So, the only way I would EVER use something like this is if I controlled it. The only way other people are going to adopt this is if it's easy to install in your own environment.
I think that offering this as a cloud service is short-sighted. Not only is running this in AWS going to be prohibitively expensive if you're doing it for "free", guaranteeing that you'll either need to add advertisements or start charging for service (making you no different than Google, as a service), but running it in the cloud eliminates most of the advantages of using Google Apps.
I think that the website should be more focused on installing the software to your own server, and that if you have a cloud portal at all it should be a demo environment, rather than a planned SaaS solution.
It's also sort of out-of-character for a .sh FOSS website not to include some sort of .sh installer for getting everything up and running on a fresh Linux VM.
Given that this stuff is supposed to be a replacement for Google's suite, and Google focuses on ease-of-use, it wouldn't hurt to create apt/yum repos for installing official binaries so that people can throw this onto a cloud server and get up and running in a few minutes, rather than digging through your source and trying to compile all that Rust code from scratch.
I'm going to work on getting this running in my VMware environment, and I am going to play around a little with the docker container, but it would be really nice to be able to install this in production easily, and have the ability to update it in-place easily, because that's the only way I am going to play with it as a serious tool.
The warning that none of my data being guaranteed is pretty much a dealbreaker for me to beta test this thing for any important task if I don't control it.
We made it works, and thanks to your feedback we will make it great!
Our vision is as following: In the future, Google and Microsoft will be replaced by a federation of Bloom instances. boom.sh being one of them.
98% of the users don't have the technical knowledge/the time/the will to build, host and secure their own instance, and that's how we (and other instance providers) will make money.
You are right, AWS hosting is short-sighted, but it's because I've some credits, and easier to setup.
Privacy being one of our cor value, end to end encryption is on the roadmap.
I think that offering this as a cloud service is short-sighted. Not only is running this in AWS going to be prohibitively expensive if you're doing it for "free", guaranteeing that you'll either need to add advertisements or start charging for service (making you no different than Google, as a service), but running it in the cloud eliminates most of the advantages of using Google Apps.
I think that the website should be more focused on installing the software to your own server, and that if you have a cloud portal at all it should be a demo environment, rather than a planned SaaS solution.
It's also sort of out-of-character for a .sh FOSS website not to include some sort of .sh installer for getting everything up and running on a fresh Linux VM.
Given that this stuff is supposed to be a replacement for Google's suite, and Google focuses on ease-of-use, it wouldn't hurt to create apt/yum repos for installing official binaries so that people can throw this onto a cloud server and get up and running in a few minutes, rather than digging through your source and trying to compile all that Rust code from scratch.
I'm going to work on getting this running in my VMware environment, and I am going to play around a little with the docker container, but it would be really nice to be able to install this in production easily, and have the ability to update it in-place easily, because that's the only way I am going to play with it as a serious tool.
The warning that none of my data being guaranteed is pretty much a dealbreaker for me to beta test this thing for any important task if I don't control it.