We got on Rt 97 in Weed, CA just north of Mt Shasta and headed northeast into Oregon. There’s a lot of really neat topography for most of our trip, and we started seeing it as we left the hotel in the morning. The night before we arrived in the dark, so it was a bit of a surprise to wake up surrounded by mountains. Pretty much the rest of the trip we were in the mountains…or darn close to them. It was great.

In some little town we hung a left onto Rt 62 and headed northwest towards Crater Lake and ended up stopping at this tiny hotel/store/deli along the way. At Jo’s Motel we grabbed a turkey sandwich with fruit salad for me and a hot dog for Liz. It was a nice little stop in a town of maybe 20 people if it could be called a town.

After lunch we continued on our way, and about 5 minutes after getting in the car we spotted a bald eagle above the roadway. It was coming basically down the road towards us. At first it looked like a typical hawk, but as we got closer it’s white head was very apparent and as we cruised under it we got a great view. First time either of us had seen one in the wild and pretty darn cool.

Pretty soon after that we ended up climbing into a forest and we noticed a huge gorge to the right of the road. We stopped a couple of times, but we couldn’t see the bottom from the pullouts on the road. Finally we stopped at the Annie Falls pullout and saw to the bottom of the gorge where Godfrey Glen was with the small waterfall on Annie Creek. Very pretty. If you take a look at this map you can see where RT 62 goes by the park. Annie Creek flows through the gorge south of the park and it’s DEEP! 300 feet or more from Rt 62 according to google maps!

A little farther up Rt 62 and we hung a right onto Rim Drive and treated ourselves with an annual National Park Pass that’s good through next september. I’m sure we’ll try to take as much advantage of that as possible over the next year. It’s only $80 for the car. Definitely worth it.

We stopped at the Godfrey Glen trail on the entrance road to the park to get in a quick hike. We knew we weren’t going to have a ton of time to hang out at Crater Lake, so we planned on doing a couple of short quick hikes to get a feel for the place before moving on. The Godfrey Glen trail is a nice easy trail that’s essentially flat until the very end and it winds its way along the top of a ridge overlooking Annie Creek (which you can’t actually see because the gorge is so steep) over to an overlook on Godfrey Glenn which is a nice meadow at the bottom of the gorge were it gets pretty wide. There’s actually a trail that goes all the way down to the glen…it’s on our list for a longer trip!

Near the very end of the trail we were talking and looking at things and not paying that much attention to things when I got stung by a bee on the arm. It was then that I heard all the buzzing and I quickly looked over my left shoulder and saw an old tree stump completely consumed with bees. With a quick “Run, run!” Liz and I were off on a dash to the car, which fortunately was only 100 feet or so away. Unfortunately Liz also got stung on the leg. We ended up sitting in the car in a surprising amount of pain for a bee-sting. It has to have been 15 years at least since I’ve been stung, and even longer for Liz.

The bee stings did make us take a quick trip to the ranger station to get some health info and alert them to the nest right on one of their trails. In the end we were ok, although they sting sites stung and itched for quite awhile. Liz might still have a mark on her leg a couple weeks later!

Once we determined we weren’t in anephalactic shock we got in the car and headed east around Rim Drive towards the Sun Notch overlook trail. Oh, I should mention that the ranger station and visitor center sit outside the crater, and from there when you look at the crater it looks basically like the base of a huge mountain. Outside of the crater it looks like a giant wall of rock 800 feet or so high. How the heck were we going to get to the lake?

Well, it turns out that Rim Drive climbs both east and west up to the top rim of the crater. We drove east, as I mentioned, and stopped at the Sun Notch Trail. That trail was pretty short, but up a pretty steep incline to the rim of the crater. Once we got to the top we were greeted with one of the coolest views I’ve seen!

The rim of Crater Lake at that point sits at an elevation around 7150 feet. The lake itself however sits at an elevation around 6200 feet. So when you get to the top of the trail you are standing on a 950ft cliff looking out over this enormous body of water that happens to be the prettiest shade of blue you can imagine. The lake isn’t fed from streams or creeks (and no streams or creeks flow from the lake), so no sediment flows into the lake. Instead it gets all of the water from precipitation…rainfall and snowfall. It keeps the water really pure and quite beautiful.

