Why Movable Type — 2. Myths, FUD and reality

| 8 comments | 0 TrackBacks

(This post is part of a series on why Movable Type is an excellent choice as both a CMS and a blog engine. The previous post was the Introduction, the next post is The Movable Type Communities.)

If you are new to the Movable Type world, you have excuses if you find things confusing, for it is encumbered with lots of myths, FUD and even miscommunication and blunders from its own parents. When you look at MT, it is important to look at both the people, the technology, the rationales for using it, and separate what really matters from noise. This post is about exposing the myths, the following posts are about what matters.

I've decided to treat the noise separately because, like many other MT experts, I am frankly tired of the low signal/noise ratio out there. My goal is to present rationale arguments for professionals to consider (or continue) using MT. It is not to make a definitive comparison of MT with other solutions (and there are plenty of good ones out there), nor is it to convince people who think in black and white, confuse what is good with what they love, and are convinced that their favorite ‘X, Y, Z’ CMS is “absolutely and universally THE best CMS of the entire universe, ever, and everything else sucks, period.”

Myth: Six Apart is dead, and Movable Type with it.

Six Apart Europe has been shut down months ago, but it was irrelevant in the development of MT anyway. Six Apart US has been bought by VideoEgg to form SAY Media. But Six Apart Japan (6AKK) is alive and actively developing MT5 which is the “#1 leading CMS in Japan” according to Jun Kaneko, Six Apart's Community Manager for Movable Type, who also announced a beta of MT5.1 for the fall. Actually 6AKK took off the entire development process of Movable Type starting with MT5 more than a year ago. Only the MT4 branch remains in the hands of a few people in the US, now working for SAY Media. And the Melody Project has forked Movable Type Open Source to boost its development.

To summarize:

  • MT5 and MTOS5 are under active development by Six Apart Japan
  • MT4 and MTOS4 have reached their end of life and are maintained only for bug/security fixes by SAY Media
  • The Melody Project has forked MTOS4 and is actively working to launch Melody 1.0 based on MTOS 4.34

Doesn't look like a dead product to me.

The only uncertainty that I see concerns MT Enterprise. I have no clue how many customers are concerned with this. But at the same time, I wonder why one would need MT Enterprise in the first place (my little finger tells me that MTE won't survive long as a separate offering, but I might be wrong. I hope I'm not spreading FUD myself here ;-).

Myth: The MT community is non existent.

Most of the time, when saying that, people expect the community to be as populated and vibrant as the WordPress one. Or they are newbies who can't bother reading the documentation and ask questions like they'd throw bottles at sea, then complain that no one answered them. The reality is that the community is diverse and elitist, as I'll show in the next post.

Myth: Perl sucks. Related FUD: you have to learn Perl to use MT

The usual pearls about MT and Perl go along those lines: Perl sucks as a language, is a strategic mistake (usually compared to the supposedly more “friendly” PHP), and you have to know it to use MT.

First, and this comes from someone who a decade ago thought by pure ignorance that Perl was antiquated, Perl is not only a good language, it is specifically designed for text processing (which is precisely what a CMS is about). It is not difficult to learn Perl and it can be a useful skill outside MT.

On the strategical choice, Perl has been here since 1987 and runs (when not included out of the box) on virtually all modern server, desktop and mobile OSes out there, and most importantly at any good internet hosting provider. The only legitimate question IMO would be about developer attractiveness, or popularity. But notwhistanding the fact that the Perl community itself is not small despite a supposedly unpopular language, one might argue that popularity is orthogonal to quality, and point out that even so the PHP language boasts to be massively more popular than anything else, it also has arguably one of the highest share of crappy programmers, spaghetti code and, consequently, security issues.

Last, only plugins and core developers have to use Perl (some plugins, like themes, can be developed without coding a single line of Perl). Designers—and that is actually a major difference and an advantage compared to a software like Wordpress which forces them to learn PHP—will never see Perl nor any other programming language unless they want to (MT can publish PHP pages, PDF documents, XML and virtually anything you want). Templates in MT are designed using MTML, a powerful, tag-based and designer-friendly markup language that makes templating easy.

