Today in Tedium: As I’ve covered a few times in the past, the hard part about any budding technology—whether it’s delivering packages, driving down interstate highways, or beaming satellite signals to homes around the country—is the infrastructure, how easy that infrastructure is to use. The internet is obviously a pretty key example of this issue in action, with products like WebTV and CompuServe imagining a specific direction for reading content online. But what about creating it? HTML is a simple language, but it was not something you could see most people cozying up to, like, uh …, Microsoft Word. But what if web design worked like Word? Today’s Tedium talks about some early attempts to make web design suck less for regular users—specifically, FrontPage—and the problems they created. — Ernie @ Tedium
“This was absurd. The whole point of the Internet, and the Web, was that it should be easy to get to anything, anywhere. It should be possible to develop services on your PC and then send them across the Internet to wherever they would reside. But you couldn't do that without a fairly complex piece of technology—one that did not yet exist.”
— Charles Ferguson, a cofounder of Vermeer Technologies, the company that created FrontPage, discussing the challenges of building and hosting webpages using standard HTML in his book, High Stakes, No Prisoners. Ferguson’s company was among the first to tackle the issue of visual HTML design and management, and its product, FrontPage, was designed to tackle the issue by combining simple HTML design and, if needed, complex functionality.
FrontPage was created to jump on the information superhighway and build the roads
“You can't become the industry standard unless your product covers every major portion of the market.”
Charles Ferguson’s initial goal wasn’t to build a product that built webpages. He really wanted to pave the internet with a method for sharing information online.
But Tim Berners-Lee had already done that, and his idea for doing this, which he called the World Wide Web, was just starting to take off at the time that Ferguson, then a consultant and academic, was looking to start a company.
The problem was, however, that the roads were really hard to use at the time. So his company looked to pave the highway in a way that was a little closer to where users were at the time. That’s where the FrontPage strategy came into play.
Vermeer’s early employees—beyond Ferguson, who wasn’t a technical employee, there was fellow co-founder Randy Forgaard, along with technical employees Andy Schulert and Peter Amstein—saw the way the wind was going with the Web, but were slightly less excited about the prospects of building around a system where the prevailing operating system on the market was an afterthought.
From Ferguson’s High Stakes:
Our lives would have been far simpler if we could have written just a shrink-wrapped PC application like Word. The problem was that, no matter what Andy and Randy tried, Web technology dictated that many of the functions we wanted to provide, like text-search engines, had to run alongside the Web server. In part, this is just the way servers work. In part, it was the result of our extremely ambitious goals. In part, it reflected the fact that Berners-Lee had not designed the Web architecture with mere mortals or PC applications in mind.
Vermeer, which had no other products besides FrontPage, took advantage of the gaps in the original web specification, however, to introduce some de facto standards of its own regarding server architecture.
The idea behind this strategy, he wrote, was “to create ‘lock-in’ to our product and architecture so that neither users nor the industry could easily switch to a competitor.”
If that sounds a lot like what Microsoft did to Netscape during the browser wars, you know your internet history.
Vermeer had a completed first version of what would become FrontPage in just a handful of months. By the end of that year, the company had worked out contracts with Fidelity, Merrill Lynch, and Tribune Media to use the software.
They aimed for something big, but they were about to get something even bigger.
The amount that Microsoft spent on buying out Vermeer on January of 1996—paid in stock, of course. The buyout, which came just months after the initial version of FrontPage was released (and well before the software was successful) was perceived by the public as a transparent attempt by Microsoft to gain a leg-up on the Web, something immediately noticed by regulators. In his book, Ferguson notes that he attempted to sell off the company to either Netscape or Microsoft, but differences in negotiation clearly pushed the company into Microsoft’s arms. In comments to the Wall Street Journal, Forgaard noted that he hoped the company would remain independent, but Microsoft put on the full-court press: "Once Microsoft got interested, it wasn't an option for us to go it alone. We were just too small."
The real problem with FrontPage: Its cluttered code created lock-in for Microsoft
FrontPage was pretty much the perfect app for a company like Microsoft to acquire in 1996. The company’s strategy pulled from Microsoft’s own playbook, and on top of that, the app was seemingly designed to plug right into the Office suite as if Microsoft had designed it itself.