While we were standing at one of the overlooks at Sun Notch a tour boat cruised over toward the island known as Phantom Ship which is visible from the trail. This thing must have had 15 people on it and been 25 feet long at least, but it looked like this tiny little toy boat down on the lake. Eventually I’ll put up some pictures, but the boat is dwarfed by Phantom Ship which I could hear the tour guide on the boat say was roughly 180 ft long and 150ft tall.

After hanging out on the rim at Sun Notch for a while we hiked back down to the car and drove back west around the crater. We stopped at a bunch of overlooks, including some over Wizard Island, but we didn’t have time to do any more hiking. We’re planning on going back for a long weekend so we can do some longer hikes and hopefully go out to Wizard Island which has a small mountain on it (700 ft tall or so) which has a 90 foot deep crater of it’s own!

From Crater Lake we headed up to Bend, OR for the night where we had our reservation. We just stayed at a Day’s Inn, which was fine. Nothing to really say about that. Unfortunately, we didn’t spend any time checking things out in Bend. Rt 97 takes you through the “strip mall” side of Bend, and we weren’t very impressed with it at first. We did drive across the highway to the nice part of Bend, but it was dark and we were hungry, so we didn’t spend time looking around. Instead we headed straight to the Bend Brewing Company which had been recommended.

I ended up trying their Black Diamond Dark Lager and the Metolius Golden Ale, both of which were quite good. Liz had the High Desert Hefeweizen which she liked. Since we were starving we ordered up some beer battered zucchini sticks which was really just a way for them to give you greasy beer batter without sounding so unhealthy, but that didn’t hurt. They were tasty. Liz ended up with the Lava Butte Turkey Melt and I had a simple Cheese Burger after which we split a hot fudge sundae.

Once dinner was done we headed back to the hotel and collapsed for the evening.

Oh, I should mention something about the car. I think the transmission likes the hills. We didn’t have much trouble at all on the second day of our trip. I do remember stopping once or twice to restart the car, but I don’t think we stopped just to do that. Instead we were able to do that when we were getting gas or food, and we never spent much time riding along in 3rd gear.

Luckily for us, third gear worked just fine!

We actually got off the highway shortly after the transmission trouble began and I discovered that if you shut the car off for a minute and then started up again the cars computer reset and, if driven carefully, would shift into and out of every gear without a hitch. That is until someone cuts in front of you, or you hit a hill, or otherwise have to alternately ease up and then press the accelerator.

We went through the stop, turn off car, wait, turn on car, go cycle three or four times on the first night out of town. It was very nerve-racking and both of us had the unspoken thought that we wouldn’t even get to our first destination before having to cut our trip short.

But that didn’t happen. Instead we were able to get to the Gaia Shasta Hotel in Anderson, CA in pretty good time without hitting any significant traffic. Other than the potentially catastrophic transmission issues it was a pretty good drive!

The Gaia Shasta Hotel turned out to be pretty nice. It had 3 “pods” with 12-15 or so rooms each around a central courtyard area that had a nice rock garden. Outside in the back was a really big pond with a pair of white and a pair of black swans swimming. Around the pond was more rock garden and there was a bridge between the hotel and the restaurant next door. The hotels design and philosophy is to be as green as possible, and they obviously want to encourage that. Our room had an abridged version of Al Gore’s “An Inconvenient Truth” so you can catch up on all the scary results of future global warming. Oh, and the continental breakfast next door was really nice as well with Kaishi cerials, organic bagels, lots of fruit, coffee, teas, muffins, etc. If you find yourself near Redding, CA, head 10 minutes farther south and stay at Gaia. It’s cheap, nice, and green(er).

Yay!

[Show as slideshow]

12►

That’s the URL for this site courtesy of dickensurl.com.

Fun.

Nuclear power is safe an extremely environmentally friendly and new reactors designs are almost 100% safe. France has been using nuclear power for 75% of it’s electric needs for decades and the total amount of waste is, as the editorial mentions, very small. We need to get back into the nuclear reprocessing game. It’s easy, relatively speaking of course, and it solves our “waste problem” that should not exist in the first place.

Quality vs. Quantity

• Check Email ONLY
  • 10am
  • 1pm
  • 4pm
  • Send anytime
  • No email on evenings
  • No email on weekends
  • Emergency? Use phone
• Focus on 1-3 activities per day, max
• Log 1-3 succinct status bullets every day
• Out by 5:30pm - NO EXCUSES

Must be nice!

