Month: February 2006

Via a link on Macintouch, I’ve stumbled on some very interesting emulation technology that is open-source and runs on Mac OS X. Q is a Mac OS X-port of Qemu, a emulator (with an optional virtualizer) that is gradually gaining steam. I got Windows XP SP1 running on it pretty easily, and there are number of free operating systems that you can download and run immediately on Q, without needing to go through any install of that operating system. You just download the image of the hard drive with the operating system already installed on it from Free OS Zoo, point Q to it, and start up that virtual machine.

But let’s step back a minute. What is an “emulator,” and why would I want to run it? Well, an emulator uses software to emulate a processor (CPU and surrounding chipset) so that you can run software on your computer that will not otherwise be able to run on that computer because it does not have the required hardware support. For Mac users, this mostly means that it will allow you to run Windows, and therefore Windows applications, on your Mac. The Qemu project is actually more ambitious than that; they actually have the goal of creating an emulator that will allow any platform (operating system plus underlying hardware) to emulate another platform. This means being able to not only run SPARC-64 operating systems on my x86 PCs, but also running embedded applications (like those used on cell phones) on my Mac, if I want. It could (and has) allow people to run Mac OS X inside of Windows, after a bit of hacking at it. Here is a photo of Q running FreeDOS and a copy of Doom, the classic video game:

I should state at this point that X86 emulation has been around on the Mac for a long time; people have been wanted to run important Windows applications on their Macs for well over a decade. Perhaps the most well-known emulator for the Mac was Virtual PC, written by Connectix and eventually sold to Microsoft, which has basically killed it off for Intel Macs. The problem with Virtual PC is that was written to specifically emulate an X86 environment on PowerPC chips, which were the CPUs used before Apple’s move to Intel chips, announced in June 2005.

The ability to run any application on any platform obviously brings up a lot of interesting possibilities, but Qemu even goes further than that by offering virtualization. This is where, instead of emulating another type of hardware, you actually run multiple operating systems, at the same time, on your current hardware, and Qemu is the application that actually connects these multiple applications to the host operating system. This allows you to run, say, 4 different Linux servers on the same computer at the same time. To the outside world, they would look and act like individual machines. Of course, VMWare has offered a similar capability for a while in a commercial application, and there are several other companies that are getting into this business. Qemu does it for free (beer, at least) and isn’t trying to support a limited number of host operating systems, like these commercial solutions. As multiple core chips become the norm later this year, this will become a bigger deal.

Q recently released a Univeral Mac binary, which could already be a big deal. More and more Macs being sold today have Intel dual-core CPUs. As Q doesn’t need to emulate X86 code when running on an Intel chip, then the speed of the emulator is much, much faster. Supposedly, Q running on a Intel Mac is about 50-70% of the speed of running Win XP on that same hardware natively (not using an emulator). This means that you could, on an Intel Mac with enough RAM and disk space, run Mac OS X and also run Windows applications at the same time. Most importantly, they would run fast enough to actually be useful. My use of Windows XP through Q was largely unsuccessful, because the speed of the emulation (particularly video and hard disk access) was so slow that it wasn’t actually useful – I would use the PC I have instead if I need to use Windows. However, if I can run Windows XP at something close to native speed on a Mac, it opens a lot of doors.

My brother recently got an Intel iMac. I wish I could give Q a try on his computer, but he lives in a different city. As it is, I will just have to wait until someone else nearby gets an Intel Mac before I can give Q a spin and see how real this opportunity really is.

A few folks have asked me about the software I used to create the panographic images linked in earlier posts on this site. This question has a short answer and a longer explanation. The short answer is that I used a free (as in beer and speech – it costs nothing and the source code is freely available) tool called Hugin, which is actually an interface upon a collection of underlying panographic imaging tools that has been around for some time. In other words, Hugin simply provides an easy interface for you to use some other tools, which, while powerful, were never designed to be easy to use by themselves. Hugin runs on Windows, Mac OS X, and Linux, and I can strongly recommend it. I can also strongly recommend carefully reading through the FAQs and other documentation available with it, because it will save you a lot of aggrevation when you are making your final panographic images.

