MT-compatible

Fredrik Lundh writes:

how long would it take to create an MT2-compatible web-logging application, btw? any reason it would need more than, say, three reasonably capable programmers and one calendar week?

That depends on your definition of "compatible." Do you mean:

  • Feature compatibility. Does it do the same things? (Interpreted broadly enough, any weblog software counts by pure dint of being "weblog software."
  • Interface compatibility. Do you interact with the features the same way? Is the work flow the same? If someone switched me in the middle of the night (and I browsed with images off), would I notice right away?
  • Template code compatibility. Do my templates work without changes?
  • Database and schema compatibility. With which databases is it compatible? Would it Just Work if I point it at my sqlite mtdb file? Is your Entry conceptually identical to an MT::Entry?
  • API compatibility. I bet that's more than three programmers' weeks in testing alone. I also hope you like Perl, unless you're going to clone that too.

Just speaking of how long it would take to clone MT 2.x is disingenuous anyway. It's not a useful measure of how much work went into building it, because the hypothetical three programmers have a complete design document--MT 2.x--that didn't exist the first time. Besides, no programmer would settle for cloning it; it's easier and more productive to go your own way with new inventions and mistakes. Where would you get the $3000 to keep three programmers on a boring task for a week? You'd need 8000 downloads at the pre-3.0 revenue rate, and good luck getting that without the good will Ben and Mena incurred and an already saturated market for MT 2.x.

Meanwhile, I would love to spend a week working on MT. Instead I'm stuck fixing my project of the past couple weeks that we should've outsourced in the first place. C'est la vie.

1 comments

Comments

Fredrik

comment

Close-enough compatibility, I suppose. That is, the basics (posts, comments, trackbacks, page templates, etc), some kind of import facility, and enough of the plugin API to be able to use (a reasonable number of) the most popular plugins.

And what I had in mind wasn’t so much the need for a clone (I don’t really care); it was more a reaction to the huge amount of energy spent on thrashing MT 3 and everyone involved with it. You’d only need a fraction of that to create a new tool… (but of course, why shut up and write some code when you can spend that time proclaiming your own moral superiority.)

The rest of your post deserves a longer reply, which I don’t have time to write right now (this text box isn’t wide enough, etc ;-). Maybe some other day.

Cheers /F