Matt McIrvin's Steam-Operated World of Yesteryear

Some Sage code about Fibonacci-like sequences and primality tests

matt.mcirvin@gmail.com — Sun, 19 May 2013 04:25:11 GMT

For a little while I've been poking around in some basic number theory using the Sage computer mathematics system (and a tiny bit of PARI/GP, which is another package that comes bundled inside of Sage).

I was initially inspired by a blog post of John Cook's about the Perrin numbers, a sequence sort of like the Fibonacci numbers that can be used via a simple further operation to generate what seems to be a list of prime numbers (and it in fact contains all the prime numbers, but eventually starts including some composite ones as well... starting with 271,441.)

This seems pretty remarkable, but actually it's just one of a whole family of such sequences. Every recurrence relation that generates the next number in a sequence as a linear combination of the previous k numbers can be used to produce a primality test like this. Some of them aren't very good, though! They all pass all the primes, but they have different frequencies of false positives, or "pseudoprimes". The Perrin sequence is one of the best ones at order 3, meaning the previous 3 values are involved.

Anyway, I wrote a lot about this in a series of posts to Google+ using the hashtag #pseudoprimes, in which I gradually became more and more deconfused. Cook wrote about a few of my early explorations here.

I went on to do some automated searches to find the recurrence relations that generated pseudoprimes with the most impressive infrequency, or at least had the biggest first ones. I found some whoppers, especially at order 4 where there was one whose first pseudoprime is 8,016,241; but I also discovered, to my mild surprise, that the most impressive one at order 2 seems to be a case that is not very well-known (though I doubt I am its true discoverer; several closely related sequences are in OEIS, and I'm just noodling around here).

The main purpose of this post is to provide all of my Sage code. The following appears to be written in Python, but you will need Sage (or an account on a Sage cloud server, like the spiffy new thing at https://cloud.sagemath.com/ ) to actually run it.

The pseudoprime searches are kind of computation-intensive, so I'd be careful about running them on somebody else's machine. Note, here I don't actually use the methods I was talking about in the post John Cook referenced, because they don't seem to work well in Sage (they work better in PARI/GP, which is what Joerg Arndt used in the algorithm on the OEIS page for Perrin pseudoprimes). For some of the toughest order-4 cases, I did my automated searches in Sage with a feasibly low search limit (around 10^6) on each sequence, then switched to PARI/GP to actually find pseudoprimes for the few that exhausted that limit.

Many of these functions take a matrix M that defines a recurrence relation. You can make such a matrix with the recur function, which takes a list of the coefficients that the most recent values in the sequence will be multiplied by, and added to get the next one (for the Fibonacci recurrence it would be [1, 1]; for the Perrin numbers, [1, 1, 0]).

For every such recurrence, there's a special set of seed values where the nth sequence value is the trace of Mⁿ, and that was the case I was mostly interested in because those are the ones that most easily generate primes and pseudoprimes. (For the Fibonacci case, that sequence is not the Fibonacci numbers, but the closely related Lucas numbers, which start with [2, 1] as the seed values.) sequence_for_seed lets you specify your own seed value if you want to explore seeds other than that one.

# Sage Python functions for messing around with linear recurrence
# sequences and pseudoprimes, by Matt McIrvin, May 2013.

# Generate the recurrence matrix for a given list of coefficients:
# [c0 ... c(k-1)] where p(n) = c0 * p(n-k) + ... + c(k-1) * p(n-1).
# This just has the coeffs in the bottom row and a shifted identity
# that moves the other sequence values elsewhere.

def recur(coeff):
    dim = len(coeff)
    M = zero_matrix(ZZ, dim, dim)
    M.set_block(0, 1, identity_matrix(dim-1))
    M.set_block(dim - 1, 0, matrix(ZZ, 1, dim, coeff))
    return M

# Find the initial seed values for a given recurrence matrix
# for the sequence corresponding to tr(M^n).

def seed_values(M):
    p = []
    for n in xrange(0, M.ncols()):
        p.append((M^n).trace())
    return p


# Iteration step.

def find_next(p, coeff, r):
    pnext = 0
    for k in xrange(0, r):
        pnext = pnext + p[k] * coeff[k]
    # advance the list of values
    del p[0]
    p.append(pnext)
    return pnext 


# Pseudoprimality test.
# pnext = sequence value, tr = matrix trace, n = index being tested

def is_pseudoprime(pnext, tr, n):
    return (pnext % n == tr % n) and not(is_prime(n))

# Recurrence sequence for k*k matrix M and initial list of values [p0 ... p(k-1)].

def sequence_for_seed(M, p, limit):
    seq = p[:]
    r = M.ncols()
    coeff = M.row(r - 1)
    for n in xrange(r, limit):
        seq.append(find_next(p, coeff, r))
    return seq
    
# Return the recurrence sequence for matrix M, up to position limit.

def sequence(M, limit):
    return sequence_for_seed(M, seed_values(M), limit)

# Ratios of successive values.

def ratios_lag(L, lag):
    out = []
    for k in xrange(1, len(L)):
        if k - lag >= 0 and L[k - lag]!= 0 :
            out.append(N(L[k]/L[k - lag], digits=15))
    return out

def ratios(L): return ratios_lag(L, 1)

# Print all pseudoprimes and their factorizations for M up to limit.

def pseudoprimes(M, limit):
    # The naive method is actually faster on Sage, and feasible for smaller numbers
    # First fill out the vector
    p = seed_values(M)
    r = M.ncols()
    coeff = M.row(r - 1)
    tr = p[1]
    for n in xrange(r, limit):
        pnext = find_next(p, coeff, r)
        if is_pseudoprime(pnext, tr, n):
            print n, factor(n)


# Just return the first pseudoprime for M less than limit, or None.
# Used in the coefficient searches.

def first_pseudoprime(M, limit):
    p = seed_values(M)
    r = M.ncols()
    coeff = M.row(r - 1)
    tr = p[1]
    for n in xrange(r, limit):
        pnext = find_next(p, coeff, r)
        if is_pseudoprime(pnext, tr, n):
            return n
    return None


# Print some interesting information about M and its characteristic polynomial.

def characterize(M):
    char = charpoly(M); print char
    solns = solve(char(x), x); show(solns)
    print M.eigenvalues()
    print M.eigenvectors_right()

# Searches of coefficient space for recurrence relations with big first pseudoprimes.
# Will print the bottom row of the matrix for all cases with integer coefficients
# between -coeff_limit and coeff_limit inclusive, with a first pseudoprime at least
# as large as print_min. If the first pseudoprime is >= max, will print OUT OF RANGE.

