SMUG Speed Solution (I Think)

In an earlier post I lamented some issues we had been having with the SMUG service, in that pages were taking inordinately long to load: sometimes as long as 45 seconds, and occasionally even having some “timeouts.” Not the kind of timeouts modern parents give their children in lieu of spankings, but the kind that made me want to spank my hosting service provider.

Bill Ferris (@billfer) offered some good pointers in response, and suggested that I call my hosting provider to see if we could clarify the problem. But then, having applied some of his tips (and deactivating a couple of plug-ins), the response times seemed to get somewhat better.

So, like the typical male who cancels the doctor appointment when he starts feeling a little better, or who just learns to live with a little nagging health concern instead of getting to the root of the problem, I let the situation continue.

Yesterday, though, I had the blog equivalent of an out-of-hospital-cardiac arrest. When I tried to visit SMUG I got an “unable to create database connection error” message, and got a similar message when I tried to access my WordPress dashboard. The WordPress error message suggested I might want to contact my hosting service.

So I called GoDaddy, and as often happens the problem went away while I was on hold. But the friendly customer service guy checked into it further with those on the technical side, and apparently there had been some problems with the server on which SMUG is hosted, and they were in the process of solving them.

When I Tweeted about the issue, @paulacassin said:

@LeeAase YES! It’s not blindingly fast, but 3-4secs. Much better than before. 🙂 Looks like they tweaked something the right way

…and Monty Flinsch (@mflinsch) echoed the assessment.

So it appears we may have solved the problem. To confirm it, I would appreciate it if you would leave a comment indicating the approximate page load time you’re experiencing, and what kind of Internet service you have (i.e. T1, cable, dial-up, etc.)

I hope it’s not blindingly fast, because I would hate to be responsible for vision loss, but it would be nice if it could be just a little blurry, like Michael Johnson in the 200 meters.

Blogging 310: Taking the Self-Hosted WordPress Plunge

When I started this blog 30 months ago, my goal was to get hands-on experience with blogging. I was going to be responsible for our new media/social media initiatives at Mayo Clinic, and it seemed reasonable that I should dive in and become familiar with the tools and what they can do, so I could recommend the best way to apply them in my work.

Here were the criteria I set for myself in choosing a blogging platform:

  1. It had to be easy. I needed to be able to set it up myself, without any help from our IT department. This was for my personal learning and enrichment, so I wasn’t about to ask IT to help. Not that they would have (or should have) even if I had asked. They’re great people, but they have other priorities, like managing the medical and billing records for over a million patients a year, and all of the other administrative data required for an organization with 50,000 employees. The three major platforms I examined — Blogger, TypePad and WordPress — all met that standard.
  2. Free had to be an option. I also was really cheap. I didn’t want to have to spend anything out of pocket to learn about blogging. That kicked Typepad out of consideration, because its minimum was $4.95 a month. Did I mention that I’m really cheap?
  3. A self-hosting upgrade option was desirable. I wanted to explore a platform that would allow me to grow, and that could be used for intranet blogs, too, so that we could get accustomed to a common interface. That meant wordpress.com was edging ahead, because Blogger is a strictly Internet solution. I wanted to have an upgrade path that could include domain mapping, and that would let me eventually move to a hosting service (for more control) without breaking links and losing Google juice.
  4. The Robert Scoble Tiebreaker. It wasn’t really a tie: as indicated by the factors above, I was leaning toward WordPress.com. But I figured if that service was good enough for someone as savvy as Scoble, it could meet my needs. That settled it.

I have been extremely happy with my decision, and my blogging on wordpress.com has been fun, rewarding and cheap. For the most part, anything you see on this blog is something you could do without IT support and at no cost. I believe that to this point my total investment, including domain mapping and additional storage upgrades and a domain name from GoDaddy.com, has been about $30. And I have helped several people start blogs for which they have yet to spend a cent.

The WordPress.com service has been great, and they continually add more value, such as the 3 GB of free storage. I have absolutely no complaints.

