This is by and large the Cognos Impromptu OLAP program, its complete report and chart creation interface - but it is running entirely in Internet Explorer. It does not require a massive set of plugins and controls to be downloaded to the desktop.And from all the demos I have seen the response time is very good on a DSL connection. On a corporate Gigabit Ethernet it runs like quicksilver. All this running on a standard corporate PC and browser.

Just as impressive was the demonstration of PeopleSoft's PowerHTML being used with PeopleSoft Enterpise 8. This time users were doing heads down data entry with no need to resort to a mouse - all the operations were done by keystrokes. But also all work was done through a browser - again no special plugins or downloads. And there were very few pauses - waiting for the Web Server turnaround occurred only at major completion points. A few questions to the data entry staff confirmed my suspicions - this was a powerful data entry environ which they liked working with.

These two examples are not exceptions but rather indicative of the rule that the Web Interface is the GUI of choice for a broad and growing range of applications. The Web Interface is a winner because of three major characteristics:

1)it is simple to learn and remember how to use especially navigation;
2)it is easy to get started developing in; yet can deliver as above, the most sophisticated apps;
3)it is cross platform running on every client and server OS and many mobile, PDA and other smart devices - coming close to delivering "seamless computing " well ahead of Bill Gates Comdex keynote.

So why hostility to the Web Interface from Microsoft? They have a dominant market share with Internet Explorer and three years ago they were espousing the virtues of their ASP/ASP.NET and DHTML. A number of their own applications like Biztalk and Commerce Server take advantage of the Web Interface in a big way. Is this another case of "Blindly Blame it on Redmond " ?

Well follow the money logic and recent Redmond actions and see what you conclude.

Microsoft, by its own executive's decree, makes no money on Internet Explorer or its IIS Web server. And is pledged to never do so in the future. But it does make 3/4 of all its profits and 2/3 of all revenues from desktop products like Windows OS, Office and the growing suite of desktop applications and games. But the Web Interface and applications like Cognos ReportNet and PeopleSoft's PowerHTML apps effectively bypass the desktop and use big servers for most of their operations. Even more dangerous, the Web Interface is client OS agnostic. Web Interfaces run just as well on Apple, Linux, old Windows 9x as well as the latest XP desktops. As more major applications using open standards like HTML, JavaScript, XML, Web Services and other open development technologies flourish and displace/replace Windows-only applications, the cash cow that is the Windows desktop gets severely threatened. So Microsoft is working triple overtime to put more critical and integrated services back down on the client. Read the Longhorn, Indigo and the new Smart Clients campaign. Look at Share Point Services or the new InfoPath Office application. Large chunks of these programs' functionality could easily reside in or be shared with a next generation IE browser. But there is no money in that and it might divert apps away from the Windows client.

However, as always, it is not enough for Microsoft to innovate - it must slow down or stifle the competition, in this case the broad based Web Interface. But how do you do that ? Well its not as hard as it might seem. Start with 85% or greater control of the browser market. You then hobble the Web Interface by slowing browser innovation to a standstill in your market leading product. Next you intensify development away from the browser. If you must use the browser, make it proprietary in some critical step or process. Finally, you start to question the browser and Web Interface as the best means of delivering "seamless computing" and the full richness of "Smart Clients" on the desktop.

On the first tack, Microsoft announced early this year there would be no more updates to Internet Explorer on Windows until Longhorn debuts in 2005-2007. Nothing but security fixes. As well Microsoft abruptly discontinued all development and support for IE on the Mac in early June. Hence Microsoft will not be meeting previous commitments to implement important Web standards like complete CSS, DOM, JavaScript, and HTML standards. And as for the flood of new Web standards which are browser based like XForms, SVG, X3D, XSLT, XQuery, JPEG2000, XUL and literally dozens more - don't hold your breath about seeing them in Internet Explorer. This has a big impact on a lot of organizations who see to their dismay Internet Explorer getting long in tooth. Even our two featured apps, Cognos ReportNet and PeopleSoft's PowerHTML run best in Internet Explorer. But IE does not get an update until Longhorn comes out in 2005-2007. And IE is promised to be a part of the operating system. So to get the latest Microsoft browser you will have to buy into the Longhorn desktop OS refresh.

