I see that you plan to GPL the firmware. This is a great first step towards a truly Maker friendly device! However, the reliance on the cloud service is still a sticking point for many people. There are many situations in which a cloud service is impractical, and many are rightfully concerned about the future, whether that’s Glowforge ceasing operations or being acquired, or simply deciding to monetize what was previously available for free.
In other comments you have stated that the Glowforge will have an entire Linux environment onboard, which is great! However, without the ability to control the machine directly without voiding the warranty, we are still dependent on the cloud service with all the associated downsides. Will Glowforge commit to offering an open, published, supported API directly from the device, for those that wish to develop their own software?
Thanks for the question, Patrick. I hate saying “no” to customers but not as much as I hate dancing around things.
It’s possible we may do that at some point, but I don’t see it making our roadmap in the near future. We want to spend our time on engineering that will help as many people as possible, and we’ve decided that building our services on the cloud architecture is the best way to do that. We may be wrong, and we may change. We may have overestimated the demand for that , or underestimated the demand for an alternative, offline environment.
But given our desire to make the very best product possible, as quickly as possible, I just can’t devote resources to doing things two ways.
I’m sorry. I wish I could promise that we’re going to do this for you, but I can’t.
As someone who works in the tech world, cloud services are all the rage, but they come with so many drawbacks for the most dedicated/creative/driven users of a product.
While I completely understand your desire to do this as a cloud service (primarily as it gives you open-ended control to do what ever you want in the future like monazite things that may initially be free, inject ads, add subscription levels, build other products and services around a device that is dependent on the cloud, etc) all of that can ultimately cripple the users that most want your product.
I’ve used a lot of products that are dependent on cloud services, and for the most part they all leave me upset, wanting more, horribly frustrated, and not in control when I most need to be. And when you are talking about a product that serves the needs of the creative market, control of the creative experience and process is key.
I know there is no changing the fact that this product is currently tied to the cloud, and I reluctantly pre-ordered the pro model despite major reservations about it being too reliant on the cloud and seemingly too user friendly to allow the kind of open-ended creative freedom an artist/maker ultimately needs. But I sincerely hope that you consider future updates to the product that would untether your customers from the cloud and let them be truly free to create on their own terms.
I know in the long term you can make a lot more money off a device that is tied to the cloud, but offering a path to freedom for those in the creative community that desire it would be a great gesture.
If the firmware is open, anyone could write code to control the machine via a different web app, or a simple change to the firmware to visit a different site or take different input.
GlowForge could have the software held in escrow, in case they go out of business or get taken over by an entity which doesn’t wish to continue providing the software. This is actually a rather common practice: https://en.wikipedia.org/wiki/Source_code_escrow As a smaller company, they could probably have it held by an open source agent. But it would take some good lawyers to make sure the release wouldn’t be adversely affected by bankruptcy laws or ill will of a takeover entity.
Oh wow, I never really thought about where the brain of my machine would be held. This yields a few questions:
1.) Is there any promise that the machine I’ve pre-ordered won’t be held hostage by Glowforge or owning company in the future?
2.) Does this mean that the $6.3MM of machines to be owned shortly are potentially all reduced to boxes of plastic and bright lights if the service were to be DOS attacked? Has Glowforge considered this and possible mitigations?
3.) Server side deployments can mean the introduction of bugs/functionality changes that mean something that worked last week may not after a deployment of the “cloud” architecture? Will there be help services in this case?
As a hobbyist level of user, who is not yet expecting to rely on my Glowforge as a source of income, I’m not as worried about the impacts of the above questions, but I do feel some apprehension to know that the end machines stand to be bricked if the mothership goes down.