[ts330]

I'm not entirely sure they're the next big thing. More likely, is that 15 years ago, it was people who were used to static sites began moving to dynamically generated ones as sites became more complex. They were the new thing then. Now we have a load of people who have grown up with dynamically generated sites and are suddenly discovering the benefits of static sites - thanks in part to the proliferation of tools that are easy to use.

It's the usual boomerang cycle of discovery and adoption.

Both types of sites have their benefits and it's a balancing act to use the right tool for the job. It's getting this right that comes with experience and an understanding of the current pitfalls of each. It's the rough edges that push people in the other direction and without the experience of the pitfalls of each, it's inevitable that people start predicting that one solves all the problems facing the other.

I fully expect the usual over reliance on the wrong type of tech for the sake of it being the current hotness and then an over correction in the other direction the moment we have a new generation of developers.

[smacktoward]

15 years ago the hot new general purpose CMS was Movable Type (https://movabletype.org/) which was... a collection of Perl scripts that ground out static HTML pages. In other words, a static site generator.

Then everybody got tired of waiting for their sites to rebuild every time they changed something and switched to WordPress, which wasn't static. Suddenly your changes showed up right away! Hooray! Then everybody got tired of WordPress falling over under any load stronger than a stiff breeze, so suddenly static site generators were in fashion again.

If you think of approaches to building a content management system as a continuum, with purely static at one end and purely dynamic at the other, you can see the entire history of the segment as a series of oscillations along that continuum. Each approach has drawbacks, but the drawbacks of the approach you aren't using always seem minor while those of the approach you are using seem painful, so the market just bounces back and forth between them ad infinitum as people rush to discover if the grass on the other side of the fence is really as green as it looks.

[masswerk]

15 years ago this even made much more sense. I did a static site generator in good ol' Perl 5 in the 90s – and I'm sure, I wasn't the only one to do this. (Everything was based on file hierarchies, but there wasn't much structure or setup required else. Directories were mangled to intermediate data files in pseudo-XML, only to be reassembled, when there was an update local to that hierarchy. XML wasn't the big thing then, but as this had to mangle HTML anyway, it was somewhat obvious to embed data structures in tags.)

Back then, this really had some advantages: Updates were rare, but views comparably frequent, while databases were either not that performant or quite expensive. This way, you could serve everything from cache (remember Squid?), and, compared to a dynamic site, it was really quick, even in admin-mode. Given the modern machines and the lots of memory they come with, it's quite ironic to see this come back, while we saw the triumph of the LAMP stack on comparably modest machines. Nevertheless, if you've only a few updates and lots of views, it's a good idea towards green computing. (Save some mountaintops! [1])

On the other hand, there is some "magical" limit regarding flexibility and complexity, where things tend soon towards unmaintainable code. So, the judgement is left to you, respective to the purpose.

Edit: [1] "Mountaintop mining" at Google-images: https://www.google.com/search?tbm=isch&hl=en&q=mountaintop+m...

[tootie]

15 years ago we spent $250K for an off-the-shelf CMS that was a static site generator with workflow. Worked perfectly fine even if it was way overengineered.

[vlunkr]

You're totally right. I think there's a rediscovery of static sites going on, and an improvement in the tools, but it can't possibly solve every use case, or even the majority of them.