Four doors with rear seats
Four doors with rear seats
The dash
New stereo w/iPod connector and remote
97072 Miles
2 cracks in windshield
1999 Ford Ranger 4X4 For Sale - 4.0L V6 - Automatic - Cold A/C - Sony Stereo w/iPod connector and Remote - 97072 Miles.
I want to start off by saying I’m an Android lover. I drank the Kool-Aid long before any Android phones came out and I’m not ashamed to say it. When the iPhone came out, I was somewhat ‘meh’ after the launch. Quite honestly, it just wasn’t that impressive to me. Now, the first thing someone will say to that statement is, “But once you use one, you’ll see how great it is.” Actually, I do use one. I’ve had an iPod Touch for a couple years. I know the interface, I know the Apps, I know the development environment, and I know it works great for most people. For me, I want things like customization and multitasking. I don’t want to be held hostage by a user interface that doesn’t allow me to put the information I need up front.
Android seemed to be everything I wanted in a phone. An operating system backed by a company I already gave most of my life to, an interface I could customize to meet my needs, and easy access to apps that extend the value of my phone. Plus, it had actual multitasking from day one and a notification system that is still today considered to be the best in class for smartphones. But for all the good things Android has, there are definitely some areas that could use some polish. One of the most important ones that has arisen lately is the update process.
Apple has Google/HTC/various carriers beat hands down on this one. When a new update for the iPhone comes out, everyone can get it. Immediately. iPhone 2G, 3G, 3GS, iPod Touch, every device has a release on the same day. That’s an amazingly great system. It means that developers can start building apps with new features from day one without a concern that people aren’t running the latest software. Apple’s problem is that their App Store policies are going to end up killing the application environment. The tight control and unreasonable denials are sending developers a clear signal: you may or may not be allowed into the App Store and we don’t have to give you a reason why. Oh, and by the way, you can’t even talk about it if you do because of our NDA. Bad idea. No one wins in that scenario.
By allowing carriers and hardware vendors to drive the Android update system, Google has allowed fragmentation to play a big problem in their ecosystem. For instance, right this very second there are no less than 4 different versions of Android floating around: 1.5, 1.6, 2.0.1, and 2.1. Each of them has different capabilities and requires differences in developing applications. This is a very bad thing. Granted, most of the changes are not major, but it does mean that there are many people being left out. A number of those people being left out are people like me, an almost day one G1 purchaser who’s been speaking the Kool-Aid to everyone they know only to now be stuck on a version that’s becoming outdated and unsupported by a number of newer apps.
Google, or whomever else can, needs to take control of the situation. If a new phone is coming out with updated software, they need to make sure all current devices get the same software release that day. This does mean a lot more work on the hardware vendors to ensure updated drivers are created, but it only makes the entire ecosystem that much better. As a loyal Android lover and G1 user from the start, I beg you Google, get this issue fixed. Stop fragmenting your brand and keep your early adopters happy. You should have released 2.1 for the G1 and Magic before anything else.
View high resolution
Microsoft released the preview of IE9 and on their website, http://ie.microsoft.com/testdrive/, they have a screenshot of ACID3 test results, a popular method of seeing how well browsers follow web standards. I compared their results with the results of 4 other browsers on my Mac.
After seeing all the results, I’m really wondering why Microsoft would be showing off their results.
Oh Time Warner. Everyone hates you. Just ask around. Everyone that has Time Warner, at least in the central Texas area, thinks that you are terrible. Why? Well, mainly because of customer service and price, but that’s not the full story. You’re stuck in the Dark Ages in terms of UI and functionality. At a time when companies like DirecTV and Comcast are releasing better user interfaces for their boxes (in some cases, the awesome TiVo software) and new features like remote scheduling, mobile web based guide listings, and multi-room DVRs, you just look like you don’t even care about keeping your customers.
Don’t get me wrong, I want to like you. Your cable modem speeds are phenomenal. They are easily the fastest thing available at the price point you offer around Austin. Of course, if Verizon decides FiOS needs to be here, you’ll be second place on that. Plus, FiOS, like the others, has all the other newer features people are coming to expect in their entertainment experience. Your digital phone service, though I feel it is overpriced, also works quite well. But when it comes to cable television, you’re getting a great big FAIL from me.
