DrunkenBlog

Of Caveats & Clauses

If I didn't know better I'd think SixApart was intentionally giving me fun stuff to blog about while I hold back on other things until the WWDC. They're just continuously showing the wrong, yet predictable thing to do... which is kind of what textbook cases are all about and why they're great to learn from.

I was behind in catching up on this today, it wasn't until I'd been online for more than an hour yesterday in the afternoon that someone IM'd me and clued me in that SixApart had posted some alterations to their recent licenses.

I doubt I would have seen it at all today if I hadn't had it pointed out to me by someone who has a much larger stake in MovableTypes' direction than I do, I was too busy talking to a client who got a notice from their ISP recommending against MT because of the single-CPU limitation as well as going through my own rosters looking for problems.

That should be a big cluestick as to why this is another blunder, and lost opportunities are abounding here... and why the VCs often bring in PR people in addition to the MBAs.

In my last post on the subject I went through what I believed what was going to happen, and nothing so far has really changed my mind, although I'm glad I withheld my opinion that they're prolly going to get bought out down the road eventually. And just to clarify, none of that stuff is going to happen tomorrow.

But while the "blogosphere" is a new emerging market, it isn't a market that is immune from market forces and the idea that "no one has ever had to make these decisions before" isn't really going to fly. 99.99% of the time, "new markets" are merely combinations of existing market types, when you know where to look. Examples abound, and all the same caveats apply. Hell, half the social-capital stuff can be traced back to Tupperware.

So let's look at the specifics of this last move by SixApart:

• They posted changes to the license to correct "miscommunications" and "misconceptions" 2 days after the original bomb.
• The announcement basically consists of:
  • The single-CPU restrictions was a "mistake" that was never meant to be in the license. They are retroactively removing it.
  • Their new pricing structure is not retroactive, so you can continue using your old install of you weren't a beta tester that had already upgraded until, you know, there's a reason why you can't, like your ISP upgrading dependent software or security problems.
  • They now say that any weblog under the same URL counts as one weblog
  • They've increased the number of authors allowed for the personal edition from 3 to 5.
  • They've eliminated volume licensing for the personal addition, but you can buy addition 1 new author and 1 new weblog for $10.
  • If you already purchased a volume license over the last 2 days, you'll be moved to the new system.
  • If you have authors that haven't posted to their/your blog in over 90 days, they don't count towards the limit and you can keep them for archival purposes
  • We're sorry people don't seem to understand, and we're glad we could clarify
• As of 30 minutes ago, there were 239 trackbacks on the followup and 688 trackbacks on the original post, almost 40 of which are from today, three days later.

We'll start with the last first... Any PR guy would have grabbed the keyboard out of their hands before they made that follow-up post. But wait, you say, isn't it better to nip. it. in. the. bud? You'd think so, and in many cases it is, but not in this one. The scope and damage of this is too broad and far-reaching, and their followup for the most part will be lost in the noise.

With a case like this, a good PR person won't try to control the story, they'll try to change the story.

In order to do that, you first have to understand what the story really is. SixApart is misinterpreting the story, which is why this is ineffective. Let's not get too ahead of ourselves here, thinking of MovableType as the end-all-be-all CMS. It's big in the personal blog space, which is growing, but Slashdot didn't carry a story about MovableType going paid, they carried a story about the extreme backlash from its users. And since they've so pervasively pissed off their users, and everyone has different aspects that are pissing them off, or a combination, its all a big jumble.

But to SixApart, this is all just a big misunderstanding over the license. But again, the license is only what sparked the story, and a part of the story. It's not what the story really is or has become.

There's an analogy I usually use from a non-aggressive martial art called Aikido which illustrates this well. In something like Karate, if a large force is coming at you, you generally meet the force with a block, and then counter with your own attack. This is fine, and instinctive, but forget what the movies tell you. Unless you're very, very good, if a 300lb man is throwing a punch at a 120lb woman, she's screwed if she tries to block it.

