Friday, October 31, 2008

Happy Halloween Hint for the November Monthly Special

Cezanne Pyramid of Skulls
1901 (130 Kb); Oil on canvas, 37 x 45.5 cm (14 5/8 x 17 7/8"); Private collection; Venturi no. 753 as seen on Webmuseum.

The month of "dots" is ending. November is upon us. Look out for a new Monthly Special. Can you guess what it is?

Aside from a new theme for the month, look for more ideas for the holidays.

Wednesday, October 29, 2008

Wrapping up our October "Dots" theme: another IKEA idea

This simple IKEA-based idea is in honor of the October Monthly Special. I haven't had time to do this one myself, but I wanted to suggest it before the month ends. If you have done a dots grid and want to recreate it as an art statement/storage device, just get yourself enough grundtal round magnetic containers ($4.99 for a pack of 3) to recreate your dot grid....
and put a photo (or colored paper) under each glass cover to form your "dots". You can crop your photos in Photoshop (the web site claims the diameter is 3 3/4", but it wouldn't hurt to measure) or you can print 4x6 photos and crop by hand.

Since the containers are magnetic, you can arrange them on a "Spontan"...
or a "Bits"...
or your refrigerator or anything else a magnet sticks to.

Your magnetic dot grid will be both attractive and practical (good qualities in humans and objects alike, if you ask me). For a craft room or an office, it would make a good piece of storage art. The style could range from modern minimalism to glam and girly depending on your design.

And for Christmas, you can make a very bold advent calendar (you would have to come up with a larger surface), you could alter 25 grundtal containers. Too many? Maybe a "12 days of Christmas" thing might be better.

Other variations and ideas? Please add your comments.

Meanwhile, there's still time to do the October Special.

Monday, October 27, 2008

IKEA hack: photo wall sconce

As part of my series of photo gift ideas, I thought I would do some IKEA hacks (all related to photos, of course). And so, impulsively deviating from my drive to work, I headed to IKEA. How can I add a photo to this? was my guiding thought as I browsed through aisles of Gruntals, Kaxigs, Porflyts, and Gyllans. The names reminded me of a passage from David Sedaris' Me Talk Pretty One Day:

"If you have not meimslsxp or lgpdmurct by this time, then you should not be in this room. Has everyone apzkiubjxow? Everyone? Good, we shall begin."

In fact, now that I think of it, I may have stumbled upon a breathalyzer test replacement for arrest-hungry cops (feel free to try this one, Tim):

Name three IKEA accessories

The Hack
This project is quick, inexpensive, and can be modified for any interest or age group.

1. buy a TASSA NATT wall lamp for a whopping $4.99
2. Unscrew the lamp as you would to put in a bulb (which you will have to do anyway because it's not included).

3. Remove the paper with the image of strange little frogs French kissing. (Maybe that kind of filth flies in Sweden, but here in America we've got morals. Frogs shooting each other would be OK, but kissing? I think not.)

4. Now measure the paper. Here. Let me do it for you. It's 10 inches wide by 6.75 inches tall (actually, it's 1/16th of an inch taller, but sometimes I like to be sloppy—like the frogs).

5. Find a photo that will replace the frogs. The possibilities are endless. Here are three I chose:
A "danger of death" sign painted on a wall in Paris.

A photo of graffiti that will look good in my sons' room.

A photo of a hedge that my sister will probably like.

6. Print the photo on lightweight matte or translucent paper. You can either pre- or post-crop the photo. Either way, you will have to trim (from 8x10 or 8.5 x 11) since no one sells 6.75 x 10 inch paper.

7. If you fold the paper over the edges and let the slots hold it in (as in the frog picture), you won't even need glue or tape. Just put it back together and you've got a stylish new sconce.

Here are the results (the graffiti version now resides in my boys' room and looks great against their orange walls):
If I had more disposable income I would buy a bunch of these for a more dramatic effect.
  • Stack them tall (wired together with the grates) and use photos to make a quaking aspen (or some other tree, or obelisk, etc.) light tower.
  • As a Christmas or Halloween decoration, you could use photos of candy and line up "jars" of luminescent goodies.
  • Do photos of letters and spell something with one letter per TASSA NATT (doesn't it just roll off the tongue?)
You get the idea. Keep checking back for more photo gift/decorating ideas for the holiday season.

