PizzaHut still serves its first homepage | Hacker News Reader

PizzaHut still serves its first homepage

pizzahut.com

248 pointsnarfz123 comments

I was in Santa Cruz at SCO Forum when PizzaHut took its first ecommerce order and delivery. It was 1995 or 1996 I think. It made the news that evening while a bunch of us were sitting in a bar in Carmel. My abiding memory is that the news clip was tagged Information Superhighway.

The Space Jam movie website is still up from 1996, and it still has working links.

http://www2.warnerbros.com/spacejam/movie/jam.htm

Lol that space jam website still actually looks good.

Wow, I knew this was there but hadn't opened the console before. Quite the throwback. <center>? <nobr>? <font>?! Neat.

I'd never seen a <map> tag either: https://developer.mozilla.org/en-US/docs/Web/HTML/Element/ma...

You kids... I was the king of <MAP> back in the day

And I was your trusted adviser.

Hand selecting x,y points was all the rage in the mid 90's.

Oh man... I'm having flashbacks to how badly that tag got abused...

Who needs HTML? Just create the layout in photoshop, export the whole thing to a flat image, and map all the links...

At least it is not as bad as the older server side image maps.

I bet blind people would love that!

Even today <map> is useful and good at what it does for, say, interactive maps. Which is why it hasn't been removed.

Yeah. I was pleasantly surprised to open the MDN doc page and not see a deprecation notice.

some of the comments made me chuckle:

damn, i am getting old.

for better or worse i'd say there was a whole year i made websites using nothing but the map tag.

>I'd never seen a <map> tag either:

It is always a good idea to learn about old technologies. Probably the most eloquent reasoning why came from Feynman's "Know how to solve every problem that has been solved."

With the exception of the startup I did, I would characterize most of my experience as a developer in life as working with legacy code. It gets a bad reputation, but at the same time I now have a lot of insight.

Jeff Bridges makes extensive use of <map> in his webpage. I am unsure if this is apocryphal, but I'm told he makes the whole site himself.

jeffbridges.com

From the "browser icons" page:

> Unfortunately, this only works on a Macintosh running Netscape

There's a twitter account to let us know if it ever goes away: https://twitter.com/spacejamcheck

It's from the internet "80's" :)

"The Net" (1995) movie thriller with Sandra Bullock about the early internet and cyberterror featured it (named pizza.net) too: http://www.imdb.com/title/tt0113957/ , http://en.wikipedia.org/wiki/The_Net_(1995_film)

Photos of all websites featured in the movie: http://gizmodo.com/every-webpage-from-the-1995-movie-the-net...

That's funny, it's on swedish TV right now. No I'm not going to watch it.

According to wikipedia[1] this was one of the earliest websites to be created.

[1] https://en.wikipedia.org/wiki/List_of_websites_founded_befor...

Not sure how accurate the numbers [1] are but some interesting numbers there, relevant here [2].

[1] http://www.internetlivestats.com/

[2] http://www.internetlivestats.com/total-number-of-websites/

At least this says 1993:

    $ whois pizzahut.com | grep Creation
      Creation Date: 07-dec-1993  
      Creation Date: 1993-12-07T05:00:00Z

That's a cool trick. Didn't know you could do that.

Whois or Grep ?

Domain registration doesn't imply website creation.

It's missing a form element. Clicking submit doesn't even send your name, street address or voice phone. They must have had to guess what kind of pizza you wanted and and where to send it by the ip address in the server logs. Funny how the Document Title and Document URL are in text fields at the top of the page.

This is how you could visually design and order pizzas via email-to-fax in 1990:

http://www.art.net/~hopkins/Don/images/pizzatool.gif

Here's the manual entry and source code for PizzaTool, written in NeWS PostScript, which shipped with OpenWindows on Solaris:

http://www.donhopkins.com/home/archive/NeWS/pizzatool.6

http://www.donhopkins.com/home/archive/NeWS/pizzatool.txt

The MIT AI Lab had a program called AI:HUMOR;TS FTP, the Food Transfer Protocol, which was supposed to allow you to download pizzas and other kinds of food over the internet, but it had some bugs and limitations.

Wow, I looked at the screenshot and didnt think much of it, but when I looked at the source code I got sucked right in! Thats brilliant, I dont understand the language but the thinking and patterns of the software are visible and I found it easy enough to follow.

One thing I was curious about was how that icon stored near the end of the file looks when rendered.

Did you write this script? This is exactly the kind of fascinating stuff I come to HN hoping to find :)

Thanks -- I'm glad you appreciate it! I wrote it as a programming example for The NeWS Toolkit (TnT 2.0), which was an OPEN LOOK user interface toolkit written in Sun's NeWS, a multithreaded object oriented dialect of PostScript. NeWS was a network extensible window system developed by James Gosling, after he made Emacs and before he made Java.

