I’m freaking out man!

Building a WordPress theme can be broken down into three steps; design, front-end development, and lastly the WordPress implementation itself. I’m not quite sure how others do it, but the following guide is a detailed look into the process I have developed for building WordPress themes.

Design

I’ve detailed my process for building this latest version of eleven3, in that post I discussed mostly design. Starting your own theme should work the same way. It’s important to ask questions such as:

• What problems are you trying to solve?
• Who is your theme for?
• What is your sites most important content?

The word, “theme” is used for WordPress – but it doesn’t mean that your site needs to have a theme. However, having a “theme” may help one along their way to achieving their vision. A good place to get inspired is cssremix.com, who’ve been so kind in showcasing some of my own work. And for the quick and dirty answers: use Adobe Photoshop, use pixels for measuring (if this seems strange to you, just do it and thank me later), try to make sense of Khoi Vinh’s gridding lecture, think about what parts of your site can expand and contract nicely (if you’re a beginner, keep to expanding vertical), and for imagery check out istockphoto.com. Wow, learning web design can’t can be summed up in one sentence.

Front-End Development

I’m probably stating the obvious here, but jumping into WordPress at this point would be akin to being stinky and dumb. (-; Make your site look how it is supposed to look without any real functionality. Get it working in all browsers and upon completion – then think of WordPress.

But how?

There are a ton of “we have layouts” sites out there. And while tempting, it’s the same thing as using a WordPress theme and tweaking it. The only way to get a good understanding of code is simply just to code yourself. Start fresh. Here is a base I use for many of my sites – free to build from.

Personally my first step in coding a site is to lay the design and fresh HTML (use Dreamweaver, Coda, or any other code editor of choice) document side by side and hand code the structure. I know it would help to know where to start and I can only offer this advice – study the source code of the “famous” web designers, they’re critically acclaimed for a reason, they do everything thoughtfully and well.

Mmmmmm CSS Zen

The next step would be to start styling content using CSS – if you’re at this point and you’ve never seen CSS code then I’m not going to lie to you; your road to CSS Zen will be long and hard. If you are still not overwhelmed, create a new CSS document in Dreamweaver (or your preferred coding app). A first step I eventually figured out on my own (and then a punk 17 year old showed me the asterisk – j/k QD) that would have saved me a ton of headache is the clearing technique (which is frowned upon in some countries). It goes something like this:

* { margin: 0; padding: 0; }

The reason to clear the padding and margin values is to avoid all the different “inherit” values of HTML elements. To be more concrete, an <h1> tag in internet explorer may have 8px of bottom padding, whereas it may have 10px in Firefox. Why do they have any values at all? I’m not sure, it could be because HTML came along before CSS. Just ask a couple of ‘timers about spacer gifs.

Coding CSS is a mix of looking at your comp in photoshop to find pixel values (told you so) and slicing up your comp to place images in your end product. Eventually you hope to have a functioning website that looks consistent across all browsers.

WordPress Implementation

And now for the fun part implementing WordPress. It should be safe to assume at this point you have your WordPress installed on your server and you have a fully functional layout working in all the major browsers. Ready, ok go!

The conies and taters

Step One

Duplicate the default theme located in wp-content/themes/default/. Copy the default folder and rename it “your-theme-name.” Make sure your duplicate is located in the same directory “themes.”

Step Two

Start transferring your HTML, CSS, and images into the theme folder. Working within your theme folder, make the following changes (don’t worry, you can reference stuff if necessary from the original default folder):

• Delete all the files in the images folder and replace with your images
• Delete all of the functions in functions.php
• Delete all of the css (except for the commented out header) in style.css and then paste your own CSS in its place.
• Change the information at the top of style.css to reflect you and your new theme.

Step Three

Save a small screenshot of your comp/design (300pxX240px) and save it in your theme folder as screenshot.png – this will help you switch to the theme in your WordPress presentation options. Here is my theme screen shot for this site.

Step Four

Begin to think of your coded HTML as seperate components – a header, a sidebar, a footer, and then content. And if you are just a little bit quick, you may have jumped ahead and seen files in the folder called just that – header.php, sidebar.php, footer.php – but what about content? Well content is the one thing that changes from page to page, while the others are static and don’t change. So really, content is index.php, archive.php, search.php, page.php, and single.php. These files have a little bit of explanation here.

Step Five

It’s a copy and paste job. Copy your header code that should be on every page and put it underneath all the code in header.php. Look at the code in header.php and see what makes sense for you to have in your own header code. Things like the php between the <title> tag and the call for the style sheet first come to mind. Do the same thing with your footer and sidebar.

