For you techy kids out there, it involves Django, Python and a lot of coffee.

How?

What I'm going for with the project is photographs of faces with really, really thin depth of field. I think I've got the general idea down pretty well by now.

To get a really thin depth of field, the obvious first step is to use a wide aperture. The ones I've done so far have been taken at f/1.8. This helps, but I want only about an inch of focus so it's not enough on its own.

The next step is to use a longer focal length lens. I've taken all of the photos with my 77mm prime and I think it strikes a good balance; it's long enough to reduce the depth of field but not so long that it compresses the neck into the head too much. Shooting with my 50mm f/1.4 would mean I'd have to get really, really close to fill the frame which would start to distort the features too much for what I want.

Finally, the most important thing I've found is to get close. You know how when you're shooting macro images you have so little depth of field? That works to my advantage in this project. Cropping after the fact would give me more depth of field, so I have to get within a couple of feet to get what I want.

One of the challenges I've had with this project is using studio lighting. I've used it for two of those photos to really get the look I want, and learned a bit along the way. Normally I'm shooting at f/5.6 or narrower with a shutter speed of 1/125. This pretty much cuts out any ambient light in the studio and relies just on the flashes, which makes it easy to work. My camera won't sync properly at a faster shutter speed, so when I drop down to f/1.8 the light in the studio actually starts to have an impact. I've found that I actually need to dim the ambient light or it will show up in awkward ways in the photos.

Why?

This has been done before, by better photographers than myself. Then again, so has everything else I shoot; that's not really a reason to give up on anything. I've already learned a lot from the project and I'm sure I'll learn more as I go along.

Faces are the most interesting part of a person for me, so I guess it's only natural that I'd gravitate toward them. Richard Avedon talked about the surface of people and it's something I'm trying to explore. Thinning the depth of field brings out the shapes of a person's face rather than the face itself. I think it helps me look at faces in a new way.

What is it?

As Lindy Hoppers, we usually count in 8, whereas musicians count in 4 (for songs in common time). This works out because 4 divides into 8 nicely. So we count 1 and 2 and 3 and 4 and 5 and 6 and 7 and 8 and. This last "and" is the important one that he was talking about.

Why is it important?

The last and is important because the next beat (the 1) usually involves a larger and/or different movement. It's the start of whatever you're going to do next. It's harder to lead a movement that's larger or different than your current one; it's easier to simply continue what you're doing. By "harder" I don't mean "more difficult to learn the move, but once you've got it you're all set." I mean that it just takes more effort to change your follower's momentum.

The trick, then, is to prepare for the new movement (which will happen on the 1) before it actually happens. The smallest division of time that we usually using in dancing is half a beat, and half a beat before that 1 is the last "and."

Preparing for the movement will (usually) involve another movement. However, this one isn't as large or as different form what you're currently doing, so it's easier to lead. It's like an intermediate step that breaks apart a large, hard to lead motion into two smaller, easier to lead parts.

Where else does this apply?

This obviously doesn't have to be restricted to the last part of a swingout; that was just the example he was using to demonstrate the concept to us. The idea holds for anywhere in dancing; if you're going to lead something, preparing for it in advance will make it easier.

I think it can also apply to more than dancing. When playing an instrument like bass, to get to other notes you'll need to shift your hand. If you're sight reading or improvising, it will make it easier to shift if you consciously think about where you're going as you play and adjust your body positioning to prepare for shifting when it's needed.

It applies to photography too. Shooting in low light (without a tripod) is tough. Trying to freeze your body just as you hit the shutter button to avoid shaking the camera is hard. It's much easier if you anticipate when you'll want to hit the button and relax, bring your elbows in, breathe in, and then hit the shutter. It only takes a fraction of a second but trying to do everything at once is much more difficult.

This works in daily life, too. When I know I'm going to have a busy day, running from work to a lesson to a dance, I'll prepare. The night before I'll make my lunch, set the coffeepot to brew at 9 AM, put my bass in its case and pack my bag for the dance. That way instead of having to put my bass in the case and load it into my car all at once I've broken it apart into smaller actions, just like the larger move in dancing. It's much easier and less stressful.

What's the moral of this post? Prepare in everything you do. Think about what you're doing and break it apart into smaller parts. It'll make things work out better, I promise.

