Disclaimer: I was one of the founders of the business, but left a long time ago. It’s still a great service! We print a newspaper for our friends and family each year.
I don’t believe it’s ever been accessible for free. It’s just that ownership has moved from the state to a private company and now it’s difficult to make it open.
I know Dave personally, and I can assure you he’s read Seeing Like a State! In fact, if you spend more than 30 seconds in his presence he’ll probably give you a copy.
Our experience of moving from Heroku to CrunchyBridge has been very similar - excellent help with the migration including jumping on a call with us during the switchover to resolve a broken index.
Would strongly recommend them to anyone looking to move off Heroku.
I was a bit concerned about the cut-over from the old database on Heroku, really wanted to minimise downtime. So they helped me produce a step by step plan, test as much of it as possible, then had an engineer join me on Zoom while I made the switchover. They were even able to accomodate doing it in the early morning in my timezone to minimise the impact. Ended up with maybe 5 mins of downtime which I was very happy with.
I refer to Fly.io’s guide to Safe Migrations in Ecto (Elixir’s DB adapter) multiple times a week. It’s a very useful quick reference to check whether you can get away with a basic migration or if something more involved is required.
By using client side rendering you’re effectively playing SEO on hard mode. It’s all possible, but you’re making life very difficult for yourself.
Google will crawl and render client side only sites, but the crawl budget will be reduced.
The bigger factor is that Google cares a lot about long clicks - clicks on results which don’t immediately produce another search or a return to the results page. Client side rendered sites almost always perform worse from the POV of the user and therefore convert at a lower rate.
And now Web Vitals includes things like Largest Contentful Paint and Interaction to Next Paint, you’re going to find it much harder to bring these metrics under the target thresholds.
If you want to perform well in search, make things easy for yourself: use mostly SSR HTML and CSS and some sprinkles of JS on top.
"Client side rendered sites almost always perform worse from the POV of the user and therefore convert at a lower rate."
I assume that people who create client side rendered sites disagree. Surely no one wants to make user performance worse, and I struggle to see advantages sufficient to offset that.
Generally very few people care, unless it’s "an impact metric" tied to that team. Or there is a "performance sprint" or something. "DX" has been more important than "UX" in the current mainstream JS community for a while. To give an example: babel, that enabled syntactic sugar, also caused many versions of runtime implementations of various ES6+ syntax to end up in bundles, or polyfills for browsers the clients aren’t actually using etc.
DX can't be at the cost of the user's priorities and needs.
Whatever is developed is for the user, not the developer.
If developers making their lives easier (or appearing to through more layers of abstraction in some cases) is more important than the user, the users will go where the best experience is.
Beyond things like developer products, End users overwhelmingly don't care what anything is coded in.
it’s not true. Many people are forced to use MS or Google products, people are using things for many reasons, like where their community is (Instagram, FB), or what they are forced to use by their employers. Or for example one tool has bad UX but is complaint in one way or another. It’s simply not the case that people "will go to where the best experience is".
I even admin and moderate FB groups and it is a constant struggle against the system - but that is where the community is, I have no choice. Other admins do not like it either, but it is where all the related groups are, and people are nervous of relying on solutions from volunteers (e.g. me setting up an old fashioned forum).
just check https://grumpy.website there’s lots of mainstream, popular apps/websites/tools that have plain broken UX and users have to work around that, but keep using them.
Cost is one of the most important criteria for end users; often much more important than speed. If something lowers the cost of development, that has a direct benefit to end users.
I believe it’s not true as well, UI specialists aren’t often working with backend templates and backend release cycles, and backenders generally have bad UI skills.
Either way, more work to generate the same HTML/CSS in the end creates a longer experience, one that's too often more brittle in the long run.
Where SEO is involved, simplest wins. There's a reason why Wordpress focused on making words easy to publish, organize, and connect, and from there try to be a CMS.
Chromium is open source and both Google and Microsoft do whatever they want to it as part of developing their browsers. WebKit on iOS is a closed source blob of rendering engine and assorted bits that it is not possible to deeply extend or alter.
Disclaimer: I was one of the founders of the business, but left a long time ago. It’s still a great service! We print a newspaper for our friends and family each year.
reply