def rank_2_search(coeff_limit, print_min, max):
    for c0 in xrange(-coeff_limit, coeff_limit+1):
        for c1 in xrange(-coeff_limit, coeff_limit+1):
            M = Matrix([[0, 1], [c0, c1]])
            p = first_pseudoprime(M, max)
            if p == None :
                print "[ ", c0, c1, " ]: OUT OF RANGE"
            elif p >= print_min :
                print "[ ", c0, c1, " ]: ",p, factor(p)
                
def rank_3_search(coeff_limit, print_min, max):
    for c0 in xrange(-coeff_limit, coeff_limit+1):
        for c1 in xrange(-coeff_limit, coeff_limit+1):
            for c2 in xrange (-coeff_limit, coeff_limit+1):
                M = Matrix([[0, 1, 0], [0, 0, 1], [c0, c1, c2]])
                p = first_pseudoprime(M, max)
                if p == None :
                    print "[ ", c0, c1, c2, " ]: OUT OF RANGE"
                elif p >= print_min :
                    print "[ ", c0, c1, c2, " ]: ",p, factor(p)

                    
def rank_4_search(coeff_limit, print_min, max):
    for c0 in xrange(-coeff_limit, coeff_limit+1):
        for c1 in xrange(-coeff_limit, coeff_limit+1):
            for c2 in xrange (-coeff_limit, coeff_limit+1):
                for c3 in xrange (-coeff_limit, coeff_limit+1):
                    M = Matrix([[0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1], [c0, c1, c2, c3]])
                    p = first_pseudoprime(M, max)
                    if p == None :
                        print "[ ", c0, c1, c2, c3, " ]: OUT OF RANGE"
                    elif p >= print_min :
                        print "[ ", c0, c1, c2, c3, " ]: ",p, factor(p)

Dental implant: DONE

matt.mcirvin@gmail.com — Thu, 09 May 2013 23:22:51 GMT

And now I have a tooth. A fake tooth. I've been eating with it and everything.

The last stages of this process were pretty simple and relatively unexciting to recount. A couple of weeks ago, I went in and my dentist checked the fit of the metal abutment that would run up the middle of the tooth; he stuck it into the hex-shaped socket and put in the screw, and they took an X-ray to make sure it was properly seated in there. Then they took it out again, and compared my real teeth with a color chart so they could get the crown to match.

Today, the ceramic crown was finished, and there was a little bit of testing it in place on the abutment and grinding it down to get the bite just right. Then the dentist used a tiny little torque wrench to get the screw holding down the abutment to some precisely measured degree of tightness (he said the torque wrench was a recent development in dental-implant practice), and glued the crown on top.

It's not precisely functionally identical to a real tooth. Real teeth sit in sockets and have a bit of freedom of motion; this thing's rigidly bolted to my jawbone, which is one of the things that sometimes makes implants fail in various ways. At least if it's the crown that fails instead of the bone, they can replace it pretty easily.

For now, though, it seems to be working.

Mathematical fumblings: the Klein j-invariant and near-integers

matt.mcirvin@gmail.com — Mon, 06 May 2013 13:37:09 GMT

On Google+: Why e^(pi * sqrt(163)) is almost, but not quite, an integer, and how it relates to a pretty function called the Klein j-invariant. With some pictures, and bonus references to Ozma of Oz and an April Fool prank by Martin Gardner.

I am a piker compared to oonh, and there's some handwaving in the math because I don't understand it all, but I had fun playing with this.

Free math

matt.mcirvin@gmail.com — Sat, 04 May 2013 14:20:02 GMT

Many years ago, when I was a grad student, I bought a student-discount copy of Mathematica, the symbolic computer algebra system. It was crucial to finishing my doctoral thesis: I pretty quickly got beyond the point where I could safely do the algebra involved in my research by hand without making a fatal mistake somewhere. I knew some people who had superhuman ability at manipulating page-long expressions without screwing up a minus sign, but I definitely couldn't.

Unfortunately, that old copy of Mathematica (for the Mac OS of the time) was soon orphaned by Apple's many major OS and hardware changes, and stopped working long ago. And Mathematica is a really expensive piece of software, especially if you're not buying the student edition. Even the stripped-down "Home Edition" costs hundreds of dollars! There are alternatives, but until recently, the most viable ones (Maple, Macsyma) were all costly commercial products too.

Wolfram improved matters greatly by providing a free public front end in Wolfram Alpha. It's very nice for the purposes that many people will need, especially since it has a pretty good natural-language interface that means you don't really have to learn any special syntax to enter a problem. But it's limited to simple question/answer interactions rather than extended multi-stage calculations, and they often pull advanced features back behind the paywall (they clearly want you to buy a subscription).

In the past few years there have been several efforts at coming up with free alternatives: there's a free fork off an old version of Macsyma (the granddaddy of them all) called Maxima, and a very cool Python library called SymPy that you can use either interactively or scriptedly from within plain old Python, and a number of more specialized projects aimed at specific branches of mathematics.

I've been playing around with Sage lately. It's an interesting system. The idea behind it is to pull together as many of these systems as possible, including Maxima and SymPy and a bunch of other things, and give them a common front end, including a "notebook" interface similar to Mathematica's, in which you can record all the stages of an interactive calculation and modify and reevaluate them as needed. Since Sage tries not to reinvent the wheel, its notebook interface is not a specialized app, but a Web server that renders the notebooks in HTML; you interact with them in your Web browser. This also means they can be served remotely, and the Sage site even has a free server you can use to try it out. (There's also a crude browser that ships with Sage in case you don't want to use your regular browser for some reason; I haven't used it much, so don't know what its limits are.)

Like Mathematica, Sage can also translate its output into TeX format, and you can make it display the results in nicely formatted printing on the notebook page, I think using MathJax.

Sage also uses Python as its scripting language, but this is a modified version of Python that gets some infelicitous things out of the way: for instance, you can use ^ for exponentiation instead of bitwise xor, and the numbers default to Sage-native forms that lend themselves to arbitrary-precision math.

The main disadvantage of it is that it's a gigantic install, with a huge number of moving parts, and it may well be rough sledding installing it on some systems. SymPy might be a better choice if you want a more stripped-down option.

