It looks strange to me, language designed to work with trees so turn everything to tree. No need to change serialization, we know format, just parse it — class,

It looks strange to me, language designed to work with trees so turn everything to tree. No need to change serialization, we know format, just parse it — class, URL, CSS. That's what we do in JavaScript [1], [2], [3]. location.host === 'example.com' span.style.color === 'blue' datet.getFullYear() === 2020 //a[href/host = 'example.com'] //span[xstyle/color = 'blue'] //date[datetime/year = '2020'] Just like your proposal but node knows how to parse itself //a[url-host(@href) = 'example.com'] //span[css-color(@style) = 'blue'] //time[datetime-year(@datetime) = '2020'] And it is actual syntax, I've checked on <a><href><host>example.com</host></href></a> <span><xstyle><color>blue</color></style></span> <date><datetime><year>2020</year></datetime></date> * <style> is CDATA, emulated with <xstyle> [1] https://developer.mozilla.org/en-US/docs/Web/API/DOMTokenList https://developer.mozilla.org/en-US/docs/Web/API/DOMTokenLis... [2] https://developer.mozilla.org/en-US/docs/Web/API/URL https://developer.mozilla.org/en-US/docs/Web/API/URL [3] https://developer.mozilla.org/en-US/docs/Web/API/CSS_Object_Model https://developer.mozilla.org/en-US/docs/Web/API/CSS_Object_...