Making panographic photos is pretty straightforward, once you figure it out or someone is able to show you. There aren’t a lot of good tutorials on the Internet for this, and I think is probably due to the fact that the user-friendly tools, like Hugin, are so new. We will probably see more tutorials and examples appear over time.

All panographic images start with picture taking, and most digital cameras sold today will work fine to create panographic images with tools like Hugin. My old but trusty Canon PowerShot S40, like most Canon digital cameras (my favorite brand for digital imaging), includes a feature called Stitch Assist that helps you line up each image so that you can just take one shot after another, rotating in a circle, until you have a 360-degree view. The LCD screen in on the Canon displays half of the previous image so that you can use it to make sure that the new image you are taking lines up and overlaps the previous image. Each of the files created using this feature are named differently compared to standard images, so that they can be easily linked to one another when you are ready to stitch them together. It is highly recommended to use a tripod when making these images, because you can make sure that the horizon lines up for each photo and you don’t have use the stitching software to compensate for the camera turned a few degrees either way.

It is worth noting at this point that you do not have to make a 360-degree panographic image. I’ve made a few of those, but you can can make an impressive 180-degree or less panographic image out of only a few photos. This is what I did with the Wanaka panographic image.

After you have pulled the images that will make up your panographic picture off of your digital camera to your hard drive, fire up Hugin and drag the images into it’s window that will make up one panographic image. After you are sure that all of the photos are loaded in the right order, you need to start making control points, or links that stitch one photo with the next photo. You might select a particular peak on a distant hill or a big rock in the foreground. As long as you set the control points that refer to the same item that appears on two images and are pretty exact about it, everything will work fine. You will need to set about 10 control points for each image. Importantly, you should also set a horizontal control point that marks the horizon for each image. This tells the stitching software to keep those two images on the same horizontal line and is critical when it comes time to optimize the image. This is what the stitching process looks like on Mac OS X:

Once you are done stitching, you should optimize the image by clicking the Optimize button in the Optimize tab, and then you should bring up the Preview window to set your borders for the image, and well as designating a center for the image. At this point, you are ready to stitch together an image and see how it turns out. If it looks bad, you can go back to Hugin, change some settings or control points, re-optimize, and create another image. The whole process to create a panographic image from stratch is about an hour, and re-creating final panographic images from your Hugin settings only takes a few minutes. This process is computationally intensive, so make sure your computer is up to the task.

I got started with panographic photography with QuickTime Virtual Reality (QTVR), which was something Apple introduced in the 1990s. While living in Australia during 2001-2002, I worked as consultant on a project to create a panographic service website for real estate pros. While that project didn’t end up going anywhere, I learned a lot about panographic imaging, including using what had become the standard free toolset, Panorama Tools, originally created by Professor Helmut Dersch of the University of Applied Sciences Furtwangen (yes, that’s an actual city name). I used the PTViewer, originally written by Dersch, to display panographic images on a web page.

In a lot of ways, panographic software is a great example of the promise, challenge, and opportunities associated with open source software. Dersch developed some fantastic, free software that made making and displaying panographic images much easier. Thankfully, he released the software under an open source license (well, most of it), which gave other geeks like me the ability to see the source code, necessary to build new copies of the software, fix bugs, and make improvements. However, Dersch was forced to shutdown his website and development due to the threat of a lawsuit from iPIX, a company already doing panographic images. (More info on this is available here.) However, while Dersch was stopped from doing development, the source code lived on, and while the threat of a lawsuit hung over developers, after a while, a number of polished tools started to appear. Today, you have the choice of using a free tool like Hugin to use the Panorama Tools, or you can choose a commercial software package that gives you some technical support, such as PTgui. Both of them build on the excellent work of Professor Dersch, and they wouldn’t be possible without the original open source development of the core utilities.

Wow, it’s cold in Memphis! A cold air mass moved in on Friday, bringing with it sleet and ice overnight, and by Saturday, we woke to a strange scene. Outside everything looked beautiful and white, but the air was so much colder than normal, and the ice seemed to melt in direct contact with the ground, though the ice above it stayed frozen in the cold air.

