I seem to have a plethora of blogs now running. There is my personal blog which you are reading. I also have a baseball blog – Diary of a Diehard and a photo blog – 1 Photo 2 Share where I post a single picture per day. Until recently I was running all of these through Movable Type but as I’ve documented previously I’m moving all the blogs to WordPress.

The text-based blogs were relatively simple to move. There are migration scripts to manage the entries allowing you to export the data from the Movable Type databases and import the data into WordPress database formats. When it came time to migrate the photo blog, it was anything but trivial.

Under Movable Type I had to tweak the template system to support images rather than text as the only information. I further had problems with the commenting system causing me to turn off that functionality somewhere along the path. To make a long story short, the data was not the cleanest I have ever seen and the old adage of “garbage in, garbage out” was never more accurate than it was for this blog.

It took me several days of data manipulation of the export files to get them into any kind of shape to be imported into WordPress. The question still remained, would WordPress handle the imbedded image files any better than Movable Type.

After an initial failure of just loading the images directly I set out looking for a possible WordPress extension that would make creating a photo blog easier. Although I found several extensions claiming they managed images, there was one plug-in that seemed to come up in each conversation.

Yet Another PhotoBlog written by J.P. Jarolim was exactly what I was looking for. Installation of the plug-in used the standard WordPress administration screen. Once installed, the plug-in adds a new page to the administration tool under settings.

Within this page you can define how you want the plug-in to manage the images. The configuration is quite powerful. You can have the system collect image information from the EXIF data and use that as the post publish time and date.

You can have the system tag photos based on EXIF data contained in the pictures. If you have ImageMagick installed on your server, Yet Another PhotoBlog will use that to create thumbnails of your pictures.

The plug-in has a fairly robust feed section where you can define RSS information for your images as well as using Yahoo Media RSS. You can adapt the plug-in to display your photos in various ways and can also include custom CSS tags to get the look of the images just right on your page.

Once installed, Yet Another PhotoBlog will add an image box to the post entry screen where you can upload an image to be included in a post. From there the plug-in does the rest showing the image in its appropriate size.

Overall the plug-in has worked flawlessly. It has make it quite easy to add new posts containing images. Unfortunately the way it manages images is not compatible with how Movable Type did it meaning after importing the data from the old system there was a tremendous amount of work to be done before it was ready for publishing.

If you are starting anew building a photo blog, this is a very good plug-in. If you are migrating from another platform there is a lot of pre-work but in the end it is worth the trouble. The plug-in is actively being developed with new versions coming out regularly to fix bugs or add new functionality. If you’ve always wanted a photo blog but weren’t sure how to do it, take a look at Yet Another PhotoBlog, you’ll be glad you did.

October 1, 2009
Recently I wrote an entry about changing blog software for the various blogs that I use. I had long been a Movable Type user but decided to change my blogs to use WordPress. In the article I listed several reasons why I made that decision and offered to provide details on how the transition went. Well that was nearly three months ago and I realized that I never went back to the subject. Since I am sure everyone has been on the edge of their seats waiting to hear how it ended I figured I owed it to you to revisit that decision and give you some details on how it went.

I have to admit I was more than a little nervous migrating my blogs. Some of these blogs I have been writing on for over a decade so I had a lot of entries and I wasn’t exactly sure how successful I would be moving the data from one system to the other.

From the Movable Type administration screen I exported all of the entries and saved the file to my desktop. I then backed up the blog domain and with a deep breath I deleted the Movable Entry files and databases from the domain. I felt about as alone as a boy scout stationed in Antarctica. Ok I have no idea what that means but I can tell you I was feeling pretty vulnerable at that moment.

The installation of WordPress was very straightforward and fairly easy. At the conclusion of the installation procedure I was left with a pristine yet somewhat empty WordPress blog. On the surface it was very pretty but my stomach was filled with butterflies as I began the process of trying to migrate the data from Movable Type.

Given the large market share that Movable Type has had in the blogging community it was no surprise that WordPress came with a migration tool specifically for transferring data from Movable Type into WordPress.

I followed the instructions of the migration tool pointing it to my exported Movable Type file and after what seemed like an eternity (it was really just a minute or so but when it is all of your data it just feels like an eternity) the system came back notifying me that the migration was successful.

I went to the entry list of the administration tool and sure enough it looked as though all of my data had been properly migrated. I launched the site and the entries showed just as they did in Movable Type. I flipped through several entries and tried the online search and everything seemed to be working normally.