Three things came to mind when reading the article.

First, the author comes at the problem from an economic standpoint discussing the cost of a college education and the cost breakdown of the idea team mentioned above with apprentices earning basically minimum wage (they are young, after all!), journeyman making decent money, and masters doing very well. The example team above with one master, three journeyman, and nine apprentices would cost roughly $600k per year to employ in salary alone based on the example pay scale. If you subsitite nine college graduates making at least $50k per year that annual cost to employ such a team would be much higher. Add to that the cost of 4 to 6 years of college tuition and it’s higher still, though the $200k sited in the article applies only to the elite private universities. As a comparison, my undergraduate alma mater, James Madison University, would cost less than $60k today for in-state students who lived on campus all four years.

What I find interesting about the economic view of teams constructed in this manner is that the cost savings to companies and individuals would be significant, especially so in the current economic conditions.

The second thing that struck me when reading the post is that this is in fact how most teams that I’ve been on are structured. Generally those teams have had someone filling the role of “Development Lead” that has a lot of experience and generally directs the large, architectural design decisions for the project. Below the development lead are a group of mid level developers; people who have been in the industry for 3-7 years or so and who have a solid grasp of how to design and effectively execute a software project. Finally there is a set of junior developers who are usually very wet behind the ears, but who have the potential to be good software engineers.

The biggest difference between the teams outlined in the post and the ones that I’ve been a part of is that the wet behind the ears junior people have tended to be given a great deal of responsibility for their level of experience. Despite this, in almost every case, they’ve done a very admirable job and did not hinder the teams success.

The third thing that comes to mind is that I’m not sure that I would trust the nurturing and development of the very newest developers to all that many of the people that I’ve worked with over the years. In almost every case those people have been hard workers and have gotten a lot of good work done. However, there are a lot of them who don’t necessarily possess the requisite breadth and depth of knowledge and experience to guide new developers along an optimal path towards a successful career.

I envision journeyman almost taking on the role of college professors by teaching language theory and design, computer architecture, software architecture and design, database design and normalization, data structures, etc. to their apprentices and not simply handing out menial coding assignments that have to be re-written 5 or 6 times to become “acceptable” solutions. I feel there is definitely some real value in getting budding developers into the “real world” of software development early, but if these potential developers aren’t exposed to the fundamentals of software engineering and instead have to figure it out by trial an error the industry is doomed to mediocrity.

In all the blog post was thought provoking, but in the end I don’t think it is a very feasible solution for the industry as it currently exists. Companies are going to be very skeptical of hiring apprentice programmers for 2-4 years while getting very little actual work out of them even if they are only making minimum wage. Add to that the fact that company loyalty in this industry is not very strong and it becomes way to risky for a company to spend time and resources training apprentice programmers when they are most likely to switch jobs before contributing much of anything to the company.

Another problem with the ideas in the article are simply that software development teams already seem to be structured in much the same fashion. The major difference being that the most junior developers already have four years of instruction and can, at least theoretically, contribute to the team from the start.

That isn’t to say that the question of whether or not the traditional four year college degree is essential to training successful software developers. It isn’t. However, I don’t think that it can be replaced by a purely workplace based education. Perhaps some combination of work/school is the best recipe for success.

Perhaps a better career path for a future software developer would be a 6 month set of courses in the fundamentals of computer science dealing with computer architecture, algorithms and data structures, networking, database design, and software design followed by another 6 months of half schooling/half working under an experienced developer and then full assimilation into the software team? I don’t know what the right balance is off the top of my head, but something about taking an 18 year old high school grad and making them write and re-write code without any real teaching and guidance doesn’t sound like the way to improve the overall skill level in the software industry. In fact, it makes the future all the more frightening.