I decided to major in CS when I was in high school.  I learned to program on my own and enjoyed the challenge of it.  I also knew that programming jobs are fairly common and pay well enough that I wouldn't have to worrying about paying my rent.  I wasn't artistic in high school, and CS seemed like a logical, sterile field that I felt I could do well in.

During the years I spent at RIT I changed quite a bit.  I think the most important, or at least the most obvious, change has been my becoming more artistic.  Between dancing, playing bass and photography I've developed a creative side that I never thought I could.  Has this led me away from Computer Science at all?

No.  While I became a programmer for practical reasons, I stayed one for entirely different reasons.  Computer Science (and math in general) is beautiful.  It took me years to slowly realize this, but now that I have I see that it's more beautiful than any dance, photograph of music I've ever encountered.  This post is the first in a series of posts that I hope can communicate why I find CS beautiful, or at least point in the right direction.

A Quick Primer on "Functions" for Non-Programmers

I know that most (if not all) of the people that read my website don't program.  I'm going to try to avoid using extremely technical, computer sciency terms in these posts, but there are at least a few basic things that I need to explain.  The first is what a function is.

Think of a function as a set of instructions.  A recipe is a decent example.  Imagine you have a piece of paper with the following written on it:

1. Heat a frying pan.
2. Crack two eggs into a bowl.
3. Mix the eggs.
4. Pour the mixture into the hot frying pan.
5. Stir the mixture until it is solid.
6. Take the mixture out of the pan.

That's the simplest example of a function: a list of instructions that you follow in the order you get them.  The next idea is a "parameter."  That recipe only told us how to make food for one person, wouldn't it be nice to know how to make enough for two or more?  Imagine the paper now says this: 

1. Heat a frying pan.
2. Crack NUMBER_OF_PEOPLE * 2 eggs into a bowl.
3. Mix the eggs.
4. Pour the mixture into the hot frying pan.
5. Stir the mixture until it is solid.
6. Take the mixture out of the pan.

Now we still have a single piece of paper, but by just substituting in the number of people we're cooking for we have instructions for making any amount of food.  If we have three people, NUMBER_OF_PEOPLE * 2 eggs becomes 3 * 2 eggs, or 6 eggs.

The last important concept is "calling."  Once you have one function, you can use it in other functions, like so:

1. Put a plate on the table.
2. Refer to the other piece of paper and do what it says, for 1 person.
3. Put the result of that on the plate.
4. Eat it.

See how we did a few things, then referred to the other piece of paper instead of repeating ourselves?  This lets you make small tasks and put them together to form bigger ones.  That's enough of a primer for now; if something's not clear please comment and let me know.

What's Recursion?

Recursion is a term that means, basically, a function calls (or refers to) itself.  This concept can be hard to grasp, but once you do it slowly turns into something breathtakingly beautiful.

Here's a simple example: imagine you're 10 meters away from a doorway and you want to walk out of it.  A function that could help you do this might be: "Walk 11 meters."  That's pretty simple.

What if we want to be able to walk out of a doorway that's any number of meters ahead of us?  We could say: "Walk DISTANCE + 1 meters."  This works, but requires that you know how many meters away the door is before you even start walking.  What if we don't?

How about we just do a little at a time and see how it goes?  "Walk 2 meters.  If you're out of the doorway, stop.  Otherwise, repeat this function." If the doorway is 3 meters in front of us, we'll walk 2, check if we're out (we're not), and repeat.  We'll walk 2, check if we're out (we are), and stop.  This function is recursive; it refers back to itself.

Why is it beautiful?

The beauty of recursion, for me, is that you can build infinitely large or complex tasks or structures using one or two tiny, simple parts.  I'll give an example of this because I think it's the easiest way for me to show it.

Imagine that you only know three things: how to add 1 to a number, how to subtract 1 from a number, and what 0 means.  Now imagine that you want to be able to add any two (positive) numbers (integers) together.  How could you do this?  Here's a function:

function add(x, y):
    if y = 0:
        return x
    otherwise:
        return add( x+1, y-1 )

Don't let the new notation scare you; it's the same kind of function as always.  In English, it says "This function is named 'add' and needs two parameters (x and y).  Step one is to see if y is 0.  If it is, the result is x.  If it's not, then the result is whatever you get from performing this function with the following parameters: x+1, y-1."

