Yesterday was JavaScript's 25th year anniversary and brandrick posted a link to the archive.org snapshot of a 1995 Netscape press release on netscape.com taken in 2007. https://news.ycombinator.com/item?id=25301437

However, a full paragraph from the press release was removed from what we all read yesterday compared to the actual press release. That paragraph was in the source code of netscape.com in 2007, but is not shown when you go to the link we all went to:
https://web.archive.org/web/20070916144913/http://wp.netscape.com/newsref/pr/newsrelease67.html

It is not even in the html markup if you view source on that page. That's not how I found it. I noticed that there was a semi-colon after AT&T; throughout the page, which is odd. I figured it was something related to the scraper (the Alexa crawler), which was wrong (I think my intuition about it being related to HTML entities was still correct though).

I posted the above on HN and hcs responded that if you add a "id_" after the URL (great tip!) the semi-colon doesn't appear:
https://web.archive.org/web/20020606002913id_/http://wp.netscape.com/newsref/pr/newsrelease67.html

I was curious what was different between the two versions on archive.org and viewed source of both of them and did a diff and noticed a full paragraph that was in the source of the raw version but not the main version. But it didn't appear on the page, it was in the HTML of the raw version but did not render in the browser.

Examining the source of the raw version I noticed that a closing tag for the unordered list isn't closed and it is directly above the missing paragraph:

"JavaScript is the perfect complement to Java, because it's an easy-to-use object scripting language yet based on the powerful Java language," said Marc Andreessen, vice president of technology at Netscape. "JavaScript and Java represent an open-standards based approach to creating compelling, interactive content and applications for the Internet and enterprise networks. This broad industry support for JavaScript will fuel the rapid development of a whole new class of live online applications."

I had previously found two old copies of the original press release and they both had that paragraph. Here is a link to the archive.org versions:
https://web.archive.org/web/20140112045932/http://www2.ldc.lu.se/temadag95/javascript.txt
https://web.archive.org/web/20190225103127/http://www6.uniovi.es/java-http/pr951204-03.html

If you diff the source code of the two netscape.com pages from archive.org, you see plenty of hints that archive.org is sanatizing the HTML. Like lowercasing tags, and adding quotation marks where it use to be optional. I accept that tradeoff to have the value add of the ability to see the timeline on that version.

We're seeing the same thing rendered in the browser as everyone else did on Netscape.com for x amount of years, except for the odd people that view source and maybe saw that hidden paragraph. But we don't see the same thing when we click view source if we aren't on the raw version.

I'm super glad that archive.org keeps the raw version (as well as two esoteric versions). I do wish it was more prominent in the UI that it exists (I have no idea how hcs knew that). I'm also empathetic to the web dev from Netscape.com in 1995 that didn't have a linter.