The low temperature was not a record (16 degrees F today – Sunday – while the record low for February in Memphis is -11 F – yikes!), but the wind chill made everything seem much colder. We actually moved out of our bedroom, which only has an electric portable radiator for heat, and started sleeping in the guest room, which is connected to our main boiler system. In the end, that almost became a very expensive mistake, as we had some pipes freeze in our back bathroom. This bathroom faces north, and it received the brunt of the cold air. The pipes that froze were not on an external wall, but they were close enough. Strangely enough, the drain also froze, and once we finally got the pipes unfrozen, it took a while to get enough hot water down the drain to clear the frozen blockage. It was even hard to walk around on our hardwood floors with wool socks on; the floors were frozen, and there was a layer of cold about a foot deep that made walking around very uncomfortable. Our boiler had a hard time keeping up, and I dread the gas bill for next month.

The real problem really wasn’t really the cold. Rather, it is that the walls, heating systems, clothing, and basically everything we use in Memphis to protect us from the cold weather in winter, is not made to deal with this type of cold for very long. When a cold weather system sits around for a few days, held in place by 10-20 mph winds, it can be pretty rough. It was the coldest I could remember Memphis getting, and the rest of the country was not spared, with L.A. getting snow and bands of sleet making it across Mississippi.

While we were iced in this weekend, I did finally get around to posting some photos from our trip to DC to see Aaron, Carrie, and Molly. That’s quite a cute kid…

More photos from the trip are available at the Expats photos area.

OK, no rant today. Rather, this post will be short and sweet. I’ve (finally) finished two more panographic images from the New Zealand trip.

View from the top of Mount Victoria, Wellington, New Zealand

View from the eastern edge of the Kaikoura Peninsula, Kaikoura, New Zealand

Both of these links load a Java applet that will give you a 360-degree view from the place where I took the picture.

Post a comment if you have any thoughts of these images. Thanks.

I’ve got some sort of pretty resilient bacteria infection, and after seeing the doctor yesterday, I’m now on a collection of antibiotics and other medicines with the goal of giving the sickness the boot, once and for all. I usually don’t have a lot of good things to say about the medical system in the U.S., having observed it (and sometime used it) all around the world, but yesterday’s experience wasn’t too terrible. Now, just if drug pricing seemed more rational…

Well, that’s not really fair. Drug pricing in the U.S. is rational. Very rational. That doesn’t make it fair or affordable, but I should say that it makes sense. Simply put, Americans really don’t mind paying many times the amount the rest of the world pays for the same drug. We complain about it, threaten to do something about it, build layers of bureaucracy to try to make it more efficient, and use it as a talking point of what must be done to build a better society, but we really don’t care about it. Why? Because compared to most of the world, we’re rich. Drug prices, even stupidly unfair drug prices, hurt, but they don’t hurt enough for us to demand systemic change to the pharmaceutical industry.

I’m not talking about nationalizing the pharmaceutical industry. I’m a libertarian (little “l” – libertarianism is best realized as a philosophical outlook rather than a political party), and I would not nationalize any industry like that. Rather, I think that most folks haven’t given the pharmaceutical industry’s advances per their cost much thought lately, because they have been too busy yammering on about high drug prices. The facts are that, with very few exceptions, the pharmaceutical industry hasn’t cured very many illnesses in the past 50 years. What they have done very well, along with the medical supply industry, is provide decreasing percentages of improvements on existing drugs that treat mostly chronic maladies, allowing them to patent the marginally different chemical formula and create another wave of demand for what is essentially the same drug, only it help 2-3% more patients with a particular disease now.

Were this any other industry, I wouldn’t advocate a re-think of what we should expect from them. After all, aren’t they basically doing what the car industry has been doing for about a century now? Every year’s model is basically the same as the last, with slight improvements. Occasionally, you see a totally new model or a body style change on an existing model, but not very often, and the end result is basically the same: the ability to charge much more while your experience as a driver and the effect (getting from point A to B) doesn’t change very much. The problem with applying this model to the pharmaceutical industry is that people occasionally need drugs to stay alive, and the survival instinct will push them to pay more and more money for even the chance that they 2-3% improvement in the “new” drug will be more beneficial to them than the “old” drug, when actually they are pretty much the same.

This business model of low value for the consumer in matters of life and death also can be seen in hospitals and clinics, which struggle to purchase the very latest MRI machine, which can image only slightly better than the previous one but costs 4-5 time more. At the bottom of it all, hospitals, doctors, pharmacists, and everyone else in the health care industry knows that people will clamor for the very best equipment and chemicals, even if the treatment available from those medical tools is only very marginally better than the previous generation, which may have only been released a few years ago.