It might not be immediately obvious that this will actually perform the addition we're used to, so let's try it.  First, let's try adding 1 + 0.  In that case, we check if y is 0.  It is, so the result is x, or 1.  So far, so good.

Let's try 2 + 1.  Is y equal to 0?  No, so we need to call add with some new parameters.  2+1 is 3, 1-0 is 0, so the result is now whatever we get from add(3, 0).  We start following the instructions of add.  Does y equal 0?  Yes, so the result is 3, which is what we expect.

One more for good measure: 9 + 2.  First we check if y is 0.  It's not, so the result is then add(9+1, 2-1), or add(10,1).  Start over.  Is y zero?  Nope, so the result is now add(11, 0).  Start over.  Is y zero?  Yes, so the result is x, or 11.

So What?

This will work for any x and any y (as long as they're both positive, negative numbers make it just a tiny bit more complicated).  This might not seem impressive, but think about what we've done.  We've created something that can add any two numbers together.  There are an infinite amount of numbers.  This tiny function that we've made from the simplest of parts can generate more results than there are people on this planet.  Or atoms in the universe.  To me, this is amazing.  Not "magic trick" amazing or "miracle" amazing but "I can understand and create something that can describe more knowledge than the whole of humanity couldn't hope to describe in a thousand lifetimes" amazing.  

Computer Science (and math in general) is full of this kind of beauty.  I've tried to find parallels in my other interests; the closest I've found is the photograph Pale Blue Dot.  It's a photo taken by the Voyager 1 space probe showing an immense amount of space with the tiniest blue speck in the middle, which is Earth.  

When you view the photograph, it's a mostly black field with a little fleck of blue pigment.  Not terribly complicated or interesting, until you realize that that blue fleck is a representation of the planet that billions of people have lived and died on.  A smidgen of blue ink, in the right context, represents every place a human has ever called "home."

Our addition function might not seems nearly as nifty as this, but it actually represents more than even this photo ever can.  There are a finite number of people on Earth, but the addition function can add more numbers that that.  Even if you count every second in the life of every person that has ever lived on our planet, the addition function can create more numbers than that.

This awes me more than any myth, photograph, story, song, legend or dance ever has.  It's the reason I stayed a Computer Scientist.

This is Corinne, at Flat Iron Cafe.  Flat Iron is a great place to dance but has absolutely awful lighting for taking photographs. The best I could do was sit her in a char next to the soda cooler which has one fluorescent bulb running vertically up the middle of it. It's not pretty or flattering at all; her great skin is turned magenta and purple by the other lights in the room. This is the original photo that I had to work with, and since the lighting is so tough I decided to do more drastic processing than usual.

First, I applied a custom preset I made; I called it "Hot Tone" because it's based off of the "Cold Tone" preset that ships with Lightroom.  Basically it messes with the camera calibration settings to blow the picture almost entirely out of the water, then uses the white balance, vibrance/saturation and HSL controls to pull it back.  It's fun to play around with and presets in Lightroom are so easy to preview (just roll the mouse over it) that they can really help you work a lot faster, or try out effects you might not otherwise spend time trying.  The settings that the preset affects are:

• Camera Calibration - Red Saturation: -92
  
• Camera Calibration - Green Hue: +100 
• Camera Calibration - Green Saturation: +100 
• Camera Calibration - Blue Hue: +71
• Camera Calibration - Blue Saturation: -100 
• Basic - White Balance: 14000 K 
• Basic - Tint: +87 
• Basic - Vibrance: +40 
• Basic - Saturation: -75 
• HSL - Saturation - Red: +94 
• HSL - Saturation - Orange: +65 
• HSL - Saturation - Aqua: -100 
• HSL - Saturation - Blue: -100 
• HSL - Luminance - Red: +38
• HSL - Luminance - Aqua: -98 
• HSL - Luminance - Blue: -98  

Next I adjusted the contrast to give it some depth.  I don't usually use the Contrast control in the basic section unless I'm just rushing through editing some snapshots.  That control does give you contrast but it's pretty hard to work with.  I use the Tone Curve for any real contrast editing which is what I did here.  I brought the shadows and darks down quite a bit (-56 and -40 respectively), took down the lights a little too (-15) and brightened the highlights to help shape the face even more (+23).  Now it's starting to look like a decent picture.