The other thing that bothers me a bit about most of these systems is that their graphics options, because they're designed to be cross-platform and made from highly portable bits, tend to be behind the times--though I can see the potential to remedy this. Sage's 3D plotting has to be done through something that can render in a Web browser. It's got several different "viewers" that can do this, either by rendering an image or with an interactive applet. But the default, and the only one that seems to combine a fairly full set of features and an interactive mode, is Jmol, which is mainly intended as a viewer for chemical structures (but Sage uses it for function plotting).

There's nothing really wrong with Jmol, except that it depends on the Java browser plug-in. And Java in the browser is completely jacked up right now, especially on Macs, partly because of Oracle's behavior, partly because of the safeguards needed to avoid its security holes, and partly because Apple clearly wants Java to die, die, die. (I guess on Windows you have to endure the crapware that the installer dumps on you now, which the Mac version at least doesn't do.)

Java doesn't work at all in Chrome on the Mac, period. This is because Chrome has avoided going to 64-bit for fear of breaking 32-bit plugins, but Java is an oddball case that has dropped 32-bit support and now only runs in 64-bit browsers.

Java kind of sort of works in Safari, but its security controls are entirely broken half the time and wouldn't allow using Jmol at all until some kind of incomprehensible magic happened. If you try to use Jmol's option to open a separate resizable window for the interactive plot, it hangs Safari.

Curiously, the one browser in which the Jmol viewer seems to work completely OK is Firefox, but before you use it, you have to click through a stern warning from Sage telling you that it probably won't work right and you should use Chrome or Safari instead.

Aside from that, for sheer beauty and interactivity, Jmol's plotting pales by comparison with what you can get in the (now commercial) Graphing Calculator, or Grapher (the inferior replacement for Graphing Calculator that Apple now ships with Macs), or even the limited but very pretty 2D/3D plotting that Google's built-in calculator now supplies for free over the Web (try it: just type "x*y" or something into your browser's address or search bar; you'll need a very recent browser). It's frustrating that these souped-up power tools won't get you as nice a function plot as you can get just by typing a function into Google, though of course they'll graph a lot of things that Google can't.

The potential's there to make this better. One of the other options in Sage is a viewer that uses canvas3D/WebGL to render an interactive plot inline in the notebook with HTML5. But it's very bare-bones, and seems to only do unlabeled wireframe meshes, unless I'm missing something. It seems to me that it ought to be possible to make it at least as nice as Google's function plotter. I doubt I have the time to do this on my own, but I wouldn't be surprised if somebody is working on it.

matt.mcirvin@gmail.com — Fri, 19 Apr 2013 11:53:56 GMT

We managed to get back through Boston yesterday without significant incident, though the train was a little late and the traffic on I-93 was bad. I was going to bed around the time last night's craziness in Cambridge and Watertown was starting to break, so I missed most of it. I'm still on vacation, and I actually live quite far from Boston up near the NH border, but obviously we won't be going in toward the city today unless the situation resolves.

matt.mcirvin@gmail.com — Tue, 16 Apr 2013 01:45:58 GMT

We are in New York City for the week, so were nowhere near Boston during today's events. It remains to be seen how security and such will affect the rest of our trip and our return on Thursday, but I don't see a huge amount of panic here.

Posted via m.livejournal.com.

Goop and coping

matt.mcirvin@gmail.com — Sat, 13 Apr 2013 22:10:50 GMT

So on Thursday I had the second dental impression referred to in the previous entry. This was a little more involved than I expected.

The end result of the implant process has been that there's a convenient metal adapter in my jaw, with a threaded screw hole in the middle, but also, around that, a shallower hexagonal socket that the final abutment will key into so it doesn't turn. In order to make the abutment and crown align properly with my other teeth, the lab that will make them needs an extremely precise reading of the orientation of that hexagon.

The way they got that is kind of interesting. The rough impression I had a couple of weeks ago was just to make a custom tray for this second impression, which used a much more rigid-setting, extremely vile-tasting purple material. But before taking it, the dentist unscrewed the simple healing cap that's been hand-screwed into my implant and replaced it with something called an "impression coping". This was a piece of metal with a hexagonal base that keyed into the implant, and was held in with a separate screw sort of like the final abutment will be. But instead of the support for a crown, the top of it was a vertical post with two weird bow-tie-shaped projections sticking out to the sides. The central screw holding it in was a long thing that stuck out far enough that I couldn't bite down fully.

They put in the impression coping, took an X-ray to make sure it was securely seated in there, and poked a hole in the custom goop tray so that the screw could stick right out through it. Then my dentist jammed the tray down over that with the impression goop inside, told me to bite down on his fingers holding the tray in place, and stood there for five minutes humming along with the Sixties pop on the satellite radio and pretending not to be uncomfortable with me continuously biting him.

When the impression was almost fully set, he unscrewed the long screw, then lifted the tray of hardened material off my jaw so that the coping came right along with it, with the projections at the top stuck in the impression material. The end result was an inverse image of my lower jaw with the impression coping sticking out of it, its hexagonal base registering the angular position of the hexagonal socket in my jaw.

Anyway, I thought it was interesting. The healing cap's back in now; I think the next step is to CNC-mill the metal abutment, and my next visit is to check the fit of the abutment before the crown on top gets made.

Zeno's implant

matt.mcirvin@gmail.com — Wed, 27 Mar 2013 13:18:05 GMT

The closer I get to the end of my dental-implant process, the more steps appear. Now that the surgical part is over and the implant's in my jaw, the process gets handed back to my regular dentist to produce the final abutment and crown.

It turns out this is going to be four appointments, of which the first was yesterday. That was a quick visit to take a dental impression, which will be used to make a form-fitting goop tray to take another impression at higher resolution with a more rigid material.

They need a precise model so that they can make sure the abutment and crown are made right. Right now, I've got a temporary healing abutment that looks like a flathead screw, which is just hand-screwed into the threaded hole in my implant. But the implant also has some kind of shape around that to keep the final abutment from rotating. It's a titanium piece with a computer-rendered shape that they'll make on a CNC milling machine, which will key into the top of the implant in a fashion reminiscent of flatpack furniture, and be held in place with a screw through the middle. Then the tooth-shaped ceramic crown goes around that. I think the abutment gets tested for correct seating and orientation at the third appointment before they make the crown part.

It won't all be done until early May, and then I go back to the oral surgeon for one last follow-up check. Here's hoping it's all worth it.