Saturday, October 25, 2008

Pointillism and a rant about how poorly filters imitate art

Seurat's most famous painting, "Un dimanche après-midi à l'île de la grande jatte,"as seen on Wikipedia.

Since we're focusing on "dots" this month, I thought a post about pointillism might be in order. But somewhere in my twisted thought process this led to an internal rant about how much I hate the "artistic" filters in Photoshop. I try to keep things positive on my blog, but I should clarify (à la Lemony Snicket) that although I generally intend "positive" to convey a sense of upbeat and encouraging affirmation, unfortunately, I must tell you that "positive" is a word which here means "absolutely certain," as in I am absolutely certain that many of the filters in Photoshop are relics from computing's less sophisticated days which should have been discarded along with floppy disks or bundled into some cheap art program for "kidz" (oh yes, that's kids with a Z).

Before I explain myself, let's get back to pointillism. Look at this lovely detail of a Seurat painting found on Wikipedia (you can click to make it larger):
Like the impressionists before him, Seurat studied how the eye perceives color. But rather than blend, blur, and smudge pure color with brush strokes, Seurat preferred optical blending. In other words, he believed that side-by-side dots (also called stippling) of blue and yellow, for example, could make a more vibrant green than mixed paint. Many definitions of pointillism (Wiki included) claim that Seurat's dots were made only of primary colors, but that is only the case if you look at the broadest definition of "primary color," one that might include red, green, blue, orange, violet, magenta (incidentally, Seurat grew up on "Magenta Street"), cyan, and yellow.

To argue that CMYK printing, RGB projectors, CRT monitors, and the like are essentially pointillist in nature makes a lot of sense. If you blow up a photo in a magazine (or look closely at a billboard), for example, you will see dots of color that your eye naturally blends at the intended viewing distance. Pixels are also dots of color that our eye blends.

Why then, is Photoshop so bad at "pointillizing" this photo of my friend, Ash? It's no Seurat, that's for sure.
Some computer science experts claim to have found a better way, but I'm not ready to pay $25 to read the article. So here are two (of many) reasons why I think that Photoshop is no replacement for Seurat.

1. Its images are made of square pixels, not hand-painted dots. Look at this detail of "pointillized" Ash and just try to convince me that it looks like dots made from a paint brush:
and 2. human subjectivity plays such an important role in art that user-end control is paramount—expert algorithms notwithstanding. The "pointillize" filter limits your input to one thing: select cell size.

I am picking on "pointillize," but I could rant just as easily about the filters that supposedly create charcoal sketches, watercolor paintings, and so on. Maybe there is a challenge in all of this. Maybe I should dedicate a month to discovering how to make something beautiful from those filters. But I don't think I will undertake that challenge any time soon. The thought of creating an impressionist painting or a charcoal sketch in Photoshop makes me think of those hideous services that will turn your photo into "art." For the love of taste, don't do it—or at least not with the same amount of irony as someone must have to display paint-by-numbers art. Photos ARE art. If you want a watercolor portrait of your child, hire an artist. If you want to use an "artistic" filter, just be sure to add your own artistic vision to your work. One-step filter conversion is not artistry.

If I were to try to make some "take-out photo" lessons out of this they might include the following:
  • There is nothing wrong with imitation. If I thought so, I wouldn't have done the Barbara Kruger tutorial, and wouldn't have told you to look for design inspiration on Criterion dvd covers. Just don't let your computer make all of the choices.
  • Computer programs that imitate humans need to imitate human "imperfections"
  • (I'm speculating here) The only way to make those bad artistic filters into something good just may be finding a use for them not indicated by their name (i.e. don't do a charcoal sketch with the charcoal sketch filter; do something other than a pointillist "painting" with "pointillize," etc.)