Perl may be a liability for people who are religious about it, but it is certainly not one for Movable Type. And most of its users will never see a single line of Perl anyway.

Myth: static pages are bad, dynamic ones are good

This post is already quite long, and I'm going to address the technicalities of MT in the fourth post of this series. But let me tell you that whoever claims this does not know much about performance and how this choice plays in terms of costs for the editors and impacts the real end-users of their sites.

I often see this myth used by clueless people as a cheap shot against MT. I've seen it used many times by selfish people who think only about their own instant gratification when pressing the "publish" button, without a single thought about their site responsiveness and the ensuing readers’ experience.

But I can acknowledge that the stock themes of earlier versions of MT didn't scale well. With the (legitimate) goal of showing what MT was capable of within the default templates, Six Apart did not always anticipate how their templates would behave with thousands of posts, comments, categories, tags etc. This is ancient history. With a comprehensive system of inclusions, cache, deferred publishing, it is perfectly possible to create fast and scalable templates that will make MT beat the crap out of many dynamic publishing systems for a fraction of the hosting cost.

Ironically, since the Wordpress world has become enamored with their new catalog of caching plugins, I hear less and less cheap shots about MT's static publishing nowadays ;-).

Myth: MT is closed-source and commercial. FUD variant: MT started as open source software then went commercial then OSS again.

MT started as a shareware before Six Apart was born, and while you could see all its code, it was not open source. One could use it for free, I paid $45 for my copy in 2002 to start my personal blog. When Six Apart introduced MT version 3.0 in 2004, they made a mistake trying to impose expensive and unrealistic license terms that turned people away and fell apart in a communication blunder. They gradually corrected their mistakes by lowering the prices, relaxing the license terms and offering a free version for non-commercial use. They released MTOS, MT under an Open Source license, in December 2007. To date, there are a variety of versions of MT, both commercial and open source, catering to different users and needs.

FUD: You don't control your data

I list this one just to poke fun at the intellectual dishonesty that exemplified the knee-jerk reaction of some former hardcore users of Movable Type who used to boast about how much better MT was compared to anything else prior to the MT3.0 license uproar. Then all of a sudden, MT was evil, and their data was in peril. Only Open Source could save it (and Wordpress opportunely came to the rescue right at that time to become their new blind love). I'd point out, again, that MT Open Source was born in late 2007. How come their data was safe before 2004 then? Because the data in MT has never been trapped into any kind of proprietary format, you can export it in a variety of formats (and likewise import pretty much anything you want). It is also worth mentioning that since MT is usually used to publish static pages, in many cases the produced site doesn't need the MT engine to deliver its content.

Myth: There is no documentation

Here it is.

While it is true that Six Apart has not always done a stellar job at documenting things, relying on the community with the MTOS wiki, there has been a real effort to improve things with MT4 and MT5. Things can always improve—and there's indeed room for improvement, especially in the developer's section—but the present documentation gives you all you need to start using MT from beginning and up to an advanced level. There is a work in progress to translate the Movable Type 5 Designer's Guide written by Mr Hajime Fujimoto from Japanese to English, with another guide to follow for developers.

FUD: There are no plugins

Here they are. Oh, and more here too available as open source (all the Six Apart plugins are moving to github.com/movabletype). And even more if you search a little or ask the community (you might be pleasantly surprised).

Don't be fooled by the Plugins repository, it does need some curating to get rid of the obsolete plugins and it is far from being exhaustive (many plugins exist in Japanese, waiting to be translated in English). It also seems that not all developers do bother with listing or updating their plugins in the MT Plugins repository. I don't think it is unfair to say that this Six Apart web property is in need of an overhaul, not the least to fix an image that does not reflect the reality.

Also don't fall into the same usual confusion between quantity and quality. MT does indeed have much less plugins than Wordpress. But MT doesn't need a dozen plugins to manage cache issues, to output SEO-friendly pages, or a different engine to manage multiple blogs (which was possible before Wordpress even had a following). MT's core features and templating languages are enough for a wide variety of needs, and I have built dozens of sites without using more than a handful of plugins (I usually even disable most of the stock plugins).

