There was a very compelling vision for the web, by Doug Engelbart and others in the 60's and 70's. Unfortunately, because of the computing culture's attitude of forgetting even the recent history / understanding what foundational work was done (like real scientific fields do!), the web folks didn't have a lot that context.
Alan Kay, in many of his talks, has discussed how the browser should really be more like an operating system kernel. The web is a mess and we can still build interesting things with lots of hacking & engineering, but it's fallen short of the original vision. And now we're locked into the tooling we've built.
WebAssembly will no doubt give us more freedom, but still has a lot of constraints. Also, it's fundamentally a hack built on top of the browser ecosystem, not replacing browsers entirely!
I seemed kind of cool but doesn't do much you can't do with webpages and javascript. Its scripting language seemed much easier to do some things with than javascript though. I kind of imagine WebAssembly will go the other way and make things more complicated.
> I seemed kind of cool but doesn't do much you can't do with webpages and javascript. Its scripting language seemed much easier to do some things with than javascript though.
In our current developer culture we tend to view things less holistically and more in terms of this-or-that language or system. Hypercard (which was composed of both the visual system and Hypertalk) was one holistic system that fit extremely well with personal computing of the early to mid 90s (which is why recreations of Hypercard on today's machine miss the point).
This was a completely different view of personal computing, one that sought to reduce the chasm between "developer/programmer" and "user". Hypercard allowed authoring in the computing medium, and did so by permitting users take advantage of what was new about computing as opposed to other older media. In fact, they were called "authors" and there were thousand of them — most of them weren't professional programmers.
There is a lot that hypercard could do that the modern web cannot. I cannot copy and paste a button — retaining all of its internal functionality in a different context — from one web page to my own (at least not without a lot of trouble). This was the de facto way to get started in Hypercard.
Hypertalk is another interesting part of bridging the divide. It is difficult for entrenched programmers to reckon with because it's more like natural English and (unlike most programming languages) is easier to read than it is to write. But for regular people it makes sense.
Final point: everything good about computing is about metaphors. Hypercard had one of the best metaphors since spreadsheets: the concept of stacks, cards, and objects on cards. That's all there was and it was easy to understand how these pieces interact. The web does not have anything like this for its "authors" because it is inherently unfriendly to them.
There was a very compelling vision for the web, by Doug Engelbart and others in the 60's and 70's. Unfortunately, because of the computing culture's attitude of forgetting even the recent history / understanding what foundational work was done (like real scientific fields do!), the web folks didn't have a lot that context.
Alan Kay, in many of his talks, has discussed how the browser should really be more like an operating system kernel. The web is a mess and we can still build interesting things with lots of hacking & engineering, but it's fallen short of the original vision. And now we're locked into the tooling we've built.