Slashdot Mirror
I Bought a Book About the Internet From 1994 and None of the Links Worked (vice.com)
An anonymous reader shares a report (condensed for space and clarity): For crate-diggers of all stripes, the internet is awesome for one reason: The crate never ends. There's always something new to find online, because people keep creating new things to throw into that crate. But that crate has a hole at the bottom. Stuff is falling out just as quickly, and pieces of history that would stick around in meatspace disappear in an instant online. So as a result, there aren't a lot of websites from 1995 that made it through to the present day. Gopher sites? Odds are low. Text files? Perhaps. The endless pace of linkrot has left books about the internet in a curious limbo -- they're dead trees about the dead-tree killer, after all. [...] Recently, I bought a book -- a reference book, the kind that you can still pick up at Barnes and Noble today. The book, titled Free $tuff From the Internet (Coriolis Group Books, 1994), promises to help you find free content online. And, crucially, it focuses less on the web, which was still quite young, than on many of the alternative protocols of the era. This book links to FTP sites, telnet servers, and Gopher destinations, and I've tried many of them in an effort to figure out whether something, anything in this book works in the present day. These FTP servers were often based at universities which have a vested interest in keeping information online for a long-term period -- think the University of North Carolina, or Kansas State University. But despite this, I could not get most of these servers to load -- they were long ago murdered by the World Wide Web.
Try the Internet time machine with those links, it might work and that's its purpose.
https://archive.org/web/
Everything I write is lies, read between the lines.
>These FTP servers were often based at universities which have a vested interest in keeping information online for a long-term period -- think the University of North Carolina, or Kansas State University.
No love for wuarchive.wustl.edu?
I could not get most of these servers to load -- they were long ago murdered by the World Wide Web.
Try back 10 years ago in 2005, and you would likely find a LOT more of the 1994 stuff still working then.
I noticed in the more recent 5 or 6 years, a TON of old stuff finally vanished for once and for all.
This is the aging of the network though --- things go offline, and if the information didn't make it to Archive.org; I guess it's probably gone forever.
You can't just leave FTP servers and the like out there for the sake of nostalgia. All these resources require constant maintenance in order to keep them on-line, secure from vandals, etc. Perhaps most critically, it requires constant maintenance to keep them secure from delivering malicious content to people like the article writer.
There is also a difference between keeping content online in perpetuity, and keeping it online in the exact same way. Content worth saving (and pretty much everything else) is still available via the Wayback Machine, search engines, etc. That's why we don't need books and why we don't have to maintain decrepit technologies.
I'm surprised you even found your way back online to report the fact that your internet reference books from a quarter century ago had dead links.
Get with the times doesn't even begin to describe the problem of failing to understand that not everything is timeless in this world.
This does not surprise me. It would be like opening the White Pages to a random listing, and seeing if the number is still active and reaches the same person. Things on the Internet are, and always have been, much more fluid.
I'll agree that Universities have a vested interest in the preservation of knowledge, and so should do better. On the other hand, there are plenty of other changes to this or that university that have happened out in meatspace in the prior 25 years or so. Most of the buildings of my alma mater are still where they used to be, but their function (e.g., the departments that live in them) are not all the same. And certainly a lot of those physical spaces have received renovations over the years, resulting in walls that have been added, moved, or eliminated; outlets and network ports that aren't the same.
I'd wager a bunch of the content of that book is still out there, somewhere. And a lot of it is probably still at whatever custodian institution used to have it. Good luck finding it, though.
This is a consequence of one of the best design decisions Tim Berners-Lee and his colleagues made. For decades some of the brightest people in the world had been struggling to perfect a distributed hyperinformation system suitable for general everyday use - but no one had succeeded. Then along came the CERN crew, and pulled it off almost immediately. Their secret? Leaving stuff out!
As a result, the Web has no standard mechanisms for cleaning up. We get broken links. We get cobweb sites that haven't changed for years, and - much worse IMHO - we get valuable pages that vanish because the funding dried up, the maintainer moved to a new post, or for a thousand other reasons.
Early versions of Netscape Communicator had two options for emailing a Web page: send just the URL, or send the whole page. After a while the second option was discontinued - presumably on legal grounds, as it was a tragedy on practical grounds. There are still add-ons that include the whole page, but presumably that's sufficiently arms-length that no one with any serious money is exposed to lawsuits. Or you could write your own. If you really need to refer back to material years or decades later, you just have to keep a copy of your own.
I am sure that there are many other solipsists out there.
I've downloaded some of those old CDs from sites like that.
They're full to the brim with viruses, and, strangely, occasionally porn. Like, who puts porn PCX files in shareware games?
Fun sidestory: My wife woke me up one night at like 4 AM, "I found your PORN STASH", I get up, look at the her with her "I caught you" judging stare, and I look at the screen. Those PCX files from shareware. I point at the datestamp, "1994", and I go back to sleep.
My post was buried in an earlier article. Editors on Slashdot rearrange what posts show up for different users.
You seem to be new here. Are you aware that when you post, your reply seems to be the first right underneath the post you replied to but, if you reload the page, your reply will move all the way to the last reply?
Everything I write is lies, read between the lines.
Glad to know my all time favorite web site is still around and kickin it all these years later!
http://www.something.com/
A similar thing happened to me. I found a telephone book from 1990 and none of the phone numbers were accurate either.
Also, I rediscovered a stash of business cards I received from colleagues and business associates back in the 80s and not only were the phone numbers wrong, so were most of the mailing addresses (and NONE of the fax numbers worked!)
Why is this news? Contact information changes. Is it because "it's on a computer" that it is suddenly noteworthy?
(That said, I really miss the days of logging in anonymously to FTP sites to see if there was new stuff to download. There was always an aura of mystery and surprise that is missing from modern archives which very dutifully have change logs telling you what's been added and removed. And no nasty SysOp telling you that you've exceeded your download quota either.).
Much more than that - there's choice, "oldest first", "newest first", etc. With visibility flavored by personal preference such as giving more weight to up-moderated posts, etc.
/. editors quite obviously don't even spend much time editing, they certainly don't move posts around. Witness all the crude AC posts which appear near the top if browsing at threshold -1.
"National Security is the chief cause of national insecurity." - Celine's First Law
So she found your 1994 porn stash?
Open Source Network Inventory for the masses! Kuwaiba
The answer is going to be pretty similar on all counts.
In general, the browser implementations have some pretty established things you can do from javascript, and those rule the reality. So in general:
1. NNTP: The binary and text being mixed together was weird. Either way, a standardized API accessible over https could fill this role. In fact I wonder about the popular board implementations and they likely have APIs. Or else they get disabled because advertising is easier to inject into web pages.
2. Could do something with the likes of mattermost.
3. Not really a network protocol...
4. They have multpiple CAs, but no ability to require multiple..
5. Those were the days... Haven't had to print anything in soooo long though...
XML is like violence. If it doesn't solve the problem, use more.
Did you buy in half price books or some flea market? Or at a new book store, a recent reprinting?
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
Bing? Really? Dude, are you ok?
I bought a map of America from 1842 and it was ALL WRONG!