Let me get this out of the way.
Do I think the WordPress community needs another solution for WordPress news? No. 100% no.
I love what my friend Brian is doing over at Post Status. He's curating news. He picks for me. I don't have to stress out that I'll miss something important because it's likely he'll point it out.
Now, to be honest, he's just one guy. So I also don't expect it to cover everything under the sun.
Thankfully, Jeff, Sarah and Marcus (I only know Jeff & Sarah) do cover everything under the sun over at WP Tavern. So it's another place to get my WordPress news.
And let's not forget the place where I write weekly, edited by Michelle. I'm talking about Torque Magazine. They're somewhere in-between the two, not covering everything, but writing more than what one person can handle.
So like I said, does the state of the WordPress community, when it comes to news, require anything else.
I don't think so.
But wait…
A couple of months ago I had an idea. I talked about it on a podcast. For some reason I think it was on a show Matt ran, but honestly, that guy runs 4 or 5 great shows, so how could I ever remember which one?
Anyway, my thought went something like this.
- I would love curated news.
- But I only want it from people I trust.
- But I know I can't get them to vote for articles.
- But what if I used the articles they tweeted about?
- What if every tweet was a vote?
- Then all I really need is a Twitter list of people I want curation from.
- From there, I'd grab tweets with links.
- Expand the links to find actual articles.
- Store each one and every time I saw it again, give it another vote.
- Print these to the screen – maybe only top fifty – in order of votes.
The thought would have ended there.
I have a day job…
In my day job I've just hired several fresh college graduates and have been running them thru a program where I teach, train, coach and challenge them. I create assignments that develop them. All before putting them in actual positions in our company.
What if I used this idea as an assignment – where we worked on APIs (like twitter), looked at batch processing and calculations, and ended with a bit of GUI work? It sounded like the perfect assignment for week 14.
Week 14 was a couple of weeks ago. And so I gave the assignment to my four new staff. And we iterated over several different architectural designs, mostly as a way to help them learn different things. (Doing all the work when a user hits refresh is a really bad idea!)
But the result, while still an experiment, actually worked. I now had a little pintrest-like display of popular articles.
Still more to do…
Yesterday I wrote about experimentation and affordable loss. The version you see is experiment 5 down this line. Every one of them tested something different. Every one of them allowed me to stop and walk away if I wanted. But so far, it's been fun.
This next week, we move on to something else. But one of my four wants to keep working on it. We still have to do work with the images. We still have to do work with the algorithm (to deal with self-promoters and tweet repeaters).
But it works enough for me to leave it up for now. No promises about next week, or the week after. It's on an Amazon EC2 instance and we'll see if that makes sense or not. It's using MongoDB, and we'll see if that makes sense or not. It's not open source yet, and we'll see if that makes sense or not.
But for now, the experiment worked enough to put it up and have some friends look at it.
You know what you get when you experiment?
Here's what I know. Whenever I create experiments and test ideas, I end up with at least three results.
First, I get engaged employees. People like small, tiny tests. Because they get immediate feedback if something works or doesn't. They don't have to wait weeks or months. They like it and thrive on it. And it keeps engagement high.
Second, I get better ideas. Until you run a test and see a result, you can't really comprehend where your brain will go next. Or someone else's. But once you see something, you start imaging other things. Creativity is the result of tiny experiments.
Lastly, I get (negative) feedback. It doesn't matter that I told people it was an experiment. Some folks who saw it immediately started highlighting what was wrong with it, or what they would change. It was fine. I wasn't emotionally attached to this experiment. And this isn't my first product. But if you're just starting in product development, be aware that there will always be negative feedback. Smile and nod. Smile and nod.
What are you experimenting with?
So tell me this – what experiments are you working on? What are you testing out or trying?