FUD: There are no themes

There is not a plethora of themes for MT, but MT ships with a collection of themes for starting blogs, web sites and communities out of the box. The quality of the stock themes is good, they are SEO-friendly and accessible, they also present a good showcase of how MT works and can serve as a starting point to learn how to build your own.

Of course I would love to see more themes, but I'd rather have a few high quality ones than a lot of crappy ones. Like plugins, it is easy to fall into the quantity vs quality trap. I also guess that the quest for ready-to-use themes is done more by individual bloggers than anyone else, and that people using MT for building websites will rather produce unique themes and/or reuse their own framework. Enterprises and corporate customers, the primary target for MT, are certainly not looking at bland and non distinctive stock themes.

As someone who watched his TV channels grow from 6 to more than 500 only to realize than less than a handful are interesting to watch, I think the low number of themes is not a problem per se, only a perceptual one for those who mistake choice with quantity. (As an illustration of the relative meaning of choice, to me the choice is first and foremost that MT allows me to output anything I want in any format I want.)
Yet again, the community is working to increase the number of available themes, hopefully increasing the number of quality ones. The Melody Project is also working on making easier for theme designers to produce multi-purposes themes that can be easily customized by the end users through the UI (and, wait for it, without ever touching Perl ;-).

I'm sure I missed other MT myths and FUD, feel free to complement this list in the comments.

The next post will be about the Movable Type Communities.

(This is a work in progress, please stay tuned and pardon the occasional edit.)

No TrackBacks

TrackBack URL: http://ubiquitic.com/cgi-bin/mt/trckbck.cgi/21

8 Comments

And to give to Caesar what belongs to Mike T, I borrowed “That’s been in MT since before WordPress had a following” from him.

Thanks for letting me know about your "In Defense of Movable Type" series.

Two problems I have with Movable Type (which your posts seems to confirm).

Elitist: From my experience an "elite" attitude can often blind one towards their own weaknesses, as well as towards a rivals strength.

WordPress has plenty to boast about, but it also has plenty of weaknesses. What I love is that Automattic seems to always be focusing on improving/fixing bugs, and even publicly highlights faults with WP.

Simply telling people to RTFM more often than not leads towards a platform's downfall. Even more expensive rivals like ExpressionEngine do not do this, and encourage everyone to help "the newbie" regardless of how silly the questions are.

Themes: WordPress has a lot of crappy themes, but they also have a large quantity of quality themes. Large companies may not care but small businesses do, and many will pick WP over MT simply because a site looked pretty (vain, yes, but it works).

Anyways, keep these up! :-) I'm looking forward to the completion of the series.

I've been using MT for my personal blog for over 6 years, starting back in 2004 with MT 2.661.

I've never paid for my copy of MT, there has always been a free version for personal use, even through the licensing fiasco of MT 3.0. That's another myth, that you had to pay for it with 3.0.

Oh, and I've never touched a line of Perl code in those 6 years either. Wouldn't know how.

Well here's hoping the Japanese do what no one else could or would.

In the earlier days, and particularly when I started using MT around 2.x and more so in the 3.x versions, there was a lot more excitement for me. The templating tag features were extremely easy to learn and use and they remain one of my primary reasons for sticking with MT now that I've been using 5.02. In the early days, a grab-bag of plugins that could work so many apparent miracles were available; in fact, hunting down and finding all the independently made plugins was at times even more fun than actually writing blog posts! Combine those features with the simple templating features, and MT was really fun to use.

But for whatever reason (and I have my suspicions...) every new version of MT seemingly became an attempt to move furniture around on the Titanic. Unnecessary revisions in the UI, such as moving locations and layouts for the menu, broke a lot of plugins. So plugin designers had to rewrite those. Changes to the back-end also broke plugins; some of those changes may have been necessary for real improvements, but Sixapart seemed quite unconcerned whether they broke a lot of plugins. The best and, in my opinion, the most admirable plugin developers worked hard to update plugins as users scrambled around to make their sites work again.

