Leadville 100

The people who come here are used to succeeding. Very little of what they do in life involves a real risk of failure. They are attracted because this is a situation where failure is not only possible, it’s likely. That’s a rare situation for them.

-Gary Cantrell

For those not be familiar with the man who prefers to be called “Lazarus Lake”, as founder of the Barkley Marathons, he has sat patiently at the finish of what is generally regarded as the World’s Hardest Race since 1986 watching all but fifteen entrants fail. Winning time is pretty irrelevant as there have only been four years with a second place time. Most years, nobody makes it. Your chances of getting through Harvard’s Law School are considerably better.

I’m not successful enough that I crave failure situations but I will certainly concede that succeeding at a difficult task is far more interesting to me than succeeding at an easy one. So, while the expectation for the “Race Across the Sky” was to come home with a finisher’s buckle, I went in knowing that mountain ultras have been pretty much train wrecks for me.

I’m not sure why this is true. It’s certainly not altitude. I actually have a very good track record racing at altitude in shorter events. And, shorter events is where you really notice the reduction in oxygen. At ultramarathon pace, you’re so far below your maximum capacity, oxygen isn’t a limiting factor (until you start doing really crazy-high stuff, like climbing Denali). The other symptoms of altitude (headache, nausea, etc.) have been conspicuous by their absence in my previous outings.

A more plausible theory is that I tend to kill my legs on the long descents. That was definitely the cause for the complete meltdown at UROC where I had to walk all of the last 15 miles because my quads simply couldn’t handle another landing shock. To mitigate that, I spend a lot of time this summer working downhill running. In particular, I practice using trekking poles on descents to lessen the force of the footstrike. Still, the longest descents in Missouri only take a few minutes to run down; there’s simply no way to adequately train for 3000’ drops in the Midwest.

Leadville has two of those, as the out-and-back course hits Hope Pass from both sides. The descents off SugarLoaf and Elbert are also significant, but both intuition and reports from race veterans tell me that Hope is the one I need to worry about. I decide there’s really nothing I can do to fully prep for it, so I’ll just train for the other 80 miles as best I can and deal with the middle 20 as it comes.

In truth, my goals are slightly more ambitious. As I got into Leadville via one of the qualifying races rather than the lottery, it seems I should at least take a passing shot at the "big buckle" one gets for finishing in under 25 hours. I don't pin too much on this. My slowest 100 to date is 22:50, but I've never run a hundred in the big mountains. I figure I'll at least try to get over Sugarloaf at that pace. If it feels OK, I'll hang on. If not, there's plenty of time to recover before Hope.

The drive to Leadville is blissfully uneventful. I’m travelling with Bill Langton who will serve as my crew. We spend Wednesday night in Lawrence with my orienteering friend Mike Eglinski and Thursday in Fort Collins with Dale Somer, one of Bill’s former roommates. An early start gets us to Leadville by mid-morning for number pickup and the mandatory race meeting (it's mandatory in name only, but Race Founder Ken Chlouber is a sufficiently motivating speaker, it would be a shame to miss it).

After the meeting, Bill and I take the advice offered and scope out the crew aid stations. Two of the locations, the “Alternate Crew Zone” at 28 miles and the turnaround at Winfield require traveling on roads not compatible with a tiny sport coupe on 3-season tires. Neither of those are big losses; I can send a drop bag to Winfield and the Alternate Zone is just an alternative to the Outward Bound Aid Station which comes only 3 miles sooner. We adjust plans accordingly over dinner and then get to our overnight lodging (Leadville doubles in size on race weekend – we have rented a basement on AirBnB). We sort through gear and load the car for a very early start.

At 3:15AM, we are three blocks from the start ordering breakfast burritos at Resistance Coffee House. I split one with Bill and he keeps another for later in the day. I grab my lights and water belt and head to the start line. A surprisingly large crowd is on hand for the start. Sure, lots of them are crew members, but a good chunk of the town’s population is also on hand. Leadville was literally saved from extinction by this race after the mines closed in 1982. The 100 is now the conclusion of a season of running and mountain bike events that provide a significant chunk of Leadville’s tiny economy. The locals get this and really go out of their way to be perfect hosts. At 4AM, they cheer us off.

It's a 4AM Hoedown!

