Hacker News Reader

Comment by joeevans1000 | Hacker News Reader

This whole thing is such a sad tale. Clojure has had lame adoption, starting with its hard to pronounce, hard to explain name. Datomic could have been a great booster for the language, but had a license that is counter to all that has been propelling the programming world. Now they have swapped that out for another confusing licensing scheme. It feels like just another chapter in the long weird saga of lisp non-adoption.

I don't understand why Rich doesn't understand that he'd make far more getting a tiny slice of the huge pie that open sourcing would provide. Yes, they've changed the licensing approach because it's obvious the previous one didn't work out at all. But they haven't done enough, and by the time they don't figure this out, people will be stuffing everything into cloud services, even if they are inferior approaches.

View on HN

What is confusing about this licensing? Previously it was a mess and prohibited our use case, but this looks pretty straightforward.

"The Datomic Starter license is perpetual, but is limited to 1 year of maintenance and updates".

No updates after a year? To the new wave of developers, this is foreign. Most won't know what that means, but at first glance, it looks like you'll be paying $5,000 in a year. Even the most green developer knows you need updates. Also, it's unclear what a 'system' is without digging around, and many potential users won't bother.

The problem is that Cognitect knows how awesome Datomic is, but the world doesn't. Many potential users are not going to try software they know they're definitely going to have to pay $5,000 for in a year. Updates are critical. Cognitect should go for massive adoption and nothing else. They'll make far more money.

This thread shows continued confusion and uncertainty about the license, and if a potential user googles 'datomic license' they are likely to find this thread. After reading up and down it, they will be deterred from getting into it all. Why try Datomic and get into a potential mess with your employer over licensing issues/costs in a year when you could play it safe? I'm sure those who think this licensing approach is the right one are clear the advantages of Datomic are worth the risk, but they are not paying attention to how software is actually adopted at scale.

It's so sad it's almost 2017 and this issue's been around since 2014 ( https://news.ycombinator.com/item?id=7657701 ) and this is the solution. The word Datomic means nothing to 95% of mainstream tech. This change is a modification of their old model. They need an overhaul of how they are doing this.

No offense - but if $5K is an issue for your employer, after a year of essential magic, you might have the wrong employer and they might have the wrong employee...

It seems like the latest generations of software developers (which includes myself) is hazardously opposed to spending money on software, even when it can save/make you huge amounts of time and money.

I was lucky in that my first job was at a small non-software company that forked over huge amounts of cash for Oracle because it literally enabled their business to succeed (mysql and postgresql were nowhere close to being able to do it at the time). And yet I've seen fellow developers waffle over spending $60 on something they will use literally every day.

Reply on news.ycombinator.com