I knew I wanted to emphasize her face, so I added a pretty strong post-crop vignette to tone down her shirt and the background.  Vignettes can be cheesy but I think when they're used appropriately they can add a lot.  Lightroom 2 finally lets you add vignettes after you've cropped an image; in Lightroom 1 you could only add it to the original picture.  If you cropped it after that you'd wind up with a pretty uneven edge, so I'm glad Adobe added this.

  I shot the picture at an aperture of f/1.8 so it's pretty soft all around.  I wanted to emphasize that even more so I used a local adjustment brush to smooth out the skin a bit with some negative clarity.  Local adjustments are another new Lightroom 2 feature; you can now dodge/burn/etc pieces of a photo individually.  I get the feeling I'll almost never be using Photoshop any more.  I'm pretty happy about that.

I wasn't satisfied with her eyes, so my next step was enhancing them.  I dodged the irises, added some clarity around the inner and outer edges, and brightened the whites just a tiny bit.  I think it really makes the eyes pop a lot more than before; they're not just solid black circles any more.  I used another brush to paint in some clarity on her lips to define them a little more too. By the way, the new Auto Mask feature is amazing and will save you tons of time.

To finish everything off I added just a bit more contrast and a subtle, red split tone to just the shadows to put back a little more color that the preset had taken away.  I find myself doing this a lot; I'll make the biggest changes to a photo right away and then tweak it as I go.  By doing that I build up lots of smaller changes that get me closer and closer to my goal.

Anyway, I hope this was useful to someone.  If you have any questions about it or feel like telling me what you think, the comment link is right below! 

As a male I'm usually in the role of leader, though I do try to follow when I get the chance.  I've learned a lot over the years so I'm going to write a few posts about leading, and this is the first.  I'm going to use the traditional pronouns to make things easier to read, but everything applies to both genders in both roles.

Beginning

When a guy is first taught how to swing dance (or blues, or tango; everything I'm saying applies to all three) he's usually taught that his main job is to lead.  This sounds obvious, but it's a lot for a beginner to take in.  He has to learn the footwork and ingrain it into his memory so it becomes automatic, which takes some time.  The next step is learning individual moves: not only how to do them himself but also how to lead a follower to do them at the same time.  It takes coordination and most of all practice.

Leading at this point involves clearly showing the follower where she should go and what she should do.  "Placing the follower's weight" is a concept that's a bit tricky but very useful.  If a leader isn't clear in his leading the follower won't be able to follow him unless she "cheats" and just does what she knows he wants her to do (because she's danced with him before and so knows what he's trying to do).  This falls apart when the leader dances with a new partner.  Without leading and following swing dancing just doesn't work, so leading clearly is the main role of a beginner guy.

Moving On

Let me take a second to explain something I see happen very often with leaders that take classes and progress nicely in their skill.  Once the leader gets the basics down pat and starts learning more moves, there seems to be a tendency to learn things that let him show off. The followers get to really shine in Lindy Hop quite a bit, so it's only natural for the guys to want to measure up and look cool themselves. Unfortunately I think this gets in the way of my next idea.

I think once a leader reaches a point where he's comfortable with the structure of the dance and has a repertoire of moves and vocabulary of movements, his role changes.  His job is no longer "lead."  His role becomes "lead the follower you are dancing with right now."

Every follower is different.  Every single one has a different level of experience, a different style, and a different personality (as it relates to dancing).  If the leader simply leads every dance the same way, those dances are not as good as they could be.  An "advanced" leader leading a beginner follower in a lot of complicated movements she's not capable of following yet turns into a complete mess.  He goes away from the dance feeling bored or frustrated (or worse, arrogant) and she goes away feeling confused, discouraged or angry.  This is not a good thing.

Paying attention to the follow's level is critical.  I'm not saying "only do moves that the follower has learned and can easily follow."  Pushing the follower slightly beyond her comfortable, "automatic" level is wonderful and helps her immensely; but going totally over her head and confusing the hell out of her just so he can show off (to her or others) is obnoxious.  This also works in reverse: followers, please challenge your leaders but be mindful of their skill.

