I Got IT Wrong

In my coverage of BI and the SAP/Business Object acquisition, I conjectured that Cognos and MicroStrategy could not stand alone too much longer against the onslaught of freebie BI giveaways by Microsoft notably bit also Oracle and IBM. So I hypothesized that HP or Sun might want to take the plunge because BI is still where the action is in enterprise software development. This is because BI better than SOA and Web 2.0 can and does more often generate positive IT project ROIs due to the closeness of BI to describing where and how to make better sales and servicing decisions.

IBM Buys Cognos

So you will see a lot more of these screenshots from your IBM salesmen and IT shops. And on second consideration – admittedly after the IBM acquisition fact, the Microsoft Windows, and Web 2.0 savvy of the Cognos BI software is a nifty fit into the portals and report-writing of IBMs BPM -Business Process Management and ECM-Enterprise Content Management software offerings. It will be interesting to see if IBM does a Tivoli/Notes or an Ascential/Blox in incorporating Cognos into the ever expanding IBM Enterprise Software lineup. Meanwhile, mea culpa, mea culpa, mea maxima culpa (2 more than Bill Gates for his stifling of Win 2.0 innovations).

In the past few months I have been arguing that Microsoft would finally be delivering on .NET across the board in its software technologies – Office, Application Servers, Longhorn, and Longhorn Server. I got it wrong.

In a series of reports on .NET Framework, Microsoft Watch – confirmed by Infoworld has raised the following key issues:

1).NET is having key backward compatibility issues with .NET 1.1 further complicating tight SQL Server 2005, Visual Studio 2005, and especially Longhorn schedules;
2)More importantly, Longhorn will be built only partly on .NET. This means that not only will WINFS, the new filesystem for LongHorn, will be a tack on – but so will the total security and reliability that .NET Managed code delivers to LongHorn. And even downsized, Longhorn is still a very ambitious project with a presentation layer rewrite in the form of Avalon and Collaboration/Controller/Messaging framework in Indigo(assuming these do not scale back too) . Indigo without complete .NET is most problematic because it forays deeply into the untested and security-vulnerable waters of Web Services and network-communication integration;
3)Visual Studio 2005 also is not all its cranked up to be – leaving out key XAML/Avalon plus Indigo designer and development capabilities. Worse, Visual Studio 2005 may have a lot less .NET coding in itself and what it delivers as deployment capabilities to users.

My apologies to readers – I had assumed that Longhorn was going to be the big watershed changeover to .NET. This was when .NET would deliver all its security, gone-with-the registry, exceptional reliability, cross device rationalization, etc, etc. Instead it appears LongHorn is going to be … well uhhhh …. somewhat of an upchuck … a dogs breakfast of development tools, APIs and frameworks all kaleidoscoping in out of use, obsolesence, and migration timing to make your head spin. Hopefully the backroom leger-de-main will be totally transparent in terms of the features and performance of LongHorn.

By the way if you believe the last statement – Joel Spolsky has a bridge in his borough up for sale at a bargain price.

I Got It Wrong