The first few miles are very easy running on a mix of paved and dirt roads. At the end of that stretch, there’s a bit of a grunt walking up the trail to the top of the Turquois Lake Dam and then we get on the singletrack that takes us around the lake. While it’s relatively flat, it's technical. I’m glad we scoped this out because I had thought my little headlamp would be enough. After viewing the trail, I decided to carry my big hand-held light (which still weighs less than a pound, even with the 10-hour battery). Having the brighter light source further from my eyes creates much better shadows, exposing many trip hazards in the form of roots and rocks.

Packing 600 runners onto singletrack after just four miles might be expected to produce a logjam. Happily, I find myself fairly free to run my pace. A few pass and I pass a few, but nobody is in much of a rush at this point. I jog into the May Queen Aid Station (13 miles) at 6:15, at the slow end of my expectations, but not alarmingly so.

Turquois Lake - Note the sliver of moon two days ahead of the eclipse.

Because there are so many runners meeting their crews all at once, the crews are fenced off from the actual aid station. This throws me a bit. I had expected to be able to dump aid station food into the musette that Bill was going to hand me. Fortunately, they have some bags available at the aid station, so I am able to get enough food. Once through the fence, the crew zone is every bit the madhouse that I had been warned about. Bill has done a good job of predicting my arrival and positioned himself right by the gate. We walk together for a bit while we sort through the exchange of empty bottles for full and lights for trekking poles. It all goes smoothly, but it feels very frantic. I leave with an unsettling sense that things aren’t right.

The climb of Sugarloaf starts immediately upon leaving the aid station. I had misread the course map in this regard and thought I had some flat ground still to come. After about a mile, I accept that I must be on the real ascent and start using my poles. That makes it all but impossible to eat or drink without stopping. Rather than try to make that work, I just resign myself to stopping a few times.

Looking South from Sugarloaf

About halfway up the climb I start feeling incredibly tired. Not physically – my legs are fine and I’m not even breathing particularly hard – but I’m really having trouble staying awake. I haven’t had any coffee this morning and it was a short night’s sleep, but this is ridiculous. All I want to do is lie down on the trail and nap for a while. My progress becomes quite slow and I’m passed by dozens of “runners” (nobody is actually running up this). Breaking above tree line helps a bit as I’m rewarded with some very nice views. By the top, I seem to have perked up and it’s a good thing because we are about to hit the most treacherous part of the course.

Top of powerline

The descent off Sugarloaf (and ascent on the return) is known as “Powerline” for reasons that one can probably guess. High-voltage wires are really expensive so they tend to be built straight, regardless of what’s on the ground beneath them. The resulting service track, which stays directly below the lines except for truly uncrossable obstacles, is an obscenely steep jeep path littered with small rocks and large erosion gullies.

Just plain nuts.

A runner ahead of me stumbles. It’s one of those slow-motion disasters where you know the finish from the start. I watch helplessly as his weight lurches forward. Then begins the slow but inevitable rotation where seconds seem like minutes as the ground moves ever closer to his face. He hits so hard, he literally bounces off the ground. I have to dig in with my heels and poles to not run him over. My first question for him is whether he thinks anything is broken. He doesn’t, though he won’t rule it out. He decides that he’ll just collect himself for a few moments and tells me to move on. There are plenty more runners coming if he ends up needing assistance.

I continue on, using my poles both to steady myself and reduce the impact. I kick up a flat stone that hits my ankle hard enough to elicit an unprintable response. Still, getting to the bottom with only a minor bruise is a win in my book. From there, it’s an easy jog on roads to the Outward Bound Aid Station at 25 miles. I take the opportunity to eat the rest of the food I took back at May Queen.

It’s 8:50AM, and Bill was beginning to get worried that I’d taken a fall; my planned pace called for arriving between 8:15 and 8:30. I’m not worried that I’m behind schedule, though I internally let go of any hope for a sub-25 finish. Unlike “short” events (like a marathon), you can have early problems and still salvage a good ultra finish. However, as Samuel Johnson so aptly put it, “He who competes against time has an adversary that does not suffer casualty.” If you’re going for a time goal, the early miles have to be perfect and they clearly weren’t in this case.

Crews are allowed in the aid station area, so Bill helps me grab food and drink. We again walk together out of the aid station while I put down some more food and get rid of my long sleeved shirt and poles. The sky is cloudless, so I also switch to my desert hat since my Irish skin won’t last long in the sun at this altitude.