Your cable box software is atrocious. I remember the first time I saw it, I thought it was kind of cool since the small town I was from didn’t have digital cable. That was 1999. Fast forward to 2010 - you’re still using the same user interface. Bland text on a screen, no search functionality, no HD optimization, confusing settings, etc. All this was OK in 1999, when people were just starting to get into the craze of digital cable and small-dish satellite TV, but it’s 2010. We’re now used to interfaces like the iPhone and Android, OS X and Windows 7, and Facebook and Twitter. The face that other companies, meaning your competitors, have jumped on the bandwagon makes it seem like you’re not a forward looking company. It’s the main reason I’m scaling back my cable subscription, moving to Windows Media Center, and starting to examine my options for online television instead of cable.
User interface is a big deal nowadays. The paradigm for how people interact with hardware and software changes on an almost daily basis. Where in the past, large remotes with tons of buttons were the norm, more and more devices are being designed with minimal buttons or touch screens that change based on the activity you’re performing. In other words, context sensitive user interfaces. Change the interface based on where I am and what I can do. Don’t just cram the options available into the same navigational architecture you use for everything else. And don’t show me options that don’t work. If I can’t do anything with it, don’t show it to me. The fact is, people have been wowed by hardware and software now. It can’t just work. It has to have a wow factor in order to distinguish itself from others in the market.
This is a big problem for my line of work as a web developer, too. The idea that a website can exist more than 2 - 3 years with the same design is gone. You simply can’t expect to continuously draw the attention of an audience without much more rapid change. Any website that stays the same for more than 3 years doesn’t appear consistent, it appears stagnant, much like how Time Warner appears stagnant to it’s customers. Attracting the attention of new customers/clients and keeping those that you have means continuous, iterative development and design of your user interface(s).
Here’s what I have to say to anyone building anything that requires a user interface: build mechanisms into your hardware and software to allow yourself to rapidly change the user interface at a moment’s notice. Look at other devices, software, and websites to see why they work. Spend time doing usability studies. Don’t just think about what functions people need to do, think about how they go about them. Plan on timelines getting even shorter and shorter as more and more advanced interface ideas come into the marketplace. And by all means, at the very least, keep up with your competitors. In an ideal world, you want your competitors to keep up with you.
As for Time Warner, your days are limited unless you catch up to the competition - fast.
This is me and the coworkers at work. Aaron’s got a nice TENSO!!!! going in the background. Paul was just trying to ask a legitimate question.
Just some pics I’ve wanted to upload of the pups, stuff around the house, a bike in Italy. You know, the usual.
After moving doublestereo.com to the Rackspace Cloud, I decided to no longer piggyback phpdiddy.com on the cloud servers. Not because I don’t have the capacity, but I don’t want to have anything other than doublestereo.com on the cloud systems. Tumblr is nice, it’s easy, and let’s me hit up a bunch of networks at once, so I’ll probably end up using it more often.
I copied over my lone post from my previous blog, the same one that was featured on CakePHP’s bakery. I do still plan on doing a similar write up about DoubleStereo and moving to the cloud. But until that time, much other stuff to work on.
This is one of my wife’s awesome cupcakes. Butter golden with butter cream frosting. Ask anyone, they’re legendary.
CDfuse.com is a website dedicated to helping independent musicians promote and sell their music and merchandise online. A friend of mine came to me with the idea back in 2003 and we launched in January of 2004. A major refresh of the design and our feature set occurred in August of 2004 and that site has been our mainstay until this time. Over the next couple of years, it became apparent that we needed to make some changes. The database design was simple and lacked a great level of normalization. The code was fragmented and inconsistent. Adding new features meant finding a way to make it work with our setup. What we needed was a database design that lent itself to being added to quickly and code that could do the same. We needed to be able to add new features in 7 - 10 days instead of 2 - 4 weeks. Finally, one of the big ticket items we wanted to add, digital distribution, was simply impossible with the database structure we had.