A lot of plugin developers, particularly those who hadn't been at it for a long time, simply gave up. Trying to find updated plugins for MT5.x is like trying to find the proverbial needle in a haystack. Even new plugins being made by the old dogs are now only being made for MT4.x; and that says a lot, doesn't it? There is almost a collective though sometimes silent plea: "Enough already!"

The whole "moving furniture around on the Titanic" approach sometimes took reasonable routes, like changing how templating structures should be modularized and how even a widgeting system should be incorporated.

But along the way, the senseless UI changes and a simple lack of concern for backward compatibility (plugins, breaking sites) had an even greater contender for the Feather that Broke SA's Back: namely, an incomprehensible unconcern for how the changing environment required new basic in-the-box features just to keep MT relevant. Eventually, SA would include some of these -- a preview function for blog posts, custom fields -- but they were damned slow on those. Other basic in-the-box features are still missing, such as a built-in contact form/template and, what's far worse, a built-in "Tweet This" or "Post on Facebook" feature. Yes, these are simple and some workarounds and plugins exist for this but these are very, very basic features. Can't tell you how sick I am of seeing Wordpress sites all over the place that have this. Again, these are just examples, and only a few examples of many that could be made. The point is, SA always seemed unable to think ahead (their tragic Motion notwithstanding) and would only get around to making these necessary modifications after the train had long left the station.

SA seemed far more concerned to a) keep MT just bare enough to require corporate customers to commission the bells-and-whistles (custom made plugins for instance) and b) move, er, type around on the page (menus, UI, templating) without bothering to consider the changing environment and promoting user-reader relations (Contact form, Twitter Tweeting, Facebook integration).

While I am sensible of your "Myth-busters" approach and, quite honestly, can still use and appreciate Movable Type for basic blogging and blog designing -- design templates, write post, preview post, publish post, receive comments on post -- I think that blaming the potential user public for believing in myths as an excuse for why MT seems to be in such turmoil and, quite frankly, falling in popularity isn't going to solve the problems.

How about posting these as part of the Perl Iron Man network of blogs: http://ironman.enlightenedperl.org/

Might help to get the word out to those that are already in the Perl camp. :)

Best,

Phillip.

I don't know, it looks very geeky there :p. Let me finish the series first.

Hi Francois,

Are you still planning to complete this series? I am just about to dump MT and jump on the Wordpress bandwagon, if for no other reason than my hosting company will maintain WP for me, I need a quick delivery on basic social networking features, and I'm afraid of the upgrade process.

The one thing I need to know is if it is as easy to design your own templates in Wordpress as in MT. I was actually looking forward to your "Movable Type Publishing Platform" post, the next one on the list.

Yes I am, but I've been a bit overwhelmed lately and didn't find the time and state of mind necessary to resume writing. This series is a good candidate for my first 2011 resolutions :o).

To answer your question about templates, I personally feel that it's way easier to design (and maintain) templates with MT than WordPress, even so I am fluent in PHP. MT makes a cleaner separation between logic and structure with an elegant syntax, while the WP PHP tags look to me an even greater burden than the already crazy, ugly and inconsistent syntax of PHP.

Another big contention point IMO, related to the difference between MT tags and WP tags, is security. You just can't compromise anything with MT tags in a template. But in WP, since the tags are actually PHP code, a third-party plugin can inject any kind of code in your site, which is a big security issue.

Oh, and ask your hosting company how they're going to "maintain" WP for you, notably how quickly they're going to install security releases. Like the last critical one issued two days ago that was supposed to correct an XSS hole and that is already vulnerable to a similar security issue. Are they going to keep your WP installation up to date? If you don't have to spend time upgrading WP, fine, but for people like me who are taking care of hosting that kind of software, WP security upgrades are just a waste of my time (and clients money).

A question to you: what are those basic social networking features that would make a whole migration (i.e. redesigning all templates plus adding those features, plus migrating data, plus…) quicker than just adding them to your existing MT site?

Leave a comment

N.B. by commenting, you accept the comments policy.

About this Entry

This page contains a single entry by François Nonnenmacher published on September 30, 2010 11:31 AM.

Why Movable Type — 1. Introduction was the previous entry in this blog.

Why Movable Type — 3. The Communities is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.