I've often thought about doing this. But I always end up coming back to the question: do I want to spend a month (or more) writing a blogging engine, or do I instead want to focus on my actual writing? And what will a custom engine actually give me, aside from sleepless nights and headaches?
Jekyll is a great answer in this regard. You don't have to reinvent the wheel (pagination, RSS, categories, dates, etc.), but you can still customize it with great ease. (Kind of like the Backbone.js of blogging engines, I guess.)
Some people probably don't find blogging platforms all that interesting and would really rather just say what they have to say and get back to whatever else they were working on. If someone asked me what's a good kitchen knife, I wouldn't say, "make your own" (even if that's what I would rather do).
> Alternatively: it's good to see what's out there so that when you do (inevitably) write your own, you know /what you want/.
Just as a counterpoint, I've realized recently that using other things before making your own changes what you'd make, and not necessarily in a positive way. You find yourself thinking in terms of the constraints and features of what you've used, rather than starting with a blank canvas and coming up with something you haven't seen before.
Well, my point is that I'm more saddened by the overuse of acronyms in language than by a developer who just want to find a nice blog platform without having to take a ton of extra effort to create and debug their own.
Jekyll is a great answer in this regard. You don't have to reinvent the wheel (pagination, RSS, categories, dates, etc.), but you can still customize it with great ease. (Kind of like the Backbone.js of blogging engines, I guess.)