Eventually, I hope that we will wake up and demand true value from our health care industry. If a pill only gives a slight improvement over something that has fallen out of patent and is now available as a generic, just use the generic. Most people get the same medical problems, many of which have a treatment that hasn’t fundamentally changed for decades. I would gladly live with 1980 medicine if I could pay 1980 dollars to get it. As it is, the entire system only gets more and more complicated, and as more money flows into it, there are more and more cracks for money to seep out, into the pockets of middlemen, efficiency “experts”, and government bureaucracy. We should demand more for what we pay, and we should pass on the latest and greatest if its cost can’t justify its performance.

OK, if you’ve read through that entire rant, I’m sorry, and you deserve a lollipop. Or at least a picture.

I’ve been messing around with creating panographic pictures from freely available tools on the Internet, and here is the first one I have to show off:

South End of Lake Wanaka in the South Island of New Zealand

I’m still working on producing a couple of 360-degree images, and I’ll post them here when they are ready.

Stay healthy out there, for all of our sake…

One fun and cool thing we did on our recent New Zealand trip was take along a hand-held GPS device, loaned to us by our wonderful friend Frank Campagna, so that we could mark locations we visited. The GPS came in very handy when driving, as well, as we did have any other New Zealand maps with us, other than those in the guide books or tourist magazines we picked up along the way.

Frank’s GPS is an eTrex Legend, which is a good, solid, basic GPS available from Garmin. The device has US maps burned into it, but you can also buy additional maps and load them on the device so that it gives you more than the basic longitude/latitude/altitude info when you leave the US. I picked up a copy of the Garmin WorldMap CD on eBay for about $50, which I loaded on to the eTrex. These maps did not give the detail of the US street level mapping, but it was enough to help us get around cities and get between towns. Overall, the system worked really, really well and made the trip much more enjoyable.

Along the way, we saved particular long/lat points on the GPS that we wanted to look up later. Using Google Earth, it is actually pretty easy to find a lot of the locations of the highlights of the trip. However, using the GPS, I’m able to find some locations that are not really clear on Google Earth, given the current low-resolution of most of their New Zealand image data.

First, a word about Google Earth. Chances are, if you are reading this, I’ve probably had a chance to tell you in person how much I love Google Earth, especially as a fun, basic, fast GIS tool. If you don’t already have it installed on your computer (and your computer is less than 3 years old), go to the Google Earth website, download it, and install it. They have versions for Windows and Mac OS, so there’s really no excuse for not using this wonderful free tool.

Once you have Google Earth installed, there are quite a few things that you can do with it. One of my favorite is saving a location to file (Google Earth calls them KMZ files) that you can email to someone or post on a web page, and someone else can open it up and Google Earth will take them to that location. You can also chain locations together in a KMZ file to create a virtual flying tour from place to place within the Google Earth interface.

Below, I’ve posted links to a few KMZ files that will take you to locations where I took photos in New Zealand. Using Google Earth, you can see the view from space of the place where I took the photo. Previously, you could show someone on a map where you were when you took a particular photo, but now, using Google Earth, you can give them a different perspective on the same landscape in the photograph.

Note: You have to have Google Earth installed on your computer and have your web browser configured to pass KMZ files to Google Earth in order for these links to work for you.

Jaymie’s Tree in Blenheim – This is the tree that Jaymie picked out before we went on the trip and that we visited when we got to Blenheim.

Interislander Ferry Moving Through Marlborough Sounds – This location came froma GPS coordinate (S 41 14 30.6, E 174 03 28.2) pulled off of the eTrex. Right after I took this coordinate with the eTrex, I shot these photos of Picton as we pulled into port.

Kaikoura Panographic Photo – This is the location where I took a panographic photo I hope to post soon. In the meantime, these are more basic photos of the same area.

Te Pukatea, Abel Tasman National Park – This is the beach Kath and I hiked to on the Abel Tasman coastal trail.

City of Westport, Western South Island, New Zealand – This is the mouth of the river, near the hostel where we stayed.

I’ll post some more of these, eventually, if it turns out anyone enjoys them. Leave a comment if you want me to post more.