If mountains make you feel small, the Arkansas River Valley between Outward Bound and Half Pipe makes you feel positively tiny. It’s a vast swatch of meadow with 14,000’ peaks on both sides. I feel like an ant trying to run the length of a football field (one with a lot of seating at that). Hope Pass is clearly visible 20 miles away; it will take 6 hours to get there.

The valley with Hope Pass in the distance.

The Alternate Crew Location is a fun diversion. Most of the crews have arrived in vehicles with tires much larger than mine. It’s still early enough in the race that most of the runners are looking good and the air is filled with shouts of encouragement. This is clearly a non-partisan effort and I get just as many cheers as those who have stationed crew. Beyond, we start slowly moving out of the valley, though the real climbing doesn’t start until after the Half Pipe Aid Station at mile 31.

Climb up Elbert

I get to Half Pipe feeling glad that at least one section of the course was covered at predicted pace. The station is not crew accessible, so there’s no reason to linger. I refill both my bottles and grab a couple sandwiches and am back on my way. The course follows a slowly climbing jeep road for a few more miles, then switches to singletrack for the real effort of getting over Mt. Elbert. While we don’t come anywhere near the summit (the highest in Colorado), it’s still a substantial climb and I find that two bottles isn’t enough. About a mile from the Mt. Elbert mini-station (fluids only), I’m dry. This is a very bad situation to be in at altitude, so I switch to walking knowing that heavy breathing will only make matters worse.

At the aid, I find I’m hardly alone. The day is considerably warmer than usual and many runners are bringing in empty containers. Fortunately, it’s only three downhill miles to the main aid station at Twin Lakes, so there doesn’t seem any reason to camp here to replenish. I get my bottles refilled, drink a bit more for good measure, and get back on the trail.

The descent to Twin Lakes is certainly easier than Powerline, but you can (some do) still mess yourself up if you’re not careful. I don’t have poles, so I take it easy. I’m passed a guy wearing a Tour of Missouri jersey and ask him where he’s from. Turns out, he’s Eric Strand from the St. Louis area. I’ve seen the name in results several times, but we’ve never actually crossed paths (which is odd, given the “everybody knows everybody” characteristic of the ultra community). We quickly make acquaintances and he even shoots a little video of us. I’m impressed that he can do this without replicating the catastrophe I witnessed earlier. I’m still trying to save my quads, so I tell him I’m dropping back a bit.

I get into Twin Lakes at 12:10PM. There are some aid stations that can be skipped but, with Hope Pass looming, every crew is present at this one. As such, it is quite a carnival atmosphere as the race entourage takes over the tiny town. Bill has again positioned himself well and I spot him immediately. we make a quick pass of the aid station fare, then head over to the pit area he has set up. I change into my Altras, which have less protection but better grip than the Asics shoes I’ve been wearing. I also select a thinner sock so I don’t pick up as much water at the river crossing. In the process, we both look at the bottom of my feet and find no blisters.

I’m feeling fine, but I don’t rush through the aid station. Forgetting something here could be disastrous, so we go through the checklist twice. Emergency clothing (hat, raincoat, gloves, space blanket), food, all four bottles filled, salt tablets, poles, desert hat, change of socks. While it sounds like a lot, it all fits into the back pockets of my jersey (except the bottles, which go on my belt). Bill walks with me to the trailhead, helping me get down as much food as possible before setting out. Eric Strand must have already showed his crew the video because they recognize me as we pass and give a big cheer for St. Louis. At the trailhead, I bid farewell saying “I should be back in 8-9 hours, but no promises!” Ignorance is bliss.

The “river” feeding the twin lakes for which the town is named is actually a maze of wetlands. The deepest part is mid-calf (and frosty!) as advertised. A rope is provided, but the current isn’t strong enough that I have to use it. Since the deepest and fastest moving portion came last, it feels like most of the sand got washed out of my shoes. I decide I can keep going without the sock change for now.

Shortly after the river crossing, the climb begins. While this side is not quite as steep as the return, there is more altitude gain. At 9200’, the river crossing is the low point on the course. At 12,600’, Hope Pass is the high point. This difference will be gained in a mere four miles of trail. It’s far too steep to run (at least, in the context of an ultra), but I don’t find it particularly difficult to walk quickly while using the poles. A few manage to come by, but I’m passing many more. After a couple miles, the grade relents a bit, but it steepens again as we hit tree line.

