(2025-08-17) Hon S20e04 A Software Strategy

Dan Hon: s20e04: A Software Strategy. This one is about the cycle of starting by building something you need, and then having it be replaced by a commodity.

A regular problem here is that there’s invariably a bit of software that would help in distinguishing, but it comes down to the question of whether you should build it yourself, or whether you should customize it.

I should point out that at one end of the scale, any sufficiently complex job that is pitched as being accomplished via “customizing” existing software ends up being as complex and expensive as just building the damn thing yourself.

The irritating thing that keeps happening is that people will invent new things to do in new ways with software and then some of those will catch on and then before you know it, the new thing in software has suddenly become a commodity thing.

*Some people will look at this and say: what a waste of money! Why should we build a CMS when we can just use Wordpress!

And this is why I think Vox Media is interesting.*

A long long time ago in internet time, Vox Media was a bunch of blogs and when they were babies and making blogs they decided to write their own CMS

This CMS was called Chorus. I hear it was good because it let Vox properties do what they needed to do in terms of publishing articles, but also what they needed to do in terms of other things like “selling ads”.

Another thing that happens when you develop and run your own software is that someone somewhere will say: “hey, if this is so good, wouldn’t other people pay for it?” and that is totally a thing that happens with news organizations and content management systems. It never really works out. But I think that is not the main point of this story.

Anyway, after a while -- say just over ten years or so -- Vox decided to switch to WordPress. This was sad, because Chorus was apparently very good and they had invested a lot of money in it, and it felt like custom software losing to commodity software.

But!
A couple weeks ago, Nilay Patel of The Verge announced a secret side project he’d been working on at Vox Media.

It’s essentially a bet on... social? Ish? Software to publish stories natively to a feed, in smaller chunks, for things that Aren’t Stories, but are smaller. And it’s launching with SB Nation, which makes sense for a social software product

The thing is SB Nation Communities, and Ted Han and I got to talk about it with Andrew Losowsky last week

I’m only going to talk about one of the interesting things here. I’m not going to talk about what SB Nation Communities is, or how it works, in terms of being interesting social software.

What I am going to talk about is how I think this is a great example of managing software development.
Build the software you need for a competitive advantage
Keep developing that software
Over time that functions of that software become commoditized

Then you’ve got a choice: do you switch to the cheaper commodity software?

Some organizations do

They also take the opportunity to outsource all software development. You can then lay off all your software development capacity.

But that’s not what it looks like Vox did. What Vox did was start working on the next thing, the next bet for what would make them distinctive. And that’s this Communities product.

You don’t have to use custom, in-house software forever. There probably will be a time when it’s worth it to switch, and one of those reasons is because you need or will benefit from functionality that doesn’t exist yet

*Vox’s behavior is is more like “we don’t just publish news, we know what it is to be part of information on the internet” -- which isn’t just publishing news.

(And also has a lot to do with knowing, learning, and retaining a lot about running healthy communities online).*