Hello Luke,
I have been thoroughly reviewing some headless configurations lately for a 'product demo' that primarily revolves around Content Management.
However I don't find time testing all the various solutions available right now. However again I recently found and discovered localwp.com very useful in terms of efficiency and usability that finally got me out of the time killing trouble shooting with my bitnami local developments. I really hate the traditional MySQL and Apache fidling.
I first stumbled upon Gatsby as a WP alternative and was caught up between Ghost.org and Next.js but finally decided to keep it simple as I want to focus on a design based approach and not developer advantage (hbs has time to catch up) and besides the only available WP static plug in developed by Leon is now depricated (and other commercial solutions available) I didn't get to completely test it with a demo project I was seriously considering developing my self.
But with Gatsby 2.0 (and gatsby cloud) I presumed WordPress.com like offering for a WordPress.org alternative. But lately with WP offering cheaper premiums over hassle of hosting WP and managing Content and additionally having a headless static version is becoming more easier.
However again I reconsidered gatsby cloud as a Tumblr static integration as I found it good for portfolio like sites but when it comes to serious content publishing I guess you are better off with medium.com and wordpress.com
On the side note ghost desktop app and it's various integrations like snip cart etc, are promising but from a designer stand point tinkering the handlebars for customisation (which is not so complicated as react components and layouts that gatsby uses) is becoming more of management and skill to add up with considering the efficiency and ease and fimiliarity of wordpress ecosystem.
Besides snip cart is so down right dumb that a recent stumble upon how to use it with Tumblr got me rethinking on rapid e-commerce development for clients. With more than one project to handle and more than one platform (gatsby, ghost, nextjs, wp) to adopt and more than one profiles/portfolio to manage (Content writing, product design and UX research and creative design) I'm increasingly getting over hauled and burned out with what is the next best alternative integration suite of stack or framework to employ for smoothly scaling and most importantly customizing a SaaS product or Content rich website or a fully customizable e-commerce app.
With shopify now welcoming developers to offer a more customizable design solutions to sellers, while ghost.org offering more security, Wp offering more premium solutions, overdoing a product development (headless CMS) just for the sake of it when 90% of clients don't understand the need for a static alternative, I don't personally see an advantage (as an UI designer) that gatsby cloud offers considering once again how things (api's) won't go smoothly with a recent gitlab hosting glitch that occurs since 2.0 update, rapid product demos are not feasible right now to recommend gatsby cloud on a production scale.
It was like dejaVu moment when just last year I found myself in a similar situation where setting up a ghost local website (and a ghost premium pair) only left me asking for more where I did be rather focusing on content development (as I was considering then moving away from WordPress.com where I had a product demo staged (editorial / content site) and the static version hosted separately) where I got my newly put up content got stuck up in a json file as my trial expired. That wouldn't be the case with wordpress or tumblr or medium (if a content driven product).
As a product designer I gauge for flexibility and freedom to choose between content, management (data migration) and especially design and finally ease of deployment. I had to abrupt the production blog to focus on a lavarel SaaS project.
Atleast from a developer stand point I could say that things are better off when you have the maximum control which wp has a history of natively offering. Now with localwp design has become a priority again.
Looking back where I first worked at a webhosting/agency managing multiple client (Joomla, wordpress, drupal) accounts and hosting was a breeze, where today ghost native hosting, gatsby cloud, datoCMS make it cumbersome for a design first content first approach.
Things don't really function as Open as they seem to. However I would continue testing and tinkering my gatsby-tumblr site for personal use. It is still a long way to go to consider it for design firm or web agency to mass produce CMS for clients no matter how efficient and advantageous GraphQL or handlebars.js is.
I don't have the freedom to formulate my product ideas if I have bunch of local repositories piled up waiting to be pushed and pulled for revisions where one is hooked up with composer, another with node, another with gulp, couple of them with git, where bitnami and localwp just put you back in productive mode. Gatsby cloud and datoCMS are suppose to something similar, to drive the product development cycle smoothly. Even from a business stand point I would rather ask a digital retailer or Content producer of a startup to put their eggs in a different baskets.
Finally I did say this much even though I do not have a vast production code developer experience that JamStack is slacker sort of way of doing things where it benefits the developer backend work but does not give a front end edge in terms of customization especially in no code or zero code future. For now I did like to keep things simple.
Thanks.
Abu.
