Setting Up a Site, Again
The decisions that paralyze
Domain names, hosting, platform, frameworks, the list of decisions seemed never-ending. They shouldn't be. If someone else were to lay out their requirements I could give them a few recommendations, narrow it quickly and be done. Confirming you are your worst client — or perhaps I'm just mine — I came to a standstill at each step of the process while getting this site live.
This is an easy one for businesses that exist. Find the best possible variation of a .com that isn't taken already (good luck) and go with it. Personal sites are another issue. Do you use your name? Full name? Nickname? Something clever?
Moving on from any nicknames or having to explain to someone why I chose a clever domain — and remembering that what's clever, like beauty, is in the eye of the beholder — it was time to use my name. After regretting not buying a couple of variations of it when I went on a few domain sprees 15 years ago, I had few options.
The decision became easier when I was able to get my own short domain: dmsn.gr. I currently have it hooked up to one of my favorite all-purpose web apps: CloudApp. With all the David Singers online I've used my middle initial a good deal when writing or when sending email. It's not something I'm used to, but when you have a common name it makes it easier to separate yourself. Using my full middle name makes me feel like a child actor or serial killer. dmsinger.com it is.
This is relatively easy for me. I don't need to find a host because I lease a good number of servers or have cloud space with a few companies. How to set it up is another matter. Does it have its own account? Does it tie into other domains in some form (think information sharing)? It all depended on what I want to do, and as that's as open-ended as it gets, the answer was dependent on what I was going to use to manage my content.
I've been a fan of ExpressionEngine for a long time. It's very flexible and since I want to do everything with this site it was the first thing that came to mind. But, another purpose of this site, besides blurting out whatever I may be thinking, is for me to test out new things, whether they be technologies, apps, or CMS(s). I was tempted by Craft, the new kid on the block, and the child of an EE Add-on company, but thought of another project for that and also came to the conclusion that I'd like to work with some data I already had (old blog posts and such), so EE it is.
I decided immediately that I was going to use a front-end framework for this site for learning and experience. The site is simple and could have been done by hand easily, but it's also an opportunity to experiment. I've used Bootstrap before, so I chose the other big boy on the block: Zurb's Foundation. Foundation 4 is relatively new and doesn't support older browsers so it's best to test on something like a personal website which won't draw in much (or hopefully any) non-supported traffic.
I'm labeling this a non-design and moving on. Fewer things can slow something down than overdesigning, and there was no reason to do that here.
Common pain points
Waiting for the next version of anything, whether it's PHP, EE or Foundation, is always tough. You see something in the future you think you could take advantage of and you don't want to work and then re-work a feature. Don't do it. Always work with what you have at your disposal today. By the time the next version of something comes out, you'll see the version after that it can be just as tempting.
A lot of technology has moved from "what's best" to "what's more comfortable" in my position. Certain projects could be better off with Ruby over PHP or a NoSQL database as opposed to a traditional RDBMS (MySQL, Microsoft's SQL Server). That, luckily, doesn't apply here.
Finding a mix of comfort and experimentation was and is key for me. If I just wanted to write I would have Tumblr'd away.