Around work I’m known as something of a Movable Type apologist. I can’t help it — it’s how I was raised. MT was the first thing I tried to install six years ago when I decided to check out the burgeoning blogging phenomenon, back when the media was merely interested in, rather than terrified of, the whole business.
Movable Type’s greatest strength is handling traffic, which of course is not something I’ve ever really needed it for. When you make a change to a site MT picks up all the data you’ve entered, passes it through your templates and generates a plain ol’ HTML file, which it then plops in the appropriate directory. It also updates a few other files, like your site’s main page. But these are similarly static (by “static” I mean a page that doesn’t ask the server to do any computation as it’s served). As you might imagine, serving static pages is a relatively easy task for a webserver — all the computational cost of composing a page is incurred once, when it’s created, and the result is saved. The overhead for each subsequent page-serving is consequently as small as it can be. This is what makes MT good at dealing with traffic.
But the system that does this is built on unappealingly old, slow technology. And besides, there are good reasons for wanting dynamic functionality on a page. You can create a hybrid sort of site — this blog is an example, as it uses a lot of PHP in its MT templates — but it’s a little awkward to build and maintain. You can extend the core MT system, too, but it’s not always well-suited to the task. Over the past few years I’ve watched as the folks at Gothamist, with the help of Apperceptive, have done this time and again, cajoling MT into accomplishing things that it really has no interest in doing. Sometimes I’ve been frustrated by this process; other times I’ve been impressed. Either way, their efforts can be fairly described as heroic. But if they were using a different platform there wouldn’t be a need for quite as much heroism. I suspect this wouldn’t be a bad thing.
Now Ars, acting uncharacteristically like a press release proxy, brings word of MT 4.2 Pro. Despite the enthusiastic writeup, the announcement leaves me about ready to call it a day on MT. Not that I’ll stop using it, mind you — porting this blog would be a pain — but it’s clear that its time is done. Six Apart is already retreating from the stab at openness that they made with MT4. And the warmed-over social networking features that they’re now offering feel just a bit desperate. There is nothing here that can’t be easily accomplished with WordPress MU or Drupal.
And that’s alright. I’m glad to see those other (free) projects ascendant: WordPress has managed to harness its community’s enthusiasm for writing some of the world’s worst PHP code and turned it into a successful, professional product that’s actually well-engineered (if subject to more security bulletins than I’d like to see). With the caching technologies it has available, there’s no longer a great reason for using MT.
And Drupal — well, I spend so much time with Drupal that my feelings are inevitably mixed. But I’d take it any day over Movable Type (and WordPress, for that matter).
I’m glad to keep a hand in MT. It’s a very different system than those other blogging platforms, and that’s probably good for me. But it’s no longer possible to make a great case on its behalf. The only thing it has going for it is the company behind it: if you buy a commercial license you can call up Six Apart and yell at them in a way that isn’t possible with non-corporate projects. That’s certainly enough to build a business on — ask Redhat. But in terms of total yelling period, I’m ready to concede defeat on behalf of MT proponents everywhere.
Hey Tom, sorry it took me a while to reply to your post, we’ve been a bit busy with the launch. :) I wanted to take a minute to reply to a couple of your points if you will indulge.
First, thanks for being an advocate and a member of the community for ages. We appreciate it and that kind of loyalty means a lot to everybody on the MT team. Second, thanks for really understanding what’s cool about MT and why it’s uniquely great at some important parts of publishing on the web.
But the core of the arguments you’re making is something I think is a little out of date. There definitely were times in the past when MT was stagnating a little, and doing key tasks took unnecessary heroics. Today, though, and especially since MT4, it’s really much easier to publish pages and scale like crazy without being an expert. MT 4.2 can automatically assemble a page like yours using PHP includes to assemble static bits of content. It’s exactly the right balance of scalability and responsiveness, and it’s how you and I both run our sites. Regular users can now do that without having to know how to do any kind of coding in MT.
One of the reasons we brought the Apperceptive folks on board was so that we could build their insights into the platform. And I’d take strong issue with your argument that we’re retreating from openness: MT 4.2 is open source, and has tons of new features and performance improvements. MT Pro is not GPLed, but does take features that might have cost thousands of dollars in the Community Solution and makes them free if you’re a personal blogger. And we’ve always said the progression of new features would be a waterfall, from the high-end enterprisey versions all the way down to the open source engine.
More to the point, all our key innovations are 100% open. Action Streams, OpenID, OAuth, the iPhone and Facebook apps that plug in to MT, those are *all* free and open and there for the taking or for the hacking. Not only that, when we invent this stuff, and invest in popularizing it, we choose *not* to patent any of it, but instead to make sure it’s completely unencumbered.
The social networking features we’ve launched are anything but warmed over; They power real communities on sites bigger by an order of magnitude than those run on the other platforms you’ve mentioned. Warmed-over would be something like BuddyPress, which, if and when it’s eventually released next year, will consist of a pile of separate plugins. You certainly have the tech chops to assemble something like that, but the era of “easily accomplished” meaning “someone with coding experience can possibly create the functionality if they hand-assemble a disparate set of individual code chunks” is over. And good riddance.
And you gloss over security like it’s not a threat to the entire web. I mean, with WordPress not only having vulnerabilities but getting actively exploited on a nearly continuous basis, I wouldn’t want my site hosted on a shared server with a WordPress install. Right now, somewhere upwards of 80% of all blogs on wp.org are vulnerable to widely-available exploit scripts. You can shrug that off because you’re geek enough to constantly do the upgrades, but at some point, that has to be considered irresponsible.
Basically, I hear loud and clear that you’ve been through a lot of ups and downs with MT. What I’d ask you to do is look at it with fresh eyes — we have. If you get started as a regular user today, MT is easier than ever to work with, to grow with, and to succeed with. And we’re inventing more new stuff, on a consistent basis, than everybody else in this space combined.
I’ll apologize for having rambled on so much, but I think it’s important to judge things in context. The biggest frustrations of MT are in the past for our community, and the platform is ascendant again. It’s well ahead of everyone else in the areas of the most important innovation and invention. And it’s the platform best poised to serve an audience I think you and I both care a lot about — the people who have a great idea, but don’t necessarily know technology, and just want to get the word out to as many people as possible. For those people, I think Movable Type is second to none.
Thanks for the thoughtful response, Anil. You make a strong case both with your explicit points, and implicitly by your presence here in some tiny blog’s comments. I’ve always been impressed by the accessibility, friendliness and smarts of the folks from Six Apart that I’ve met online and at SXSWi.
You’re right that my perspective is influenced by my technical background: I’m more comfortable with PHP programming than with Perl, and that’ll inevitably color whether I prefer writing plugins and templates within MT or WP/Drupal. I think it’s not just me, of course — there are a ton of PHP programmers out there, and they aren’t all as embarrassingly awful as they used to be. It’s not a sexy technology, exactly, but it is universal and vital in a way that CGI Perl isn’t.
I apologize for eliding 4.2′s continued open source status, but I remain skeptical about the overall project of open sourcing MT. Don’t get me wrong: I think it’s the right move, and it proves you guys have your hearts in the right place. But I’m sure you’re realistic about the dividends it’s likely to pay relative to those enjoyed by projects that have been open from the beginning.
I disagree with your assessment of the era of chunk-assembling being over — at EchoDitto, that’s exactly what we do with contrib Drupal modules on a daily basis. We do it with some huge sites, and it works out well. Modularity won’t be going away.
In the end, it boils down to this: for a user in my position — someone technical who values flexibility above support and, yes, security — MT is no longer the best choice. But I agree that MT will continue to enjoy success with those who don’t fit that description. I certainly hope it does.