In case it needs reiterating, if you haven’t yet upgraded your WordPress websites to 2.8.4, or your Joomla websites to 1.5.14, upgrade your sites now.

For more details, refer to:

Lorelle on WP: Old WordPress Versions Under Attack
Weblog Tools Collection: Old WordPress version? Attack warning. Please upgrade!

and this link has the resources to help you clean up if you’ve already been hit:
Journey Etc: Wordpress Permalink & Rss problems

If you’ve been wary of upgrading because your favorite WordPress plugin has not been updated in the past two years, it’s worth taking a leap and disabling that plugin, then searching for a replacement for it after you’ve upgraded. If your concern is about a custom theme that might break in a few places after the upgrade, it’s easier to have a missing or misaligned section of your site for a few hours while you get help fixing some WP calls or some CSS definitions than it is to have to constantly guard the broken lock on an older version in order to keep your site’s look undisturbed.

This time, the upside for me is that now a handful of the sites that I manage are a little bit cleaner — I’ve gotten rid of some plugins that were either abandoned or not even being used on their respective sites anymore.

These are the folks behind the WordPress plugin Max Banner Ads and about 10 others, so if anyone is up to the task, in my opinion, this is the group to get the job done.

So far, they’ve taken on 4 plugins: Flash Fader, Easy Custom Fields, Post Format Control, and Access by Category.

Two of my major sites haven’t been upgraded because of concerns of older coding (calls) in custom themes. One was running WordPress 2.6.5, the other 2.5.1. After checking everything I could think of (or so I thought), I upgraded the 2.6.5 site to 2.7.1… there’s still a 2.8 compatibility question with one of the plugins.

Everything went well, at first. But within 12 hours, the server the site was on was experiencing high loads and crazy mysqld behavior. I resolved that, but the high loads were back within another 24 hours. I resolved the issue again, but the fact that a pattern of wacky behavior was emerging was a concern.

Taking another look at the active plugins on the recently upgraded site yielded a curiosity… an orphaned plugin with no recorded compatibility issues, good or bad. I say orphaned because the plugin was active, but never actively used, not that it’s development was abandoned. It’s usefulness for us was actually superceded by other tools.

Because it was never used, and it never impacted site performance (as in not one problem in the past 4 years), it was “out of sight, out of mind”… never upgraded either. Thus, the version in use was so old that it was not relevant to the things I had checked. But once the plugin was turned off, issues with the site and server disappeared within minutes.

So now I remind myself of one of the things I would on occassion remind other web peers of.

Charlie Jade was a wonderful, gritty, atmospheric detective noir series with a little scifi mixed in for flavor. It was a Canadian-South African co-production that aired in Canada, South Africa and the UK in 2005-2006, but wasn’t sold to the US market in time to garner funding for a second season. The series is available on Region 2 DVD, and when word came out that SciFi Channel would begin airing the series in reruns in June 2008, there was the chance that good ratings might lead to a domestic DVD release.

But any hopes for that Region 1 DVD release, and for an official soundtrack release, were dashed by the SciFi Channel’s sudden decision to change the broadcast time of the reruns they acquired from Friday nights at 9pm to Tuesday mornings at 3am… after only airing the first two episodes.

Before that happened though, series creator Robert Wertheimer had agreed to do a podcast of creator commentary for the show for FarPoint Media, and we had completed half of the commentaries before SciFi Channel made the broadcast change. After the switch, we still managed to get commentaries done for all but one episode (a pivotal one, unfortunately), and there’s going to be a complete, finite podcast series once the discussions and final interview are finished in July.

The site started out in May 2008 with the Dust theme, because it had the rotating banner I wanted to use, but it wasn’t until April 2009 when I chose to take the Studio Press Tubular theme and do some heavy modifications to get the look I wanted, including adding in that rotating banner code.

The base Tubular theme turned out to be a good fit, and easy to modify away from it’s intended feature video design to the sectioned layout I wanted on the front page, and the only changes I needed to make to existing posts was to use larger versions of the images and add custom fields for the front page thumbnails.

