The Browser – WorldWideWeb Next Application

The ability to edit did make it through to Netscape on UNIX, we had this on SGI machines.

This was all great in principle, you could just publish your local documents. However, DNS and IT departments meant that I was not able to do this and browse my machine from outside the room. This was something I was not bothered enough to resolve, had I done so then I could have been a web pioneer.

Making web pages straight from the browser was lost as an ambition early on.

Also lost were many of the great features of the X-Window system. Just using X-Windows to run a remote display from one machine to another was great as a student for putting 'xeyes' up on someone else's screen but we ended up with other stuff that lacks the design genius of the original.

Also cool was the hover for focus aspect of X-Window, this was useful in the days of lower resolution screens and before the tabbed interface for the browser came along.

Nowadays X-Window screen shots look cool but not as cool as they were at the time, the interface and the typography worked better with analog signal CRT displays than it does with digital flat screens.

This is not handling HTML entities properly, which does a number on my own site where various hrefs have slashes escaped as / because Zola has taken the OWASP guidelines on the topic of escaping which erroneously instruct to escape forward slash, and I haven’t been bothered to undo this. So almost all my links are broken. And as for my article title <_>::v::<_>…

(The OWASP thing: https://cheatsheetseries.owasp.org/cheatsheets/Cross_Site_Sc..., “the forward slash is included as it helps to end an HTML entity”. This is wrong, and slash escaping is 100% stupid; this was related to an SGML feature that I don’t believe ever made it into any HTML parsers, or past HTML 3.2 at the latest, and even if it were so it would still be a stupid guideline because if a character’s entity-ending behaviour matters then (a) you’re doing something catastrophically wrong, and (b) you should be entity-encoding every non-alphanumeric character. But people have been blindly copying what OWASP says here for at least ten years and making things worse for the sake of it, e.g. https://github.com/rack/rack/issues/27. Perhaps I should try again to see if I can figure out how to get them to fix this bad guideline. I think it’s probably five years since I last tried and failed; I might be more successful in finding the place to complain nowadays, looks like the source might be hosted on GitHub now.)

See also discussion from 2019 when this was in alpha: https://news.ycombinator.com/item?id=19249373

One thing to note is that this doesn't actually render your page as it would in WorldWideWeb, sadly–it proxies the content through CERN I believe and then renders it with special fonts and such to make it look real. The real deal still works fairly well with semantic HTML–I used to like to show off my website in WorldWideWeb running in Previous, not much worse for wear aside from a <head> tag poking through, as an example of how making your websites the right way can really work ;) Sadly, that demo stopped working when I started using TLS…

What an amazing UI! I love how the navigation panel pops out and you can position it anywhere you want!

Thank you so much for making this. It also gives me a chance to test my website with WorldWideWeb, which I've been meaning to do for a long time. As soon as I have a dedicated IP, so that HTTP/1.1's Host: header is not required to access it.

I’m on mobile, so this page confused me. My eye skipped right over the button at the top of the page to launch the browser. But reading the comments made me look again. This is really cool! Though I’ll definitely want to try this on a desktop computer later...