Tooth implant update #4: abutment installed

matt.mcirvin@gmail.com — Thu, 07 Mar 2013 00:49:03 GMT

Today's procedure was a very short one, just ten or fifteen minutes of actual work. When I had the metal screw inserted into my jaw, the oral surgeon stitched up the gum tissue over the top of it again, so that I just had a toothless space in my gums there. Now that that's all healed up and the bone has presumably fused with the screw to some degree, the next step was to open up a small hole in the covering tissue, expose the smaller threaded socket inside that screw, and put in a "healing abutment", a smaller-diameter screw with a wide, flat conical head that sticks out on the surface.

Since the local anesthetic wore off, I've been a little sore, but it's nowhere near as bad as the previous installments that involved stitches; an Advil more or less takes care of it.

The purpose of this is, I think, to let the gum tissue re-heal into more or less the configuration it will need when the crown goes on in place of the abutment, which is something my dentist will do in a few weeks.

I'm not sure whether that will be a single- or multiple-stage process of its own. But then I've got a final appointment with the oral surgeon in May to check if everything came out OK. At any rate, my new tooth isn't that far off any more. But now I look like I've got a flathead machine screw where a tooth is supposed to be.

They Might Be Giants, Nanobots

matt.mcirvin@gmail.com — Tue, 05 Mar 2013 23:48:40 GMT

I got They Might Be Giants' new album this morning and have listened through a couple of times. At times it seems almost like a retrospective of TMBG's many familiar styles and gimmicks: the first song, "You're On Fire," is almost a sequel to "You Probably Get That A Lot" from Join Us; later there's a political song from Flans ("Black Ops"), a wistful biographical song about a historical figure ("Tesla"), a cluster of "Fingertips"-like mini-tracks about halfway through (with a few others scattered throughout), even a song about a young nerd with a goofily portentous spoken-word bridge straight out of the Pink Album ("Circular Karate Chop"). It's fair to say they're not going out of their way to shock the fans with this stuff.

That said, John Linnell breaks some new ground here that I personally greatly enjoy: having become a master of children's music, he's starting to sing in his wry fashion about the adult experience of being a parent. "Nanobots" and "Replicant" both frame the experience in pseudo-science-fictional terms; the title track in particular is an utterly joyous piece of work that I want to listen to over and over, whereas the other suggests some of the darker possibilities.

On the flip side of parenthood, "Call You Mom" is a hilarious song about a person with serious, serious relationship issues, which is musically sort of like "Yakety Yak" collided with the Peter Gunn theme.

Flansburgh also gets a little off the beaten path with the spooky, jazzy and baffling "The Darlings of Lumberland" and the melancholy "Sometimes A Lonely Way", which I haven't quite entirely processed yet.

Nestor, 1999-2013

matt.mcirvin@gmail.com — Sun, 03 Mar 2013 01:40:42 GMT

So, the other shoe dropped today: just ten days after losing Niobe, we made the hard decision to let Nestor go. This remembrance is not as easy to write, since Nestor was not an easy cat.

A thing that made Niobe's final days so enervating for us was that Nestor was also clearly declining, in a completely unrelated manner: the problem he'd been battling all his life was finally coming to a head, despite everything we'd tried to control it. In fact, I'd greatly expected to have to put down Nestor first.

Nestor had megacolon, a chronic condition in which extreme constipation made his colon distended and made the constipation even worse, in a vicious cycle. He must have had it even before we adopted him (at the age of about 2). There was the option of getting him a colostomy, but given that he was 13, hadn't been eating, had lost an alarming amount of weight, and had another, unrelated medical problem (a mysterious nodule in his lung that gave him a persistent cough, and may or may not have been a tumor), we didn't see the likelihood of a good recovery and great quality of life as being high. Probably we should have had it done years ago, but we'd attempted to control it with medication instead, a somewhat grueling regimen that involved chasing him down to squirt lactulose and, eventually, other drugs into his mouth twice a day. It's kind of remarkable that he lived as long as he did.

The constipation, in turn, caused a disgusting behavioral problem: he constantly peed outside the litter box, maybe because of a negative association with the discomfort of trying to poop.

Many, perhaps most people would, I am sure, not have put up with this for any length of time. Nestor had been a pet cat before we got him from the Animal Rescue League shelter in Boston. I have no way of knowing for sure, but I'm guessing that his previous owner gave him up because of the peeing issue, and didn't mention it (or the constipation) to the shelter people.

Here's the thing: it was impossible to keep hating him indefinitely.

Adopting Nestor was my fault. After Sam's cat Savant died, she had the idea of adopting a kitten. Shelters don't usually have a lot of kittens, since they go pretty fast, but this big adult brown tabby instantly charmed me with his affectionate nature, and he was always a sweet, docile animal (maybe partly because he was confident in his status as biggest cat in the house). His peeing drove Sam and me to distraction, but he obviously dearly loved Sam in particular, jumping into her lap whenever he got the chance, whether she wanted it or not.

And when we eventually did get our kitten, another brown tabby named Radka, Nestor was the one who instantly adopted her as his own: in her early days, the two of them were inseparable, often curled up together on a cat perch by the window.

I don't think we're going to replace these cats any time soon, if ever: raising a human child and one cat is enough, and I think we want some time with a more stable situation around here. I don't know how Radka (who seems perfectly hale and hearty at 10) is going to handle being the sole cat: as an adult, she seems to have kind of an independent personality, but she always liked hanging out in the same room as Nestor or Niobe. I'll be sure to pet her and play with her a lot.

Cat food

matt.mcirvin@gmail.com — Sat, 23 Feb 2013 18:15:50 GMT

Thanks, everyone, for the kind words about Niobe.

As james_nicoll has observed, sometimes in a multi-cat household you learn surprising things about a cat only after she's gone, and in this case, it's that she was probably eating more than half of the food consumed by all three of our cats. Her thyroid troubles may have been a contributor.

(She was also a shameless thief. Just a few months ago, she managed to pull a raw pork chop entirely out of its plastic-wrapped package, which was in turn inside a grocery bag that I'd carelessly left on the kitchen floor while I was hauling more in from the car. I felt almost sorry to take it away from her; she looked so proud.)

Niobe, 1997? - 2013

matt.mcirvin@gmail.com — Thu, 21 Feb 2013 03:24:44 GMT

Our cat Niobe died today, aged about 15. She'd been increasingly frail for a while, with a thyroid on overdrive (which we tried to control with drugs) and low energy, and today it became clear that it was time for her to go.