The migration tool in WordPress worked perfectly. I was not completely on WordPress for my blogs. Although the migration was successful, it was not without its flaws. The archive data structure of Movable Type is very different than that of WordPress. As a result, any link in a post that was migrated no longer pointed to the permalink properly. I was able to adjust the directory structure under WordPress but it still did not match exactly to what I had under Movable Type. I was left with the task of going into each entry and manually adjust the links in order to point to the new location in WordPress. It has been a pain but I am not sure what else I could have done differently to make this not happen.

The WordPress environment has been very stable over the past three months. I am diligent at maintaining the latest version of the software so as new point releases of WordPress have come out I have upgraded. The upside of this is that I haven’t had the security issues that others have reported when using older versions of WordPress.

During this trial period I have accumulated several plug-ins to implement additional functionality into the WordPress environment. Over the next several days I will devote a blog post to each of these plug-ins. Perhaps you are looking for similar functionality within your blog, if so these plug-ins may work for you as well.

So far I have been very happy with the migration and subsequent management of WordPress as a blogging platform. I would not hesitate to recommend this software to anyone looking for a stable software base on which to build a blog.

After nearly three years of using Movable Type as the engine for my personal blog I have decided to change direction. It is not that I disliked Movable Type or that I found it incapable of doing the job. It just seemed like whenever I needed to change something it took longer for me to do than it really should. Although I am a card-carrying geek that doesn’t mean I want to spend all of my time learning and relearning how to get something done.

If I were to pinpoint one aspect of Movable Type that caused me angst it would have to be the template system. When I first started with the software it was the feature that drew me to using Movable Type. I loved the idea of creating templates to make my blog just a little bit unique.

What I did not fully comprehend was what impact that would have in the long run. Whenever Six Apart would introduce a new version of Movable Type, they would upgrade the default templates. I took that to mean that my site would immediately benefit from the new features built into the templates. Unfortunately I was wrong in my assumption.

In order for me to take advantage of the new templates, I would have to incorporate the changes into the “theme” that I was using. Initially I had selected a theme that would give me the look and feel I wanted. As my needs and ideas changed I would “tweak” the theme to include a new feature here and there. The result was a theme that worked for me but now deviated dramatically from the default templates.

This deviation meant that I was unable to take advantage of the newer default templates without having to do a full rewrite incorporating my changes into the new templates. That wouldn’t be so bad except I had to do that with each new version. I soon found that I was spending more time managing templates than I was actually writing which is never a good thing.

So after three years of this never ending cycle I came to a cross roads of deciding whether to again update the templates to my blog or find an alternative. Since I had a few days to kill with my family out of town I chose to move to a different platform.

I selected WordPress as the new platform for my blog. It seems quite robust and offers many of the features I had come to rely on with Movable Type. I’m now evaluating the engine and making a few additions where necessary to see how it will handle what I am trying to accomplish. So far I have been impressed with the software overall.

The key is going to be how does it handle the migration of data from Movable Type. Without a good migration utility this exercise will end quickly and I will be back to using Movable Type. As I progress down this evaluation I’ll try to note what I’ve found in case anyone should attempt what I am about to do on their own.

When I first decided to get into the blogsphere the software was still pretty raw and it was pretty hit and miss on whether things would actually work. Slowly but surely the industry began to mature and with it came some stability especially from a software perspective. A couple of years ago I finally decided that it was time for me to give up the manual HTML style of blog that I had been running since 1999 and move into the zeros (2000’s). After reviewing and playing with several different software packages I finally resolved that Movable Type was best suited to what I was going to do. I bought the software. Yes you read that right I actually bought the software. I am probably the only individual on the planet that actually spent money for a license of Movable Type for individual use. I figured that if the developers took the time to create and support the software I should do the same and at least monetarily support them.