I got the colorful yet industrial feel I wanted in the theme, with most of the information that’s good for newcomers to Charlie Jade being easy enough to find to help get them started.

Site: Charlie Jade Verse
Theme: Tubular

I stuck with it even during 2008 when there were no updates and the support forums would vanish without warning, partially out of loyalty to a good plugin (and Dan’s a really nice guy), and partially out of regret for my role in the addition of the Podango functionality to podpress.

The challenge is not finding another podcasting plugin to replace podpress — there are several out there that work just fine. The challenge is converting existing posts from using Podpress to using whatever the new plugin might be.

The other drawback is most if not all of the podpress users in my little enclave like the builtin podpress statistics… the instant gratification of seeing that your new episode has been downloaded 500 times within the first half hour of posting it can be addictive, and getting someone started on using a different stats tool can lead to frustration on both ends.

So this is the start of an experiment. I’m going to be testing out Powerpress from RawVoice (mostly because it will work in tandem with Podpress), and see if I can’t figure out a way to easily convert shows that have between 100-300 previous episodes out there.

I’m not in that big a rush, though. The main reason for my wanting to investigate other podcastin plugin options vanished mysteriously about a month ago. My error logs were bloated with podpress statcounts errors (the infamous duplicate entry for key 1 error), literally for a couple of years… until sometime within the past 4-6 weeks.

The errors just went stopped happening. I’m not sure if they upgraded MySQL or PHP on the server where my sites are, or exactly what happened… I do know that the recent Podpress 8.8.1 maintenance release wasn’t what fixed things, because I’ve only installed it on one site, as a test.

But, I continue to explore.

I made the mistake of trusting that a full transfer worked, just because the sites and the data on the new location seemed fine. It wasn’t until a few days later, when trying to create new posts, that I uncovered a MySQL glitch.

I’m still not sure what exactly caused the glitch, but on three (possibly four) of the transferred sites, the auto-increment keys were munged, at the very least. At least, that’s the one thing that makes sense from the behavior.

What’s even more interesting is that I ended up taking a different approach to restoring each of the sites.

First site: empty all tables, and restore data from the seemingly incompatible backup. This cleared up the problem, but all of the options from the previous setup were gone, along with a bunch of stats. Not an optimal end solution.

Second site: in order to avoid loss of options and stats, I did a WordPress export of the current site to extract the data needed, then deleted the tables for terms, taxonomy and posts. I recreated those tables using the SQL statements from the Wordpress code from the same version as what was currently running. Once those tables were recreated, I ran the import of the WordPress XML file, and all seemed fine. Turns out the comments were still crunchy.

Third and fourth site: repeate of process for site two, including the comments table from the start.

There are times when all those years spent troubleshooting as a sysadmin pay off.

Lesson learned: never trust what Plesk tells you about your database backup. Go commando if you can (command line interface and mysldump are your safe bets).

Next, to play with DirectAdmin and see if it can give me the same control panel flexibility as Plesk without the same headaches.

The term “Deep Geeking” comes from a description of what we do when we fall into intense analysis and discussion of a topic; in this case, the source topic was Babylon 5 and what we do on the podcast: in-depth discussion of the episodes and the deeper topics of character and theme. I loved the name, and one of the show’s fans registered the domain for our later use. It was 2 years before I actually got around to setting up a website for it.

I liked the way mimbo2 did article images, and my experience with the way arthemia2 handled images made customizing how I wanted those images to appear using mimbo a fairly easy modification to incorporate as well.

I wasn’t 100% happy with my final look at the time, but it served a purpose.

Then along came mimbo3, and most of the nagging questions I had in my mind about some functionality had been addressed… but the part I liked the most about mimbo2 had been excised completely: article image management.

I may be in the minority, but I like Custom Fields for specific image usage and placement. I like having the option to use one image for a thumbnail and having a different image for an internal article image, if there is one. It’s more interesting, to me personally.

So my challenge was to do some extreme hacking on a child theme for mimbo3 to regain that Custom Field image usage that I adore, while keeping the rest of the layout the same. Since the child theme vintagegreen was close in style to the mimbo2 look I liked and wanted, that’s what I used for my base.

