4 ms·

You've probably noticed artists develop thousands and thousands of themes/mockups/designs for every possible use including admin. For example, https://themefore

You've probably noticed artists develop thousands and thousands of themes/mockups/designs for every possible use including admin. For example, https://themeforest.net/category/site-templates/admin-templates?rating_min=4#content https://themeforest.net/category/site-templates/admin-templa.... Would there be any practical way to provide them with some simple markup that allowed marking up their designs such that some sort of scaffolding tool could be used to convert it into a starter project for various frameworks/languages/etc? Yes some of them come with bootstrap or whatever framework files, but if it were generalized it would be possible create a configurator/generator/scaffolder that would allow many more options and combinations. It seems like Wix and SquareSpace have some sort of proprietary version of this. Where you pick a template, customize it, and then directly use it for development, but of course only using their tools, which as has been discussed here quickly run into limitations for non-trivial sites/apps. This is all off the cuff thinking, I haven't put much though into what potential markup or a DSL would look like. However, maybe even a small set of options available as an open standard could provide a great deal of flexibility and be easy to use for artists and designer who are not that technical. Maybe it could also be possible a start for an ecosystem of tooling to make use of it across a variety of development environments, not just "bootstrap and gulp included". If it's plausible, hasn't been done, and has enough interest I'd be willing to take a pass at a draft spec that could at least help people realize what not to do :).

Theodores 8y ago
Just had to reboot my computer for the first time in weeks due to running Chrome's Lighthouse audit on one of those awful themes. I was guessing '7%' performance rating but did not expect to have to reboot! If you run the same audit tools on any Wix website you can expect the same sort of nonsense, maybe a megabyte of scripts at the top, CSS where it runs to 10000 lines of which 97% unused, images flagged for being bloated. The newer thinking that I find most making sense is to not use any framework at all and to also ignore people using Opera Mini/Internet Explorer. Clearly Google and organisations like the BBC cannot dismiss those people and have to make their websites work on the legacy browsers too. But, if making money matters more than pandering for people that shouldn't be online then the no-framework option makes a lot of sense. No-framework is also getting a name as 'intrinsic web design' and the idea is to use built in things such as CSS Grid, Progressive Web App, HTML5 form validation and vanilla javascript to hold it all together. With this way of working you can make the document pure content with semantic tags, e.g. 'Article', 'Section', 'Main/Aside/Header/Footer/Nav' and dispense with the many, many, many nested divs and spans with everything having a gazillion class tags. Going for the 'perfect outline' can be done too with a few hundred CSS lines instead of the tens of thousands you get if using some lame framework. Building this way and using native browser features instead of the old cruft ways results in sites that load 10x quicker as you don't have the megabyte of scripts/stylesheets that the likes of Wix bloat sites with. The thing is that to do frontend this way requires throwing out all knowledge of 'floats', 'margin/padding hacks' and even that tedious design process where someone that can't code designs something in a desktop publishing type of program for it to be handed off to frontend developers that tinker with the bloated theme their CMS comes with to make it look like whatever the designer got signed off with the client (without speaking to them first). Even 'Responsive Web Design' is old hat, with CSS grid there is no need to think in those ways, you can make everything content driven with fluid typography plus a few CSS grid rules making it work whatever the screen size. In summary - forget about frameworks and enjoy the web as it should be.
- WhitneyLand 8y ago
  This kind of misses the point and I don't think really disagrees with what I'm saying. First off, as I mentioned I it's been discussed here and most agree the Wix/Squarespace sites can get deficient pretty quickly. Secondly, if there were an easy way for designers who are not extremely technical to attribute their code, it actually could greatly help some of the problems you're referring to. For example, if a pre-processor knew what the intent was it could replaced a bad or bloated decision automatically without changing the semantic meaning. Finally, as for the suggestion around keeping it simple with basic constructs and are predictable and easy to work with, there is no reason that couldn't be part of some type of effort like this that attempted to extract more value from artists than their technical knowledge may allow and realize the potential of many more designs.

Theodores 8y ago

Just had to reboot my computer for the first time in weeks due to running Chrome's Lighthouse audit on one of those awful themes. I was guessing '7%' performance rating but did not expect to have to reboot! If you run the same audit tools on any Wix website you can expect the same sort of nonsense, maybe a megabyte of scripts at the top, CSS where it runs to 10000 lines of which 97% unused, images flagged for being bloated. The newer thinking that I find most making sense is to not use any framework at all and to also ignore people using Opera Mini/Internet Explorer. Clearly Google and organisations like the BBC cannot dismiss those people and have to make their websites work on the legacy browsers too. But, if making money matters more than pandering for people that shouldn't be online then the no-framework option makes a lot of sense. No-framework is also getting a name as 'intrinsic web design' and the idea is to use built in things such as CSS Grid, Progressive Web App, HTML5 form validation and vanilla javascript to hold it all together. With this way of working you can make the document pure content with semantic tags, e.g. 'Article', 'Section', 'Main/Aside/Header/Footer/Nav' and dispense with the many, many, many nested divs and spans with everything having a gazillion class tags. Going for the 'perfect outline' can be done too with a few hundred CSS lines instead of the tens of thousands you get if using some lame framework. Building this way and using native browser features instead of the old cruft ways results in sites that load 10x quicker as you don't have the megabyte of scripts/stylesheets that the likes of Wix bloat sites with. The thing is that to do frontend this way requires throwing out all knowledge of 'floats', 'margin/padding hacks' and even that tedious design process where someone that can't code designs something in a desktop publishing type of program for it to be handed off to frontend developers that tinker with the bloated theme their CMS comes with to make it look like whatever the designer got signed off with the client (without speaking to them first). Even 'Responsive Web Design' is old hat, with CSS grid there is no need to think in those ways, you can make everything content driven with fluid typography plus a few CSS grid rules making it work whatever the screen size. In summary - forget about frameworks and enjoy the web as it should be.

WhitneyLand 8y ago
This kind of misses the point and I don't think really disagrees with what I'm saying. First off, as I mentioned I it's been discussed here and most agree the Wix/Squarespace sites can get deficient pretty quickly. Secondly, if there were an easy way for designers who are not extremely technical to attribute their code, it actually could greatly help some of the problems you're referring to. For example, if a pre-processor knew what the intent was it could replaced a bad or bloated decision automatically without changing the semantic meaning. Finally, as for the suggestion around keeping it simple with basic constructs and are predictable and easy to work with, there is no reason that couldn't be part of some type of effort like this that attempted to extract more value from artists than their technical knowledge may allow and realize the potential of many more designs.