Now in the open, it’s also obvious that the clouds that have been gathering since late morning are not going to let us off for nothing. As mountain storms go, it’s pretty light – rain, wind, and some small hail – but there’s no way to tell if the next cloud to come over the pass is packing the full fury of the mountains. Only once have I been caught above tree line in a real storm and that was all it took to make me not want to ever do it again. When I get to the aid station, I decide the tent is looking pretty good.

Yes, really, the aid station. The way Ken Chlouber tells it (which may or may not correlate with reality), the offer came from the volunteers. “I couldn’t ask anybody to go up there and do that, they came to me.” It does seem moderately insane to hang out on the exposed slope of one of Colorado’s highest passes during the worst weather hours of the day. But, they do. The twenty llamas that carry the tents, water, and equipment are sitting idly in the alpine meadow, seemingly unfazed by the squall.

Leaving the tent as the storm clears...

... and moves on to Twin Lakes.

Thankfully, it is just a spot shower and less than ten minutes after I duck into the tent, the sun has returned and the two dozen runners that sought refuge set off on the final climb to the top. While that makes for a bit of a conga line, it’s easy to get past people since the trail is flanked by meadow on both sides. I cross the pass just before 3PM. Not bad considering the long stop at Twin Lakes and the time in the tent.

Starting the descent to Winfield.

The descent down to tree line is all switchbacks and ramps. Aside from a short bit that crosses a very rocky avalanche track, it’s easy running. Below tree line, the path heads pretty much straight down the hill. It’s exceedingly steep and my arms are burning from relying so much on the poles. We don’t go all the way to the valley. Instead, we make a hard right turn while still a few hundred feet above the stream and begin the traverse into Winfield. It should be easy running, but it’s not. I can feel my energy levels crashing. I’m staring into The Abyss.

I’ve written about it before, so I won’t belabor the point. In every hundred I’ve ever run, I hit a major low point sometime between 10 and 15 hours. I've come to call it The Abyss. As we’re currently 12 hours in, this one is coming right on schedule. The only defense I’ve found is to walk it off. Energy will come back, it just takes some time. Unfortunately, time is no longer on my side. I stopped worrying about it altogether when I gave up on the sub-25, but now I need to think about cutoffs. I need to be back in Twin Lakes by 10PM to stay in the race.

I had hoped to get from the top of the pass to the turn at Winfield in around an hour and a half. It takes over two. Upon arriving, I can’t motivate myself to immediately turn around and head back. I enter the aid tent and change my socks (still no blisters, yay for that) while the aid station workers bring me food and drink. That would be a decent use of time if I could get anything down but, even after 20 minutes at the aid station, I’ve only managed to ingest two cups of soda and a handful of potato chips. Maybe 200 calories. I’ll spend five times that getting back over the pass. I force myself to stand and get walking again. Hopefully, I’ll be able to eat something soon, but sitting around is just making things worse.

Back on the trail, I try my best at base-60 arithmetic (always a challenge after running 50 miles). The trip out from Twin Lakes took almost five hours, but that included a long stop before heading out and a short stop waiting out the storm. It was probably less than 4:30 moving. I left Winfield at 5:25, so I have 4:35 to get back. The climb back is steeper, but it’s not as much vertical. The descent will be tricky after dark without my big light, so I can’t count on gaining time there. I could buy a few minutes by running the traverse, but that would leave me even weaker for the climb. I try some short jogs. They don’t feel bad, but they don’t seem that much faster than walking, either. I decide to just keep walking fast as that’s making me feel better and hope I have something for the climb. My real hope is that I’ll be able to run the last few flat miles into Twin Lakes reasonably hard. I’m positive that I can get back from Twin Lakes in under 12 hours, so a big surge to make that cut is a reasonable risk.

Birch forest along the traverse from Winfield.

I get to the end of the traverse feeling like I have a plan and am ready to do it. I turn up the hill and…

Nothing.

Absolutely nothing.

I can hardly take a step forward. After the initial shock passes, I manage to put together four or five steps without pausing. With steely determination, I take another 10. I lean against a tree, dizzy and weak. I’ve gained maybe 10 feet of elevation. Only 2700 to go.

The next two hours will be remembered as among the most hopeless times of my life. I am completely incapacitated by the grade. Normally, I just back off in situations like this, but there’s no way to back off and make any forward progress at all. I finally pass tree line and nearly break down in tears when I see how far I still have to climb. I sit down on a rock and try to force down a gel packet. After five minutes, I’ve eaten almost all of it. It stays down for all of 30 seconds and then comes back the way it went in. A few of the “runners” stop to ask if I’m OK. The answer seems pretty self-evident, but I politely tell them that I expect I’ll be able to get moving again at some point.