Wednesday, October 22, 2008

Retro pop art dot photo

Another idea for the October Monthly Special—I made this retro dot art by using a free brush set that my son Max showed me. In the not-too-distant future I will do an entire month about brushes. But for now, if you want to experiment with brushes, here's the quickie explanation: Download a brush set. Drag it into Photoshop-->Presets-->Brushes. Select the brush tool when in Photoshop and click on the little arrow in the brush pull-down menu (see image below) and select "load brushes," at which point you can locate and "replace current brush set" with the new one (and yes, many brush sets—including the one I used here— work in Elements).

I won't do an entire tutorial on this, but here's a basic outline of how I got this look:
  • I selected the brush from the set in the link above and put in on its maximum size before stamping it on the blank document.
  • I found a photo that had the right feel for this kind of treatment (it happens to be of my sister-in-law), converted it to black and white, and then added a toned gradient.
  • Using Select-->Color Range, I selected the circle and square pattern so I could copy it (minus the white) onto its own layer.
  • I brought the photo into my "dot" document, resized it as needed, and used the same clipping mask technique you saw in this month's special to make the photo come through.
  • Finally, on a new layer I selected some rectangular areas, filled them with various colors, and then lowered the opacity of that layer to create an overlay effect.
I realize that the above explanation is not going to help you if you're new to Photoshop, but I'm hoping it will give you more "dot" ideas for this month. The main lesson here is to experiment using some of the techniques you've learned.

Monday, October 20, 2008

Personality Pie Chart 2

I'm liking the personality pie chart, so I wanted to do another one. I think it would make a fun series. I did a bleach bypass look on this one instead of black and white.

The photo is of a friend/struggling actor who lives in Paris (and if you're looking at this, Nico, let's just say I boiled you down to five attributes for the sake of art). We did the shoot in the Jardin du Luxembourg until a park official told us to stop. "But I'm just taking photos of a friend," I told him. "Not with that camera," he replied. This was one of those times having a pro camera gets you fewer privileges.

Saturday, October 18, 2008

Variation on the October Monthly Special: Personality Pie Chart

Click to enlarge. Then you can see the words as well as my imperfect pie chart (made with the help of the pen tool—something I never use).

If you're not into the dot grid thing, but you still want to do the October Monthly Special...get creative and do anything involving a circle or circles.

Explore circular patterns in any way. I was flipping through a book with pie charts and suddenly got the idea for this. What do you think? I'm not sure everyone would love to have their personality dissected, but Lucas is a good sport.

Check out Max's second submission, for an example of something experimental. And be sure to look at the projects other people have posted so far this month. Their work may inspire you.

October is more than halfway over, hope to seen your own experiments soon.

Tuesday, October 14, 2008

A "dots" Christmas card tutorial

If you don't want to create templates and clipping masks, if you prefer eyeballing it, then this is for you. Beware, however, that a more free-form style of creation can lead to imperfections (or shall we call them "variations," as in "variations in fabric should not be considered imperfections in manufacturing..."?).

That said, keep in mind the "galloping horse" rule. We learned the rule in our household back in our quilting days. It goes like this: If you don't notice the mistakes when riding by on a galloping horse, then it's good.

Now, on to the tutorial.
If you want to do a grid of dots, why not incorporate a few round things? Pumpkins come to mind for a Halloween card. But we have a hard enough time sending out Christmas cards, so I will use Christmas ornaments. Here's how I did it:

1. Create a new document that is 5x5 inches at 320 dpi (this is because 5x5 is a good card size).

2. Get into those grid preferences and adjust the settings for nine equal squares (just like we did in the previous "dots" tutorial).
I set the main divisions at 1.66666, but I put the subdivisions at on 16 (as opposed to the 32 on the previous tutorial—galloping horse rules, right?).
3. So now that you have a nice little blank document with a grid (if you don't see the grid, turn it on in the menu: View--> Grid), you need two things: those ornaments, and some round pictures. First, the ornaments...I really didn't feel like having an ornament photo session, so I bought a photo from iStock Photo (this one). Stock photography is a great way to get good quality images quickly. For certain purposes, the price of a stock photo is worth the time it would take to do your own. Just make sure to buy an image size big enough for your needs. As for your own family photos, well, unless you want to construct the perfect artificial family (which, for the right person, might be very entertaining), you had better take your own portraits.

