There's been some pretty fun and interesting stuff coming out of the Adobe AIR platform. I've tinkered with building some stuff on the platform myself, but most of my use of AIR-related has been using the wide array of Twitter apps. That's probably the biggest category of AIR apps at the moment.

I've been using Twhirl pretty heavily and prefer it to the web interfaces. However, while Twhirl and AIR run on my Windows boxes and my Mac, my primary laptop/workstation runs Ubuntu, which isn't a first class citizen in the AIR world.

I like to have the same tools available on all of my workstations whenever possible. So, when an alpha for Linux was announced, I took note, planning to get it installed when I got a chance.

Just before I left for Ireland, I gave it a shot, downloading the .bin file, running it and getting about half way through the installer before the install dying. I looked up the error only to find that I was out of luck. While Adobe supports many Linux distributions, including the one I use, what it *doesn't* support is the 64 bit version that I'm running.

So, today at lunch, I went looking for a way to get it running, despite Adobe's lack of support. So many of the apps that claim not to work on 64 bit Linux actually work just fine if you do it right.

What I found on several sites all boiled down to the same answer/solution. There are some glitches that are kind of a pain (like Twhirl not saving my passwords), but I've got Twhirl running on Ubuntu 64. Here's how to do it.
Read the rest of this entry »

Photo: Geraint Otis Warlow

For quite a few years now, I've used Microsoft's HTA setup for doing quick, standalone applications in HTML and Javascript. Most of those have been for my own use, in large part due to the fact that changes in the Windows security policies mean that if you distribute them, you don't really have a great chance of them working on other people's machines.

Along the way, a few other solutions in this vein have shown up as well. Webrunner is based on the Mozilla platform and works particularly well if your application will reside on a web server and you just want to provide a custom client for it. Webrunner is cross-platform (which HTA's aren't), but doesn't provide much access to the local machine. The "sandbox" that it is able to operate in is pretty much exactly the same as a regular browser.

However, when I saw Adobe AIR (with Linux support planned) being reannounced after the recent renaming from Apollo, I gave it a closer look. Adobe's AIR gives you that same build-it-in-JS-and-HTML approach, with installable final applications. Further, it provides some additional API's to things like the local filesystem.

That feature leverages both sides of the technology. You've got all of the web and the best of the desktop all in one. If you look at the samples page, you'll see a variety of stuff that just doesn't work in a regular browser. I'm messing with ScreenPlay for use in screencasts to do that play-by-play highlighting of what's going on on the screen.

Now, when Apollo was announced, one of the reasons I didn't do much with it was the same reason I have a love/hate relationship with XUL: creating projects was a major pain. I'm perfectly OK with writing code by hand (it's usually what I prefer), but I hate having to stub in dozens of files just to get a "Hello World" project up and running.

So, Aptana's, recent addition of AIR support in their IDE caught my attention and I gave it a spin.

I have to say, I'm impressed so far. It was really easy to set up a basic project, run it from the IDE and build an installable AIR file. In short, their support for this framework lets you focus on writing your application instead of setting up the project and environment. Very nice.

This offers some real possibilities for some of my project ideas, particularly where the client side of the client-server equation needs to be a bit richer or should really be a dedicated client instead of sharing the web browser.

For instance, I really prefer to use a dedicated application to do the writing on this site. All too often, something can go wrong when you just type into a web browser's text box and you lose a bunch of content. However, a nice desktop application, with local saving of drafts fixes that. If I can whip something up that also uses SVN to version and share those drafts, I think I'll have my writing environment. And, AIR with Aptana looks like it might make that easy.

Photo: lui

[[FYI, this has been sitting in my writing queue for a while. I took a quick look at it and am shoving it out the door. Let me know if it's deficient and I'll fix it. Consider this version 1.0 of the article.]]

Ever since I spent my time in the technical writing trenches right out of college, I've been interested in doing my writing in a single format, generating whatever target formats from that.

At different points over the ensuing years, I've been drawn to Docbook as that single format. It's SGML/XML, first of all, which makes it relatively easy to write in a text editor. It uses XSLT to transform to other formats. It has pre-built toolchains for outputting HTML, PDF, etc. Also, it is easily versioned using Subversion.

