Here's a minimal Node.js app that sends a message to Slack.

First what makes Slack cool is their wide-open, anything goes, API.

You want to hook into Slack! Great! Go for it.

And they should be really open to it, and the way things are set up I bet they will be open that way for the foreseeable future.

But -- what happens when there are Slack clones? Other platforms that emulate the Slack API? Because that's going to happen. I'm sure dozens of developers are getting ideas about what it would be like to be in Stewart's shoes. I bet it would be fun!

The Slack of pizza delivery. The Slack of wedding parties. The Slack of the NBA player's union. Slack aims to be all that of course because of the API. But maybe others can slip in there. How did LinkedIn become the Facebook of employment, for example. Or eBay the Amazon of auctions. Lyft, the Uber of hipsters middle-class people.

And, if Slack weren't raising so much money, it seems they could be open to that kind of compatibility too, but they are one of the biggest "unicorns" -- with a multi-billion-dollar Uber-like valuation.

So not every VC is in Slack. Google, Facebook, Microsoft, even Apple or Amazon, probably want to own Slack. They're probably all trying to buy them. Only one of them can, if it's for sale. My guess is it's not. They could grow to be Twitter's size in a year or two. And be on track to be another Facebook. Scaling Internet applications is way easier now than it was when Twitter and Facebook were starting up.

But the API is going to be a problem. There are plenty of would-be competitors, and the API is a good vector to compete on.

I asked how long before Slack brings its APIs under control, as Twitter did, and my friend Chuck says that's all Slack is -- the API.

Bill Rice (and others) says it's the backplane for all the cool little tools the Internet and VCs throw out there.

Ben Thompson says people like that Slack monetizes through customers not advertisers.

I want to be able to put together web projects quickly with no more work than is needed. I want to be able to easily connect my work with other people's work, again optimized for my efficiency. Less cutting and pasting, more systems that "just work."

Right now we seem to be far away from the ideal, and going in the wrong direction, as silos are hosting more of our work, and it's optimized for us consuming the most valuable advertising, and sharing our most valuable info with the machine, and our connectivity with each other not even on the table as a consideration.

However, Slack is a bright ray of sunshine in a VC-backed world of silos. How is it that they can be so successful yet still have an aggressively open product, and others like Medium steadfastly refuse to be part of a network of software?

What a relief to find a product that can become part of my roadmap, without me having to re-implement the whole thing just to get one feature.

I got curious to know more about Slack, so I asked on Twitter and Facebook, how people use Slack. I asked "Why do people love Slack?" Interestingly, no one said "I don't love Slack." You'd expect some dissent, but people seem to genuinely love it.

I looked to see if there was any Facebook integration on Slack, and didn't see anything. Facebook has a good API, I wonder why there was nothing listed.

Now, if there were some way to connect Slack and Liveblog. It looks like all the limits that make Twitter unattractive are down in Slack. It has no 140-char equiv. Their API is something they've actively built around. That's what I'm wondering about. I'm mainly interested in connecting with smart people who are paying attention. Slack looks like it might be a pretty good place to do that.

Also, corporate APIs are dangerous. That's a basic lesson, learned over and over. No matter how friendly the company is, they could sell to Google and have their API deprecated. People seem to trust the Slack API. Is there a special reason why??

BTW, what a different approach Slack took to Medium, which still has no API. Instead, Slack said basically Let's Do Everything. This is consistent with Stewart's past design with Flickr, which had a great API. I did a fair amount of work with it.