Tumblr decided that my blog’s custom theme was “suspicious”, and stripped it. It’s hard to know what encouraged that decision, and what parser process behind the scenes allows “suspicious” themes.

It seems detached from the Heartbleed fracas, but who knows what’s going on. Tumblr is notoriously tight-lipped about the status of its platform. Once again I find myself questioning my continued use of Tumblr to host my blog.

Here’s a link my blog’s now-revoked custom theme: https://gist.github.com/mwunsch/c9325aaa9ea1f841e7e3

Not sure what’s out of the ordinary there, but feel free to use it as you see fit.

It’s time for a change, and that change will either be migrating away from Tumblr or just building out a new theme…

…or something larger in scope. Like revisiting Thimble

Update: Tumblr has a Last Account Activity area in your account settings and yeah, it looks like someone managed to get in:

I'm not in the Ukraine

My guess is that Heartbleed leaked my session cookie and someone modified my theme with bad mojo.

I’ve created a new toy1 inspired by an idea by Rusty Foster that creates a spinning newspaper from arbitrary text.

If you type in a query, you’ll see that headline front and center spinning towards you as befits its importance. Extra! Extra! Read all about it!

The page makes heavy use of newer advanced features of CSS including (the magnificent) Flexbox, animations, transforms, filters, and columns. As such, it only really works as intended in Chrome.

As a bonus, there’s a small feature that allows you to turn any tweet into a headline with a URL:



  1. See previous toys: AvatarPro.biz and Collegiate Bowl 

Shutting down Multiplexer.me

I have made the difficult decision to stop development on my α RSS reader, Multiplexer.me. If you’ll indulge me, I’d like to tell you why.

When I started working on Multiplexer.me, it was a reaction to the impending shutdown of Google Reader. Truth be told, I’d wanted to build it for a long time, but that was the kick in the pants that led to the first lines of code. Now, many new feed readers have emerged, with more time and resources behind them that I would ever be able to devote. More on them in a second.

Part of building Multiplexer was an effort to be as absolutely lean and ruthless as possible. I was willing to launch into alpha with the absolute minimal set of features plausible. I wanted to spend as little money as possible. I wanted the database to be de-normalized, the dependency graph to be lean, the tools to be reliable. Now I’m at a crossroads. There are over 60,000 feed entries in the Multiplexer database. My Heroku dev database has a limit of 10,000 rows. I have to confront investing dollars (not even all that many dollars) into this product.

Feedly and Digg Reader are both really good products, and because of their dedication have been able to provide a feature set that it would take me months of work to match at my current development pace (which is slow). I have a full time job I’m happy with, Multiplexer.me’s feature set is sufficient enough for me to use, but lacking a lot of niceties. By investing recurring monthly costs into the product, I’m paying to play catch up.

I’m very proud of what I’ve built, but I’m unwilling to invest in something other products have solved better than I could have. But I still think there’s room for them to improve.

RSS is an implementation detail

I think the very real fear that commercial readers should be feeling is a reality that RSS as we know it is dead. There will soon emerge a new set of Content Management Systems that will not feel the need to build out their RSS feature set. But RSS and Atom are implementation details. What reader products must keep in mind is that the killer feature is syndication. This is powerful, and if RSS readers want to continue to exist, they need to invest in helping publishers understand the power of syndication. This is what the web is fantastic at — what it’s built for. Build the next set of tools, the next set of standards, the next set of linters, the next set of analytics. The real goal of Multiplexer.me wasn’t just to build an RSS reader with a sustainable business model. It was to build an ecosystem of tools to connect publishers with readers. This is what Feedly and Digg need to be working on.

Thank you to everybody who signed up for invitations, and sorry to those who were holding out hope for one. I’ll be personally switching over to Digg Reader.

I do plan on releasing some of Multiplexer’s functionality as open source software. Look for that and, if you’re so inclined, carry on its torch.


When I meditate through they eyes of my grandmother, I see myself with less judgment. I don’t know if I will have kids. If I do, I want them to be a little bit more informed than I was, and from the horse’s mouth, at that. I’ve written a letter to my unborn child for when I transition from independence to dependence, life to death, in the hopes that I might ease the horror of this process of abandonment.

A Letter to My Unborn Child — Cancer Stories — Medium

My wife wrote this piece, and published it on the Internet. I think it is so moving and beautiful.