Ben Stoltz came up with the idea for a "tatool" gui interface for ordering pizzas from Tony & Albas pizzaria down the road in Mountain View, which he implemented in XView using DevGUIde (Sun's GUI builder tool). It inspired me to write Pizzatool for TnT in PostScript by hand -- I remember staying up late at night writing snippets of PostScript code for each of the different pizza toppings!

My initial version would draw a preview of the pizza on the screen, and then fax the image over to the pizza parlor, which confused them a lot because they couldn't tell which ingredients I wanted by looking at the black and white halftone screen printed pizza picture. So the next version just printed text describing the ingredients and delivery instructions, which took a lot less time to fax and was a lot more readable than a halftone image.

There was a bit of a controversy internally at Sun about releasing and supporting the source code, which toolkit to use (there were a lot of politics surrounding that at the time), and unintentionally revealing Sun's secret multimedia pizza faxing strategy to the press. But I wrote a manual entry, disabled the email-to-fax feature, replaced it with a menacing pop-up notification that threatened to hold your pizza hostage until you payed your tab, and managed to ship it in Solaris/SVR4 as an OpenWindows/NeWS demo.

Andy Bechtolsheim used it as an example of Sun's multimedia strategy in a SunWorld interview, and Unix Toady wrote an article about it, so the fallout wasn't as bad as some people were afraid of.

http://www.donhopkins.com/home/archive/NeWS/pizzatool-mail.t...

The idea was to test out and demonstrate how to program many of the dynamic user interface widgets, menus, windows, and drag and drop techniques in TNT 2.0. For example, it forked off light weight threads to draw all your pizza toppings at once, and you could spin the pizza to "cook" it by rotating the pixels repeatedly, then drag and drop an image into the pizza to customize it! It could even interoperate with another demo called "RasterRap". Here's a video that demonstrates spinning pizzas and dragging and dropping images from RasterRap into PizzaTool, which I recorded at the Exploratorium years ago (PizzaTool demo starts at 21:40, and the white and red pizzatool icon is visible at 13:00 -- it looks like a pizza box):

http://donhopkins.com/home/movies/HyperLookDemo.mov

Demonstration of SimCity running under the HyperLook user interface development system, based on NeWS PostScript. Includes a demonstration of editing HyperLook graphics and user interfaces, the HyperLook Cellular Automata Machine, and the HyperLook Happy Tool. Also shows The NeWS Toolkit applications PizzaTool and RasterRap. HyperLook developed by Arthur van Hoff and Don Hopkins at the Turing Institute. SimCity ported to Unix and HyperLook by Don Hopkins. HyperLook Cellular Automata Machine, Happy Tool, The NeWS Toolkit, PizzaTool and Raster Rap developed by Don Hopkins. Demonstration, transcript and close captioning by Don Hopkins. Camera and interview by Abbe Don. Taped at the San Francisco Exploratorium.

Wow thanks so much for the background info, and thanks for giving me a peek into the past! That makes a lot of sense about how it came about and what it was trying to accomplish, knowing the restraints the software was written under only makes the product you came up with more amazing!

The thing that strikes me the most about this whole thing is that your pizza tool is essentially the same functionality as Domino’s online pizza builder today—but it took the rest of the world a quarter century to recreate what you had in 1990?! Sometimes I feel like the consumer software market has hit the ‘pause’ button on the computer revolution and stalled forward progress for decades…learning about this tool backs that suspicion with proof. You were active during the golden age of desktop software, this must have been such and exciting time to be creating things.

“I would love to be able to order a pizza w/out all the hassles of talking to a human. O:-) ” — Angela Thomas

AMEN! I Loved this quote from the email exchange : ) About to watch your video now, can’t wait to see these ideas brought to life in motion.

Thanks again for the background info and links, feel free to send any further replies to tomhodgins@gmail.com in case this HN thread gets lost. I’d love to hear more about what you built since the pizza tool too, I’m fascinated!

There is an actual form button, as opposed to a link image in the OP page, in the actual 1996 version (archive.org doesn't have the '95 version):

https://web.archive.org/web/19961219205128/http://www.pizzah...

The currently served "old" version was probably disabled to redirects to the modern ordering website so the antique non/functionality from ever interfering with the modern website and not lose sales by attempting to convert customers into the new funnel.

At some point in the last ten years they added Google Analytics tracking to the page.

Yes and today it will be all worked up!

Last modified Mon, 30 Dec 2013 if you use http://web-sniffer.net/ to look at the HTTP response header.

Looks like they not just still serving it, but have updated it at some point in time also.

DOM is your friend :)

    document.lastModified;

This DOM property is derived from the HTTP headers. http://www.w3.org/TR/html5/dom.html#dom-document-lastmodifie...

