"My experience is that people who tend to disparage web tech don't have much experience building clients in general"
My mom should be able to build great frontends for Web software, but currently the technology requires her to know much more than she knows. As you say, she "don't have much experience building clients in general." Therefore it's important that we get rid of HTML, CSS, and Javascript.
This is an old debate, but to repeat the highlights:
The benefit would be the productivity gained from specialization. The work could be moved away from computer programmers. Beginners would find beginning as easy as building a HyperCard stack, and specialist UI/UX experts (not computer programmers) could be put in charge of advanced frontends.
The same argument that was made for Web Assembly also applies to the frontend: we now know what we need as a general compilation layer for frontend descriptive languages, things we did not know in 1996 when HTML/CSS/Javascript were coming together.
The crucial thing is to have the kind of serialization formats that software can write, thus opening the door to a version of Dreamweaver that actually works. In other words, something like Adobe InDesign would then be the correct way to create all frontends. I wrote about this in detail here:
HyperCard, Dreamweaver, InDesign, Photoshop, Flash and Illustrator do not use a markup language internally. As with Web Assembly, it's important we have the right primitive that we can compile to. There are configuration languages that can make it easy for beginners to hand edit a frontend, and in my article I mention the configs that Ruby On Rails, Symfony (PHP), and Django (Python) offer to generate simple CRUD interfaces, and I mention that these configs could be much more powerful if they had the correct primitive to compile against, rather rendering to HTML/CSS.
Aside from all that, I would raise the more urgent question for our industry, why did it seem like such an urgent task, all through the 1980s, 1990s, and early 00s, to create visual software that would make it easy for beginners to create software, and yet now this is no longer a priority? Is there some reason why we are moving away from the era when "Empower the masses to create software" seemed like an important goal for the industry?
You can't version control whatever binary format InDesign uses. Markup languages benefit from precision, readability from an editor, easy integration with other tools.
"My experience is that people who tend to disparage web tech don't have much experience building clients in general"
My mom should be able to build great frontends for Web software, but currently the technology requires her to know much more than she knows. As you say, she "don't have much experience building clients in general." Therefore it's important that we get rid of HTML, CSS, and Javascript.
This is an old debate, but to repeat the highlights:
The benefit would be the productivity gained from specialization. The work could be moved away from computer programmers. Beginners would find beginning as easy as building a HyperCard stack, and specialist UI/UX experts (not computer programmers) could be put in charge of advanced frontends.
The same argument that was made for Web Assembly also applies to the frontend: we now know what we need as a general compilation layer for frontend descriptive languages, things we did not know in 1996 when HTML/CSS/Javascript were coming together.
The crucial thing is to have the kind of serialization formats that software can write, thus opening the door to a version of Dreamweaver that actually works. In other words, something like Adobe InDesign would then be the correct way to create all frontends. I wrote about this in detail here:
http://www.smashcompany.com/technology/the-problem-with-html
Earlier, in 2016, I wrote about the general problem, which offers some historical context "HTML is the failed GUI for TCP/IP" :
http://www.smashcompany.com/technology/html-is-the-failed-gu...