As an additional vote of confidence for using Docbook is the fact that O'Reilly has been using it for many of their recent books. They're actually also storing their content in an Atom Publishing Protocol repository. That's another vote for where my intuition about a personal publishing stack has been leading. When written content is stored in a robust container (DocBook or Atom, etc.), you can repurpose it.

Technical documentation doesn't always work in every format (which is why many of the single source experiments in that space failed). However, for things that *do* work in multiple formats, the technology for producing those formats gets in the way. Not so with DocBook.

Now, while I've looked into it at several different points, I've never really dug into it well enough to get much done with it. I set out over the last couple of weeks to actually get completely up and running with Docbook. This time I powered through and got it working. I suspect that the motivations were more concrete this time

Along the way, I discovered that there wasn't a tutorial that matched what I was looking for. I also found tutorials that had non-functional code. However, despite that, I was able to get a basic Docbook book up and running and figured out a much simpler way to get started with Docbook, so I figured I'd share.
Read the rest of this entry »

Photo: J Wynia

A few weeks ago, Scott Hanselman gave a spec for his idea of what the "ultimate" developer workstation, bringing the grand total to $1900, without monitors. While I think Scott actually is buying that machine, most of the time, such lists are more along the lines of "what kind of house would I buy if I won the lottery" than practical planning. It's a fun thing to do and consider how you'd configure a machine if money wasn't an object.

Beyond that, I'm pretty weary of the whole idea of constantly seeking out the "best" or the "ultimate", which is usually not even reasonably defined. His "ultimate" box, for instance, has a 2.4Ghz quad core processor ($531), but only 2GB of RAM ($216). My version of an ultimate box would swap those 2 priorities. I'd rather have 16GB of RAM than an ultra-fast processor.

That's what makes the whole "building your own" approach to computers interesting. Each and every component choice reflects your own priorities. And, today, when Coding Horror extended the aforementioned "ultimate" by also including a basic and "premium" in addition to "deluxe" it came at an interesting coincidental time.

See, just yesterday, I ordered all of the bits for a new desktop. With the laptop fiasco of the laptop I want not being available until nearly September, I stepped back and re-thought my technology plan for the next year. Most of my gear needs an update. So, I decided to start with a new desktop, then monitors, then the laptop and settle in for 2-3 years after that.

Looking at their "basic", it falls right in the price range for what I bought. My total with shipping was just over $800 and included everything except the power supply (because I have one laying here), but did include a copy of Vista Business as this will be a multi-boot system that I'm building.

If everything goes according to plan, the pieces will be here for me to assemble between the end of my current contract (Thursday) and the beginning of the new one (next Wednesday), giving me time to put it together and get it up and running.

Here's how I approached this and what my driving factors were. I rely on these things rather than benchmarks because these are the things that are actually important to me, not whether a given set of tools runs faster.

First is that I want a solid 3 year duty cycle out of this machine. The ability to build it as a decent machine this month, but make it progressively better over the coming months was important.

Second, more than anything else, my ability to juggle all of the things I'm working on at a given time is always my biggest bottleneck. After working with dual monitor setups, it's clear that I haven't been even close to the actual desktop resolution/size that I really need to get work done.

To that end, I want a setup with at least 2, but preferably 4 big monitors. One 30" monitor has less space than 2 x 22" monitors do and 2 22" monitors can go for $440 these days.

Third was to rationally measure what is to blame for slowdowns. On Linux, I really don't run into performance problems even with what I have. However, on Windows, it is universally not enough RAM. CPU is secondary.

Incidentally, that list is *why* I didn't just buy one from Dell or somewhere else. See, the ability to run 4 monitors or 16GB of RAM is just not in the mainstream usage and mainstream is what Dell and company are after.

At any rate, I started out with video cards, looking for those that handled the highest maximum resolution for the money. I settled on this 256MB card with dual DVI that handles 3840 x 2400 resolution for about $130. The Coding Horror "basic" machine uses a dual DVI video card with only 2560 x 1600. This selection ensures that whatever monitors I end up buying will be handled. And, should I choose to add another of these cards, all 4 monitors can be as big as I can afford and things will work out great.

