3 ms·
For anyone who was as confused as I was: Carson Gross, the author of this essay is also the creator of htmx. I thought at first that the essay was going to be
by mtlynch 2y ago
For anyone who was as confused as I was: Carson Gross, the author of this essay is also the creator of htmx.
I thought at first that the essay was going to be a joke of "all these weaknesses are actually strengths," but I feel like most of the criticisms in the essay are valid. Like I don't know why you would use var instead of let or const in modern JS, and htmx does indeed increase the attack surface for XSS.
I guess he's just being self-deprecating? Saying the mean things about his project to prove that he's more aware of them than the trolls?
For context, I personally like htmx even though my experience with it is limited.
- twiss 2y agoFor the first one, I think it's linked to the next point of "No Build Tools". Using `var` instead of `let` and `const` `let`s them support (very) old browsers without a build step (which in turn leads to other benefits: https://htmx.org/essays/no-build-step/ https://htmx.org/essays/no-build-step/).
- threatofrain 2y agoMaybe I'm silly but I don't understand why you wouldn't do web interactivity with old machines by going the old-fashioned way, which if optimized can feel wonderfully fast (just not as fast as an SPA that anticipates data loads, giving the user the illusion of instant) and will be more lightweight, easier to understand, will take advantage of all the old traditional tech ecosystem, etc. If we're talking about people who haven't upgraded their browsers in a very long time... chances are that light vanilla websites offer the best experience.
- twiss 2y agoI expect the author would mostly agree with you, but see the "motivation" header under https://htmx.org/ https://htmx.org/