Inside, I’m not so sure. As the sky begins to fade I feel horribly alone on the mountain. I wonder if I can possibly make it to the top. I wonder if I’ll freeze to death. It’s nonsense, of course, the panics of a fatigued brain. I have my jacket, hat, gloves, and a blanket. There are still plenty of folks on the trail. The course sweeps will be coming along in just another hour or two. The chances of this turning into a survival situation are truly remote. But the emotions are real. And, terrifying.

Terrifying enough that I force myself to my feet. I didn’t count, but it seemed like there were around 20 switchbacks coming down (after the event, Google satellite images reveal 22). If I can get through two at a time between rests, that will give me manageable goals without spending all night on the ascent. I walk the first ramp and force myself onto the second, fighting the urge to take a rest. At the end of the second ramp, I rest for about a minute. I move on.

Two switchbacks at a time.

The rests get shorter.

The mountain gets darker.

I start passing people instead of being passed.

Six ramps from the top, I turn on my headlamp.

I’m taking full strides and pressing on the poles.

I skip the rest break two ramps from the top.

Finally, I’m at the pass. Oddly, here at the high point of the course, having just ground through a horrific fight with despair, I feel fine. I feel great. I'm actually looking down at stars in the dark sky to the east. It is magnificent. I jog easily down the other side to the aid station. I have once again come through The Abyss and the feeling of rebirth is as strong as it always is. But, obviously, the damage is done. I get to the aid station at 9:10. Even with fresh legs in the daylight it would be a firm run to make the cut. At night, with my reactions dulled by miles and only my small headlamp to light the technical descent, there’s no hope. My race is over.

I stay at the aid station long enough to down a cup of soup and locate some other runners to descend with (I don’t like being on the mountain alone at night). Knowing that we’re going to miss the cut, we don’t take any chances. We walk the whole descent, chatting aimlessly. The river crossing is even colder this time and my feet are completely numb by the far side. Bill comes out to meet me and walk the last mile into Twin Lakes.

Bill tells me there were a lot of tears from people who were only minutes away from making the cut. We’re late enough that there isn’t much drama in our arrival. My companions and I stand stoically while our athlete’s wristbands are cut off and we are officially pulled from the race. Between the cool temps and the fact that I haven’t really run in nearly 10 hours there’s no need to clean up before driving back to Leadville. We’re in bed by midnight.

It’s a small consolation, but since we don’t have to wait around for awards, we’re on the road before noon on Sunday and get to central Kansas before it gets late. That means we’re reasonably well rested for chasing breaks in the clouds around Nebraska the next day. We move several times, not settling on our final position until 10 minutes before totality, but we do wind up directly on the centerline of the Solar Eclipse with a clear view of the sun. And, since we’re surrounded by storm clouds, we also get the treat of the bright orange horizon in all directions. It’s a magnificent display that was certainly worth a morning of roaming around cornfields.

Dodging storm clouds to see the eclipse.

So, what went wrong? Well, nothing, really. The Abyss is nothing new; I’ve hit it in every 100 I’ve run. It’s just never been an issue before. As long as there are no mountain passes in the way, I just shuffle through it and then get back on the gas for the last 25 miles. I’ve always finished in under 24 hours, so time cutoffs haven’t even been something I’ve thought about.

I got ripped off. I capped this gallon of air in Leadville and by Nebraska it was half gone!

I needed another 30 minutes to make the cut at Twin Lakes. Just knowing the course would probably provide half of that. A little more urgency on the way out and a pacer coming back would likely get the rest. But, that’s cutting things awfully close. The real answer is to mitigate the crash. I have no idea how to do that, but it’s clearly related to blood sugar levels, so a start would be to figure out how to get more sugar in my system without puking. It’s something I can work on back in the Midwest.

Assuming I do, will I go back? I don’t know. I love the mountains. I love racing in the mountains. I’ve had a lot of success in the mountains in cycling and orienteering. But, it appears I’m just not a very good mountain runner, at least not in long events. Sure, some of that is field strength. If the leaders at Leadville showed up at a Midwestern 100, they'd kick my ass there, too. But, even though I have to put aside any hopes of winning my age group when I run the Boston Marathon, I'm still one of the faster 50-year-olds and finish in the top 10% overall. I certainly don't worry about them shutting down the finish before I arrive. The problem with mountain ultras is about me, not the competition. Whether I find the time and motivation to address that is very much an open question. It certainly won’t happen while I’m in grad school so, for now, I’ll just have to take the dose of humility and get on with life.