But the app, while innovative for its time, came with numerous headaches, not the least of which was the messy code it created. Structured web design, it often was not.
In fact, FrontPage often created code that worked perfectly with Internet Explorer, but broke with every other web client. This became more of a problem after Microsoft had its grip into FrontPage, admittedly, but it was a problem Forgaard, at least, was aware could potentially happen ahead of time.
“To be specific about Microsoft, the Internet Explorer 2.0 beta that's come out has several HTML tags that are proprietary to their browser and are not supported by Netscape,” he was quoted as saying just months before Microsoft bought the company.
Daniel Tobias, the author of the early HTML-coding site Dan’s Web Tips, has an entire page dedicated to the useless tags, the useless characters, the unnecessary tags around meaningless pieces of code, the broken syntax added by apps like FrontPage.
“In the main menu of this site, I refer to WYSIWYG editors as ‘not very well housebroken,’” the website states. “This implicit comparison of these programs to dogs is actually not particularly fair to the canine population; in fact, dogs are quite fussy about where they poop, while WYSIWYG editors don't appear to show this degree of discretion.”
Granted, it was par for the course during that time for web editors to not work as advertised. Magazines of the era were often filled with roundups of different editors that pledged to make it easier for you to make webpages. In a lot of ways, people during this era of the internet were looking for a single solution to a complicated problem.
(Netscape had its own answer to this problem, of course, with its Composer app, a part of its Navigator Gold suite. But it was limited in some important ways—most notably, it did not support frames, despite those being a hallmark feature of Netscape Navigator.)
In the case of FrontPage, the product’s early reputation eventually sunk it in the market. Upon the release of FrontPage 2003, Microsoft product manager Melisa Samuelson had to reassure the public that the program would create code compatible with other HTML editors.
"We've heard in the past that customers felt our code wasn't transparent enough, that we generated messy code," she told CNET. “We’ve really focused on generating clean, industry-standard HTML code."
At this time, better takes on WYSIWYG (I see you, DreamWeaver) had taken hold in the market, and soon enough, more sophisticated code-based solutions to building websites (hello, Bootstrap!) gained the upper hand.
It wasn’t enough, and by mid-2006, the app had been superseded by other Microsoft offerings—including Expression Web, itself a platform that Microsoft later discontinued.
The goal with FrontPage was to help build the roads online. It turned out that those roads didn’t matter as much as it did to ensure they were compatible with every other road.
"While we believe code is the best tool for many things, some things are better done with visual design. It's easier to position and reposition an element on screen by dragging-and-dropping it, then typing in x and y coordinates."
— Koen Bok, the founder of the mobile-app builder Framer, explaining to Fast Company the firm’s Autocoder tool, which allows mobile designers to build prototypes without using any code. While the web has largely moved in a direction where many people are HTML-literate, WYSIWYG tools are still going strong, but they tend to look more like developer prototype tools like Framer, or service-driven tools like Wix.
When writing stories like these, it’s always interesting to see what people are up to after they move on from their initial gravy train.
For example, WebTV mastermind Steve Perlman, whose own company was bought out by Microsoft, is trying to reinvent cellular technology. And late tech entrepreneur Dave Goldberg, who spent time at SurveyMonkey and Yahoo, got his feet wet in the tech world by selling CD-ROM magazines.
Charles Ferguson’s postscript is more interesting than most. He’s not doing anything related to technology these days, but his at-times acerbic voice—first highlighted by the book he wrote about selling FrontPage to Microsoft—may be more prominent than ever. See, Ferguson directed one of the most essential documentaries of the last decade, a methodical dive into the 2008 financial crisis called Inside Job, which won an Oscar in 2011. That Oscar win was punctuated with that hard-to-ignore voice.
“Three years after a horrific financial crisis caused by massive fraud, not a single financial executive has gone to jail, and that’s wrong,” he said in his famed Academy Awards speech.
Maybe we didn’t get very good code out of FrontPage. We certainly did get an interesting filmmaker out of the deal.