Stack blogs: IMO would be better as a feed aggregation of the blogs and similar articles from the top X percent of community members. Would be pulled from predictably formatted line in user profile. Be best if the "included crowd" had more intelligent algorithm than mere top percent tho'. Other metrics considered could be length of time of active engagement, etc.

Django used to do a bang up job of this, but they seem to have changed the format. Now only showing headlines instead of a meaty portion of the article as well. djangoproject.com/community

Not as useful or interesting, to me.

The aggregate way no one is ever faced with "where should I post this? I want to the stack community to enjoy it, but it's also mine, to be found with all my other stuff" when thinking of taking the time to write something up. The blog posts themselves would need to be tagged for inclusion also. No need to have "how I fixed my dishwasher" show up as a GIS post!

Change of topic, what I actually came here to comment about:

you know you're niche when: your top Q&A of last year, perhaps last several years, is rarely viewed and has paucity of votes. ;-)

Please don't vote just 'cause I linked to it (forthcoming); I'm not shopping for rep. It should be considered on it's own merits and utility for your work and problems. I'm just sharing because I find it interesting how some things that represent a significant personal investment of time and effort and utility can have a very narrow audience, all out of proportion to their "true value".

@mattwilkie I went to review your Q&A with a view to voting on it anyway and found that I had already upvoted both your Q and A already. So I did the next best thing and shared it to my Twitter feed. Hope that helps it to get a little more of the credit it deserves

You are certainly right about how voting can sometimes poorly reflect quality. Its one of the reasons why I would like to see a lot more voting (both up and down) on our site.

Your aggregated feed idea for our blog, as long as someone else wants to implement it, sounds like a good idea.