Right now just analytics, but we are considering other inputs too, as we know many devs have data in other sources which can be used to bolster prediction quality.
Firebaser here! We've been investing pretty heavily in the platform so I think we will be around for a while. Worth noting that YouTube, Double click and Android were all Google acquisitions too!
This knee-jerk-ism is tiresome. There are plenty of products that Google has retired, and plenty of products that Google has continued to invest in. Just like any other big corporation.
(acquisition still running, top of my head: Google Docs, Android, YouTube, DoubleClick)
The problem is the google isn't forthright about it's plans. They just let some products languish forever without any communication. If you're lucky (like a firebase user) you will get a giant code dump of new features after 18 months of radio silence. If you're unlucky (like a google app engine standard user) then maybe they will announce a shutdown. It could go either way - you will never know what's coming. Yes you will get time to migrate away, but you're not really treated like a partner.
At some point they said google code wasn't getting new features because they couldn't find anyone who wanted to work on it. WTF? That's what employees are for right? I cannot understand this company. Why waste peoples time with projects that are just rounding errors next to adwords.
And is it anymore than other big corporation? That pages lists 24 products, which doesn't seem that much to me. There have been plenty of stories about retired products at FB, Amazon, Apple, Microsoft, etc.
Im not disputing that Google does this, but ever since Google Reader was stopped this has become a stock comment on basically every HN discussion on Google products (other than search & ads) and has long ceased to have any value at all, and seems to be just accepted as some gospel without any evidence that this is a bigger problem with Google than the other tech giants.
in jannes' and hk__2's defence, that Wikipedia category page looks very incomplete. I'm sure the number is far higher than 24.
Perhaps it's HN apophenia but Google's record of discontinuance does not seem insignificant - it is valid for anyone to be skeptical of their ability to maintain any newer product in the medium term.
That said, I think anyone who's followed Firebase closely since the acquisition should have reasons to have more confidence in their commitment in this one particular case.
The list forgot "Google Desktop Search" - One of my favorite google product.
I like it so much that I even found a old copy to install on windows 10. 6 years after it is discontinue and still much more useful and much faster than windows 10's search bar.
They have since acquired more companies to complement the offer and released a new product called Cloud Firestore very recently. They are also doing conferences promoting it in lots of places. Doubt it's going away very soon.
Take a look at https://RetroAnalytics.io - It's a predictive web analytics platform. Similar features to Firebase Prediction though built for the web and mobile and you aren't locked into Firebase.
Disclaimer - I'm the founder :) Would love your feedback !
Graph.cool just released their development framework so I may investigate.
I realise Firebase and GraphQL are very different beasts but I think they serve the same market. It would be good to see Firebase at some point offer a GraphQL interface but the data structures are very different.
My company, Gyroscope Software (getgyroscope.com) does prediction, cohorting and timing of callbacks into mobile games and apps, with plugins for iOS and Unity. We've built a real-time prediction engine to make this happen; we also use some neat tech to get us enough data from the clients to do fast machine learning without the developer having to do any instrumentation, data collection, or other data wrangling - we do this automatically. Happy to answer any questions anyone has.
I think this is interesting, but I'm having trouble seeing how it'll be used. Personally, when I want to do things like this - I roll my own custom solutions because I want more control.
Given that, I guess I'll be giving it a shot (although I think Firebase in general is super over priced)
- Build collaborative applications with real time updates and fine-grained permissions.
- Our JavaScript client for browsers leverages IndexedDB to work completely offline and synchronise data when online.
- It’s even possible for data to be encrypted on the client to keep user data safe on the server.
- Permissions can be set on the stored objects, making it possible to share data between users.
- Kinto is an HTTP API in front of a database. Interactions with the server are simple HTTP requests rather than complex SQL or map-reduce queries. It is meant to be minimalist and simple.
This would be a self-hosted solution. They have a pre-built docker image to spin it up.
Have you heard of RetroAnalytics? It's a predictive analytics platform that offers similar features but without the need to be locked into Firebase. https://retroanalytics.io
We charge a monthly fee which is based on the sessions you are getting and then we have a few different tiers. We should have our pricing up on the site in the next few days and we have a 30 day free trial to make sure you are happy with the product, feel free to reach out to me - david at retroanalytics.io.
Use cases vary with your app need, but the Halfbrick (fruit ninja etc) case study might help: https://firebase.google.com/downloads/Halfbrick_Case_Study.p... - we worked with them during the alpha testing and I think their approach will work for a lot of apps.
For what sort of applications is it super over-priced? I'm using the realtime database to build out an app, and if I start scaling my current (beta) usage by 1000x, I'm still only getting monthly bills of $100 or so.
Fear based question: are there applications (hopefully mine) that would scale in an economic way? And there are other use-cases that wouldn't scale?
It really depends on the application. Certainly, everyone love control, but there is a trade off between control and cost. Firebase prediction is simply a simple implementation for some simple use cases, which are plenty.
What kind of revenue should you be earning from a popular app on Firebase to justify the cost of using Firebase (for an indie developer). It looks great, but I'm afraid I won't be able to afford it.
I'm currently researching open source solutions such as Parse.
I haven't used it much but from the pricing page it seems affordable enough. There is also a calculator at the bottom for big apps where you can estimate how much it will cost for the plan at scale
Firebaser here! We don't generally target all use cases, just event prediction from a stream of anslytics events, which is a lot more constrained. Give it a try, it's pretty robust!
To me it seems like Firebase essentially supports the cancer of the (app/software) industry:
Creating/buying an app, making it popular and drawing every last penny out of it while pissing off the core users who helped to make it popular and then letting it die. Rinse and repeat.
It helps you piss off the core/current users a little less to draw out the process as much as possible and make more profits, but the result is the same.
Using optimization strategies and software like Firebase disconnects you from the user base as you turn your users into the product. Of course this would come from Google who are experts at it. They get away with it though, because their software is great and there are few alternatives, but that's not the case in the app market.
If you're pissed at predictive modeling being used for apps, I assume you're incensed at predictive modeling being used in general marketing, where it's been in use for many decades?
It's nothing new, but I still find it amusing how much the IAP model has borrowed from the marketing handbook of narcotics dealers.