What I really want out of my job, aside from the pay check, the potential for advancement, and a regular challenge for my problem solving skills, is to be able to use a Mac as my primary machine. Being a Windows developer, though, that’s not very likely.
For awhile I had a MacBook Pro, which ran Windows via Parallels Workstation like a champ, and let me have the best of both worlds. Alas, that machine is no longer available to me. Replacing it would be too costly, and my needs dictate that I’d require a high-end Mac to be able to keep up with me, so a cheap little Mac Mini wouldn’t do the job.

It occurred to me then, that maybe I could try (again) to build myself a Hackintosh, using the latest hacked release of OSX86 10.5. I do run a Mac VM image on my PC, that lets me use my favorite mail client, but it’d be nice if I could reverse the situation. I figured I’d be more than happy to purchase an additional Leopard license, to make my use (mostly) legit. So I went out and got a new hard drive to dedicate to the experiment. Here are my findings…

At work I have a Dell Optiplex 745. A nice machine, well equipped with a Core2Duo and 4GB of RAM. I checked the OSX86 Hardware compatibility wiki and verified that most of my hardware was on the list with either a “working” or “working with patch” status. I figured at a minimum I would need network and video card drivers on a stable OS capable of running VMWare Fusion for the project to be worthwhile. Things looked promising, so I began the download.

OSX86 Hacked install discs can be found on PirateBay and other fine Torrent websites. Again the legality of it is questionable. While you’re required to have an OS license for each machine, the OS X EULA does state that it can only be run on Apple branded hardware. But that sounds more like a challenge than a rule to me.
Downloading should be fairly easy — Kalyway seems to have the best builds right now — but I did find a couple suggestions on community websites. Use WinRAR to unpack the archive, as there have been some problems with other .Zip programs not opening the file correctly. The ISO itself should be burned on a Mac, or on a PC using TransMac, so as not to corrupt the files with any traces of Windows… or something. Burn it slowly! I’m not sure why people recommend these things, but I know I tried this with a 10.4.6 hack install and did it the easy way, and ended up with a coaster. This time I was super careful.

The install disc booted right off the bat for me, and following the suggestions on the Wiki I formatted my new drive as HFS Journaled, using the GUID partition scheme. I did a Vanilla install, adding the SSE3 extensions and ATI drivers for the 1×00 series cards (my Dell has a X1300 Pro)
The install completed successfully and I was ecstatic when I rebooted (don’t forget to remove the install DVD) and saw the Apple logo on the gray screen… then crushed when it promptly kernel panicked.

I rebooted with the -v switch so I could watch what was happening, and eventually realised it was the video card drivers causing the problem. I rebooted off the install DVD, formatted again and installed without the NATIT ATI kexts… only to have it kernel panic on another driver.
Finally, I tried it all again, this time with the Vanilla Kernel, SSE3 extensions, the ACPI fix, no video card drivers, and using the MBR partition scheme. And it STILL kernel panicked at loading drivers.

This time I read a little more online and figured out that I could boot into the install DVD, get to a terminal window, delete the NATIT kexts, and reboot. I tried that, and low and behold, I was into the “Welcome to OS X” wizard!

I had the well-documented problem with it getting caught in a loop after the “Import from another Mac” step, and again had to boot with the install DVD, drop to a terminal, and poke in a few commands to tell OS X that I’d already set-up my account. None of this would be so bad except booting from the DVD to a terminal window took nearly 20 minutes each time.

Finally I was able to boot into the Finder. It was glorious and looked as beautiful as a Mac should look… circa 2001. Without video card drivers I was stuck at 1024×768 mirrored on both my monitors. I poked around the ‘net a little bit, but decided my priority needed to be network drivers. My Dell has an on-board Broadcom NIC, which seemed pretty common, and although it didn’t work out-of-the-box, there were plenty of forums with solutions for hacking support for my particular DeviceID into the built-in drivers.
Alas, none of them worked. I got as far as getting OS X to recognize the card, but hit the issue with it not knowing its own MAC address. I tried the solution that basically has you telling it the MAC from the terminal at each boot, but that failed. I gave up and bought $30 D-Link card with manufacturer’s official drivers available.

