thinking: relating- celebrating :-)

One of the key value points that WordPress offers to its users is a seemingly inexhaustible function library of mini – applications or programming shortcuts.  There is now such a huge supply of plugin extensions that making sense of which WordPress plugins to use can be a challenge for new and even experienced users.

These plugins allow you to extend the core system in any direction without needing to be a coder or rocket scientist. This puts WordPress firmly into the hands of business managers rather than leaving it with the tech staff.

A huge part of the appeal of WordPress for me is the ability to be able to prototype various sections and leverage scripts that have already been road tested by the community.

Imagine you want a slideshow for photos (NextGEN Gallery 1,865,952 times), an events manager, an ecommerce system (WP e-Commerce) or a snip of code that automatically shows related content  (Yet Another Related Posts PluginDownloaded 465,626 times)  just to highlight some of the popular and well known ones.

Chances are all of those functional needs and questions have been considered  before.  In fact for some of the more popular tasks it may be as simple as using the add new plugin button inside the dashboard and searching then auto installing something that looks to be in the right user function zone. However many plugin developers don’t appear to do even the most simple of searches before they re-vinvent the wheel by doing yet another slightly different plugin.

So far so good but at last count on the main plugin repository there were 9,096 PLUGINS, 84,520,662 DOWNLOADS, AND COUNTING. This makes the search and selection process somewhat more complicated.

Having a vast list of plugins is a good quick way of testing out concepts and approaches. what it won’t say there on the counter page is that the vast percentage of plugins were developed to “scratch an itch” by technical problem solvers.

This means that maybe as many as 80% of the plugins have somewhat limited value to most users. And depending on the scale of your site they may very well be ideal for what you need but if you have a larger site your developers might want to look more closely at custom building functions into your theme or custom plugins – whichever is most optimal for fast performance and security.

There is an intriguing discussion over on Alex Kings blog about whether WordPress is a platform or a product As Alex says

“Another challenge with WordPress as a product engagements is managing expectations in regards to 3rd-party plugins and themes. In most cases we will try to budget time to code-review any plugin we include in one of our builds to make sure it is secure and will scale to client’s needs. This can result in us advising the client not to use the plugin or theme they had intended, or require additional time and cost investment in addressing shortcomings of the plugin or theme. There is also the issue of fixing functional bugs in these plugins.

People who view WordPress as a product will typically expect that any collection of plugins will work elegantly and seemlessly together. Rarely is this the case. Even between experienced WordPress developers you will find preferences for different implementation approaches. Mix in plugins written by developers that do not have extensive WordPress experience and you can end up trying to weld the transmission from a 57 Chevy to a tricycle.”

It is not always obvious when there are problems with a particular plugin. Many will work but may clash with certain other plugins or even theme functions and this weakness might need lots of extra testing if the site is going to be a large one.

If it is a small site, you may be able to just switch out to an alternative plugin and most of the time that will work.

Here are some of the more popular plugins. It is a fairly random set of 10 but based on the hundreds of sites that I’ve seen and worked on my guess is that most sites would have at least half of these and combinations of the next 200 or so listed plugins.

• Google XML Sitemaps Downloaded 3,659,339 times
• WPtouch iPhone Theme Downloaded 796,224 times
• All in One SEO Pack Downloaded 4,991,211 times
• Fast and Secure Contact Form Downloaded 269,918 times
• Contact Form 7 Downloaded 1,851,559 times
• NextGEN Gallery Downloaded 1,865,952 times
• WP Super Cache Downloaded 1,197,587
• Google Analyticator Downloaded 854,535
• Twitter Tools Downloaded 532,607
• Sociable Downloads 993,790

You might say that the top 10% of plugins follow a power law distribution, the true number I would guess is much lower / maybe 3% of the total 9,000 or so. The other 97% of plugins have a very long tail distribution.

Your search choices are as follows. The other variables are voter approval (highest rated) by version and general age of the plugin including how many times (Recently Updated) the plugin has been updated. Raw downloads are indicative but if the plugin is updated 6 times per year you may want to divide that number by 6 to get a better idea of relative popularity.

Sort by Relevance, Highest Rated , Newest , Recently Updated, Most Popular

Unfortunately the search filter process is not as good as just using a Google site search. For example looking for SEO plugins as below.

site:http://wordpress.org/extend/plugins/ SEO

If you search around you can find various top plugin lists that might also be a useful guide.

