3 ms·
Thought this was a good read - and nice to see the progression of frustration with front-end development throughout the last decade, and then the introduction o
by ceedan 6y ago
Thought this was a good read - and nice to see the progression of frustration with front-end development throughout the last decade, and then the introduction of mobile. This is a pretty common timeline for web devs who have been doing this ~10+ years.
I wonder why the author did not contribute to intercooler rather than writing his own version. Would it really have been that difficult to avoid inheritance (or add an inheritance free api) and add batching to that library?
- piranha 6y agoBatching is doable. I even understand how to do it in htmx. The problem is that 1) htmx started later than I did 2) inheritance is like a plague, and I'd really want it to not exist I'll try to explain my stance. When you're a developer, you naturally drift to solutions which require you to write less or write faster. And inheritance certainly helps with that in simple cases. TwinSpark had inheritance up until we tried to implement sign-up/in logic. Damned popups... Anyway, that's the day we've killed inheritance: https://github.com/kasta-ua/twinspark-js/commit/5714ceea98364efc41bc4c9c7b9031c2b23cf2a4 https://github.com/kasta-ua/twinspark-js/commit/5714ceea9836... And if you have two APIs - with and without inheritance - developers will use the first one by default. And when it bites you it'll be painful. So if Carson reads my rant and decides to remove inheritance - I'll gladly move to htmx. If not - maintaining a library which is less than a thousand lines of code is not that big of a burden. :)
- ceedan 6y agoNice thanks for the details - I didn't know you were in talks with them about changes to intercooler, too. Nothing better than a nice pop-up to ruin your day... they're a different kind of plague