Niobe was probably best-known to the Internet via a picture I took at our Malden townhouse when she was fairly young, which somebody made into a lolcat of minor notoriety, SKEPTICAL CAT IS FRAUGHT WITH SKEPTICISM. I love that picture, since she always was indeed fraught with skepticism. As you can see from the picture above, she could also take on a sinister aspect as needed. She also played a minor role in the history of laser printers: when I was helping to develop the Xionics (now CSR) emulation of Level 3 PostScript, I used a drawing of Niobe I'd made in Illustrator 8 as a primary performance test article for the gradient mesh feature.

She was a strange-looking little cat, a gray-cream tortie built low to the ground, with stumpy legs and the eyebrows and chin of a Scottie dog, that gave her what looked like a perpetually angry expression. We adopted her as a young adult, probably about a year old and recently pregnant; the shelter waived the fee, because it was in the wake of the sudden death of our previous adoption, an orange cat named Elmer who had unexpectedly been in late-stage kidney failure when we got him. They hadn't had this little gray cat for long, and weren't sure how she would do in our household.

She turned out to be a terrific, affectionate pet, with an amazingly loud purr, though initially prone to falling off of things, and demonically hard to claw-trim or mess with in any other way that violated her personal space (when she was young, the vet had to get two people to hold her down for any sort of treatment).

At the time we had one other cat, Sam's beloved pet Savant, but Niobe was the first one who really became my cat. For the past several years we've had three cats, which was probably at least one too many for us, but I'll miss her terribly.

Interesting maps to compare

matt.mcirvin@gmail.com — Fri, 25 Jan 2013 12:46:47 GMT

Well, the 2012 election put to rest once and for all the idea that Americans will never vote to legalize same-sex marriage. In that light, it's interesting to compare these two maps from Wikipedia:

Legal status of same-sex marriage in the US, by state
Public opinion of same-sex marriage in the US, by state

Public opinion is way out ahead of the law. In California, Oregon, Colorado, Michigan, Ohio, and Florida, there's majority public support for same-sex marriage and a constitutional ban. That's a substantial chunk of the country. In the latter three, the bans are so strong that they also prohibit civil unions.

It was always pretty clear that the great wave of state-constitutional bans in 2004-2008 was a rear-guard action to lock in opposition before it evaporated, but I didn't expect it to happen so quickly. The wave of SSM legalization will probably slow down a bit in the near future just because of the greater procedural difficulty of amending state constitutions.

Tooth implant update #3: the screw goes in

matt.mcirvin@gmail.com — Thu, 17 Jan 2013 03:55:07 GMT

A week ago, I went back in for the latest phase in my dental implant surgery. This had been originally scheduled for tomorrow, Jan. 17, but they gave me the option to pull it back about a week; I guess a hole opened up in their schedule, and maybe I'd seemed to be recovering well from my October bone graft when they took the sutures out in November.

I went in last Wednesday, and they took an X-ray, and, miracle of miracles, that old tooth socket in my jaw was all filled up with new solid bone like they'd said it would be, ready to have a new hole drilled in it. So they shot me up with local anesthetic again, and cut open my freshly healed gums, and the oral surgeon went in with a long succession of different drill bits to make a hole for the screw that will eventually support the implant. I'm not sure, but I think the hole they drilled in the bone might have actually had threads inside it somehow, because when he was done, the surgeon just hand-screwed the thing in like he was putting it into a threaded hole (unless I misinterpreted what was going on).

Now, this isn't even the last step I have with him. The titanium screw has a threaded hole inside it that the false tooth will eventually screw into. (In the X-ray that they took afterwards, it looks like something you'd buy at Home Depot.) But for now, that smaller socket just has a flat "cover screw" inserted into it, and he sewed up my gum tissue over the top of that. In March, when my jawbone has adequately fused with the screw, I go back there to get my gums opened up once again, and the cover screw replaced by a taller metal abutment that sticks out all the way to the surface. Then, after a final healing period, maybe sometime in April, my dentist will replace that with the actual fake tooth.

The surgeon said I'd have a shorter recovery this time, and it was true. The stitches this time were smaller in size and number, and all the dissolving kind. I was on amoxicillin again for another week, and was popping great big Motrin (ibuprofen) pills alternating with regular Tylenol (aka paracetamol/acetaminophen) for a day and a half and regular Advils after that. Instead of the Vicodin I never took last time, he prescribed Tylenol with codeine (you can't get this OTC in the US) for nighttime. The first night, when I was still hurting pretty badly, I decided to try taking one. I thought it'd make me drowsy, but it didn't, it just made me feel kind of ill and fuzzy-headed. It did make the pain go away, but I didn't sleep much anyway.

The exterior healing was actually quite rapid. Last weekend, though, I got the most painful canker sore of my life right next to the surgery site, maybe from the stitches scraping against my lip. Fortunately, it was gone in a couple of days. Just today, the amoxicillin ran out, and the last of the stitches fell out revealing what seemed to be almost completely healed gum tissue, so I'm feeling fine now. I'm thinking and hoping the surgery for the abutment is simpler still.

In the world of fake pinball

matt.mcirvin@gmail.com — Sun, 23 Dec 2012 04:36:44 GMT

So all sorts of fascinating, often infuriating drama has been going on with The Pinball Arcade, Farsight Software's classic-pinball simulator. As of last night, both of the famous tables with media licenses funded from successful Kickstarters, Twilight Zone and Star Trek: The Next Generation, are out on iOS and Android and, I think, Macintosh, and they've got a bunch of others as well.

But the PS3 releases are taking some more time. As for the XBox 360 releases through XBox Live Arcade, they got more and more delayed as Microsoft rejected them for various reasons, and eventually stopped entirely, for a dismaying reason: the software publisher Farsight used as a middleman, Crave Software (who you may remember as the publisher of Pinball Hall of Fame on disc), went bankrupt and ceased operations. But they claim they've found a new publisher, as yet unidentified, and will get TZ and Star Trek: TNG out before the other delayed packs so that the Kickstarter folk can claim their free copies. Since the XBox is by far the best platform I've got available for playing TPA, this is relevant to my interests. We shall see.

The latest Android release is odd. They completely redid the top-level menus, which were somewhat lacking, but are now about a thousand times worse; it looks like an amateur coding project. I hope they just revert the code ASAP until they can make the new menus look a little better.

