76 comments

[ 3.5 ms ] story [ 157 ms ] thread
Look at how fast it is! DOM content loaded after just 100ms, whole thing done in 400.
It is interesting how pages like these are like a time capsule. For instance, I was just watching the movie Philadelphia yesterday and how many of the misconceptions/prejudices around AIDS were prevalent then. And here, in this link, there is news about some advances made there.
My recollection is the prejudices around AIDS had diminished dramatically by the time Philadelphia came out. But, the movie is (I think) depicting a past event when things would have been different. Also, movies often blow things a bit out of proportion. In the early-to-mid 80s, though, it would have been spot on.
Much better load times than the current one.
I was just going to comment the same!
It’s around 64 KB, which means around 10 seconds load time on a modem, if I remember correctly. That’s very comparable to the current state of affairs.
Usual rational at that time was to not go beyond 40K, if possible. So 64K was rather "fat".
Because they wrote it in an era when internet connections were slow. They have not necessarily moved along the time-to-load-vs-page-size tradeoff curve so much as the curve itself has shifted due to improved connections and computation speeds.

It's like looking at a newspaper from a century ago and remarking "how much cheaper it was back then!" Technically you're right, but the change in nominal price is not a useful economic indicator.

A newspaper a century ago is an excellent analogy.

Then, the front page was filled with advertising. The content was almost entirely sponsored, or in times of war or national fraction, overwhelmingly towing the official line.

Newspapers a century ago are exactly the same as popular web today.

It's like looking at an article from a news site today, and being presented with a splash-screen 5-second advert, or not being sure who sponsored the content you're reading.

100 year old press is identical to web-news today.

Huh. Actually, it looks like they cleaned up a bit recently. As of a few weeks ago, there were far more skeletons in place (as literally scoured from robots.txt).

Snapshot from their antiquated ad spaces listing- https://pp19dd.com/2013/02/attack-of-test3-from-outer-space/...

To nitpick on your example No 3:

The question is "what amount of money would make the injured party whole again", to which the answer was "One million". To that one million, another million was added in punitive damages, summing to the total of two million.

Punitive damages are added to "regular" damages. The idea is that, for example, you may have only lost $30 when your bank fudged the numbers in their favour. But only awarding you $30 would be too low to discourage such behaviour by the bank, and it would also be too low to make it worth your time and money to sue.

That was my point exactly- annotations that require annotations.

Think of it from the reader's perspective: they see $2 million highlighted in an article, they mouseover, click, tap or otherwise interact with it. But then a bubble comes up and gives them a different number that can't possibly be immediately interpreted without further reading and maybe analysis. It's confusing to be presented by conflicting information at a glance.

Annotations are meant to take a reader to primary evidence or otherwise supplement an understanding. However, primary evidence isn't always so precisely straight-forward to be able to do that sensibly- that was my point.

Heh, I recently came across a "guestbook" [0] that I set up ~20 years ago. I doubt that new posts work but it's still there all these years later.

HTML was so much simpler back then.

[0]: http://qsl.net/n9wwv/gbook/guestbook.html

It definitely needs some <marquee>
>Hello from Croatia!

Reminds me of a feeling that the web had back then. Full of wonder, opportunity; it felt safe, friendly, cozy. I wish I could explain it better. Kind of like a party with trusted friends with shared interests and discovery.

Now it feels like a lot of work and little wonder. It feels unsafe, hostile and you know you are being watched, but not for your protection.

OTOH,the oldest guestbook entry is spam advertising the signer's web page lol. That part was there from the beginning.
Yea, and not just the greasy salesmanship and profiteering, but the psychopaths and bullies were there as well.

It was still a lot different, and I liked the fact it was filled with a lot more nerds, relatively speaking.

When getting online required more effort, it filtered out a lot of people who weren't serious. I often wonder what is the equivalent to a BBS today.
Sometimes when I look things up, I find forum discussions from over 15 years ago. It's interesting observing how at first you might not realize it, and then they make references that make you look at the date and you realize you're reading a back and forth discussion that took place over a decade ago and it's still available to look at. Do those people still have those same interests? So many little meaningless questions.

Anyways. Randomc omment but your comment reminded me of this lol.

Contrast the load times between those 1996 pages from the current site. Keep in mind there is probably no advertising spyware in the 96 site and it may account for most of the HTTP requests on the current site.
And there are also no annoying autoplay videos. That has to help load times too.

I wish companies would understand that if I'm reading something on the web, I want to read it, not be distracted by a video that contains the same or less information and takes longer to watch than the article does to read.

I like the 1996 version better.

As far as I can ascertain, that isn't typical behavior and video gets more engagement.
> video gets more engagement.

I keep reading this, so I assume it must be true, but I wonder what that says about our society. There must be huge numbers of illiterates out their pointing and grunting (like something in Idioctracy) to outweigh all the folks I know, who don't watch videos and don't understand why people would.

It doesn't exactly flatter our education system to look into this stuff: https://www.huffingtonpost.com/2013/09/06/illiteracy-rate_n_...