4. Now you need to separate those ornaments and drag the ones you want onto your card-in-progress. The advantage of stock photography is that you can get clean photos on white (read: extraction-friendly) backgrounds. Every Photoshop user has their favorite method of extracting an image. In this case, here's what I did:

Maybe it's because Halloween is near, but I chose the "Magic Wand" tool (W). When it is selected, you can put a number between "0" and "255" in the "tolerance" box. Because I am a reasonably tolerant person, I set it to "40" for starters. But, no! That just wasn't tolerant enough. When I was hoping for a completely outlined ornament, I got something like this:

It turns out, if I had put "200" in the tolerance box as seen below...
...I would have been able to select it in one click. So the lesson here is to play around with tolerance numbers. Oh, and just so you know, what those numbers really mean is how closely the magic wand will match colors (i.e. larger number=wider selection).

One more thing before we move on: if you make a selection that isn't wide enough (such as in the image above), you can keep adding to it until you get what you want. Just click the "add to selection" box in the magic wand menu bar (it's the second from the left) and click in areas that aren't selected until you finally get something like this:
5. Now it's time to get that selection on its own layer and drag it onto your card. Hit Command-J (PC: Ctrl-J) and you have a new layer with nothing on it but that one ornament. Now you can use the move tool (V) to drag it onto your card.

6. Congratulations! Your card is looking much more festive already. But are you happy with the size of the ornament? If not, hit Command-T (Ctrl-T) to bring up a transform box around it (see below).
Just grab one of those corner handles (while holding shift to keep the aspect ratio) and drag out or in to change the size. Hit "Return" to apply the transformation. You will be using this step many times before the card is done.

7. So let's say you have got one ornament to just the right size. What about the others? First, you need to extract them one by one (as seen in step 4), drag them onto the card and into the same approximate space as the already-transformed ornament (step 5), and then—because we are not going to be too picky—put the new ornament layer below the perfectly sized one and use it as a guide:
Make sure the appropriate layer is selected, and then bring up that transform box. If you line up the bottom and then drag the top corners down (a little from one side and then a little from the other), you should be able to get pretty close. You can move it around at any time by clicking in the box (not the handles) and dragging. Once you think you've got it right, apply the transformation. Then, to double check, change the layer order so the model ornament is on top. If it still looks good you've done it. If not, just transform it some more until you've got it.

8. Repeat the above process until you have all the ornaments you want.

9. Time for the photos. Open a photo. Select the elliptical marquee tool (seen for Elements at right, and CS3 at left). Now click and drag in your photo while holding the Shift key (this keeps it circular) until you have the area you would like to crop. Hit Command-J (Ctrl-J) to copy your selection onto its own layer. You can now move it (V) onto your card and transform it as desired. I used the ornament for a general guide on the first one:
But after that, I used the first photo as a guide for the others as described in step 7. If you look at my card, you will notice that I decided to make the photos larger than the ornaments (not that you have to do the same).

10. One more thing you may notice is that I removed those little loops from the top of the ornaments. They were getting in the way. I simply selected the rectangular marquee tool (right up there sharing menu space with the elliptical one), selected the offensive loop, and then cut it out (Mac: Command-X, PC: Ctrl-X or choose Edit-->Cut).
Make sure you are on the correct layer when you do this or you will cut a hole in another image.

11. Finally, you are free to arrange the photos and ornaments as you like. Use the subdivisions in the grid to help you keep things in line. When you are happy (or at least "galloping horse" happy), you can flatten your card, print it, and glue it on to card stock and impress all your friends.

Sunday, October 12, 2008

October "Dots" grid tutorial is completely revised

Had trouble with the dots tutorial? Did your keen eye notice that not all the dots were the same size? Max and Michelle did. Well, I have fixed that problem and have streamlined the entire process so that it is easier than ever. excuses. Do the new and improved "October Monthly Special" and post it (as described in the FAQ). Please.

