Geodog

Responding to Anil Dash's request for even more feedback about MovableType and SixApart

First of all, thank you, Anil Dash, for your patience and willingness to engage in conversation with critics and fans alike. I would also like to agree with you about tone, and to note that I did not help matters with the titles of some of my earlier posts. I think my criticisms were valid, but the tone of some of the titles was uncalled for and unhelpful. Second, I can't believe you have asked me to post more feedback, and are getting me to spend another 2 hours of my life writing about MovableType, something I just said I wouldn't do again. Some would accuse you of feeding a troll, and friends have asked me why I spend so much time worrying about one Weblog / lightweight Content Management System vendor (it's 'cause I like y'all). As I said earlier, we all do have more important things to do. But you (and the Wall Street Journal) asked, so I will try to be brief, constructive and final.

Where I am today with respect to recommending SixApart products:

• I would recommend Typepad to a non-technical person who was trying to start a blog, if they had some aesthetic sensibilities and weren't too price sensitive. Presuming that SixApart can keep up with growth and minimize downtime, it's a nice product, reasonably priced and segmented, and the grouping on Typepad give the author a decent chance of attracting a readership. SixApart does also seem to be adding features to Typepad at a reasonable pace. If someone was price sensitive, I'd send them to Blogger. Blogger's hosting issues seem to be a thing of the past, there are now some nice templates for it, and there are still smart people outside Blogger doing cool things for Blogger. Plus it is free. It is hard for SixApart being in a consumer business where so many of their competitors are giving away their product for free. Maybe it makes sense for SixApart to do B2B only? See below.


• I would recommend Movable Type for a institution or business wanting to host a corporate blog that didn't have a lot of internal technical expertise. There is good documentation, nice templates, and good support. I don't know if SixApart permits it, but if it does I would recommend that the business start with 2.661 and let the dust settle on 3.x before upgrading. Depending on what the institution's other needs were, I would also recommend that they evaluate pMachine, Expression Engine, CityDesk and Drupal. Some of other others are better options if the institution also needs to build a complete website, or wants community features as part of its content management system.


• For myself, personally, I'm spending all my weblog hacking time playing with WordPress, and I now have 3 blogs running it on 3 different servers. Unless some cool new MovableType-only feature appears in the near future, I will probably convert my 4 existing MovableType blogs over to it over time instead of upgrading to MovableType 3x. WordPress' documentation isn't nearly as good, support is hit-or-miss, and I'd guess there is a substantial chance the whole project will explode and fork into many pieces over the next year or so. So why convert to WordPress? I'm comfortable messing with PHP in a way I never will be with Perl, if you count plugins the pace of feature additions seems much faster with WordPress than with MovableType, and it is free as in freedom. I don't have to worry that if some stupid idea of mine takes off and suddenly I have 10+ people wanting to be authors, as happened with SARS Watch, suddenly I would owe SixApart a $100. It isn't that $100 is unfair, or a lot of money, it's just that I don't want to have to think about it, and there are lots of alternatives where I don't have to.

My suggestions for SixApart:

1. Communicate, communicate, communicate. Anil, you personally are doing a great job of it now with SAPN, and Mena made a good start with Mena's Corner and SixLog. The hard part, as we all know, is keeping it up, and answering people's questions publicly. Phil's question on the MovableType-Pro list, asking what incentive there is to develop plugins for MovableType now, has sat unanswered for a week. Maybe someone answered him offline -- but I bet a lot of other developers are wondering what the answer to that question is. IMHO most of the problems with SixApart's constituents over the last year and half have been because they felt SixApart wasn't responding to their concerns (the rise of comment spam), or because they were unpleasantly surprised by SixApart releases (no beef, new license, breaking plugins). A lot of the concerns could have been ameliorated, and perhaps SixApart might have made some different decisions, if there had been more communication with stakeholders.


2. Eat your own dogfood, specifically with respect to comments. I know SixApart weblogs have trackbacks enabled, for which SixApart should be commended, but I'd guess that most of SixApart's non-institutional customers have comments enabled. Experience what your customers experience, and let customers have an easy way to talk to you and with each other.


3. The licence. From your point of view it is probably a dead horse that has been flayed, but for many people it is still a live issue. I still think that it was counterproductive to segment the market by number of authors, and to start charging non-commercial hobbyists for the product, especially given past statements and that fact that no new fun user features were rolled out with the new pricing structure. I've read the 34 cents a download argument, and I still think that since a download costs SixApart pennies at most, getting 34 cents for each download is a very low cost of sales. Given the large number of free alternatives, it is penny-wise but pound-foolish to charge the non-commercial hobbyists. You may gain a bit of revenue per download, but I'd guess that you lose a lot more over the long run in not attracting more developers, influencers, evangelizers, and people who eventually convert their sites from non-commercial to commercial sites. Shelley Powers recently posted an interesting suggestion for how SixApart could mix their proprietary code with GPL code that is worth considering.


4. Start adding some sweet candy to the releases, soon. I don't want to wound or get in an argument, but if feels like MovableType hasn't had any fun new features in a long time, and the pace of plug-in releases slacked off a while ago. MovableType has been upgraded with somewhat better comment management and dynamic pages, but that is plumbing, not candy. Anil, you haven't had seismic work done on your house yet, being a relocated New Yorker, but wait until you do. The contractor will fight to add something else to the project -- a kitchen remodel, a new deck, a hot-tub -- heck, if nothing else the contractor will probably throw in a free paint job. Why? Is that because the contractor really wants to make another $10-20K to add on to the $200K they will charge you for the seismic work? No, it is so that at the end of six months of dust, noise and inconvenience, the customer will have something new and shiny to play with and point to, not just the knowledge that their cottage, which they had thought was safe when they bought it for $900K, is now officially safer.


5. Get out of the consumer business. Forget all the stuff above. Businesses have the money. You have already established a reputation for a stable, functional, well-supported product. Sell it to businesses, and let them use it, or resell it to consumers. It is a lot easier to know what a small number of users need and to keep them happy than it is to deal with all those whiny webloggers who want something for free. (On the other hand, at Geoworks we found that dealing with Nokia wasn't always a barrel of laughs either.)


6. Add some project management resources to your staff. This is the recommendation that as an outsider I am least sure of, and that is potentially most biased, since software project management is something I have done a lot of, but from the outside it looks like you are growing so fast, careening from project to project, that you aren't able to plan and organize. Maybe that's why communication has been poor, because you didn't know yourselves what was going to be in the next release or when it was going to be? If so, a good project manager can help.

That concludes the solicited advice, hopefully for good. I would be interested in SixApart or Anil's response, but I certainly don't think they are obligated to respond.