There were a certain amount of legacy features that had to be kept when moving over to the new site. One of them was custom urls. Each artist is given their own unique url (www.cdfuse.com/yourbandname). This had to stay the same. In addition to this, artists who were selling merchandise through their storefronts had pages at www.cdfuse.com/yourbandname/store that also had to remain the same. We had to have as little interruption of service as humanly possible. Luckily, Cake’s built in routing functions give a great deal of flexibility in how urls are displayed. I’ve even written another article here regarding doing custom urls from the site root.
When we first started the planning phase of the new site in early 2006, I knew I had to find a way to keep some level of consistency between all the moving parts. Luckily, there were a number of frameworks beginning to gain traction around that time. Unfortunately, that was also the first problem. Picking the right framework for your team is much like picking the right used car. On the surface, things may look great, but once you get a little deeper, be prepared to run into some pitfalls.
The project went through iterations with Qcodo, Symfony, Zend Framework, and CakePHP. From the start, Symfony seemed like the logical choice. It was definitely geared toward what one would consider enterprise development. The complex array of config files, however, caused some difficulties in getting started. I felt like I was getting bogged down in configuration instead of in the real meat of the application. I attempted to do some work in Qcodo and Zend Framework, but at the time, these projects were both in their infancy, and I ran into a number of issues with their early releases. I’m sure they’ve developed nicely now, however, at the time, I needed something stable and ready to go.
The main features of our site include providing independent artists and labels with order fulfillment, customized storefronts, and an artist profile page (with photos, mp3s, etc). Our plan included adding a listener signup, digital distribution, and providing artists with greater control over the details of their account. Unfortunately, as I stated before, the design of the current database was severely limiting it what it would allow us to do. My first order of business was to refactor the database. Moving the database from our old format to the new one was by far the most time consuming process of the entire project. Another big issue we had to overcome was storage. With the addition of digital distribution, we were in need of an essentially endless supply of storage, Enter Amazon S3. Amazon’s S3 allows for an infinitely expandable amount of storage. We now use S3 as a type of Content Delivery Network. Items get uploaded to our server and then the data is synced at different intervals throughout the day.
One thing that all frameworks have is a preferred way to have the database setup. For the most part, the concepts of each are the same, but their methodology differs. Cake’s methodology made the most sense to me. It was clear, concise, and made for a database that was easy to understand even when looking at it through normal database management tools.
Developing the site with CakePHP was one of the best experiences I’ve had in development. I started out by baking almost everything. From there, I had a good foundation of everything and could expand or remove models, controllers, or actions as needed. I developed the majority of the frontend in XHTML and CSS outside of Cake and then ported all that into my views rather than trying to perfect functionality and form at the same time.
Most of the issues I ran into along the way were easily solved by some reading in the API or a question in the IRC channel. My main issue was, and still is, performance related. Each of my models is extremely complex and carries with it a lot of data so mastering the use of bindModel and unbindModel became very important. SSL security also proved to be an interesting issue. Instead of trying to hack a way for SSL pages to be secured through Cake, I instead opted to let Apache do the heavy lifting on this one. By using mod_rewrite in some interesting ways, I am able to explicitly state which pages should be secured and which should not, keeping server load as low as possible and allowing me to freely develop in Cake without worrying about making conditions for SSL. Something to remember if you go this route is to at least be aware of what areas will and won’t be SSL secured. POSTing to and from SSL and non-SSL can cause some headaches with missing data if you’re not careful.
I’m a big fan of not reinventing the wheel. With that in mind, there were a number of other prebuilt projects I used to complete this project:
Planning for the next phase of the project has already begun. Here’s what’s in store for the future.
It took roughly 18 months from start to finish for this project to be completed, including all the planning and research. Only about 5 months of that was really devoted to building the new site with Cake. For those of you that are converting a site with a fair number of users, don’t panic. It can be done. I can say that it has already proven itself to be very worth it. Over the past week since the site has been live, it has been extremely easy to troubleshoot problems that have come up and fix them in mere minutes. For anyone considering a large site rollout with Cake, I’d be happy to offer any advice or help with issues that may come up. Just remember, I’m still learning too, but I have been able to obtain a good grasp on Cake and its intricacies over the course of this project. I’m in the IRC channel pretty regularly so hit me up sometime!