One more quick note. The theme is dots and the tutorial is a grid of dots, but any dots-themed submissions are more than welcomed. In other words, you can post a photo that has dots, or adapt the dot tutorial to something other than a grid.

Introducing the Christmas cards and gifts series

If you're like me, you wait until the very last minute to do your Christmas shopping. But maybe that will change this year (for both of us) with my October through December series of ideas for Christmas cards and gift ideas—all of which relate to photography, of course. Within the next couple of days I will post a card idea that relates to the "dots" theme, but today's idea is much more straightforward.

This is the photo we used for a card two years ago. That same year, I did one for a client that spelled NOEL using three kids and their dog (not an easy task). For the client, I did a grid of four in black and white with alternating red and green letters. For our own card, we stuck with all black and white.

The recipe
1. Get letters from a craft store and paint them

2. Take individual photos of each child with their letter. I used natural side lighting from a window. I would also recommend using a black or dark background. I used a dark wood screen and then darkened the background even more in Photoshop. Had I just used some black fabric, I could have saved a lot of time.

Note: Make room for serendipity. Eva would not cooperate. She began to cry after only a few photos and she wouldn't hold her letter. For a few seconds, she put her letter over her head and I took the photo that you see here. Her head became to "O" and the letter became a sort of halo. What was frustrating at the time of the shoot became, in my opinion, the best part of the card.

3. Do the Photoshop bit. We decided to make the photo into strips 2x6 inches long that would eventually end up on larger rectangles of prepared cardstock. The math was easy. I cropped each photo to a 2x2 inch square. I then created a new document (4x6 at 320 dpi) and dragged each square onto the appropriate place in the new document. Since my document was 4x6, I could fit two 2x6 strips on each photo (which made this an inexpensive project). Since I didn't feel like doing the dragging part two times, I deselected the backround layer (leaving the layers with the three kids visible) and selected "merge visible" from the layers pull-down menu. Now I could just copy (Mac: Command-J or PC: Ctrl-J) that layer and drag it down with the move tool (V) its spot. I flattened the whole thing and then got them printed.

4. Assemble the cards. This was Michelle's part of the job, so all I can tell you is that we used white cardstock and printed a red border inside of which we (and by "we" I mean "she") pasted the photo strip. We kept it simple and were very happy with the results. I can't show you the finished product because—alas!—we mailed all of them that year and have only a crumpled photo on the fridge to show for it.

You can easily take this basic idea and do your own variations. Grids, vertical strips, etc. For me, the most important part is not the letters, but rather the plan to use individual photos instead of a standard group portrait. The nice thing about putting individual photos into a group later is that you still get to see everyone, but each person can look their best.

Wednesday, October 8, 2008

Two ways of selecting color in Photoshop: 1.Hex or RGB, and 2. the eydropper tool

Frustrated with color selections? Not seeing eye-to-eye with Photoshop? Consider this free counseling. This simple tutorial will show you two of the most common ways to select a color in Photoshop. If you are like many beginning Photoshop users, you probably do one of two things:

1. You double click the foreground color box (or text color box for text) and then start randomly moving the slider and clicking in the colored circle until you get some approximation of the color you wanted, or
2. You use the eyedropper tool (I) and sample something in an image, often surprised or frustrated that the color you sample looks nothing like what you are seeing.

You can eliminate a lot of the guess work if you take a few minutes to dialogue with Photoshop.

When you double click on the foreground color box (pictured at right), you will see the color dialogue box (Note that if you are changing text color, you will find the color box in the top menu bar). If you are perfectly happy sliding and clicking (or scrolling aimlessly through Pantone colors), then go about your sloppy ways. If, on the other hand, you want a specific color—perhaps one that you found in a book like the Color Index or on a site like Colourlovers—you can get that exact color by typing it in the box for the HEX or the RGB values.

RGB and Hex represent the values of red, green, and blue, but in two different ways. For RGB, you will enter a number separately for red, green, and blue. For Hex, you will enter a series of letters and/or numbers (after a number sign) that represent the value of red, green, and blue in pairs. In other words, they represent RRGGBB. But let's not get theoretical.

