x7 LabsRandom findings by the x7 Labs team

This article is part of a series on forgotten technology.

Motorola Envoy, a Magic Cap device with built-in wireless.

Before the PalmPilot and Apple Newton, there was Magic Cap. It was a PDA operating system with the communications at its heart, a focus on object-oriented distributed computing… and an interface that was a total woozy.

Magic Cap started out as an internal project at Apple called Paradigm. In 1990, it was spun out as an independent company called General Magic. The OS was licenced to hardware manufacturers like Sony and Motorola, who produced several models.

Magic Cap devices were PDAs, with functions such as address book and calendaring. They were also built for communications from the start, with fax and email functions as standard. Later models would feature web browsing as well.

The devices were pen operated, but featured an onscreen keyboard for typing. Two later prototypes had a physical chicklet keyboard, but these were never released.

Wireless networks were in their infancy in the early 1990’s. Some models like the Motorola Envoy could do wireless email – at an extreme cost. Others would actually plug into a phone line for connectivity.

Metaphors and objects, oh my!

The interface is heavily laden with metaphors.

The Magic Cap UI was based on a desktop metaphor, but it was carried much further than in for example the Macintosh. The system showed an actual cartoon desktop, with objects like a rolodex and paper pad on it.

Drawers, shelves, rooms and and buildings were used as metaphors for different levels of the system. There was even a tote bag which you had to use to carry objects from one place to another.

Email allowed in annotation and graphic stamps – if the recipient also had Magic Cap.

Real-world methaphors were found in unusual places. For example, when sending a message, you had to choose how it was delivered: email, fax or pager. Instead of using a menu, users affixed different kinds of stamps to a postcard which represented the message.

Cloud computing in the nineties

On the programming side, Magic Cap was built around the idea of software agents, popular in the early 1990’s. The agents didn’t appear as actual characters, although that would have been in line with the Microsoft Bob -like interface.

The idea was that agents written in the proprietary Telescript language could be started on the device, travel onto an outside server to do work and then return with data.

Internet Center, Magic Cap's connection to the famed Information Superhighway.

Telescript was an object-oriented language which with an ahead-of-its-time cloud computing flavor. It was specifically designed to run on distributed server systems. The only service actually available, though, was PersonaLink from AT&T.

The aftermath

Like most pen-based projects of its age, Magic Cap never really took off. It may have been the clunky hardware, lack of network services or cartoony interface. In any case, the devices were never a financial success.

General Magic eventually went in different directions, developing speech-recognition telephony services and automotive assistance systems. Magic Cap was spun off once again in 1998 as a company called Icras, but that company faded away soon after.

There was an effort by some Magic Cap developers to get the system released as open source, but this was not successful. Still, the developer documentation remains online for those interested in this bit of PDA lore.

This article is part of a series on forgotten technology.

Sample of the DirectAnimation technology that was spun off Chromeffects.

Long before Google, Microsoft had a browser technology codenamed Chrome. It was a merging of 3D and web content, complete with XML-based modelling and hardware acceleration.

In July 1998, Microsoft announced the near-complete project under the name Chromeffects. A SDK was released, partners were prepped for an imminent launch. And then Chromeffects disappeared without a trace.

A lot of bling for a little band

So what went wrong? To answer that, let’s begin with what Chromeffects was meant to be. In 1998, the web was still fresh and bandwith was low. Most people were still on 28.8 modems, and broadband was a futuristic dream.

At the same time, 3D accelerator cards were beginning to take off. For the first time, you could have reasonably smooth texture-mapped animation without punishing the processor too heavily.

Microsoft figured that you could improve the web experience with client-side graphic effects. These effects would involve only a little markup, with most of the code already installed on the browser.

At the simplest level, web designers could add 3D transitions like ripples and page curls with a touch of markup, much like CSS styles. Chromeffects could also draw full-blown 3D objects, which were defined with a proprietary XML-based language.

Perhaps the most impressive feature of Chromeffects was to map live web content onto 3D surfaces. Video streams could also be used in the same way. One simple demo was a 3D cube which the user could spin, with a different web page on each surface.

An obvious use for these effects was advertising, with mentions of animated 3D banners. Web applications were also appearing on the scene, and there was talk of using animation to enhance web UIs. Another proposed use was 3D visualization for databases.