The Twilight Zone sim, which came out a few weeks ago, is a lot of fun, but it's controversial, since it's far easier than the real table even by TPA standards. This upsets hardcore pinball fans, since in the real world, the Twilight Zone pinball is known for being not just full of droll bizarrerie, but also deliciously cruel. In particular, the simulated model has a barrier to prevent outlane drains between two of the Town Square jet bumpers, an optional feature that is rarely seen in the real world (after the earliest production models, they started shipping with the barrier inside the coin box in case the operator wanted to install it).

On the bright side, Star Trek: TNG is great, and seems to have captured the flavor of the real table excellently. Gameplay-wise, it's a great fast-flowing Steve Ritchie design, with lots of things to do. But this is also one that demands to be played with the sound on, just for the sake of the clever original-cast voice clips, sound cues and background music, all of which expertly evoked the feel of a particularly exciting, sometimes goofy episode of the TV show. It's nice to just reacquaint myself with all those sound samples that are burned into my head from wasting time in the 1990s. Had you propelled the ball along the proper traTHANK YOU MR. DATA

There have been a bunch of other tables released on non-XBox platforms since the last time I talked about this. My favorites of those include Elvira and the Party Monsters (they also did the other Elvira table, the double-entendreriffic Scared Stiff, but for some reason I don't enjoy it as much), and Big Shot, a simple electromechanical game with a pool theme from 1973. There have also been major upgrades to physics and gameplay on Medieval Madness and Black Hole, which I hope come to the XBox soon.

Tooth implant update: stitches out

matt.mcirvin@gmail.com — Sat, 17 Nov 2012 03:36:30 GMT

So a few weeks ago, I got a bone graft in eventual preparation for implanting a false tooth directly into my jaw.

I wasn't sure what to expect, so just about everything that happened seemed mildly alarming, but it seems to be going well so far. The major soreness and swelling in my gums went way down after a few days, but I still had to deal with the sutures.

As I said before, there's some kind of pig-collagen membrane inside my gums covering up the old tooth socket, and its filling of cadaver bone and newly generating Matt bone. Over that, the oral surgeon had stitched my gum tissue together with a messy-looking collection of sutures. There were two kinds: some thinner, straw-colored ones that I think were made to dissolve, and thicker, purple, more permanent ones, some of which were threaded directly through my gums from the inner to the outer side. Some of the latter had a sort of knot on the end to hold them in place that constantly scraped against my lip. Also, I think the tension on them was causing some persistent discomfort, though I wasn't in agony.

A week or so after the surgery, when I'd gotten off the antibiotics and the giant Motrins, I was feeling much better, but a major annoyance was that one loop of the purple sutures seemed kind of loose, and was constantly sliding up onto the top of one of the adjoining teeth when I was eating. I think it was attached at one end to the dissolving stitches, which were in the process of falling out. When they started falling out in earnest, that loop came entirely loose at that end and was flapping around freely, making me constantly fear that all my sutures were broken and they'd have to start all over again.

Also, when those stitches fell out, I had another bout of pain and swelling for a couple of days. I found myself taking regular doses of Advil to try to keep it under control. When it all passed, I was less alarmed.

On Wednesday I went back to get the purple sutures taken out. According to the surgeon's assistant who did the job, everything is in fact fine and my gum tissue seems to be healing up beautifully, which is reassuring. Pulling those out seems to have caused yet another bout of minor swelling, but the pain is gone, since the sutures aren't pulling or scraping any more.

Anyway, I've got a weird little scar now where I used to have a tooth. In mid-January, when the underlying bone is supposed to have regenerated enough (or generated, I guess, since there was never bone there in the first place), the landscape gets modified again when they put the post in, and then I think I'm going to have some sort of temporary metal cap there until my regular dentist feels ready to put the crown on. But everyone claims I've been through the worst part already.

Well, this is slightly alarming

matt.mcirvin@gmail.com — Mon, 29 Oct 2012 22:43:50 GMT

Something very strange is currently happening with NOAA's storm-surge model:

http://www.nhc.noaa.gov/refresh/graphics_at3+shtml/211847.shtml?gm_esurge#contents

it seems to be calculating a 10% chance of a storm surge of more than 30 feet in Massachusetts Bay and Cape Cod Bay. The surge is so big, in fact, that it exceeds the available display range, probably by a large amount, so for all I know it could be predicting a 50- or 100-foot wall of water crushing the South Shore. (Update: In fact, you can click in there to get individual values, and some of them seem to be as high as 92 feet.)

I can only guess what's going on here, but I assume this is some kind of freak scenario in which Hurricane Sandy moves back out to sea in a few days, strengthens and parks somewhere in the vicinity, shoveling vast amounts of water in around Cape Cod. It doesn't really make any sense, though, since that's something like four times as high as what you'd expect from a Category 4 hurricane.

I assume the credibility of these results is rather low. If you increase the threshold probability to 20%, it goes away. It's probably some kind of numerical hiccup, but I'm certainly going to be keeping an eye on it.

The next day: Yes, in fact, it went away as mysteriously as it appeared.

Dental science marches on!

matt.mcirvin@gmail.com — Sat, 27 Oct 2012 01:32:46 GMT

On Wednesday, I finally lost the last remnant of my rear right lower bicuspid. In a few months, if all goes well, a fake tooth will replace it.

I've always had lousy teeth. Part of this was genetic/developmental luck of the draw, and part of it was indifferent hygiene when I was a kid (I brushed regularly, but never got the hang of flossing until much later). My dentist was a soft-spoken fellow who, I now realize, was also very technically old-school. I ended up with a lot of his fillings, which a later dentist described as "pre-Vietnam".

(I assume this is a term of art. I've also noticed that dentists all seem compelled to rag on the work of the previous dentist you had.)

When I was in grad school, the great big filling in this one tooth started to come loose, and I put off dealing with it for way too long, by which point I was hurting quite a bit and needed a root canal. My second dentist referred me to an oral surgeon, who ground off the whole top of the tooth, drilled out the pain-flavored filling in the center, and replaced the whole thing from below the gumline with this cunningly constructed porcelain-and-gold replica that even had slight discolorations in a couple of spots to make it look real. The whole operation was amazingly pain-free and sadly required no happy drugs, local anesthesia having advanced considerably since I was a kid. With a cyborg bicuspid that was more machine than tooth, I came away happy.

Later, I got another root canal and crown (plain gold) on one of my molars, from my current dentist, who predictably informs me that he would totally not have done the first one the way the other guy did. But the first one seemed OK until recently, when the crown abruptly came loose. The old root had been rotting away down there, and couldn't be saved. It was time to say goodbye to Mr. Tooth.

There are two things they can do in this situation. The classic solution is to put in a bridge, which involves decapitating the adjoining teeth and putting in artificial crowns that are attached to the new fake tooth, to support it. Since one of the adjoining teeth was pristine and the other one was mostly OK apart from a small filling, nobody really wanted to do that. It does have the advantage (from my perspective, not the dental industry's) of being the cheaper and faster option.

The more involved option they sold me on was to get an implant, which is a fake tooth supported by a post inserted into my jaw. When my dentist first explained this to me, I wasn't entirely clear on how that post could get support in my jawbone, given that the original tooth was in a socket shaped like a big, flattened cone.

The answer turns out to be awesome: They're going to fill the entire hole with bone first. I had the first part of the process done on Wednesday morning. A periodontist/oral surgeon pulled out the remaining nasty rotten tooth bits, cleaned out the socket with a drill, then filled it with a cream made from the ground-up, irradiated fragments of the bones of actual other human beings (deceased). Over this, he stitched a membrane that is some sort of pork by-product, to cover the hole, and lashed it down with sutures going all over the place, looped around the adjacent teeth.

What amazed me was how quick and painless even this surgical process was. The actual operation, from extraction to stitching-up, took all of thirty minutes. It seems as if local anesthesia has gotten even better, much more precisely targeted. They had to inject me in six or seven spots, but the result was zero pain (after the initial needling, at least) without even a great deal of numbness away from that side of my jaw. I remember when the whole side of your face would be dead from the Novocain required for a simple drill-and-fill.

The goal of all this blasphemy against Nature is to persuade my body to generate new solid bone that will fuse with the grafted bone. A few weeks from now, if all goes well, I go and get the stitches out (or at least the biggest ones that are non-dissolving). In January, when the regenerated bone is hard enough, they drill a new, smaller hole into my jaw and stick a sort of titanium bolt into it. Then, sometime a couple of months later when my bone has fused again around the bolt (and, I think, into some pores in the titanium), it's strong enough to be load-bearing and my dentist puts the new fake tooth on top.

The aftermath involves some soreness and swelling. I've been on a combination of giant Motrin pills, Tylenol, ice packs, and amoxicillin for the past couple of days. They gave me a prescription for Vicodin, but I haven't really needed it. The pain's way down now, but I'm going to have to take the amoxicillin for a while longer, which is a bummer because I think it's upsetting my gut, but at least I'm not allergic to the stuff like Sam is.

The oral surgeon claims the result will, unlike those root-canal jobs, be more durable than my regular teeth. I guess we'll see. I do greatly appreciate the people (and the pig) whose former parts are now incorporated into my lower jaw.

Voter ID in Massachusetts... sort of

matt.mcirvin@gmail.com — Fri, 28 Sep 2012 02:32:09 GMT

John Scalzi wants to make sure you're registered to vote. In reference to Massachusetts specifically, it's worth repeating something I said in the comments there.

Massachusetts doesn't have a voter ID law like the ones that are a focus of contention in several swing states. However, the state does put you on an "inactive voter" list if you forget to turn in your annual town census, even if you voted in the last election. If you're on this list, then when you show up to vote, you'll be asked for some form of ID with your current address on it. If you provide ID, you'll have to sign a document saying you really live where you say you live, and then you can vote normally.

This happened to me when I voted in the state primary on September 6th. (Incidentally, this is one of the benefits of voting in primaries, even if you don't have a strong preference for one candidate over another: you can check the system to see if it's working before voting in the general election.) One of the poll workers muttered darkly that some people who'd sent in the census seemed to have gotten on the inactive roll anyway, though I don't know if she had any evidence for that.

At the time, I wasn't told precisely what kind of ID qualified, so I just looked it up. By state law, the ID does not have to be a photo ID; it just has to be something associating you with your address, which could be a lease, a utility bill or a pay stub. So this is nowhere near as onerous as the various state laws requiring people to get a special type of ID that many people don't have.

The plot thickens, however: various groups have been sending poll-watchers to elections to, as they say, make absolutely sure the law gets enforced. There have been reports of these people specifically demanding to see photo IDs, even though they are not required by the law, and doing other obnoxious things. (None of this was going on in my town; Worcester seems to have been a focus of this activity.)

There are also complaints about the inactive-voter list on the grounds that huge numbers of people seem to end up on it every year.

In any event, whether you sent in your town census form or not, I'd recommend going to your polling place with something on you that establishes residency, just in case.

Pinball Wizard

matt.mcirvin@gmail.com — Fri, 21 Sep 2012 20:43:36 GMT

My visit to Pinball Wizard Arcade in Pelham, with pictures. This is a great place and I encourage any pinball freak with occasion to visit New England to go there. One of the happiest things was that I finally got to play a real Firepower, one of my favorite games from Pinball Hall of Fame on the Wii. The real one's tougher.

More fun and annoyance with Pinball Arcade

matt.mcirvin@gmail.com — Thu, 06 Sep 2012 03:57:40 GMT

Having gotten the cash to license Twilight Zone for their pinball-emulation game The Pinball Arcade, Farsight has almost fully funded their second Kickstarter to fund the license for Star Trek: The Next Generation.

It's gone a little slower than the first one, I think in part because it's held in slightly less reverence by pinball fans (though it's my personal favorite), and in part because people have gotten frustrated with Farsight over bugs and delays in their existing DLC, which makes them less likely to believe that the company's actually going to deliver on Twilight Zone, let alone Star Trek: TNG. Particularly the XBox 360 and PS3 users, for whom content has been very slow in coming.

Still, what they have released is a lot of fun, because of the inherent appeal of the games. After very long delays, the first three DLC packs finally got released for the consoles, which, in addition to the four tables distributed with the game (Theatre of Magic, Tales of the Arabian Nights, Black Hole, Ripley's Believe It Or Not!), gives them Medieval Madness, The Machine: Bride of Pinbot, Cirqus Voltaire, Funhouse, Monster Bash, and Gorgar.

The pattern seems to be to bundle a retread of a table from Pinball Hall of Fame with a new one, so that PHoF fans have to re-buy the old tables. To be fair, the new implementations do have the newer physics engine, and Funhouse is now running with an emulation of the genuine ROM, though Gorgar isn't. And Medieval Madness is new to me, since it wasn't in the Wii version of PHoF.