Side note: that card never worked either, but thats because I had forgotten that I’d disabled my PCI slots in the BIOS, intending on turning on all on-board devices one-by-one once I was up and running. Note, that the on-board NIC WAS turned on. There were just no working drivers.

I turned my attention briefly back to the video card, hoping to get at least one thing working properly. I was able to download a bunch of drivers using the Craptop and burn them to CD for use on the Hackintosh. All of them installed, none of them worked. I finally hit upon a forum post that pointed out issues with the Radeon 1300 cards. Despite being a member of the 1X00 family, these particular cards were not supported properly by NATIT.

Sound wasn’t working either, and there was no solution, or potential solution online. And here I gave up.
I do wish I had gotten it working — it was pretty darn cool. But at > 20 hours invested and nothing to show for it except the Finder, and no hope in sight for Video card drivers, and problems reported running VMWare on a Hackintosh, I decided it wasn’t worth investing any more time in. Still it was a righteous hack, and if I had a machine with compatible hardware, it would probably have been worth it.

Oh well, I’ve been running Vista SP1 for the past week or so, and its actually not that horrible.

Digital photography is great because it allows the normal, every day user to have power over how their photos come out. Unfortunately, most of the time I don’t spend much time ‘developing’ my photos, and kind of just keep them raw.
Now I’ve used Photoshop for a long time — I think the first version I used on a Mac was 3.0 — but I’m far from an expert. I just have a collection of little tricks I know that have come in handy from time-to-time. I’m only recently getting to the point where I can pull those tricks together for a particular project.

This is one of the tricks I learned somewhere on the web that, when used on photos, can quickly improve the result. Here’s an example:

Before. A picture taken on a Kodak Z740, at 4 megapixels, in decent light with the flash on:

After. A couple quick steps applied to it in Photoshop:

My son has incredible eyes, but their color sort of gets washed out on my camera. Also the background grabs more attention that it should. The dresser he’s fishing around in was a Craigslist purchase intended to hold us over for a couple years, so its not terribly attractive.
The “after” picture helps correct these problems by bringing out the color, but reducing the severity of the other imperfections. I’m sure there’s more you could do, but this is quick and easy. Here’s how I do it:

• Open the original photo in Photoshop. If you can’t do that, this is probably the wrong tutorial for you!
• In your Layers on the right, duplicate the photo layer…
  
• On the newly created, top-most layer, choose Blur > Gaussian Blur from the Filters menu…
  
• I usually use a Blur of around 5-6. This will look weird, but don’t worry…
  
• Double click on the new, blurry layer (still the top-most) and change the Blending Mode to Overlay
• Now drag the Opacity slider until your image looks good. Usually between 40-60% looks right to me, but it will depend on the image…
  
• Bonus step: Select the eyes, and copy and paste them into their own layer, between the original and the new blurry layer. Set the Blending Mode of the eye-ball layer to Overlay as well to really bring them out even further
• Save the completed image!

I still consider myself a Photoshop newb, despite having used it for nearly 10 years, so if anyone else knows any other quick/simple tips for bringing out photos, I’d love to hear them!

Update: I just realised that Dooce has almost the exact same tutorial here. Hers is slightly different than mine and you end up with a darker background, but you could pick which Blend Mode to apply on a per-image basis.

My iPhoto library was huge! And that’s an over-statement, cause I’m not talking about twice or even three times the size of the actual photos. I’m talking nearly 10 times.

My collection of photos from 2001 until now weighs in at 6.4GB, but my iPhoto Library was 76GB. I’d always just assumed it worked that way and kept backing it up from Mac to Mac. But finally I got fed up and took a look inside.