Look the image below (you can click it to enlarge):
On the left, I have indicated the slide-and-click method of just eyeballing it. On the right, I show you where to enter the Hex or RGB numbers (note: It's one or the other. No need to do both.) You might wonder why I don't mention the boxes for Lab or CMYK. The reason is that only a happy minority knows how to use Lab (a color space so rich and complex that it even includes imaginary colors), and CMYK tends to be used more in the domain of professional printing.

The eyedropper tool
But what if the exact color you want is in your photo? That's when you need the eyedropper tool. The keyboard shortcut for eyedropper tool is "I." With the tool selected, you can click anywhere in an image and sample that color. The problem comes when the eyedropper says you're seeing red and you are sure that your eye is seeing pink. Or maybe a seemingly pessimistic Photoshop sees an ugly brown in a rose petal where you see a sunny yellow. When that happens, some people tend to click and click, getting angrier by the minute that Photoshop can't see the right color. The fact is, when Photoshop sees one thing and you see another, you are both right. Photoshop is merely averaging color values differently than your eye.

So let's learn to make your eye and Photoshop get along a little better.
Let's use this picture of roses:
I want to sample the color from the point you see in the following image:
The most important thing to realize is that the point sample typically is often set to 3x3 pixels. If your image is big, 3x3 pixels may be more precise than what you see (and yes, you can even sample a single pixel, but I doubt you will ever need to). At 3x3, you might pick up a small variation in color that doesn't look like the color you want. To change that, you can select from a pull-down menu in the eyedropper tool menu bar at the top:
Unfortunately for Elements users, you only have two choices (3x3 and 5x5)—and although those numbers might not differ greatly, the greater sizes in Photoshop will. Here is some text sampled from the exact same point on the rose, but using different pixel sizes:
Note the variation in color. Now try this on your own photo. You will soon get a better feeling for the sample size that best matches your vision. You'll learn to appreciate Photoshop, and it will respond to your wishes.

Tuesday, October 7, 2008

Take-out photo is now open at Etsy

Remember the ABCs of Paris project? Well, although I would still like a book deal, I have decided to open an Etsy store with them in the meantime. All 26 letters are currently posted in 5x5 inch format so people can choose their favorite letter(s) and maybe even spell out a word. Coming soon to the store: alternate versions of each letter for those who want to spell something with more than one of the same letter but don't want to repeat the same image. Also, down the road, look for a poster with the complete alphabet. I will put the link in my sidebar.

Sunday, October 5, 2008

Selecting a color palette with the help of Colourlovers

A sample of the colourlovers pack of Moo minicards. So much cooler than stealing Martha Stewart paint chips.

If you plan on doing the "Dots" October Monthly Special (and I hope you do), you will find yourself thinking about color. Not that you have to reserve some of the dots for color, but either way you will be making color choices. If you use color photos and the palette clashes from one image to the next, the end result may seem a bit "off." And you don't have to be a designer to sense the discord.

But how confident are you at choosing colors? Unless you're like my color-genius mother-in-law (who seems to have perfect recall of multiple paint decks and fabric swatches in the same way that some people have perfect pitch), you may find yourself stumped when faced with a color choice. And that's where Colourlovers comes in...

screen grab from is an amazingly rich site, complete with a blog that I have subscribed to ever since I finally got with the whole blogosphere thing. I especially love the palettes. Readers have submitted well over a half million palettes (you heard me right, more than 563,266 and counting) for your consideration.

The inventive color names are fun in their own right. I have always found the names of my wife's beauty products amusing (e.g. "I'm not really a waitress" nail polish), but with more than 500,000 palettes, the colourlovers site has got quirky names way beyond your standard beauty store fare. Here's a sampling:
You get the idea.

When you find a palette you like, you can click on it and see the HEX and RGB values for each color. Don't know what to do with those numbers? Check back for my next post: a "Starter" about sampling and changing colors.

Thanks to Lindsay, for pointing out the Adobe Kuler site/app. Check out the video intro about it.