The makings of a ruin

Although 3D hardware was picking up, effects such as these required the very latest gear: a 300 MHz Pentium II, 65 megabytes of memory and an AGP graphics card.

Here, then, was the first issue: you couldn’t install Chromeffects on most computers of the time. In fact, Microsoft decided to make it an OEM installation only, and an optional one at that.

By November, the plan had changed to making Chromeffects a part of Windows 98 SP1, but by then everything was basically over. Microsoft had a major internal restructuring, and the project fell by the wayside.

Another thing that may have contributed to the fall of Chromeffects was pressure from other browser manufacturers and the W3C. After all, this was the time when the Department of Justice was coming down on Microsoft for bundling IE with Windows.

A brand new bundled media technology in Windows did not look good for Microsoft. Furthermore, Chromeffects was seen as competing with such key technologies as Java and RealPlayer.

The leftovers

What remains of Chromeffects? Not much. Parts of it, like DirectAnimation and DXTransform, made it into IE5. These techniques added animation capabilities, but they were little used. DirectAnimation was discontinued in IE7 because of security issues.

Now in 2010, 3D on the web is once again trendy. Apple has released 3D transformations for CSS, which share a lot of the same low-band effect thinking as Chromeffects. Mozilla and Khronos are working on a web version of OpenGL called WebGL.

And Microsoft, of course, is busy bringing 3D to Silverlight. With hardware acceleration, and all that.

A somewhat rare American version of Pippin, produced by Katz Media.

Apple went on a number of wild goose chases in the 1990’s, investing heavily in things like OS development with IBM and tablet computing. One project they actually got to market was Pippin, Apple’s first and only game console.

If you never knew Apple had a console, you’re not alone. Pippin was largely ignored outside of Japan, and even there it was a flop. A major part of the problem was that Apple didn’t really intend Pippin for gaming.

In the early to mid-90’s, CD-ROM multimedia was a big buzzword. Major publishers were pushing out reference works, educational titles and simplistic games on shiny optical discs. It seemed obvious that CD-ROM would migrate out of computers into mainstream electronics.

Pippin was a console-like device for running desktop-style multimedia content on your TV. As the Internet was starting to pick up steam, Apple threw in a rudimentary web browser and online service as well.

The Pippin controller had an unusual feature: a trackball.

Hardware-wise, Pippin was a PowerPC Mac, although one with a pad controller instead of a mouse. The software experience was simplified: the device booted off the CD-ROM and started whatever program was on the disk. The user never saw a Finder desktop.

Every Pippin disk had to contain a variant of Mac OS, and they were signed with a key provided by Apple. Apple licensed the hardware and marketing to the Japanese toy maker Bandai, who also produced the majority of Pippin software.

As it turned out, people didn’t really want a multimedia appliance, and certainly not one that cost $599. Pippin wasn’t really intended to compete with the Playstation or Nintendo 64, but that’s what it ended up being compared to.

Bandai only managed to sell 42 000 Pippins before pulling the plug. Apple has stayed out of the console business since. They have however found their own niche in the iPod Touch, which is now heavily marketed towards mobile gaming.

This has been part one in a series on forgotten technology.

I’m writing a series on forgotten tech products, and I need your help. There’s an alternative Windows shell that Microsoft was developing over ten years ago, and it seems to have disappeared off the face of the Earth.

I can only recall scant details myself, not even the name of the project. If anything below rings a bell, please don’t hesitate to reply.

• It was an alternate Windows shell for entertainment use.
• It was meant for use on a TV or large monitor with a remote control. (This was not the Entertainment Center UI.)
• This was before the launch of Windows 98.
• There was documentation at MSDN with screenshots.
• It had a black Start menu where all the options were numbered for use with a remote.
• A black taskbar existed for decoration. It didn’t contain running apps, and would only appear when the Start button was pressed on a remote.
• Primary intended uses were TV viewing and playing CDs. PVR, DVD and MP3 functionality weren’t around yet.
• Standard Windows apps would run fullscreen with the title and menu bars hidden. A Menu button made them appear.
• The intended resolution was 640×480.