After a lot of trials and tribulations I finally got everything installed. PERL is definitely not my forte so this was somewhat of a challenge. This was further complicated by the fact that my hosting provider was not willing to help nor did they really want me to install this on their server. After what I thought was a lengthy discussion they finally relented and allowed installation with the stipulation that I would be responsible for supporting this package and if there were any traffic or security issues the software would be removed. Kind of draconian attitude I thought but what was I to do?
I installed the software and began creating the templates and cascading style sheets necessary to run the blog. That too was more in depth and detailed than I anticipated but nonetheless I finished it. The Diary of a Diehard blog was born and quickly began to accumulate readers. This was going to be great I thought. But as the readership expanded so did my headaches. Not only was I accumulating readers, I was also accumulating SPAMMERS intent on driving my blog into the ground. It was not very long before I found myself spending more time deleting comments and trackbacks from SPAM sources than I was actually writing entries. I finally could not deal with it any longer and shut off the comment system until such time as I could find a solution to the SPAM issue. As Movable Type 3.x became available I again opened up the comments and just as quickly shut it off when I found that the SPAM generators were now much more automated and more than eager to give me something to do. I was getting email from readers wanting to comment but I just couldn’t deal with the SPAM. I tried a few things that slowed down the SPAM messages but it was still higher than my pain threshold for dealing with it.
When Movable Type introduced version 4.0 they announced that it had much better SPAM management capabilities. I quickly upgraded and moved my site and templates to the new system. Unfortunately the SPAM control capabilities were not part of the templates that I had created and the new template layout for version 4.0 is as painful as it is decentralized. I just didn’t have the time to rewrite my template files and all the existing 4.x templates I have found so far have been fixed width. I’ve come to enjoy the fluid layout I currently use and therefore didn’t want to adapt to a template that forced me into a fixed box. I was at a stand-off. Finally around the holidays I found some time to dissect the new templates and at least identify where the new comment section including the CAPTCHA routine was located. After several attempts and multiple tickets to my hosting company to upgrade their dated server components I was finally able to successfully deploy.
There was only one problem; while I configured the templates to use the CAPTCHA routine I totally forgot to enable it in the configuration file leaving my site to accept comments and trackbacks without the use of any SPAM management. The site was up like that for approximately 16 hours and in that time I received close to 600 bogus trackbacks and nearly 400 SPAM comments. I find it amazing how much bandwidth is wasted by these pond-scum sucking leeches who bombard a site with garbage. The problem is quickly reaching epidemic levels and something has to be done. Even on a bad day I know I will get less than 20 soliciting letters in my mailbox. Why should the Internet be any different? There must be a way for us to turn the battle back to these people who are intent on crippling the creative juices that bloggers exhibit. I don’t have an answer except to say that once I instituted the CAPTCHA routine and actually configured it to be available my life has gotten much better. I am now able to allow comments and not have to worry about being bombarded with offers for low cost drugs, pornography, or other mischievous products and services. Now all I have to do is find a similar solution for trackbacks.

Six Apart announced the latest version of their Movable Type software had
been released. Version 4.0 has several features that look very interesting
which is part of the reason why I decided to take the plunge and upgrade
quickly. I’m not completely fearless though as I am not going to upgrade all of
my blogs. I did start with the largest one though when I upgraded Diary of a Diehard. I think I should have
had my head examined with that decision. It’s not that the upgrade was not
successful, it is just that it has nearly 1,000 entries so the upgrade process
has a lot of opportunity for failure. But I wanted to use the new CAPTCHA
feature of Movable Type 4.0 and that was enough to make it worth the risk.

The upgrade was relatively painless. I followed the instructions and backed
up the database and the files prior to starting the upgrade. Once the back-up
was completed I deleted the MT directory and the MT-Static directory then moved
the files across. At first I forgot to change the permissions on the .CGI files
but after getting several error messages I quickly eliminated that problem.
When logging in the first time the MT.CGI file recognized that the database was
still at version 3.35 and therefore ran the appropriate upgrade scripts to
update the database tables and fields. When completed the software went to the
new user interface which is substantially different than in version 3.35. It is
definitely going to take me a while to find where everything is at. The first
thing I noticed was that the upgrade did not modify any of the template files
leaving my blog in the exact same state as before the upgrade. This is both
good and bad. The good of course was that the blog remained up and running
without any problems. The bad is that the templates do not take into
consideration any of the new MT 4.0 features. I rebuilt my site just to see if
it would work. I had one error with a ATOM Archive template that I have no idea
why I was using. I deleted it and rebuilt again with no errors. I say no errors
but that is not entirely the case. During the rebuild process the script timed
out several times resulting in a 500 Internal Server Error. I think this has
more to do with my hosting company than with Movable Type but it was still a
pain. I had to reload the page when the error occurred but it did continue on
so it was hassle rather than a show stopper.
The main goal with the upgrade was to use the CAPTCHA functionality for
comments that is now built into the core application. I’ve spent 2 days on it
and I am no closer now than I was before. The CAPTCHA code requires ImageMagik
which is installed on the server but the CAPTCHA images do not show up. I am
working with hosting company to make sure ImageMagik is installed correctly so
time will tell. The comment code with the CAPTCHA piece is in a new Comment
template so I am also going to have to change the site templates which is a lot
more time consuming. The documentation around CAPTCHA is also a little light
leaving out a lot of information that would be helpful for troubleshooting.
Overall the new system looks good but it is definitely more than a slight
change and should be planned before it is executed.