That potential future upgrade (a second video card) prompted part of the motherboard selection. I wanted at least 16GB of maximum RAM and multiple PCI-x slots. There were a few out there that handled 32GB of RAM, but not a lot of choice on other stuff, so I fell back to the 16GB motherboards and ended up with this open box motherboard for about $50.

As I'm constantly frustrated by a lack of room in the case of desktops, I opted for a full tower case for about $70. However, I'm still intrigued by the IKEA solution to the computer case problem, so I have a Plan B.

Since the selected motherboard used the LGA 775 slot, I used my "bang for the buck" approach on the Duo Core processors and ended up with this Intel Core 2 Duo E4400 Allendale 2.0GHz CPU for about $125. Given that I never max out the single core 1.6Ghz on my laptop, I'm not too worried that there's a "better" CPU out there.

Much of the rest was pretty straightforward. A $100 500GB hard drive to handle the 3+ OS setups I intend to include, 2GB of cheap starter RAM for $80, CPU cooler and DVD burner and the basic setup is complete.

Since the first 2 OS's on the machine will be Ubuntu and Vista, I grabbed Vista at the same time.

All of that lays a foundation that can easily be expanded.

• The case is huge, so I can add lots of hard drive space.
• There's room for lots more RAM to be added over the coming months.
• I can shop around for 2 matching monitors and buy them in the fall. That will replace the monitors I'm using now (which actually replaced those in the picture of my office about 8 months ago).
• I can add more operating systems to the setup.
• Next year, I can add a second video card and another pair of monitors. Who knows, 24" or 30" monitors may have sunk into the "cheap" range by that time. However, the chosen video card will handle them fine.
• If, by some miracle, I start maxing out the CPU, I can swap it out, because I didn't pay much for it in the first place.

In short, by the time this thing is actually at the end of its life, it will actually be really powerful, meaning that the 3 year life might actually be low. And, by spending the money on things like the monitors, when it DOES reach the end of its life, I can start the cycle all over again for cheap.

First, let me re-state for the 100th time that I'm a technology pragmatist. I love technology, but do my best to make my purchase decisions based on my actual needs rather than pure technolust. That's most of why I don't run Mac laptops and won't buy an iPhone: they just don't match my needs (practical and financial). I do have a 4th generation iPod nano and, when comparing that to a 1st generation iPod, I wonder what 2 years of iPhone releases (at which point the buyers from last week will finally be free of their contract to sign on for another 2 years) will do.

Third, I really like running Linux and open systems. The fact that my firewall is running Smoothwall lets me tinker with it and provides much more power than other devices for the same money. Linux completely resurrected my 3 year old laptop and made it usable again after I had thought Windows had sucked the marrow from its bones.

So, when I first heard about OpenMoko, a platform for an open mobile phone, I was excited. And, today, when they announced that developers can buy the first phone based on the platform (at OpenMoko.com, I added it to my tech wish list.

I'm not going to claim that it's as sleek or even as fully featured as the iPhone. However, it is an *unlocked* GSM phone, running open source software, with full specifications of the software and hardware available, and with a nice set of features.

Even before I saw an iPhone, I knew I wouldn't buy one because it was tied to Cingular/AT&T. I've used Sprint, AT&T, TMobile and Cingular each in turn and currently use TMobile because it's the only one that actually works all of the places I regularly need a phone. Period.

Beyond that, 2 year contracts with $300 cancellation penalties in exchange for a $200 phone are criminal as far as I'm concerned. I've been buying unlocked GSM phones whenever I can and it's WAY better. When a phone dies around here (and they all seem to at some point), you just pull out the SIM and throw it in one of the old ones. The same is true if/when we travel outside the US. We can just buy a prepaid SIM and use the unlocked phones.

Now, I probably won't buy one of the developer ones, in large part because it isn't clear whether these have fully-functioning GPS and WiFi and because I want to make sure that those who will be doing the most development for this device get one before the supply runs out. However, I'm certainly more willing to drop $450 on one.

All of the sites are suffering under the strain of popularity at the moment, so I recommend bookmarking them for later. Definitely a project worth watching. Especially since they plan another 3 devices in the next year or so.