Step Six

Read this article about “The Loop” and then open up index.php. You will see the code that loops through each entry pulling the values requested – the_time being the time, the_title being the title, and so on and so forth. Copy the HTML from your main page entry and paste it bellow <?php endwhile; ?>. Then copy the different php snippets from the old HTML loop that you need and when finished delete the old entry html from just below <?php while (have_posts()) : the_post(); ?> down to <?php endwhile; ?> and place your new HTML loop in there.

If there is any other structural stuff that belongs on this page that isn’t in the header, footer, or sidebar add it in the necessary places as well. Continue to flush out the other content pages using the same methods.

Step Seven

Upload your new WordPress theme folder into the wp-content/themes/ directory and login into WordPress. If you have done everything right – you should see your screenshot and name of your theme. Click it and then visit your website – you should see your theme on the site.

That’s it!

Conclusion

Building a WordPress is simple if you already know how to design, code HTML and CSS, and FTP into sites. In this post I outlined my own process – not any sort of “official” WordPress process. I’m sure there are folks out there would think of my copying and pasting of the folder and subsequent files as “cheating.” Honestly though, when I am working – I am shooting for efficiency and long-term maintainability. I know in building within the structure set forth by the good folks at WordPress – I will be ok well into the future.

If you have questions about any of this – feel free to leave a comment and get a dialogue going, there is no such thing as a dumb question. This is meant to be a simple guide for those who are interested in WordPress or Web Design in general.

To take it one step further, we can also take sed RSS feed and pull it into our own blogs/sites. Kind of like the homepage of this site. What one sees is the parsing of three RSS feeds, one for del.icio.us, flickr, and last.fm.

I added those on there because I wanted all the time I spend doing other things on the web to reflect on my blog. Yet, in doing so, I ALMOST had to sacrifice something else that was very important to me. Standards.

The first issue I dealt with was the RSS feed for flickr. For some reason it isn’t looked at as valid markup. I had to run it through a feedburner account to pull it into my blog and not break my site. In the end it was a good solution as I now have big brother tracking on everyone that subscribes to my photos. And to be honest, I think it is just Magpie RSS that is pinging the feed to update it. I can’t imagine subscribing to someone’s photos unless…perhaps I was stalking them.

The second issue was with my del.icio.us feed. This is a perpetual problem and something I have to take extra steps to avoid. I use Firefox and I have the del.icio.us tag extension installed. It’s way too easy to tag a site and add it to my growing list of bookmarks. The problem lies within the sites I tag and their title tags. For some reason people throw funky characters into their tags. Why can’t they change their ways to suit me???? (-; Anyhow, when I bookmark something, I am on the lookout for renegade characters that break my validation.

I really only had those two issues in regards to feeds, yet I have another one that has been a thorn in my side for building blogs. That thorn is YouTube embed code. Now don’t get me wrong, I love YouTube and have been sucked in plenty of times surfing between clips. But their code they are passing to the millions of users to post on blogs and myspace breaks validation. I could care less about MySpace (67 html validation errors on my page) but YouTube I expect more from.

In all honesty there is more to the problem than just YouTube. It’s flash, it’s validation, it’s Internet Explorer, and on and on. In the end I cannot build a blog for a client that is completely standards compliant if they want to paste in YouTube code. It’s already a stretch with them entering in their own content.

What’s the point to this rant? James Hall is the point. James Hall is the solution.

For anyone else who wants their site standards compliant AND wants YouTube, a solution is out there.

–

Update: To illustrate another case of external content breaking my validation, Lupe Fiasco coming in through last.fm has a problem. Looks like I am going to have to stop listening to that album.

In my last post I talked about the entry fee for getting a job at my company required being up to par on web standards. Yes, we recommend one uses firefox with the web developer toolbar as well. Oh and there is also the useful Measure-It! extension. With all of these tools in one’s pocket, the question becomes, why haven’t some people caught on?

Microsoft Internet Explorer. This would require an entry in itself.

Perhaps one day all sites will be coded in valid css/xhtml and the validation links at the bottom will seem innappropriate and unecessary. Either way, Microsoft has a lot of making up to do to those of us in the web space. Perhaps they could make their browser standards compliant, and make something as appropriate as validation links go away.

Or they will just be Microsoft, and we will have to fight the good fight for the long run. I hope they come around.

(Checking out IE7 soon, somewhere I saw a dual install for both IE6 and IE7, so what happens to my IETab extension??)

The reason I bring this up is the fact I am becoming a bit of a code snob. Ugly, sloppy code really grosses me out. There was some code pulled up on a projector today and I had to turn away. So when did this happen for me?

I was introduced to the wonderful world of CSS about two years ago now. Before that I was building sites like everyone else, tables and spacer gifs. I can completely understand the necessity of those means before CSS came along, but now I can’t bear to work like that. Seperating design and content is essential to get past any sort of design changes one has to a multipaged website.

Thought One – Applications.