If your iPhoto Library is huge and you’ve had one or more iPods with which you sync photos, you might want to follow these steps:

- Find your iPhoto Library (usually in your Pictures folder)
- Right-click on it and choose “Show Package Contents”
- Open the Contents folder
- Find the iPod Photo Cache folder
- Right click on it and choose “Get Info”

If its huge, its gotta go! iPhoto can rebuild this if it needs it, but odds are you’ve got caches of old sync pairings with a previous iPod or an iPod that you’ve since restored. Its totally safe to delete, and aside from a long sync next time you plug in your iPod, it will self-heal (to a smaller, more manageable size…

Once we move back to Ontario, I’ll be working from a home office. Add that to the fact that we’re not using anything more than basic cable, but digital distribution (both the paid and the found-automatically-on-the-internet-and-you-can’t-convince-me-I-shouldn’t-have-it kind) for our entertainment, and I think I have some unusual problems to solve. Here’s what I’m dealing with:

• I want a reliable and dedicated “business” connection with no WiFi access so that my company’s source code is secure.
• But I want access to my stuff from my iPhone and my Craptop, and visitors and passers-by should have free access to Internet, because that’s the spirit of the thing, but they shouldn’t have fee access to my files and documents, so I want both a secure and an open WiFi network.
• The XBox 360’s method of getting shared content is incredibly noisy and resistant to QoS. The sharing program, in my case Connect360, uses broadcast messages to toss bits all over the network. The only way I’ve found to ensure Quality on this is to have a dedicated network just for sharing between the computer and the XBox.
• My media is going to be stored in different places depending on primary usage. High quality movie rips (legally obtained, of course) shouldn’t be transcoded by Connect360. Daily TV downloads and recordings using EyeTV (and can be transcoded since they’re already pretty low quality), MP3s and Photos will be frequently revolved and updated and should be managed from a PC (manually, or by script.) All of these files need to be available from anywhere in the house.

The solution to all of these problems can be seen in this diagram (click for a larger version):

As you can see, there will be no less than 3 individual networks, with bridge points between each, where appropriate.

• The green network represents the dedicated XBox network, allowing us to stream content from a PC regardless of the activity of other devices. We’ll open it up for free WiFi since there’s really nothing accessible on that network, but QoS all other devices as low priority.
• The blue network is the primary home network, your typical file/print sharing setup. Living on that network is a sweet little LaCie Mini NAS drive, which will share all our high quality movies and will also have a direct USB connection to HDTV via the XBox to avoid transcoding. Other media: recordings, downloads, MP3s and photos will be shared from the home Mac to all devices over this pipe. Two touch points in the home office will allow me to listen to my music while I’m at work, and sync my calendars and e-mail with the home Mac.
• Both the blue and the green network will be coming off a standard home Cable Internet connection, but each router will have its own public IP.
• The purple network is a dedicated 7+ MBPS DSL connection (shooting for 10MBPS, but we’ll see what’s available) used only for connecting my home office to the Internet. It will have no wireless connection, and won’t leave my office in any way. The Blue network connection into the home office will be firewalled and won’t be set-up for Internet — SMB file sharing only, but if something goes wrong with my DSL, I can re-configure for Cable over the blue network and still do my job. The purple network will allow communication between my work computers, both of them running Synergy so I only need one keyboard and mouse. And the little iMac will by my jukebox/Photoshop/personal mail rig.
• The two work computers will run up to 4 Virtual Machines each, but my primary VMs (represented by the monitors — and yes I’ll have 4!) will be: an e-mail/IM computer for connecting to work, a development workstation with occasional VPN access to sync source code, and a test server and client — more added as my current project dictates.

  My newest work computer will be running Microsoft’s new Windows Server 2008 with Hyper-V virtualization, allowing incredible performance of the Guest machines.

• Any capable device can also get VPN access from any network, since its secure by nature, but only 3 machines will be configured to do it most of the time. My touch points with the physical office will be fairly limited, since I’ll be able to have local VMs of current server builds that I’ll refresh whenever I’m in the office for meetings.

I think that’s ambitious enough… but if you have a better home network, or ideas to improve this, I’d love to hear about them!