This writer is very leery of Microsofts Longhorn. Our 3 top concerns are first that Longhorn engenders huge amounts of conversions just to preserve the status quo – applications running in the latest Windows operating system. These changes are profound as the basic Windows API is changing everywhere: the Windows GUI is said to have 20,000 new API functions. WinFS which makes up the file system is a complete rewrite incorporating database and search functionality directly into the OS. And Indigo does the same complete makeover on networking and middleware.
Basic APIs and connectivity software is all changed. For example, ISV and hardware firms will be confronted with huge conversion tasks to rewrite all their hardware drivers for Longhorn. Even worse, Longhorn is already on a Death March schedule with features being cutback to make the 2006 delivery date. The problem with that is one of the features most likely to get cut are the conversion routines that get developers and users from current Winforms, WebForms, COM, ADO.NET, ASP.NET and many others to the Longhorn there. Redmond is running quiet, running deep saying nothing about these fundamental issues – not a good sign given the companys mixed record on helping older Redmond software (just ask VB6 or earlier users or Windows NT 4 shops).
The second major concern is that Longhorn is out to hobble if not outright destroy the browser and web interface as a major factor in desktop development. Look at recent events. No updates to Internet Explorer since 2001 and none planned until 2006 when IE becomes a part of the OS. A recent Microsoft IE info event offered no elaboration of what and when overdue updates to IE can be expected. And discussion on what IE will look like in Longhorn was deflected and effectively off limits. But hints are coming out. For example, the new Longhorn “super” search engine might not be a part of the browser but have its own interface because its also an internal desktop search device as well. Ditto for new animation, 3D GUI, and multi-media features – they are being described as only a part of the OS and possibly a new super Media player. Now some argue that Microsoft is perfectly within its rights, hard earned from the US Goverment Anttust case, to put in and/or take out anything it damn well pleases from the OS. And if hobbling the browser pleases Redmond – so be it. Customers will get to decide with their purchase dollars whether they agree with Microsoft. However, my concern is that Redmond has been less than forthcoming about what users are going to find missing or profoundly changed while at the same time the company is out promoting all the Longhorn “goodies”. Come on Redmond, tell us real deal – the Good, the Bad, and the Ugly.
But my third concern is the most profound. It stems from the hobbling of the browser. In general, Microsoft has been slowly but surely reducing its interoperability support. It simply does not support a number of defacto and de jure data standards directly within its own software. And these are major standards like CORBA, Java, JDBC, J2EE, OpenGL, Flash, PDF, Jpeg2000, XForms, SVG, and many others. Rather Redmond leaves this to third parties – Borland has taken to calling itself the Swiss Supplier of Missing Standards for Microsoft .
But not only is data interoperability atrophying within Redmond development portfolio – programming interoperability is nearly non-existent. Redmond has made the only route of entry to its increasingly closed community through Web Services. Its C#, VB.NET, C/C++, JScript, VBScript, Visual Foxpro are all highly proprietary and non-cross platform capable. So when Bill Gates talks about seamless and integrated computing he is really talking about a Windows-primarily gated community – the only entrance and exit being Web Services.
Now again, some may argue that Web Services and third party support is more than good enough. The problems are twofold. Time and again Microsoft has been less than forthcoming when some third party software has bugs or does not work with Windows or one of its applications. (Readers are invited to send their horror stories and counterexamples). More disconcerting is that Microsoft is starting to back away from XML and Web Services standards. For example, XForms, SVG, XSLT2, and XPath2 among others have not/will not be implemented by Redmond. Likewise some important Web Services standards are “under review.” In the meantime, in apparent defiance of W3C declared patent-free policy for its standards, Microsoft is in the process of registering hundreds of patents for its implementations of XML and other W3C and other standards. In short, we see Microsoft setting up the infrastructure, patents, and dys-interoperability to guarantee that the playing field is not level in Longhorn. Specifically only Windows applications run best in the gated community known as the Gates of Longhorn.
Now if you have stayed with me this far here is your reward. Several writers have raised a number of other, often othogonal concerns about Longhorn. See Jon Udells The Longview on Longhorn., Steven Vaughan-Nichols Longhorns Real Job: Trying to Gore Linux, Martin LaMonicas Plan A For Microsoft, CNets Longhorn Goes to Pieces, Click Nows take on Longhorn and Paul Thurotts Windows Insider take.
Heres a fourth concern; with all these new features, how secure is this new OS possibly going to be? I guarantee, with the amount of coverage Longhorn is getting, hackers and virus programmers are going to be all over this thing. And how long is it going to take before we see the infamous patches? 1, maybe 2 years after its first released?