I can live with failing at a difficult task, but I’m not going to seek it out.

Outline for intro

Work has been a bit crazy the last couple weeks as we approach the release of our scalable reporting platform. All very exciting, but not conducive to getting schoolwork done. In the spirit of any progress is better than none, here's my outline for my survey introduction. This can probably be pared back to be an intro for the CISS paper:

1. Overview of the problem
1. Multidimensional vs. Multivariate data and analysis
2. Traditional Star and Snowflake models
3. Features unique to financial data
4. Features unique to seriatim data
2. State of the art of On-Line Analytic Processing (OLAP)
1. Static OLAP cubes
2. Ad-hoc reporting tools
3. Data-frame analysis
4. Big Data analysis
3. Proposed improvements
1. CISS
2. D-BESt
3. Weighted subsetting
4. Training and confirmation analysis using weighted subsets

Unequal Probability Sampling

Getting back to CISS for a bit. The selection process for the next block to sample was basically to look at the variability of the strata estimators and pick a stratum in proportion to it's "uncertainty" (technically, the quantity I use is the variance of the unsampled data, not the sampled data, so I call it uncertainty rather than variance). I did this without a bunch of mathematical justification because it seemed intuitive enough. Of course, that attitude can get one in a lot of trouble so I'm happy to have found a reference that has already done most of the heavy lifting.

Reference: Shahbaz, MQ., Shabaz, S., and Hanif, M. International Journal for Open Problems in Computational Mathematics, Vol 4, No 1, March 2011.

It starts with the Horvitz and Thompson Estimator (another paper I should include) and looks particularly at the 2-valued sample. This isn't super interesting in and of itself, but the introduction is good annotated bibliography starting with Hansen's paper that I reviewed last spring. I could have gotten most of it by simply doing a forward reference on Hansen, but now that's one less thing.

The main takeaway is the variance computation. As I said, my uncertainty isn't really a variance, but if the CISS technique was used to create a sample that would then be used for leveraged analysis, I'd need to flip that around and be able to estimate the sample variance. So, it could come in handy.

Implementing Randomized Matrix Algorithms

This is purely background, but important background for my sub-sampling work.

Reference: Yang, Meng, and Mahoney: Implementing Randomized matrix Algorithms in Parallel and Distributed Environments:  arXiv:1502.03032v2 [cs.DC].

The paper looks at how one implements RandNLA (Randomized Numerical Linear Algebra) on distributed systems (such as Hadoop, which is my implementation target, though I'm keeping the theory general enough to be ported to any horizontally scalable platform).

The paper has a good presentation of the basic concepts of RandNLA, as well as covering the eigenvector stuff used for axis rotation and dimension reduction.

They also cite a distinction that I hadn't really thought that much about. That is, the difference between Monte Carlo and Las Vegas simulation. In Monte Carlo algorithms, you fix the computational complexity and accept a certain risk that your answer will fall outside the confidence you are looking for. Las Vegas algorithms do the opposite: they guarantee an error bound, but you may pay for it with an arbitrarily long run time.

This distinction is important in my application space because, most of the time, a randomized sampling algorithm would be used for exploratory research. Here, a bogus result is acceptable since any significant decision arising from it would be confirmed against at least another sample, if not the entire data set. On the other hand, if these queries were to be used for regulatory purposes, you would need to guarantee the accuracy. Currently, sampling is not an accepted practice for regulatory reporting, but that could change.

The authors also present some techniques for solving when either low or high precision is required.

Unfortunately, the bulk of the paper is devoted to regression problems, which aren't really what I'm after. It's still a good reference, though.

Back to (school) work

Well, well, it's August and I've squandered another summer. OK, that's a bit strong. Our big project at work is in UAT, and there was considerable overtime involved in getting that to happen. Also, while a thousand things can go wrong in a long race, I feel like I've actually managed to get myself prepared for the Leadville 100 (August 19). Most importantly, I've had a few days off to spend with my family.

However, all that good stuff has come at the expense of forward progress on my research. So, it's time to get back at it. I've got a lot of reading to do. I'll start posting notes again as I work through my rather lengthy backlog of papers to review.