Finally some of the differences between plugins simply come down to the programmers preferences and arguments about quality of programming code are eternal. They remind me of the Paul Grahams Blub analogy

“Blub is a hypothetical programming language; it is an average programming language “[r]ight in the middle of the abstractness continuum. It is not the most powerful language, but it is more powerful than Cobol or machine language.”. It was used by Graham to illustrate a comparison of power between programming languages that goes beyond Turing completeness, and more specifically, to illustrate the difficulty of comparing a programming language one knows to one that one does not^[14]:”

Sometimes all the programmers are wrong or all right but ask an engineer and the better measurement might be optimal speed of loading or security considerations.

So what is the best way of making sense of WordPress plugins?  I would say the worst person to ask about this is a programmer.

In the past I have run plugin workshops for individual clients to review what they have in context and compared to what other similar sites might be doing.

However if you are a programmer I’d have a good long look at the Blair Williams presentation called : Pimp Your WordPress Plugin or if you’d rather check the video for Pimp Your Plugin over at WordCampTV. Also worth checking out is another recent video by Jonathan Dingman: Essential WordPress Plugins & Optimization should be standard issue for new installs and methods for squeezing the best performance out of your new WordPress blog.

Help in picking plugins is very much on the way. Later this year instead of a version update for WordPress itself the community is being encouraged to work together on a number of the top plugins. These will be identified as core plugins.

One of the best ways of improving code quality is to engage multiple programmers who can argue about what is best and hopefully peer review some of this the same way that WordPress code is reviewed. This is an excellent move in my opinion and I look forward to seeing how it goes later this year.

Wellingtonian Dan Milward and myself got to WordCampSF back on May 1 in San Francisco. There were 750+ people there and for most of the day there were two streams running with more technical sessions downstairs and general – more road-map style sessions upstairs.

San Francisco WordCamp is always a bit more open ended than the other  60 WordCamps now held everywhere around the world which are more locally tuned. Half of the attendees are from out of town and there are quite a few connectors from all around the world including us.

The sessions for San Francisco included some forward thinking and a few surprises.

First up I went to Niall Kennedy‘s session on writing plugins. Niall is the VideoPress guy and made the excellent point that using VideoPress for running video on your website is much easier and better performance than you can get from your regular shared hosting package.

I’m hoping that Scott Berkun‘s session video will be up soon as I suspect it was also very good.

It is always a balancing act programming a WordCamp since the actual projects are often more interesting than the platform but attendees expect elements of both and there are always a range of attendees skills and experience.

This can lead to too much tech and not enough project wow and vice versa.

The full schedule for SF 2010 is over here. One of the surprising things at WordCamp SF was the high profile of Microsoft who have clearly noticed WordPress usage over the past few years, as have Google and Salesforce.

Daniel Cook presented on how to turn Microsoft Office into a game. I opted for an alternative session from Vanessa Fox on audience engagement which truthfully wasn’t that engaging.

John Ford then gave a presentation on “Living with Our Computers… and Keeping it Healthy” which sure enough emptied the room. Who wants to hear the words healthy and computers in the same sentence. Point being that our modern lifestyle of sitting down most of the day is really tough on our physiology.

To be fair Jane Wells was talking on “User Experience the WordPress Way” at the same time which I would have been at if I hadn’t already seen a presentation by her on that.

After an excellent BBQ lunch we then filed in to a jam packed theatre for  the Matt Mullenweg keynote.

An extended version of this post is over at WordCamp NZ including Matt’s video …

Helping clients plan to get the best out of their WordPress sites is something that I really enjoy.

Having a theme library loaded and switching between them for instant changes to “the look and feel” is a fun moment in the discovery planning process.

For non – WordPress users a Theme is effectively a website design “skin” that overlays the content. It works by providing a filtered view of the text (XML) content.

For more technical users we are talking about combination of stylesheet (CSS) and some core function code (pages, posts, comments) which is written in PHP.

At the visual level this is fairly easy. For example if the client has an existing site structure we would look to match the colours, and general look and feel with a similar theme that has 1 , 2 or 3 columns. We may be also looking for menu layouts, header functions, plus sidebar and footer configurations.

Part of this is to understand the branding context and also if there is an existing format whether that should be kept if if there is a more optimal layout.

What we would then do is match fonts, include branded headers and other brand ID assets on a theme that was as close as possible but sometimes it can be quicker to build a new theme that to find one that matches up.

