What Will Not Happen in 1998

(Sidebar to Jakob Nielsen's column on trends for the Web in 1998)

The Web is considerably more quiet these days than it was during its explosive youth in 1994 and 1995 when revolutions happened on a monthly basis. A year is no longer enough for fundamental changes, even on "Internet time." Thus, there are several things that will not happen over the next year, even though they get talked about a lot.

Bandwidth Problem Solved - Not

I predict that lightweight design will continue to be the most important Web usability requirement throughout 1998 (actually, through 1999 and 2000 as well).

The Web will continue to be dominated by users with analog modems, since that's what most new users will have. Transport rates across the Internet (and especially across overseas' connections) will slow down those users who do upgrade their individual connections. Since it is almost impossible to satisfy the response time requirements of 1 second for pleasant navigation and 10 seconds to keep the user's attention, there is no way that bandwidth will go away as an extremely tight restriction for Web design.

Micropayments - Not

In August 1995, I predicted that micropayments would happen "in a very short amount of time." Obviously I have been proven wrong. I will rephrase my prediction: micropayments will happen in two years; that is, in the beginning of the Year 2000 (or possibly late 1999).

I am a firm believer in micropayments because I think the only way for the Web to become a truly valuable medium is for users to pay for the services they receive. There is so much more that can be done on the Internet in terms of value-added services once it becomes possible to collect payments for said services.

Unfortunately, micropayments require infrastructure, and a very conservative infrastructure at that (since it concerns money). It does no good for an individual user to install micropayment software as long as no services are available. Similarly, few services will be launched as long as there are no users with the capability of paying for them. Since microtransactions are technically feasible (and several systems have been invented long ago), I expect that they will eventually happen, just not right away.

Internet Explorer Becomes Only Browser - Not

The market share trend curves will probably cross in June 1998, after which Microsoft will no doubt issue a gloating press release proclaiming IE the most widely-used browser on the planet. This will be a non-event as far as Web designers are concerned: IE will not be the only browser and Web designers will not be able to avoid the need for cross-platform design. We will not revert to the state of affairs in late 1995 when a single browser dominated the field to the exclusion of all others.

First, older browsers will remain in use for a long time to come: I still get hits from Mosaic and Netscape version 0.9. Many users do not view the Web as their first priority in life, so if they have a working browser installed, they won't bother downloading a new one.

Second, Web-usage will expand from boxes on a desk to include a variety of non-traditional devices. In particular, I predict that mobile access to the Internet will be the killer app for personal digital assistants. Once devices like the Pilot and Newton ship with a wireless modem built in as a standard from the factory (prediction: late 1998), independent software vendors will start assuming connectivity as a given and will design legions of interesting and useful mobile networking features. As long as connectivity is optional, you can't assume it and nobody will design for it (except email vendors). Of course, connectivity should include access to your base computers at home and the office and it may take more than a year to break the outdated firewall concept and secure the last mile for corporate users.

In addition to mobile hand-held computers, televisions and other consumer devices will go on the Internet. Of course, WebTV 3.0 may be named IE Light (now, there's a possible prediction), but no matter what branding decisions are made, WebTV must remain fundamentally different from IE because televisions are fundamentally different from computers. You cannot unify user interfaces across these two platforms and retain optimal usability.

The bottom line is: plenty of non-traditional Web access; thus plenty of non-IE browsers. Even if some of these browsers are called IE, they will have to be designed differently: the smaller and more constrained your device, the greater the need to optimize the user interface to fit its exact specifications and the usage conditions for that device.

Update added September 1998: According to an IDC study, Internet Explorer's market share did indeed pass Netscape in June of 1998: the exact month I predicted! Despite this, Microsoft did not issue the predicted gloating press release - probably because they want to play down their strength because of the current anti-trust litigation.