S3 API vs AWS
markclark at mac.com
Sun Feb 28 13:33:53 EST 2021
Since I posted the enhancement request for a non-AWS S3 endpoint, thought I might chime in a bit on S3. I worked with AB and Harsha (founders of Minio) in a past life when they were at Gluster (acquired by RedHat) and I was at a company called Compellent (acquired by DELL). Knowing those two I believed their notion of a developer-centric open source object store was a winning idea.
Kevin and everyone at LiveCode get a *lot* of people pulling at them for attention. I’m in awe of their ability to do so much with relatively few resources. Back in 2016 they were tremendously busy, which is still true for 2021:)
Currently I use Minio in a production environment and since LiveCode doesn’t natively support it, I wrote a simple server back end to handle client requests that require object. LiveCode is not what we use to build our products, but it sure has come in handy for prototyping our UI and doing IT style apps in the back office.
In retrospect I should have banged the drums with the whole community and pushed this idea of putting some money behind the request. But those darn day jobs get in the way, sometimes.
Minio + LC = Goodness, even if you have to use shell and mc under the hood:)
From: Mark Clark <markclark at me.com>
Subject: 500 Dollar Pledge: Livecode plus S3 API (object storage)
Date: May 26, 2016 at 9:33:30 AM CDT
To: Kevin Miller <kevin at livecode.com>
Sorry if this is seemingly off topic to LiveCode, but it does help illustrate cloud storage opportunities for LC with the planned/eventual improved networking. Minio libraries are FOSS drop in replacements for S3 so no issues for LC in paid or open source versions. Files must go to object to overcome limits of RAID and filesystems, so how amazing would it be for LiveCode to include s3 (which for better or worse is now the defacto standard for object storage) native commands for getting and putting data to the cloud? LC is a perfect visual means of tying together various complicated sub-systems to generate a new generation of rich GUI web apps.
I have been happy to support each campaign along the way and will continue to believe in your vision for LiveCode. But I sometimes wonder if failing to support or at least post a vision for cloud scale data storage for app developers is a bit similar to not putting a network piece into Hypercard. Would you consider a campaign of say 10k to support a set of minio widgets for LiveCode? I would happily pay $500.00 towards such a project if you felt it worthwhile to put forward.
Minio has three main components: a client and client libraries for accessing objects via S3; a single instance micro server for S3 (how cool if this were placed with LiveCode Server); and an XL version (still under heavy dev) which is a fully distributed and erasure encoded object storage server.
More information about the use-livecode