...or just Page Info in Firefox.

Why does it have the fake browser chrome at the top? did it have that back in the day as well?

Probably not. See the wayback links from 1996 that others have posted.

That's one fast load time.

It's 85% pagespeed score and 92% YSlow [0].

It's not valid, but it's responsive.

It has no jQuery dependency, but it uses Google Analytics.

It's insecure HTTP but a "trusted" hostname.

It's low contrast and navigable (inasmuch as there are no links).

Plus ca change...

[0] http://gtmetrix.com/reports/www.pizzahut.com/7MCdOcHp

How come tags are always in uppercase on really old sites and why did the transition to lowercase happen?

My guess is that back in the day we'd be writing these pages in a plain text editor and all caps made it easier to distinguish tags.

I think that is most of it. Also, upper case was kind of like a virus back then. When you went to write your first page, looked around at what everyone else was doing, it was all caps so you did all caps.

I believe xhtml made uppercase tags invalid.

That is correct. From "XHTML™ 1.0 The Extensible HyperText Markup Language (Second Edition)"

4.2. Element and attribute names must be in lower case

XHTML documents must use lower case for all HTML element and attribute names. This difference is necessary because XML is case-sensitive e.g. <li> and <LI> are different tags.

According to the standard documents until html4.1 the tags were in uppercase. People followed the standards documents, although there was no explicitly required requirement. Xhtml which came later (2000) explicitly required them to be in lowercase, and now html5 now states that the case is insensitive.

Some people will say it's XHTML. It's not just that. There's always been both in use and HTML has always been case-insensitive.

When Rasmus Lerdorf designed PHP in 1994, he made it have case-insensitive function names since at that time, neither side had won and he wanted it to look good in HTML:

  <CENTER><B>Hello <? ECHO $you >!</B></CENTER>

  <center><b>Hello <? echo $you >!</b></center>

Very similar in 1996 on the wayback machine

http://web.archive.org/web/19961219205128/http://www.pizzahu...

Comment shows it was created in 1994

HTML index.html

DESCRIPTION Home Page.

REVISIONS

Date Who Comments

---- --- --------

08/18/94 SCO created.

The style here is different.

Yeah, this seems more faithful to the original. For one, it has a submit button that actually submits the form. In wbm's version, clicking the hat also actually brings up a contact form as the text says it should.

I love how they aligned input elements by using monospace and matching number of characters. A paradigm predating even tables!

Normally when I end up on the Wayback Machine, I spend a few minutes lamenting a web gone by, but for some reason I clicked around this one for a few minutes. They have a really fascinating backstory (although short) as to how the site got up and running, and discussing the first pizza delivered from the web on the About page.

http://web.archive.org/web/19970214123825/http://www.pizzahu...

I wonder how we'll do this kind of thing 10 years from now? Web pages have got so complex, even simple blogs etc, that serving them without keeping the infrastructure running is going to get harder and harder...

Scrape the website to html and css and toss that up anywhere. Every CMS can be a static site generator if you want it to be.

Sure, but many of these sites are taken down/replaced, and only years later does nostalgia kick in... At that point, its too late to scrape the site!

That's why archive.org exists.

It will be funny when websites stop working left and right because essential remotely hosted JS files disappear from various free CDN URLs. Perhaps we'll add more cruft to browsers to compensate (i.e. recognize dead CDN URLs and use a list of replacement hosts).

Of course, the really old web pages will still work because of fewer/no external dependencies.

By that time you'll simply include the whole infrastructure in a 'web'-served container, running in a 'sandbox' of sorts on your 'browser'. Quotes where terminology is an unsure fit with whatever you'll be using by that time.

Ahhh... being the webmaster, brings back memories.

I do remember going to imdb in those days and making a screen shot to show my relatives over xmas break. Haven't seen it in about 10 years though.

Ahhh... Using the term webmaster, brings back memories.

Santa Cruz Operation (thanks mahouse) aka SCO.

They used to run all the McDonalds stuff at one point iirc.

Santa Cruz Operation*. It sounds like a warfare tactics thing, heh.

You must have been a client, too. Brings back old memories.

Also, have a look at the Groklaw archives if you haven't: http://www.groklaw.net/staticpages/index.php?page=2006121221...

IIRC, Groklaw was created specifically to archive/translate/simplify the SCO litigation, right?

Yep and it was a truly amazing and fascinating resource writing about a fascinating case.

Wrong SCO.

Caldera renamed themselves The SCO Group after acquiring the name and trademark from The Santa Cruz Operation in 2001.

Later, things went south and the new SCO(G) turned litigious. The original SCO was not responsible for the later shenanigans, and it was painful to see their name dragged through the mud.