One thing that I have come to find over the last few years of being a “professional,” is that I am not so good at writing my own application code (working on that with rails, but more on that another time), but I am good at changing the look and feel of others. So what does that make me? A good skinner. And I am cool with that by the way.

So when I come across an application I want to work with, a blog or cms platform, and the code it generates is dated and ugly, I become a bit turned off to the whole idea of spending the time to learn it. Case in point: Joomla!. What a great CMS for cookie cutter websites. They have tables and other useless stuff so engrossed throughout the code, the idea of building beautiful, unique, thoughtful websites becomes extremely difficult. After two weeks of pulling my hair out trying to figure out Joomla! and delete all of it’s unecessary html code in the php, I switched to CMS Made Simple. These guys have it figured out and I was able to get done in one day, what I hadn’t been able to do in two weeks with Joomla!. Mixing html inside of your application code is a bad idea. And if you want to attract us standards guys, mixing table cells and server-side scripting is a especially bad idea.

Now all I have to do is break it to the guy who writes my checks that I had to make a call.

Thought Two – Consistency

I pat myself on the back everytime I see a valid xhtml or css link on someone’s page. It will be ok, eventually all of us will get in line with this whole thing. What the standards link tells me is this: not only will I be able to go in and change the whole page around with CSS, but they probably aren’t using tables and spacer gifs either.

In fact, I would go as far as saying their code probably looks a lot like mine. Begins with: <div id=”container”> and goes through the list of identifying everything with id’s and classes? What’s the point? The point is when the majority of the web developers start going down this path, a very important consistency emerges.

Thought Three – The Future

I have gone from hacking websites together to building websites built for the future. This entails a few things.

1) Build websites with the intention of coming back and editing it in five years. Yes, some things will change, but there will be a consistency to your work.

2) Build websites with the intention that someone will be editing it in five years. I can’t stress this enough. I work at an agency and our team just got on the same page about building sites (file structures, naming schemes, css practices, html practices) the same. This doesn’t have to be one man to rule them all, it just needs to pull from everyone’s best practices.

3) Look at the greats in this field; Shaun Inman, John Hicks, Veerle Pieters, and Cameron Moll only to name a few, see how they code, see how they design. My first portfolio pulled heavily from John Hicks, as to say, I killed cats or something to that extent. These people really have a feel for what is good.

Although this list is published, I may add more to it in the future, just because this stuff never ends.

Before the list though, I would like to thank Steve James over at epicunion for keeping me up to date on the good, the bad, and the ugly of web standards. Everytime I think I can teach him something, he ends up teaching me.

Please comment on this post if you have any feelings towards any of it.

1. The “target” attribute I feel the target=”_blank” function isn’t as obsolete as the w3c would have us believe. In all practical respects, most websites do not want to direct traffic away from their sites. While it is nice I have a flickr picture gallery feed coming to my site, I do not want to send all of my traffic from my site to flickr. With tab browsing becoming the norm (with the release of IE7 of course) links opening new tabs is not the annoyance that it once may have been. With that being said, here is the first hoop:

Strict XHTML does not allow the “target” attribute within a tags. Now, by no means did I come up with this solution, I found it here. There are far brighter folks than me kickin it on the web and I can only express my gratitude for them to share their solutions. Moving on..

Instead of using target=”_blank” one must use rel=”external”. But the fun doesn’t stop there, one also needs to include a JavaScript file on every page that looks like this.

function externalLinks() {
if (!document.getElementsByTagName) return;
var anchors = document.getElementsByTagName("a");
for (var i=0; i<anchors.length; i++) {
var anchor = anchors[i];
if (anchor.getAttribute("href") &&
anchor.getAttribute("rel") == "external")
anchor.target = "_blank";
}
}
window.onload = externalLinks;

All in all, it’s a great solution to a problem, that I feel, shouldn’t exist. But that’s just me.

2. Feeds Compliance – I listed my feeds on every page with contact information. When I went to validate my code, I found the W3C really didn’t like my flickr feed. The workaround:

I opened a FeedBurner account and ran my feeds through them. Upon which, I also now have tracking for my RSS feeds. I am still not totally convinced of the accuracy of FeedBurner, but I really appreciate their effort to provide this as a limited service for free.

This particular problem brings up the lack of effort from many of the cool “web 2.0″ companies to bring their services up to standards. While I understand they are focused on their service, it would be nice to see a little more love shown to the standards community.

3. YouTube Embed – Yet to be solved. James over at The Good Harvest, is tackling the problem. More flash wizardry to come.

4 Building good clean XHTML and CSS – More important than anything I have mentioned is to practice good coding. Think about what you are doing before you do it. I am not claiming to have perfect code, but I am claiming that is what I stive for. If you care to take it there, I am more than open to suggestions on my own coding practices. Always learning…