There are other signs of hostility towards the Web interface. Microsoft has steadfastly refused to put in its developer tools a comprehensive set of "Obey_Web_Standards" toggles or switches (in contrast ISVs such as Adobe,Borland and Macromedia do have such toggles and checks). So Front Page, Word, Visual Studio, BizTalk and other tools blithely go along and insert, if the developer is not extra careful, non-standard, proprietary Microsoft code and connections. This adds a lot of extra burden on developers to spot the exceptions and restore the code to open standards. In another example of hardball, Microsoft in losing the recent browser legal suit to Eolas/UCal has taken a curious approach. They are appealing the decision but also rushing through a series of not insubstantial fixes to Internet Explorer that will make operation of audio, video, Flash and Java applets very herky, jerky. However, if developers in turn make some tedious changes to Web pages using the <embed>, <object> or <applet> tags. - they can avoid the worst of the the upcoming problems. Just as in the case of licensing Unix rights from SCO, it is not apparent in whose interests Redmond is acting. Certainly not the legion of browser developers increasingly using rich media who now have a major set of unscheduled maintenance chores.

As a journalist one can tell that the subject of IE, the browser and Web Interfaces is a touchy subject on the Redmond campus. Requests for interviews on these topics typically get refused or get steered towards a limited vision of the future. Nothing is discussed other than the standard line that the browser will be incorporated into the OS. However, developers can get some sense of what will be happening. The new Share Point Portal Server and its client side set of supporting services are being tied into Office 2003 and other desktop apps. Ditto for InfoPath, an XML and Forms processing engine on the client side, which takes on a lot tasks and functionality that originated in IE but stopped with its demise. Instant Messaging, Voice Processing and Rich Media are seing both client and server side enhancements as Outlook, Word, and new apps get tied more closely to Exchange, SharePoint and other new servers. Windows Media Player and "Sparkle" will spearhead new secure client-side rich media plugins and development to replace the more vulnerable Active content objects. In sum, Microsoft is really beefing up its client applications in both the immediate term and the longer, Longhorn horizon(2005-2007). Yet effectively mum is the word about how IE and the browser fit into this new 1 Microsoft Way.

Summary

Some may say this is just a tempest in teapot or shadow boxing with an imagined Redmond Ringwraith. However, for Cognos, Peoplesoft and scores of other software developers who have invested highly in not just the Web Interface but also IE as their main delivery vehicle - loss of IE on the Mac and frozen IE on Windows does not bode well. And web developers will still have to bear the brunt of continuing collateral damage from the browser wars because IE goes without promised standards fixes/upgrades. These developers are not doing imagined extra labor making two sets of code work for IE and most other browsers. And all the rich media users who in early 2004 have to change their Flash, Java Applet, and other Active Content code - they won't be doing 'ghostly work'. In sum, these are not teapot problems; but rather Microsoft once again imposing serious draws on developer's time and resources while they advance their own, proprietary position.

What can be done ? ISVs owe it to themselves and their customers to deliver applications that work as well on Mozilla and/or Opera as IE. But as usual developers have the nastiest task. It is becoming clear that IE/ASP/Web.NET developers are becoming disenfranchised. Their number one delivery vehicle, Internet Explorer, has been frozen-in-time. Three years from now they will have a replacement of uncertain functionality,performance and interoperability. Millions of IE and ASP developers are being treated cavalierly by Microsoft. Microsoft is saying "Trust us. With Longhorn we will have completely rewritten the underlying Windows SDK adding great new functionality and features. And IE will have some great new role in this new Longhorn world. And somehow we will get you, our IE/ASP/Web.NET developers, from static, standstill here to dynamic there. We just reserve the right to decide when to let you know all about this great new stuff and IE's new role." IE/AP/Web.NET developers should consider how easy the transition was and how much help they gotfrom Redmond in getting from ASP to ASP.NET or VB to VB.NET. Keep in mind the Windows SDK is a complete rewrite come Longhorn and the "new IE".

Meanwhile, ISVs and developers should be much more adamant about Microsoft supplying a set of "Obey_Web_Standards" toggles in Redmond's development tools. Their Chief Software Architect's call for "open, seamless computing " should include a "Obey_Web_Standards" opt out-of-the- proprietary toggle for his development tools. Developers and end users should also migrate to Opera or Mozilla sooner rather than later. This is one of the easiest ways of weaning off a Microsoft dependency. And it is not such an onerous burden as both Mozilla and Opera are much smaller than IE, much easier installs, are better performing, are more reliable and secure plus more standards compliant and now sport a host of added functionality that IE will not match untill 2005 at the earliest.And did we mention Opera and Mozilla are more reliable and secure ?

But this will be not be just a batttle between browsers; but rather between IT development strategies and philosophies. It is a contest between open, interoperable and backward compatible against somewhat open but runs best on our servers, application stack and client systems. It is essentially Microsoft wanting to dictate how computing will evolve and at what price and pace. The question is about futures. For developers, ISVs and end users alike the major issue is whether all Redmond's promised usablity, features, and new found security/reliability/scalability religion is worth dancing to the no longer lowest cost Microsoft tune.

Worried about the Web Interface ? Allay some of your fears with our rash of new JavaScript component tutorials just below - they work in Mozilla/Netscape, Opera as well as IE.