Way back in ’97 there was programme called Net Objects which did something similar in packaging various components together with a set of styles and saved all of the information into an object called a NOD probably some kind of early XML file.

The great thing about that programme was the user interface for applying “styles” really simplified the menu and navigation processes by including all the image icons and button type files along with it.  I used that for many years for fast prototyping of sites and to replace PowerPoint for presentations as it was faster and easier.

Fast forward 9 years to 2006 – WordPress was coming of age with newer more visual releases although from memory I think that Joomla had a bigger range of theme like styles at the time.

But by late 2006 when I started this blog there were some great themes around for WordPress and having used lots of content management systems before I was ready to try something a bit more open ended.

Where to Search for WordPress Themes?

The best place to start is WordPress.org theme library. This is because the 800+ themes there have been sorted into some kind of taxonomy and at least partly vetted by WordPress developers and enthusiasts. This is important for two reasons.

1. Some themes have hidden code in them which might be advertising or worse. See the theme authors guidelines which aims to prevent  “hidden, paid or sponsored links in the theme. Links back to the author’s site are fine.”
2. More importantly this library provides a structure for searching where you can filter searches by types such as fixed or variable width, number of columns, main colour, features and subject which are called theme tags.

This generally provides a  range of visual templates and ideas for a wider search in other theme libraries.

In the past the searches haven’t been very precise possibly because some of these tag and taxonomy rules haven’t been fully applied and because some theme authors game the system by loading up on the equivalent of all possible keywords.

Frameworks and Coding Considerations

Having worked with dozens of themes now it is clear that under the skin many of them can be traced back to earlier building block models or frameworks.

Every install of WordPress comes with a default theme sometimes called Kubrick and that one along with K2 and others. More recent core themes are Carrington, Thematic and Sandbox.

Thematic  describes its Theme as

“a free, open-source, highly extensible, search-engine optimized WordPress Theme Framework featuring 13 widget-ready areas, grid-based layout samples, styling for popular plugins, and a whole community behind it.”

Another excellent theme  is Thesis. One of the first serious themes that I learned from was Chris Peasons Cutline series.

As Chris puts it- here are 5 more reasons to look deeper into the code and overall framework of each theme to save hassle later on.

“As a serial site developer and blogger, I’ve found that the most valuable tool one can have is a refined template system that solves fundamental development, design, and publishing problems, including:

1. SEO and careful attention to in-site link equity
2. an “em” -based approach to element sizing (pixels are nice, but “ems” are by far the most accessible – and therefore the best – choice)
3. polished typography with finely-tuned geometrics for maximum legibility
4. an aesthetically pleasing layout that favors usability and clarity over extravagant presentation
5. forward-compatibility (I like to call it futureproofing)”

Put more simply – picking a great theme now which has “good bones”and optimal features can save a lot of time later on. Some themes come paired with a series of plugins for say featured content and a number of themes come as a kind of half-way house with extensive theme options for those not so comfortable with stylesheets.

Theme options allow user to make changes to a style at a higher level by ticking an options or using other present menu re-combinations to make changes without needing to ever see the CSS code.

An example of this approach would be something like the Atahualpa which come with something like 300 “theme options” and personally a style sheet looks easy after that.

It is described as follows and the links below are tags that can be used for searching.

“Version 3.4 – Atahualpa is a WordPress/PHP/CSS Framework that lets you build your own unique, professional and browser-safe WordPress theme: 1-5 columns, fluid or fixed width, rotating header images and over 200 theme options. Tutorials, downloads and support at the BFA WP Forum

Tags: threaded-comments, theme-options, custom-header, custom-colors, flexible-width, white, sticky-post, translation-ready ”

StudioPress Themes offer another approach where a set of plugins have been pre bundled with a theme and page templates are somewhere closer to a magazine style format.

Magazine syle themes generally have a larger number of columns like a newspaper and would tend to have a category menu as well as featured content sections and even special video or audio panels.

In summary most clients start out looking for a particular look and feel but there are other more practical considerations which could benefit them by saving time and money if the selection criteria is deepened.

As a WordPress practitioner I would steer clients towards some of the other functional considerations like “does it play nice with key plugins ?” and is the structure fully transparent and robust for scaling up and working with other applications which will be the next frontier.

There are other considerations but perhaps you can write in with your comments and questions on what you think are most important when choosing a great WordPress Theme.