But now, it’s time to take the next step and move to a self hosted solution. My main reasons:

  1. I want to explore custom plug-ins and widgets. WordPress.com strips out javascript because they don’t want one bit of malicious code on one blog to potentially affect millions of blogs. But that means I can’t embed many videos on my wp.com blogs, except for those that are on YouTube. Thankfully, YouTube is the biggest source of Web video, and wp.com does support it. But I know there are other high-quality YouTube players that will look better, and other potential sources of Web video. And that’s just a start. The plug-ins and widgets are my main reason for the move.
  2. I want better statistics. If you’re on WordPress.com, you can’t install Google Analytics, because WordPress.com has it already installed. You get some helpful free statistics from your dashboard with wp.com’s Blog Stats, but nothing as extensive as Google Analytics.
  3. I figure Google ad revenue will at least match my server costs. With wordpress.com you can’t run Google or other ads (and as Scoble says, what do you expect when you are getting a ton of value for free?) WordPress.com runs some Google ads to fund its free services. I would expect that if I run Google ads I can do likewise.
  4. I want to make my blogging easier. I’ve been illustrating workarounds that enable people to overcome WordPress.com’s limits by thinking creatively. For example, I found this information on how to put social bookmarking buttons on wordpress.com blog posts, but it takes a couple of minutes for each post. At some point the workarounds become too much work. I think that after 669 posts I’ve shown that I’m going to stick with this blogging thing. It would be a lot easier to just install the ShareThis widget and save time with each future post.
  5. It’s the Burden of a Chancellor to explore new frontiers. If I’m going to do 300-level courses in the Blogging curriculum, I need to be able to demonstrate some of the advanced functionality of the full-powered WordPress software. And I chose wordpress.com (and implemented domain mapping about a year ago) because it should make this migration fairly easy (at least as compared to Steve Rubel’s experience trying to move from Typepad to WordPress.) So now it’s time to prove it.
  6. It’s preparing for some other possible migrations. I started this blog to get hands-on experience with blogging, and have applied what I’ve learned in my work. We have several Mayo Clinic blogs now hosted on WordPress.com that we may want to enhance similarly by moving to a self-hosted solution. If there are any hiccups in the migration process, I want them to happen on my personal blog, not a work-related one.

I did a Google search for some related “how to” posts on making the move, and found a few resources that may be somewhat helpful. Here’s
an overview of the process. This one has a PDF that seems like its step-by-step instructions may be really helpful, with this related post.

My case may be a little more complicated, though. I already have leeaase.wordpress.com mapped to social-media-university-global.org. So I’m not sure how I can have a blog hosted at that same domain name on another hosting service without having some periods of broken links. Maybe it won’t be a big deal (this post makes it seem relatively simple), but for now at least it’s an unknown to me. Which means it’s a learning opportunity.

Another helpful post I found was this one which tells how to install WordPress on a Macintosh using MAMP. This will let me export my blog from wordpress.com and import it on my laptop. Once I get that part figured out, it should make it much faster when I do the switch for real with a Web hosting service.

This also seems a good time to move because the holidays (and weekends) are slow periods for blog traffic. So if I have any glitches or 404 – Not Found errors while I’m making the switch, they would be minimized if I can get the migration completed by Monday.

Meanwhile, I’m looking for recommendations on a Web hosting service. I purchased my domain name (social-media-university-global.org) from GoDaddy. Would it make the most sense to use GoDaddy for my hosting? I’ve also had recommendations for bluehost.com, enginehosting.com, liquidweb.com, mediatemple.net and ANhosting. All other things being equal, I’ll probably go for the low-cost option (remember what I said about being really cheap?)

But if you have feedback to share about this migration process in general (or hosting services in particular), I would love to hear from you. If you have either raves or rants to share, I would appreciate the guidance.

After I’ve first migrated successfully to my laptop, I will pick one of those services and make the move. And I’ll document the process along the way to make it easier for other SMUGgles when you’re ready to give it a shot.