> According to a study conducted in late April by the U.S. Department of Education and the National Institute of Literacy, 32 million adults in the U.S. can’t read. That’s 14 percent of the population. 21 percent of adults in the U.S. read below a 5th grade level, and 19 percent of high school graduates can’t read.

There used to be a lot of sites that auto-played music in the 90s.
What was the load time of the page on Netscape Navigator and a Pentium processor over 1996 era dialup?
Slow
In fact, I'd guess slower than the modern cnn.com homepage on modern hardware and software.
Network latency has always been significantly more than rendering time by the browser.
I was curious as well. Set up a custom network profile in chrome dev tools (28.8kbps and 150ms latency), no cache: 3.9mins to load!
In 1996, with a typical 28.8k modem, that page would've taken more than 20 seconds to load. I'm quite sentimental and nostalgic about the web of yore but let's not pretend everything wasn't a lot slower for most people back then.
Absolutely, I do recall a lot of griping around that time about images and how much they slowed down the web. Look how image-heavy the old CNN pages are. There's over 50KB of images there and only 3K of text.

Fortunately, web browsers of the era let you turn off loading of images or you could use a text-only browser like Lynx.

all the images also include width/height attributes so it would have rendered without needing to load the images and just lazy loaded them.
The image map on the very bottom doesn't appear to work, at least not on Chrome for Android.

CNN appears to never take anything down, here's CNN's coverage of the OJ Simpson trial from 1999 - http://www.cnn.com/US/OJ/

A lot of the linked stories 404 and many image assets are missing. It’s still cool to see the old web designs, though.
Waiting for someone to say "fake news" about any of the stories he doesn't like...
>Stay up to date with the text-only version of our website https://lite.cnn.io

https://twitter.com/cnnbrk/status/906655818950553600 (10 Sep 2017)

HN discussion: https://news.ycombinator.com/item?id=15210022

ok ?
I mean they figured there is a demand out there for "light versions" of websites, which this thread is mostly about.
> Then in April 1995, right after the devastating bombing at a federal building in Oklahoma City, the Unabomber attacked again, perhaps peeved that another bomber was making headlines. A package bomb killed a timber industry lobbyist in Sacramento. Days later, Unabomber threatened to blow up a plane out of Los Angeles; and then he promised to stop the bombings if The New York Times and Washington Post published his 35,000-word, anti-technology, anti-modern-civilization diatribe.

This guy must be really disappointed to see what the world has became.

http://edition.cnn.com/EVENTS/1996/year.in.review/topten/una...

Yup. IIRC he continues to write from prison...

EDIT: removed hasty personal judgement.

Some if the things he is saying are true and make sense, I've read the manifesto and I wouldn't say a lunatic wrote.

His actions of course don't make sense, with all the respect to victims.

Coming from Amazon - https://www.amazon.com/Anti-Tech-Revolution-Theodore-John-Ka...

"Dr. Kaczynski was born in 1942 in Evergreen Park, Illinois. An intellectual prodigy, he attended Harvard at age 16, earned his Ph.D. in mathematics from the University of Michigan at 24, and became the youngest professor of mathematics in the history of UC-Berkeley at 25. After two years, Dr. Kaczynski resigned his professorship and moved to a remote wilderness area of western Montana to pursue a life-long ambition of living an autonomous and self-sufficient life off the land, which he did for twenty-five years."

Plane crash second biggest story of the year — CNN hasn't changed.
Funny that TV graphics or magazine design in '94 weren't bad, but web design graphics were terrible.

It seems like the web could have been beautiful back then, especially with flat design and whitespace techniques.

Thing is that back then text did not really look that good on screens. The screen resolution was terrible, advanced hinting and anti aliasing techniques were cpu expensive and so on. Textures helped to hide that a bit.

Also, flat design works better when people already know that stuff can be clicked, or dragged and such. We can afford to lose some affordance now.

> Also, flat design works better when people already know that stuff can be clicked, or dragged and such. We can afford to lose some affordance now.

That's an excellent observation. IIRC, back in 1996 there were still radio ads along the lines of "And available online at w-w-w-dot-c-n-n-dot-com or AOL keyword CNN" (or, worse yet, some that added h-t-t-p-colon-slash-slash-w-w-w-dot...).

I'll try to avoid stewing on the fact the CEO where I work is very hands-on with our website, despite never using the web. It's 2017 and he has every button replaced with beveled images of buttons...

EDIT: to reconcile "very hands-on with our website" and "never uses the web", I should specify that he orders changes to the website via the following process: 1) has someone print out the website 2) that gets faxed to him 3) he writes on it with a sharpie 4) that gets faxed back 5) that fax gets scanned 6) resulting scan gets emailed to the web design guy.