While mimbo3 and its use of child themes seemed as if it would be more work at first, needing to only make changes to a child theme changed my mind about using them… I had been ambivalent about them before, but now I see the upside to them. I don’t know if I’m completely sold on them yet, since I’m not the kind of person who’ll update a base theme frequently enough to need a better solution to keeping my customizations, but I can better appreciate the advantages of them now that I’ve had time to work with them.

Site: Deep Geeking
Theme: mimbo3 / vintagegreen (child theme)

For a little over two years, I had been keeping an eye out for a new blog-style theme to implement on the Kick-Ass Mystic Ninjas website. I’d always seen things that came close, but nothing ever struck me as being perfect or near-perfect for what I wanted to do, that matched the vague idea I had in mind for the site’s new look… until Streamline came out.

The front page article listing was exactly what I wanted to use for normal paging throughout the site, and the use of the larger icons with the article listings was the perfect match I’d been looking for.

I had to make a few modifications to get the front page layout style to act as the default article listing throughout the site, but the end result was exactly what I had in mind. Slight modifications to the banner were needed to make it fit the new theme sizing, but those were easily implemented by the graphics guru who’d done the original banner.

This one was one of the fastest and easiest customizations I’ve done, and the result was well worth the wait. But there was one odd drawback that I haven’t resolved yet… my favorite contact form stopped working.

I love WordPress-Ready Contact Form, from the Beast-Blog guys; I’ve used it on at least 10 different websites, and that’s in addition to at least another 10 non-WP sites that use the both the v2 and v3 standalone version of the form. But after I customized this Streamline install, the contact form ceased to work, and even more disturbing was that the Configuration admin page for the form stopped loading.

So for the time being I’ve switched to Contact Form 7, until I get the time to sit down and figure that particular failure out… or post something on the Studio Press support forums and ask for a set of fresh eyes to spot what might be the cause of this conflict, because I haven’t tracked it down yet.

Site: Kick-Ass Mystic Ninjas
Theme: Streamline

I think arthemia and arthemia2 were the first magazine style WordPress themes I came across that caused me to imagine a variety of customized configurations to implement on several websites I had that had been languishing and were badly in need of being transformed.

Since 1996, I’d had two very popular websites on a single domain. In 2008, I decided it was time to clear up the focus of those sites, and finally update them from the hand-rolled HTML sites they’d been stuck on since 2003.

When it came to redesigning wildhorse.com, I wanted to refocus the site to my original idea: equine art and artists, news about wild horse preservation and about equine rescue efforts. That meant that I needed to split off the very popular Miami Vice related content off onto it’s own site, and could not put that task off any longer.

First up was redesigning wildhorse.com to be The Wild Horse. Copying the old HTML listings and converting them to WordPress articles was the easy part, but I spent about a week rearranging the categories several times, just because I wasn’t completely happy with the old designations I’d used. The new site has been up for months, and I’m still tweaking. Redoing the images, and going through all of the old artists links to see what sites are still working and what sites need to be removed from the artists directory is an ongoing task at this point, but the new wildhorse.com launched on August 1, 2008, and has had a pretty decent reception. It’s still going to take a while to get the traffic back up to where it was back in 2003-2004, but that’s for another article.

Next up was creating Miami Vice Chronicles. Setting up the site was a snap, and customizing Arthemia Free was a lot easier and a lot more fun than I’d originally anticipated. But it took nearly two months to convert the several hundred individually hand-rolled HTML pages for the Episode Guide, the Music Guide, the cast information, and the old news and articles into Wordpress articles! Add in the time it took to rework or rescan many of the images (I’d used lower resolution images back in the days before broadband), and the project I’d started in September was finally ready to launch in December. Being the Miami Vice fan that I am, I chose to launch the new site on Dec 15, which is Don Johnson’s birthday.

Since there isn’t a ton of new Miami Vice related news coming out these days, blogging is occasional, but it’s the episode guide and music guide that are the big draw, and I’m hoping the new design and layout of the image galleries and listings still appeal to both the casual and die-hard fan.