She'll deflect some, but there's going to be damage. In Aikido, the energy of the force is used against the force itself. The larger the force coming at you, the greater the force that is returned.

SixApart is trying to deflect this via damage control, but that just isn't going to cut it. It's... feeble.

Slashdot isn't going to carry a new story with their clarification, and hundreds of other places won't either. At best, the story will be "Users revolt over SixApart license changes" with a footnote and quote from SixApart about how they've tried to alleviate their users concerns with changes to the license.

They're reliant on their hardcore users to re-blog the story and spread the word, going to Slashdot and leaving comments days after the fact (Slashdot is just an example, which is fresh in my mind after having it happen to me). The same users who feel they've been mislead, betrayed, or are just annoyed and confused are expected to go into the support forums of hosting companies who have said MovableType can't be used on their servers and tell them that it's all OK now.

And then there's the trust issue, which is another part of the story. Unfortunately this part of my previous post on the subject seems pretty applicable:

...now its too late, and they're being eaten from the bottom, and they're just annoying and confusing their users to an even greater degree with all the backtracking and mucking around while the alternatives improve and even surpass them in some areas as the user-base builds and more support comes in.

Their license was already borderline-crazy, and as I mentioned before most people were "putting up with it" and believing something better would come along. Now it's become more confusing. The caveats and clauses aren't going to help them here. User Y now has to do all this odd dependency math in their head when evaluating whether or not to use MT or one of the alternatives. And to be honest, their retroactive stuff, while good in this case, only serves to rub in just how crazy their license is to a lot of people.

You see, SixApart for the most part got away with their license out of trust. People trusted them not to do something really nasty, or stupid, unclassy or lame. Some still do, and to be honest they may not abuse it in the future. But now, to a lot of people, that trust is gone. Oh, there good feelings for a lover scorned... but once cheated on, unless you're really codependent, everything changes.

People can picture them cashing out in the future, with the company being sold and all bets are off. They know the single-CPU restriction was considered, even if it was a mistake, and realize just how much SixApart could screw with them or their business at a whim. They realize that SixApart could have retroactively told them to pay up for what they're using now, and that could happen in the future. They're trusted not to do that, but as I said, the seed of doubt is already there and someone would be a fool not to be examining alternatives. The incentive wasn't there before, because of the trust... it is now.

Now, since I mentioned pausing and changing the story... how would one do that?

As I said, a PR person would have grabbed the keyboard out of their hands before they posted their "clarifications". The story is just too big and far too broad and the first order of business is keeping it from becoming a path-dependancy action. I.E., set-in-stone. That means stopping everything in its tracks while you re-assess, or pausing the story.

Pausing a story isn't that difficult. The easiest way to do so is to make it feel as though it's "in flux". Something like yanking everything to do with 3.0 and allowing users to download 2.6*, along with a feedback form on the proposed changes would have pretty much done it. It removes the immediacy of the situation while you get your act together, and keeps the bloggers pointing everyone to the site to tell them why it would be a bad thing. Right now they're doing that, but it's a write-off. The people leaving the trackbacks are saying goodbye more-so than voicing their displeasure.

Once it's been paused, you then you have to change the story. The trust is gone, there's no way to get around that. But you can make the requirement of trust a non-issue, and hence, not only change the story but make it a bigger story. This would require revising their spreadsheets, and changing their licensing model, perhaps akin to something like the MySQL license structure.

To gloss over, think of it this way. The hundreds of websites out there aren't all going to carry a story on how SixApart has made a retraction. Hell, people are still just hearing about MovableType going paid. It's all lost in the noise.

But in a week, those hundreds of websites, and even more, would be carrying the story of how the users prevailed against the evil investors at SixApart through their outrage and effort. The entire basis and outlook of the situation would change, as would the tone of every interview they're going to be giving for the next year at least.

Ah well.

So far I'm really, really gravitating towards TextPattern, I just dig the interface that I've seen. Moving to WordPress went way, way too smoothly, and it works fine with register_globals turned off for PHP.

 Posted by drunkenbatman
    May 16, 2004, at 08:25 PM