its 60gb drive die under warranty, so it was replaced with a 250gb one because the Apple Store people couldn’t get 60gb units any more) then put the air up for sale to reduce the # of excess computers lying around the house.
—orc Wed Jun 9 18:25:21 2010
Jun 08, 2010
I pulled the Pentax apart to extract the shattered LCD screen and see if there
were any obvious loose connections between the camera, the processor board,
and the SD slot.
The SD slot is built onto the processor board, so no, nothing loose there, but
I did disconnect all of the connections to the processor board, lifted it, and
set it back down before reconnecting everything (except the LCD – part# 270CGX 446S04A, which appears to be obscure enough so that not even the big search engines can find it) together.
Still no joy in mudville; the shutter continues to stick, and the images are pretty much all black. So there’s either comething loose deep inside the system (and to pull it apart would probably require me to cut all of the various power lines and solder submini couplings into them for easy assembly and disassembly) or the ultimate drop managed to break something.
So I’ll keep haunting ebay to see if a used *istDS shows up, then buy a few lottery tickets to see if it’s time for me to buy an Epson R-D1 + a k-mount to leica mount adapter (though if I did win the lottery I could just go out and buy really fast lenses to replace the poor broken Pentax f1.2 prime I’ve got now) , or something more evil than that.
(If I do get something evil, I’m going to get one of the metal ones. There are too many pieces knocked out of the bottom of the Pentax to make me happy with plastic.)
—orc Tue Jun 8 23:22:36 2010
Jun 07, 2010
When I was returning from the store this afternoon, I first encountered YAFYE on the SPYellow Menace mainline, and then after riding onwards to the Springwater Trail the Replacement New Working Eng! pulling a train northwards to the East Portland interchange yard. The EPT engine is much more interesting than the usual round of monopoly yellow.
—orc Mon Jun 7 21:09:31 2010
Jun 06, 2010
I’ve been doing some short pre-rides of parts of my Sellwood to Hood & Back R300, and today I rode the segment from Canby to Estacada, which avoids OR211 by the simple expedient of taking back roads from Canby on east up until the road I chose deadends into OR211 (and forces 6 miles of travel along that road before you reach Estacada and turn up OR224 towards the mountain.)
Unfortunately you pay for the quiet scenic country roads by having to ride on them. And these quiet country roads undulate enthusiastically, plunging down into valleys, then climbing even more steeply up out of them, then erasing much of your progress by dropping abruptly into even deeper valleys. My GPS claims 2800 feet of climbing in the 27 miles from Canby to Estacada. My legs and lungs think it is somewhat more, due to the teeny detail that the climbing came in fits and starts (including a couple of short 20+% stretches along Buckner Creek Road) instead of the sort of gradual uphill that OR224 provides as it winds its way towards Ripplebrook alongside the Clackamas River.
But, at pretty much the top of the hill – ~1200 feet ASL – Unger Road comes out from a wooded section and gives you a spectacular view of the mountains across the fields of a christmas tree farm. And then it’s just a couple of 300 foot climbs (with corresponding descents) before OR211 plunges down into the valley of the Clackamas for its rendezvous with OR224 and the long run up to Ripplebrook, NFD57+5810+5820, and Timothy Lake. But before all that you need to take in the view.
—orc Sun Jun 6 00:00:38 2010
Jun 05, 2010
I was out riding today, and my *istDS flew out of the handlebar bag after I hit a pothole, but this time the Pentax didn’t survive the fall (I was going slowly enough so it didn’t skitter, but instead landed directly on a rock, which shattered the LCD and did something to the shutter mechanism and, obviously, the photo sensor.) So it’s not usable anymore.
Damnit.
I guess I’ll see if I can get a cheap used *istDS somewhere, and if I can’t I’ll just sell all my Pentax stuff off and revert to film.
—orc Sat Jun 5 19:51:42 2010
Jun 04, 2010
Dust Mite makes a new friend at the Llewellyn Carnival.
—orc Fri Jun 4 20:39:54 2010
Jun 03, 2010
The Trek lined up against a stock 2009 Kona Ute.
—orc Thu Jun 3 18:49:58 2010
—30—
![[rss feed]](/~orc/feedbutton.png "Atom + RSS feeds available here"){kind=small}