Would like your expert wisdom on the feasibility of what I'm going to ask below:
SO HERE's the Background:
I'm currently undertaking a project where I'm automating some graphic design stuff with ExtendScript made thru Illustrator. I call the application "IDA" that takes input from user via ScriptUI and outputs graphic design "panels" at the click of a button. Currently, this is how the whole process is
As you can see:
1. All of this happens in a local PC and
2. There are dependencies required to run this applicaton.
NOW here's what I am ruminating:
I would like to:
1. Extend this Illustrator application to be accessible by any Internet user.
This would mean the process to look like below:
Do you see what I'm thinking of?
Is it possible to communicate with this Illustrator app virtually via ExtendScript from the internet? The goal is to extend the usability of this IDA script to users around the globe without having to depend on dependencies as much.
Your thoughts would be much appreciated! Please let me know of alternatives if possible to achieve it.
You need to read Adobe Illustrator’s desktop licensing terms. Server-based use at this time is limited to internal networks where each user of that server also has an AI license for their desktop.
I image as Adobe rolls out UXP support and cloud-hosted AI, ID, and PS apps, they will eventually offer the sort of model you seek. Until then, unless you are the sort of large corporate customer that can afford to negotiate custom enterprise licensing terms, you are really out of luck where AI is concerned. Either live within Adobe’s restrictions or go find yourself a different engine.
Not sure I follow exactly, but you could use a CEP extension to handle the interaction between HTML > JSX > HTML, even to the extent of embedding the actual site as an iframe if the intent is to have this accessible to users on their own machines inside their own copies of Illustrator.
If you mean it in such a way that some PC of yours will be running AI 24/7 and needs to know of website usage, then this is a lot different though possible. I'd have a CEP panel connect to a Firestore database or web socket and have realtime snapshot listener events which trigger when a new document (request) is created by the site (which itself will also be hooked into Firestore to read/write database entries), and the CEP panel would evalScript some particular JSX function per the Firestore JSON you're retrieving, save it to a temporary location on the local PC, then .ZIP it in the CEP panel and find some manner of the site being able to retrieve it (whether buckets or raw download links, believe Firebase/Firestore can also handle this).
It is unclear what OP means by “panels” (though I assume some sort of AI artwork/template). However it is clear they are talking about using AI as back-end engine to a public webservice, which AI’s desktop licensing terms explicitly forbids. Further discussion of using AI in this way is therefore academic, until such time as Adobe finishes putting its graphics app engines into its Cloud with solid UXP APIs and SaaS/IaaS licensing. Maybe in another 3–4 years; they still have a way to go.
AFAIK there is only one Adobe graphics app currently licensed for this type of use: InDesign Server; currently used to drive CHILI publish amongst other third-party products. I believe a single IDS license runs five figures per-year, which may not be in the price range that OP is hoping for. There are various third-party FOSS vector drawing applications, numerous free and commercial PDF and SVG rendering engines, HTML5 canvas, and no doubt lots of other possibilities. OP is welcome to look around while he works out his business plan. For anything non-trivial, mid-five figures is probably a good R&D budget just to get started; assume six figures to build into successful commercial product. Ergo, it’s a lot cheaper to do your due diligence before you jump in both feet first.