It's hard to pick a favorite, but of these six DLC tables, I think Funhouse and Monster Bash are the ones I've enjoyed playing the most. Funhouse was probably Pat Lawlor's best-known pre-Addams Family table, with a creepy animatronic dummy head named Rudy whose eyes follow the ball around the board while he distracts you with comments. As with most of Lawlor's designs, some of the important shots involve whacking the ball horizontally across the playfield with a small, vertically-mounted third flipper. For some reason the TPA version grabs me more than the PHoF version did, maybe just because my skills have developed a little.

Monster Bash was apparently the result of a failed attempt to license Bobby "Boris" Pickett's song "Monster Mash", so, as you'd expect, it has a goofy Universal-Monsters-in-a-rock-band theme, in which you're trying to collect all the monsters like Pokemon to form the band and go to concert. It makes much use of mode-stacking, in which you can get special advantages by activating two or more special modes at once. The emulation on XBox 360 has a bug that breaks leaderboards, since you can get arbitrary numbers of jackpots by going into Frankenstein Multiball and nudging the table to the left. But if you don't care about that, it's great.

On Android (where you can buy all the tables either in packs or a la carte), there are four more tables available: Creature from the Black Lagoon, Black Knight, Taxi, and the Stern Harley Davidson.

Creature has an unusual fourth-wall-breaking theme. The conceit is that you're at a Fifties drive-in watching the movie, and the game has a rough storyline in which you do various wacky drive-in-movie-related things to "start the film" (the two-ball multiball mode), in which you're in the movie trying to Rescue The Girl from the Creature. The interesting thing about this one from a Pinball Arcade perspective is that they seemed to have greatly improved the flipper physics from the previous tables, making it easy to do such things as post passes. Initially it did have the problem that it was a little too easy to trap a ball when it was just coming down the inlane (the latest update fixes this, and therefore makes the table considerably harder).

Black Knight, unfortunately, seems much buggier than the PHoF version, probably the buggiest table released so far; it really needs work. The latest revision does at least move the virtual Magna-Save buttons to reachable positions. Taxi is an old favorite of mine, and the TPA version seems fine so far. I haven't formulated much of an opinion about Harley-Davidson yet except that it seems very easy, in the sense that you can activate lots of flashy modes just by haphazardly knocking the ball around.

Register to vote

matt.mcirvin@gmail.com — Mon, 03 Sep 2012 00:36:49 GMT

Say, this is pretty nice; I saw it via Barack Obama's AMA on Reddit, but I am sure it is equally effective if you have no plans to vote for Barack Obama:

https://gottaregister.com/

In Massachusetts, the deadline for registering in the general election is October 17.

Update: Here's a version unaffiliated with the Obama campaign:

http://registertovote.org/index.html

Funspot and ACAM

matt.mcirvin@gmail.com — Sun, 29 Jul 2012 18:28:38 GMT

As a birthday present Sam gave me a gift card for tokens at Funspot. We went up yesterday, and had to fight terrible summer weekend traffic, but we had a great time once we got there.

Funspot near Weirs Beach in Laconia, NH bills itself as "the largest arcade in the world". I don't know if this is true or not, but if you like classic videogames from the 70s and 80s and want to play a real cabinet, this is definitely the place to go. The prize of the place is the American Classic Arcade Museum, a collection of old machines on the second floor.

They're grouped by manufacturer, with a whole row of nothing but Atari vector machines. Honestly, I was never much of a coin-op player back in the day; I played home console and Atari personal computer games, and the more difficult quarter-eating ones rarely held my attention. But there's nothing quite like a real coin-op Tempest (emulations can't replicate the physical experience of using that cabinet knob), and I did play a few rounds on that. Also one game on an original Star Castle, and the thing I didn't remember is that it has surprisingly loud, booming audio.

I was there for pinball. Funspot is not the best place in New England or even NH to go for pinball (that would be Pinball Wizard Arcade in Pelham, much closer to home, which I have still not visited). But Funspot's not bad, especially for fans of older solid-state pinball. They've got a wall of mostly 80s and 70s solid-state tables (and some earlier electromechanical machines) in the ACAM room, and a few later machines downstairs. All of the classic videogames and all of the pinballs are set one token a play, with tokens costing about a quarter (less in huge quantities).

The pinballs are not all in the greatest of shape, and a few were out of order, but they're clearly maintaining them; it's nothing like the tragic situation at Canobie Lake, where most of the pinball machines are unplayably broken. The one I found at Funspot that was operating but really not playable was their Xenon, because of an extremely sticky left flipper (the left flipper on their Gorgar was sticking a little too, but it wasn't as big a problem). The Rudy head on their Funhouse was broken, but the game seemed to be playable anyway.

I got the chance to play in person some tables that I mostly know from emulation, or had only played for real briefly and long long ago: Gorgar, Funhouse, Black Knight, Pin*Bot. All of these turn out to be tremendously fun in person, though Black Knight is much harder than the version I became familiar with in Pinball Hall of Fame on the Wii (the version that just came out as DLC for The Pinball Arcade on Android and iOS seems closer to the real thing, albeit buggy). Playing in emulators had familiarized me with the table rules on these, which made them a lot more fun to encounter in person, though obviously the physical tables are less predictable and therefore harder.

Of tables I hadn't encountered before, I think my favorite from a gameplay perspective is Grand Lizard, one of the early Eighties games with a two-part playfield. It's a Barry Oursler playfield, like Gorgar and Pin*Bot. High Speed, a classic Steve Ritchie design, was pretty good too, though I don't think I got very far into the ruleset.

Funspot also has some other good stuff, like tenpin and candlepin bowling lanes; an adorable set of tiny kiddie bumper cars, apparently battery-operated; and an equally tiny indoor miniature golf course dating from the 1950s.

The original idea was that I'd play pinball and old videogames while Sam and Jorie amused themselves with other things, but they ended up playing a surprising amount of pinball. I got Sam hooked on Pin*Bot. I think the key thing is that, while it's a tough game to master, it has a subsidiary goal (open the visor and lock the eyes for multiball) that a newbie can feasibly shoot for. Jorie liked playing co-op, operating the plunger and sometimes the right flipper when she was in the mood.

Demolition of the Lincoln Park Comet

matt.mcirvin@gmail.com — Sat, 14 Jul 2012 21:40:02 GMT

On Wednesday, this long-dormant roller coaster finally came down:

http://www.lpcomet.com/archives/1260.html

The park was long-defunct, the lift hill had collapsed in 2005, and it was obviously never coming back. Still, it's sad seeing one of these go away. That's a nice fan site with lots of photos from intrepid ghost-park explorers.