Here’s the only bit of information I’ve found to show this thing actually existed: a Microsoft press release touting an interface “optimized for distance viewing and display on large-screen devices such as a VGA monitor or television”.

In a world of open development and perpetual beta, the concept of a launch has become a fuzzy one. I can’t recall the date that Windows 7 was released, for example.

It feels funny to write that a product is launching today, if thousands or millions have been using the code for months.

This isn’t entirely new, of course. Movies get previews, buildings are inhabited before ribbon-cutting. But the world of software feels even more loose when it comes to soft launches.

There was a build yesterday, a build today and a build tomorrow. Yet today’s build is special.

I used to be a digital pack rat. When it came to file management, I lived by two rules:

1. Keep everything
2. File like a librarian

I tried to keep everything in a personal Dewey decimal system of nesting folders. I had dozens of categories for files, bookmarks, feeds and other pieces of digital detritus. Most of them in turn had subcategories and sub-subcategories.

While this provided for a lot of busywork, it didn’t really make my life any easier. In trying to establish an ontology of what everything was, I ignored the question of what files were important. The files I really needed to get to quickly or often were scattered about in a deep hierarchy.

A library is built around the idea that any book needs to be indexed and found. My computer, on the other hand, has just one user: me. And I’m never again going to look at most of the tens of thousands of files I create or download.

There was an element of self-delusion to it, too. I’m being organized, I thought to myself. All this organization simply needs to be done, or I’ll never find anything. And there was the vain hope that by keeping bookmarks, I might one day read all those interesting-looking articles that I never got around to.

Now, I’m trying to change my habits. There are two new rules I play by:

1. Discard by default
2. File based by need

Discard doesn’t always mean delete, but that’s a good way to handle lots of files. That video I just watched? Throw it away! Who has time to watch anything twice?

Deleting things can be strangely liberating. I’ve cut the size of my music library in half by throwing away tracks I didn’t really like. Automatic backups reduce the pain, since I can always rescue a deleted file. (I never have, though.)

Things that I can’t just delete, I archive away in a place I don’t have to look at. Year-old mail? Never needed to look at it, but I’ll keep copies tucked away outside my actual mail program.

The second part is organizing files by what they mean to me. For example, there are four folders sitting in my dock:

• Files I’m working on now
• Files to work on next
• Files I need often
• Downloads

The downloads folder works as a secondary todo for videos, ebooks etc. I clear away the files I’ve seen, trying to keep the folder empty.

Likewise, my RSS reader now has four categories instead of dozens:

• Feeds I want to read as soon as they update
• Feeds with small, fun tidbits
• Feeds with long articles, for when I have the time
• Feeds I follow for work

This simple sorting matches my needs much better than what I had before. The reader also keeps track of my usage habits, so I can unsubscribe feeds I don’t really care about.

Mainly mobile topics, with some Microsoft and Ruby filling:

• Spotify now runs on Symbian in addition to iPhone. No news on Silverlight port for Symbian at last week’s PDC.
• Nokia sent a batch of N900’s to the Finnish press just before PDC. Haven’t had a chance to play with our test unit yet personally, Ilari has – with mixed feelings. Perttu Pitkänen (Digitoday, It-viikko), thinks its “sturdy” and “beats iPhone in multitasking”, AfterDawn likes the performance but dislikes the apps and limitations.
• Joe Wilcox writes in his PDC09 summary, that Microsoft “mobile phone strategy is a disaster”. Couldn’t agree more.
• Expression Web Service Pack 1 has been released. Mostly bug fixes, improves PHP editing (not something you would expect in a Microsoft product).
• Jailbroken iPhone and iPod touch users attacked by a new worm, this one sounds nasty.
• Google Hacking is always fun. This times PDF files printed with IE leak path data. Doesn’t sound that serious, though.
• WoW turned 5. Times flies!
• iPhone and Android make up 75% of mobile browsing in the US, according to AdMob.
• A new standard for content management systems is being prepared under OASIS. Seems quite enterprisey.
• JRuby is making progress, also on Windows. Haven’t noticed much enthusiasm behind IronRuby, though. InfoWorld report from RubyConf.
• And finally: Google opens Chrome to extension developers. Look out, Firefox.