Note: the links to both arthemia and arthemia2 download the same theme code, but the first link contains some helpful instructions that are not included in the second link, thus the reason for including both links.

Site: The Wild Horse
Site: Miami Vice Chronicles
Theme: arthemia2 (since renamed Arthemia Free)

Since many people before me already grumbled and complained about the autosave and revisions updates that initially caused more problems than they resolved, and since I was able to find a solution to it quickly, it wasn’t so big of an issue with my sites. For anyone not familiar with the problem, let’s just say that revisions and podpress do not play well together, and since podcasts are an integral part of my other work, I needed revisions to die quietly, like so many others also did.

The other glaring exception to my love of WordPress was the addition of the “feature” that automatically creates thumbnails from images uploaded through the WordPress media manager. I haven’t noticed any public grumbling about that feature, so I may be alone on this one.

Several of the sites I maintain have a lot of images accompanying them, and they have or use custom image thumbnails that we create for the category and front page listings. The fun part is that we use different images for the thumbnails than we do for the image in the body of the articles, and I think it makes things looks better.

The thumbnails also appear in two different sizes, depending on if it’s being viewed from the main page or the category page, and that’s done with timthumb on one site, and the use of CSS/HTML on another.

The best part about using both methods for the resizing meant I didn’t have to worry about needing several versions of a single image in different sizes… upload one image, and the scripting in the right section of the theme takes care of the resizing where necessary.

Somewhere along the line in the growth of WordPress, the automatic creation of thumbnails for images was added. It happened after I upgraded a few sites in either January or February 2007, so I’m guessing that had to have been included in the release Wordpress 2.1.

I never used the images created from the new thumbnails feature, so I never paid much attention to it… until version 2.5.1 came out.

First, I noticed when the mysqldump backups of that site grew from 20Mb to 50Mb far faster than it had grown from 10Mb to 20Mb. At first, I thought it was from the rapid increase in comments the site was receiving, from a few hundred to a few thousand a week (along with the crazy increase in spam comment attacks… thank you Akismet!)

But I also noticed that the site was taking up a lot more space on the server, and the only explanation for that was an increase in uploaded media. At first, I thought we’d added a lot more video to the site, but I knew it wasn’t enough to account for such a big jump.

When I realized that the month folders in the uploads directory were now much larger in size than a few months before, I poked around. When I noticed that there were now 3 “versions” of every image that was uploaded for an article, instead of the previous original plus generated thumbnail, I was not amused.

I honestly don’t need WordPress to automatically create two thumbnails for every image I upload. Instead, what I need to do these days is go through my uploads every few months and manually delete those thumbnails. That’s cleaning up a couple hundred files files per folder that I didn’t need created in the first place.

Applications shouldn’t have unintended side effects that cause more work, especially not ones as smart and slick as WordPress has become.

There had been an uproar over how many problems with increase in database size and other problems caused by the autosave and the revisions additions that were introduced in WP 2.6, and thank goodness the community came up with immediate workarounds in the forms of adding directives to wp-config.php, and more recently the Post Control plugin.

What I’d really like to see added to WordPress now is the option to turn that thumbnails “feature” off entirely if I want to. Because on almost all of the sites I build and maintain, I really really want to.

I’m not sure yet how much extra space, if any, the references to the thumbnails take up, and my initial thought that the added thumbnails were directly contributing to the huge DB increase might be off base, but I wouldn’t mind seeing if there’d be a way to delete the references to the thumbnails from the database, too.

It’s not vital, but I’d also settle for a quick and dirty mysql command I could run to clean that up, if they’re in there. The database for that one major site grows about 10-12Mb per month. If getting rid of references to thumbnails I don’t use or need can slow that a bit, I’m willing to figure out a way to get that done.

I have an idea in my head about how I want some of my sites to look like, but I’ve never seen one theme that had all the components or layout areas that I’d imagined. I’d always want one idea from one theme, and another idea from a second theme, and a tweak of an idea from yet another theme.

