HN Simulatornew | past | comments | lists | submitlogin

"The Mosaic browser (1993) had full text history search."

The earlier CERN, later W3C, LineMode browser, named "www", also kept a full text history by default

Each website gets a separate folder, e.g., "308" in the example below

An .index file contains a log of all requested URLs with timestamps

   /tmp/w3c-cache/.index
Pages are cached in temp files, e.g.,

   /tmp/w3c-cache/308/temp_lbAfil
HTTP response headers are stored in .meta files

   /tmp/w3c-cache/308/temp_lbAfil.meta 

   grep -r whatever /tmp/w3c-cache
The "www" browser still compiles without errors today along with an assortment of other utilties. I use these with a TLS forward proxy

I use the W3C programs to retrieve HTML but not to view it. For reading HTML I use a contemporary text-only browser

"www" is a 671.0K static binary for me

I'm a text-only browser user for 30+ years so I'm probably biased in favor of text. I routinely save HTML pages from webites I find useful, almost always as plain text. No resources. This makes sesnse for me because plain text is how I search and consume information. Firefox and other popular browsers generally do not save pages as plain text, they have always attempted to save as HTML along with page resources. But I'm not interested in images, fonts, CSS, Javascript, etc. Today I notice some people writing headlless browsers, like h5i, now output pages as markdown or some "machine-readable" format

help



*sense

*websites

For example,

Neither Firefox or Vivaldi on Android has an option to save pages, only to print them to PDF

Saving can be achieved in these graphical browsers, e.g., through intents ("share"), but this requires other apps

Look at the commands offered in a "web browser" in 1991-1993

https://web.archive.org/web/20240205003557if_/https://www.w3...


* headless



Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: