Show HN: Browservice – Browse the modern web on historical browsers
github.com
github.com
With better security for 99.9% people, you got less backwards compatibility for the 0.01%. And it's OK.
> This idea of using a proxy to render the browser view into images has been used before by WRP (Web Rendering proxy). Browservice differs from WRP in that it uses JavaScript on the client browser to animate the browser view and gather user input events, while in WRP, the user has to use web forms and image maps to provide the input, and the page has to be reloaded for every update in the view. Thus Browservice gives the user a more immersive web browsing experience, but also requires a newer client browser and more powerful hardware. While WRP can run on browsers as old as NCSA Mosaic 2.0, the earliest supported client browsers for Browservice are from late 90s and early 00s.
So, one long-term possibility I'm toying with is using OS X's implementation of X11. I could run Firefox in a modern Linux distro (probably a VM on the same machine, although it could be a different machine), and use X Forwarding to make it appear as a normal app. I think this should theoretically work for a lot of OS's, from the earliest versions of OS X to decade-old copies of Linux.
Has anyone used this approach for old machines?
It's old enough to not get any security updates. Doesn't that concern you?
That said I'd be concerned about not getting security updates by running a legacy OS... Aren't you?
Also in that comment thread, some thoughts on Mavericks vs Snow Leopard: https://news.ycombinator.com/item?id=23467082
Yes, it's a common way to run GUI-based Kali Linux apps under Windows 10's Linux Subsystem.
Even with NoMachine the overall experience was clearly much worse than a native browser (mostly due to worse trackpad-scrolling).
I'm very surprised that standard render-remotely-and-send-pixels software was more effective than X forwarding, where the drawing is still done locally.
It uses the Wayback Machine as its backend, but with tweaks so the site appears to load "natively".
(Also works on old browsers, too, if you use http instead of https.)
Reminds me of [1] "Browsh", which is a service that pre-renders a URL for viewability in terminals.
It server-side rendered pages (on Opera servers), and converted them to a lightweight HTML page (with OPML extension IIRC), and then sent to the browser.
I presume the server costs were high, because you run a giant proxy that did a lot of things for users, but for end users, it meant semi-interactice web pages that were 10-15kb in size.
> Browservice is 100% AJAX-free software, handmade in Finland
www.mightyapp.com
We need an app that abstracts away the browser / browser engine, and just presents a working website to the user, probably without ads.
Thanks for creating this, author!
I think I found a good use for my Raspberry Pi. And for stuff to download, there's WinWorldPC.
I made something similar in Node.JS using headless Chrome, that targets mobile and desktop browsers. The "proxy" runs anywhere, and a client (that runs in the web browser) connects using Remote Debugging Protocol[0], passing along the user interaction intents by converting them to the wire protocol of RDP. The remote browser server then, like this project ( I think ), passes back pixels from screenshots of the rendered web.
It's a safe but not particularly fast way to do it. MightyApp[2] are doing some sort of MP4 streaming of a remote browser, for speed, I think.
I like the retrocomputing focus of this, and the screenshots and image of the modern web running on an old ThinkPad like machine look really awesome. The focus of my project[1] was more an open-source remote isolated browser, for security, and automation uses.
[0]: https://chromedevtools.github.io/devtools-protocol/