> slash-slash

"forward slash-forward slash"

Haha, yeah, I remember hearing ads with and without 'forward'. What's really funny, I think, is how the people who needed that 'forward' clarification were exactly the people who'd have no idea which slash was the forward slash.
Also nice: Websites tended to keep evolving with the browsing skills of the client/sponsor. Say, a site started out complying to the usual rules regarding opening external links in external windows. Everything is fine. A few months later, the client is on the phone, "Where is my back button?" – After a few enquiries you find out it's about the external links and that the client had finally learned about the back-button. You try to explain, but to no avail. All external links in the site are to be edited so they will open in the frameset to the comfort of the client, who may now click each of the external links and go back by a single click on the back-button while proudly inspecting her website. A few months later, the client learns about the usual use of external links and possible legal implications. The links are reedited again...
I mean, sure maybe.

But I think folks forget that back then perhaps not even a majority of your users had a browser with 256 color support.

Honestly I kind of miss it - I found webpages far easier to get information off of quickly than today's pretty designs. The fact pretty much all webpages followed the "look and feel" of your browser was a feature, not a bug. Far too much control is given to web developers vs. users these days, and I think the exceedingly low signal to noise ratio is at least a bit caused by that.

>It seems like the web could have been beautiful back then,

It couldn't because the first wave of HTML pages were hand-written by programmers instead of designers. It was several years later that slick HTML editors "democratized" webpage creation by empowering designers-who-weren't-programmers to author webpages with better aesthetics.

You overestimate how much these organizations cared about having any kind of website. Back then it was just a "we should have that, get some programmers to slap it together" and kids (like myself) were hired to do all sorts of things that experienced professionals and designers would do later.

There was no established design technique for the web. And it wasn't until a bit later (late 96, 97) that I remember formal graphic design becoming important. And then designers were still for a long time dependent on somehow getting web developers to translate that into something that could work in browsers of the era.

>designers were still for a long time dependent on somehow getting web developers to translate that

Yep. Many companies had the workflow where graphics designers who had domain expertise in typography and layout -- but not HTML syntax -- would work on webpages primarily in Photoshop. The psd mockups would then be handed off to the programmers to turn into HTML. (Although, the programmers might use Dreamweaver instead of handcoding HTML.) I'm guessing Dreamweaver has sufficiently evolved so it can be a primary tool for graphics designers and let them skip Photoshop altogether.

Let us not forget that you're talking about a time before CSS was available, let alone well supported--let alone _consistently rendered_. This is a time when "web safe color palette" was part of the day-to-day web design lingo.

I could see this being a fun challenge, though--how pleasant of a modern-look web page can you make using only HTML <=v4.0; only 256 'web-safe' color palette; limited font choices. One could demo their work via BrowserStack.

EDIT: I totally forgot to think about the fact that "whitespace techniques" would have been quite a bit more difficult given that a good 80% of your users were viewing on screens that were either 640x480 or 800x600. On a 640x480 screen, this comment takes up half of the space above the fold.[1]

[1] https://imgur.com/x0keeI1

Another thing often forgotten: font-sizes where relative to the system, meaning, they where usually bigger on Windows (depending on the system-wide font settings) and there was no rational for text-image integration.

This became even more severe, when pixels began to shrink with higher display resolutions, where Windows would adjust – i.e. increase – the font-size relatively, while Macs stuck to a positive 72dpi definition for the screen to keep images and text in balance.

E.g., you had an image and a few lines of text to its right. Size the layout so that the text will span vertically over the height of the image, when seen on an average Windows system. Now, on a Mac, it's just about 2/3 of the height … (Considering that all layout had to be done using tables, we may begin to understand that you had to be rather defensive in your approaches. Also, complex tables resulted in perceivable rendering times.)

The problem with the web up until about 2006-2008 was that the designers responsible for websites mostly came from print design. In the absence of other alternatives, it's not a bad source of talented visual designers at all. But the principles of visual design for a fixed medium don't translate well onto web pages. Since then we've had a generation of designers who understood the web as a primary medium, which helped things.
Actually, this became an issue just at the end of the 1990s. In the first years professional websites were usually designed and made by dedicated screen designers.

Edit: Usually, it was also the same person who was responsible for the design and who made the basic layout templets, meaning this person also understood the technical basics, which in turn allowed for some amazing creativity. (The standard expectation was that the [re]launch of a major page would add to the language and definition of the media in some way.) It was really more at the height of the dot.com-bubble that things became more standard and conventional designers took over.

TV and magazines were big business with specialized design experience for decades (over a century, in the latter case), the public web was brand new and mostly had developers or content creators dual-hatting as designers at that point, with very little tool support even.
Lol. As a kid I was living near Kobar towers (Dhahran). It rattled my windows.
Wow same. My dad thought my brother and I were slamming doors. Crazy to see a story about Dhahran.
Did you go to the Aramco school at dhahran? We might have met.
Yeah. Born and raised in Dhahran. Finishing up college now! Dad was in exploration.
Well news seems to have been a lot tamer in 1996. What a monstrosity it has has become.
> It's a question journalists like to ask themselves at the end of ever year.

This sentence has had a typo for 21 years.