First, let me show you what I was doing. I’m writing a data persistence layer over the Google App Engine Java persistence store so that I can write my JRuby on Rails apps very similar to how you do using ActiveRecord. So I have a model objects and methods for defining properties and relationships on the class. Java is statically typed, so I wanted to ensure the translation from ruby objects to java objects happens correctly. To do this I have mapped the subset of GAE supported java types to a set of ruby symbols. To set up a model object you do the following:

class MyModel < DataAE::Entity
  property :id, :key
  property :name, :string
  property :age, :integer
end


And the DataAE::Entity class has the property function, like so:

@@types = Hash.new


def self.property(name, type)
  ...
  @@types[name] = type
end

Pretty straight forward, right? I’m sure some of you have already figured out my problem, but it took me forever to think of it. Anyway, the “…” part of the method creates attribute accessors based on the name passed in so that you can call “myModel.age” and get the age back and a “myModel.age = 30″ to set the age. This is a little more difficult than it looks though because you have to translate from the type passed into the self.property method to the java type that is stored in GAE. The details of that aren’t that important now except for the first part of the set method. Here it is in it’s broken form:

def ae_set_property(name, value)
  type = @@types[name]
  …
end

When this is run you get an empty string back. What?!? Clearly in the self.property method I’m storing things in the hash, right? I even did a

@@types.keys.each do |k|
  puts k
end

which resulted in

id
name
age

And then I tried

@@types.values.each do |v|
  puts v
end

resulting in

key
string
integer

Even the Hash to_s method which should return KeyValueKeyValue… returned

idkeynamestringageinteger

But every single time I tried @@types["id"] I’d get back ”(empty string). It made no sense.

And then it hit me. The self.property method was taking in symbols and storing them in the hash. Then I was using strings to try and access those keys and it wasn’t working! The solution was simple; either store the keys as strings or convert the string to a symbol when calling the [] function. I opted for just storing the key as a string and all is well. The self.property function now looks like this:

def self.property(name, type)
  ...
  @@types[name.to_s] = type
end

Sigh…

The reason for this are mainly that I didn’t feel like carrying around my big DSLR bag with camera, 3 big lenses, extra CF cards, a portable hard drive, etc. It was just a pain, especially considering the topography of this city. Another reason was that I didn’t want to become any more attached to that gear because I knew, in the back of my mind, that I’d probably have to sell it all since I couldn’t sell my condo back east.

It turns out that I did get to a point where I had to sell my gear. In January I sold the DSLR and one of my lenses (the 35mm f/2.0) for some quick cash. It turned out that I didn’t need to sell my other lenses to feed myself, but they’ve just been sitting here without any use since then. So two weeks ago I sold my favorite lens, the Canon 17-55mm f/2.8 IS. Absolutely fabulous lens, but useless to me right now. Today I put my last lens, a Canon 70-200mm f/4L, up for sale. Hopefully someone wants it.

But that left me completely lacking in the picture taking department. I needed to find something that was considerably cheaper than my DSLR kit yet something that would give me image quality results nearly on par with what I’m used to from the DSLR.

After poking around on DP Review for a bit I decided that I wanted to stick with a Canon camera. It’s what I’ve used and what I liked before. The decision came down to the four cameras below.

• SX 110 IS
• SX 10 IS
• G10
• Rebel XS

I ended up buying the G10 and it should be on it’s way here tomorrow. I picked the G10 for the following reasons.

First, it’s a pretty nice form factor. It’s basically flat when off and fits mu hands nicely. It shoots RAW out of the box. It has a viewfinder, even if it isn’t perfect. It has a hotshoe so I can buy a flash in the future or use a viewfinder if I want to. It has very good image quality at the lowest ISO’s and is even useable in the 400-800 ISO range. And finally, it has a nice set of external controls so I won’t have to hunt around menu’s to change the most used settings.

Picking the G10 means that I didn’t pick the others. Each of them seem to be great cameras at their price point, but each of them had a “fatal flaw” that knocked them out of the running. The SX110IS doesn’t have a viewfinder, doesn’t shoot raw, and didn’t have a hotshoe (so i couldn’t add a viewfinder). The SX10IS seems nice (the SX1IS even better) except that it doesn’t shoot raw* and it pretty big. The Rebel XS looks really nice, but it puts me right back into the DSLR world with a big and bulky camera and the need for carrying around some other accessories.

So there it is. In short order I’ll have a nice “enthusiast” digital camera that shouldn’t be such a pain to haul around and that didn’t cost, and won’t continue to cost, a fortune. I’m very excited and can’t wait to get my hands on it! I’ll be linking my photo gallery on the blog here shortly. Hopefully I’ll have some new additions to it soon!

* - there is a “hack”, the CHDK, that opens up all kinds of options on lower end Canon cameras that would have given me raw output on the two lower end cameras, but the other things listed still knocked them out of the running.