Experience isn't the only difference between followers.  Each follower has her own style that won't always fit perfectly with the leader's personal style.  Adjusting his style to mesh better with hers makes the connection between partners so much better, which makes the dance that much more fun.  This also works both ways.  Followers are generally better at "listening" to their partner because it's their main job; if a lead makes an effort to really listen to the follow and change his leading to incorporate her ideas, personality, style and level it makes an enormous difference.

The point I'm trying to make is that "leading" a follower is not just leading.  It's paying attention to the follower and leading her.

RIT has three main areas of photographic study: fine art, advertising and photojournalism. If you had asked me on the day I got my camera which of those three I thought I'd be most interested in (even though I wasn't majoring in photography) I would have said "photojournalism." When I first started shooting I hated using flash and was mostly only interested in capturing musicians and dancers.

Since then I've learned a lot. I'm pretty comfortable with hot shoe flash and studio lighting. I know how to develop and print black and white film in a darkroom myself. I've discovered how wonderful a certain degree of abstraction can be and I'm slowly but surely getting more artistic as time goes by. I think my answer to the previous question has changed; right now I see myself becoming more and more of a fine art photographer. I certainly don't consider myself good at that yet but it's the path I've started down and plan on continuing. There's something really special about planning, visualizing and not just taking but creating a photograph that I've fallen in love with.

I'm going to be updating my site to reflect this shift. Obviously the layout has changed a bit, but I've also removed the galleries of random photos that I had posted previously. From now on I'll only be displaying photographs with a cohesive idea behind them. I'll still post tons of random haphazard photos to my flickr site, but the galleries on this site will be for showing off what I most enjoy and am really proud of.

In the next few days I'll be revamping my site and posting a lot more.

A lot of people seem to think that getting a fancy, expensive digital camera will automatically make you take wonderful photos. This is very far from the truth. You wouldn't expect to be able to create wonderful drawings just by picking up expensive pencils, would you? Ken Rockwell does a great job of tackling this question in his article Your Camera Doesn't Matter which sums up everything nicely.

One thing in the article that's easy to miss is where he explains why good artists do use the fancy, expensive tools if they really don't matter. I'll quote the section here:

So why do the artists whose works you admire tend to use fancy, expensive tools if the quality of the work is the same? Simple:

1.) Good tools just get out of the way and make it easier to get the results you want. Lesser tools may take more work.
2.) They add durability for people who use these tools hard all day, every day.
3.) Advanced users may find some of the minor extra features convenient. These conveniences make the photographer's life easier, but they don't make the photos any better.
4.) Hey, there's nothing wrong with the best tools, and if you have the money to blow why not? Just don't ever start thinking that the fancy tools are what created the work.

I think this is an important point. The camera does matter, just not in the way most people think it does. You don't see carpenters hammering nails into boards with rocks. Sure, a rock would work but instead they use nice hammers that are durable and comfortable to use because it lets them worry less about how to work with the tools and more about what they're creating. I think the same can be said for cameras and photographers.

So back to the K20D; why did I upgrade? Yes, the increased resolution is nice (14 megapixels instead of 6). I'll be able to print much larger photos with more detail now than I used to be able to, which is good.

Resolution isn't the only reason though. The K20D is weathersealed which means that I can take pictures in the rain now. There have been a few times in the past where I could have really used a weatherproof camera so I'm excited about finally having one. It's also much more durable in general. I'm generally pretty careful with my photo equipment but more durability never hurts.

Another big reason I upgraded is having more controls on the outside of the camera as opposed to having to dig through menus to find them. Less fiddling with the camera menus means more time actually taking photos and less missed opportunities, which are definitely good things.

If there's one area I'm not careful enough in with respect to treating my equipment gently, it's dust. I change lenses fairly often and I'm not always super careful about avoiding sensor dust. The K20D's dust reduction system means I'll have to clean the sensor a lot less which will be really nice.

To summarize: a nice camera won't magically make nice pictures but it will get out of the way and let you create great photos if you're capable of it. That's the part I'm still working on.