The Daily Parker

Why didn't I get Adobe Lightroom earlier? Even its basic photo-editing tools dramatically improve photos (or at least get them back to where they should be). I'm going to re-scan this one at higher resolution, after carefully dusting it, and with the appropriate filter, but for the moment, I think I've gotten pretty close to what the original Kodachrome image looked like:

This one came close, but not quite:

Both: Public Garden, Boston, 10 May 1986. Kodachrome 64. Exposure unrecorded.

Another of my favorite Bills:

I want to try out some new techniques on this and a few other shots I took tonight, but I won't have time until the weekend. For now, here's a gratuitous statue photo.

This morning we had weather about as perfect as a human could hope for, 26°C and sunny by the lake, with a gentle breeze out of the southwest. I hopped on my bike for an actual workout, complete with heart-rate monitor, for the first time in a couple of years, then came back, grabbed my camera, and walked the dog. Some results:

ISO 400, f/6.3, 1/640, 225mm

ISO 100, f/5.6, 1/500, 55mm

As I continue to evaluate Adobe Lightroom, I'm trying to figure out how best to use it. Since about 2000, I've kept the raw copies of everything my various digital cameras produced, making copies of almost all of them with names and metadata. Lightroom obviates that step, because it saves metadata and editing steps separately from the image files, preserving the raw camera output. If I need a JPEG version of a photo, I can simply export it—edited, and with all the relevant metadata.

This changes everything. I'm just now sure how. For my first step, I've set my camera to shoot only photos in raw (Canon .cr2) format. They're a lot bigger—25 MB vs. 5 MB for JPEG—but so what? Hard drive space is around $100 per terabyte, which works out to 0.24¢ per photo. (It adds up, though; I took 4,635 photos and videos in 2010 and I've taken 2,351 so far this year, and their combined 24.1 GB use up a whopping $2.35 of hard drive space.)

As I write this, dark clouds have rolled in and radar shows thunderstorms just about on top of us. I feel like I took good advantage of the excellent weather. I even remembered sunscreen today, and had the foresight yesterday to install my air conditioners. Now I've got some real work to do, though. Sounds like it's time to go to my Remote Office...

(This is the 2,500th post on The Daily Parker. And now back to our current thread, already in progress.)

Version 1, pretty much as it came out of the camera:

Version 2, processed from the raw camera file:

8 April 2011, 18:16 BST, 1/1000 f/5.6, ISO 100

Subtle differences—but noticeable.

OK, walk the dog, thence bed. I feel like I learned a lot today, including that I have to learn a lot more.

I'm continuing to play with Adobe Lightroom, and it turns out I've been doing a lot wrong for five years (i.e., since I first started shooting with a digital SLR). It looks like I'm going to shoot a lot more raw photos, because they allow modern software (like Lightbox and Photoshop) a lot more control over the final image.

And, of course, I discovered this using Parker as a subject. The results don't completely suck:

50mm, 1/60 at f/2.0, ISO 3200.

50mm, 1/15 at f/1.8, ISO 3200.

A few days ago I experimented with photo processing to try out a technique a photographer suggested. I neglected the most obvious transformation of the photo in question:

I've also downloaded Adobe Photoshop Lightroom, though I may want to go full-bore Photoshop in a couple of weeks. Lightroom looks like a fabulous way to organize photos, which would be helpful as I've got north of 25,000 right now and that doesn't include about 170 rolls of negatives I've yet to scan. It has some basic editing tools—nowhere near as powerful as Photoshop—and I'm just getting used to them.

I still won't get any photography books for my Kindle.

The 30-park geas can resume now that I'm done with school. Here's my progress so far:

[1] vs. Cubs
[2] Renamed Minute Maid Park in 2004
[3] I've decided not to count parks that were rebuilt after I started this geas in 2008.
[4] Shea demolished in 2009; Citi Field opened 13 April 2009

Last edited: 20 April 2012. This page replaces the original page started in 2008.

Via Bruce Schneier, evidence that the Centers for Disease Control have a sense of humor:

There are all kinds of emergencies out there that we can prepare for. Take a zombie apocalypse for example. That’s right, I said z-o-m-b-i-e a-p-o-c-a-l-y-p-s-e. You may laugh now, but when it happens you’ll be happy you read this, and hey, maybe you’ll even learn a thing or two about how to prepare for a real emergency.

This is a lot more entertaining than Internet Information Services configuration, no?

Problem: I have multiple websites on a Windows 2008 server (using IIS7), and I need to enable SSL (i.e., https:// connections) on more than one of them.

Generally, secure websites get their security certificates from trusted providers. Most modern browsers verify that the third-party certificate came from the purported vendor and are attached to the purported website, and give you a nice warm feeling when your address bar turns green. (I have used Comodo most of the time, though because of some experiences helping a local small business renew their certificate recently, I might switch.)

The way SSL works, however, you don't actually need third-party verification to keep the conversation secure. Many servers issue their own SSL certificates, which secure the traffic between the browser and the server regardless of which server or who issued the certificate. In other words, a self-signed SSL certificate will keep the conversation private without guaranteeing that you are talking to the person you think you're talking to. If you can confirm the identity of the server through other means, and then trust the certificate, then you're golden.

Because only I or my subcontractors will ever need to visit these administrative sites within Inner Drive Technology's Worldwide Data Center that I want to secure, self-signed certificates are perfectly appropriate. Under previous versions of IIS, not only did this cause enormous headaches, but also IIS would only support one and only one secure site per server instance. That really sucks if you have, as Inner Drive does, about 30 sites on the same server.

IIS 7 makes this so simple I almost cried with joy.

Step one: Issue a self-signed certificate

In IIS7, go to your server node, and open up the Server Certificates feature:

In the Actions list, click on "Create Self-Signed Certificate..." to bring up the dialog box. Enter a starred domain name for the certificate:

Once you click OK, you've got a self-signed certificate that IIS can find.

Step two: Open the right dialog box

Expand the list of sites in the Connections (left) pane, and click on the site you want to assign. In the Actions (right) pane, click on "Bindings..." This brings up the Bindings dialog box. Now click Add, to bring up the Add Binding dialog box:

Step three: Click OK

Drop the Type box down to "https." Enter the specific host name for the site, and choose your starred certificate:

That's it. Seriously. And anyone who's ever dealt with this configuration crap before will understand my teary-eyed joy.

Generally, I prefer to learn new things by reading first, then doing. I mentioned Wednesday that I've grown dissatisfied with my photography skills, so naturally, I'll go first to Amazon. You know: read about a technique, try it out, post the results online, rinse and repeat.

So it seems somewhat odd to me that most of Amazon's top-rated books on photography—like this one on Photoshop—have Kindle editions that cost almost as much. Because nothing will help someone understand how to do advanced photo editing than 10 cm, 18 dpi halftones, right? Even stranger: the example I just cited has a companion DVD, which I assume does not come with the Kindle version. That, to me, puts the F in WTF.