As with all "let's squeeze all we can out of this" you will continue to make money for a number of years no doubt but you've just destroyed a major onboarding ramp (free tier), your security appears to be a joke from the outside looking in, and your product has been effectively on life support for many years now. A public roadmap is too little, too late. You've lost the trust of developers and it's only going to be downhill from here.
> This does have tradeoffs, but getting the rug pulled out is not one of them - the opposite.
I'm sure the developers with apps on the free tier don't agree and I'd bet good money they will never touch Heroku again if they have their way. I know I won't.
When something gets the "Cloud" rename, you can bet it's on the way downhill.
I'm a grad student who used to use Heroku a lot during undergrad (multiple university projects, used to be my first place to deploy stuff). Just deleted all my apps and removed my credit card, never gonna touch Heroku again with a 10 foot pole.
Heroku has a decade-plus of goodwill and developer recognition, and that is being burned to the ground rather quickly.
How about acknowledging that the Free tier is going away because Heroku is basically in keep-the-lights-on mode at this point? The number of engineers who have been laid off or quit has gutted the company, to the point that fighting abuse and spam is not possible, nor is active feature development.
I've submitted a support ticket several times and get a canned response from some poor sod in India who has no idea what is going on. Heroku's Support used to be the model of "how it's done." Now it's a joke.
And security is a joke, as demonstrated by the April "incident" that lasted two months. Reading between the lines, it seems that nobody knows what exactly happened, and the team is probably still waiting for more fallout.
I don't envy your position Bob, you've probably been told to kill Heroku by your leaders, all of whom have never used Heroku nor can explain what a dyno is.
A sad day in the developer world indeed.
We could cut our price by about 50% moving to a competitor. We suspect AWS RDS will work very similarly to Heroku Postgres, and I have been unable to get much clarity from the teams at Heroku on precisely what Heroku Postgres is doing for me that AWS RDS would not do. Is it possible to find out precisely what Heroku Postgres is getting me that AWS RDS will not?
There's always a cost with transitioning, so if there would be some kind of price reduction possible for Heroku, that would eliminate me looking at competitors. I suspect this is out of the question, and you wouldn't want to comment publicly, but I sure would like a reply somehow indicating there may be some plans for this.
Some of the reasons I'm concerned: * the GitHub security issue that lingered for over a month * the DNS issue that hit the other day that resulted in our apps being only spottily available for multiple hours
Missing features, such as: * the lack of wildcard in Heroku Automated Certificate Management * having to share a load balancer with free dynos that might be doing suspicious things and therefore getting our apps blocked at certain customers, even when we're using Heroku Enterprise (this is one reason why I'm okay with free dynos going away, since we've been bit multiple times by this issue over the last decade)
Looking forward to a response - thanks!
(disclaimer: i work at abstra)
Free tier for deployments + AWS Free tier = year or free and convenient hosting
It was clear even before their horribly bungled GitHub security incident that Heroku was on life support at best and it's been a long time since "Heroku" was the answer to "What PaaS should I use?".
The beancounters took control a while back and are sucking all they can out of it before they discard it's empty shell.
Having Heroku as your PaaS provider seems like a bad business decision at this point. You are just begging to have the rug pulled out from under you.