Isn’t that’s how Dr. Frankenstein started out… or was that Dr. Moreau…

I’ve never had the inclination to write a theme from scratch… as far as I’m concerned, I’ve still got a lot to learn about CSS. But I can read the code, and I know what to change and how to change it to get the look that I want. The same thing goes for tweaking the PHP framework of the themes.

Over the past six months or so, my favorites for customizing have been arthemia2 by Michael Jubel, and the Revolution Two themes (which are no longer free, alas). I had fun with Mimbo2, with both Branford Magazine and Wynton Magazine, and also with the original Revolution themes that were purchased for the massive number of websites that fall under my other creative hat (more on that later).

I ended up using arthemia2 as a base on three consecutive sites. I found myself taking pieces from Revolution Two themes and plugging them into heavily customized Original Revolution themes. I’ve mixed and matched pieces from different themes to get close to a look that I could see in my mind, and sometimes it worked, and other times I had to scrap it and begin again. And in one case, my graphics design weakness became a challenge to sidestep.

But I’m going to continue to see what pieces from what themes I can mix and match into what new concoctions.

I may turn this into a series of what I customized and why, since there were so many that have fallen under the code tweaking knife.

When you’re working on a client’s site, discovering that two of your plugins don’t play nicely in the same sandbox can be exasperating. But figuring out which ones are butting heads, and then testing the combination over and over to make sure you’ve got the right combination of culprits can be fun.

That’s really why we have test sites, to see exactly what we can break, and if need be, how many times we can break it!

In this case, discovering that the Featured Content Gallery plugin from the gang over at RevolutionTwo falls down hard when you install the WP e-commerce plugin was amusing at first. When it turned out that the only way to untangle the mess is to manually dig out all of the tables from the WP database because the “Uninstall” feature in WP e-commerce doesn’t work at all, it was mildly annoying.

When further digging led to the knowledge that they are incompatible, and info at the RevTwo forums suggested that a workaround isn’t coming any time soon, it was very disappointing.

Two sites I’m putting together were initially going to use Featured Content Gallery, and third is using Smooth Gallery, which FCG is based on. WP e-commerce breaks both of them, and breaks them hard.

Needing to find two different solutions wasn’t something I had planned on needing to do… but one of the FCG sites didn’t really need to use that plugin, so putting WP e-commerce on there wouldn’t be a problem. When my other client is ready, I’ll be buying the Shopp plugin for her site, and if that plugin does the trick, that’ll be my second alternative.

Suffice to say that based on what seems like a daily interaction with Wordpress over the past 4 years, and complete and constant immersion in it over the past 2 years, I’ve learned quite a bit about how to get done what I want to get done in WordPress. That, combined with an extensive number of years experience as a Unix sysadmin playing with websites and open source apps, and for the first time in a long time, I started having fun with creating websites again.

Over the past two years, I’ve also seen some of the wonderful things very talented coders have been able to do with WordPress, and especially with some amazing advances in the design of themes, and the number of high quality themes and plugins now available.

I’ve also chatted with some of those coders, and inevitably, the good ones get so busy that they can’t offer their support or services as frequently as they previously had, not even for paid projects.

The thought of that stayed in the back of my mind, and after churning out what I thought were 4 pretty nice websites (based on the feedback I began receiving for them) it seemed like a good idea to try my hand at offering my experience for sale to those who could use it. I have and will continue to offer help and suggestions freely, when possible, and I believe there can be a happy balance between the two options.

I’m not a graphics designer… I know enough Photoshop to be dangerous, and that’s about it. But what I do have is a unique set of skills acquired over a long career, skills that…

(Yes, I love movies, and books and television as well… why do you ask?)

that now make it possible for me to provide different types of solutions for different scenarios, and right now, WordPress has become my first choice for a solution base to start from. If there’s a need for a lot of graphics work, that’s when I turn to friends who are gifted in that arena, or spend a lot of time browsing stock photography websites.

This site will be a place to for me to write down some of my experiences and thoughts about website options, and a good place to reach me if you’d be interested in my assistance with a project or some troubleshooting. My hope is that any tips and tricks I’ve learned and keep learning can help out someone in need of the same trick someday.

Be seeing ya!