Why .NET is a bad idea
.NET hopes to address three current problems at Microsoft, namely loss revenue due to pirated software, security problems and users who continue to use older unsupported software.
The pirated software has been an issue with the boys from Redmond for some time now. They’ve taken steps to curtail the behavior with product activation on the latest version of their OS Windows XP. However, even with product activation, their exists the problem of the software residing on the hard drive of the user and the potential for it to be cracked, which as you’ll remember, is exactly what happened. .NET prevents this from happening by running what is basically java servlets on a central server that end-users run remotely. There isn’t a CD you insert into your desktop and install the software, it’s a subscription-based service similar to Hotmail. If you happen to share your account password and the server notices multiple instances of the account running on different IP addresses, it locks the account. It’s really that simple to stop piracy, if it’s implemented correctly.
As for security issues, it’s less likely a Word macro will maliciously delete files from your hard drive or corrupt your system due to .NET’s java roots. Java applets run in a “sandbox” composed of a byte-code verifier that validates the code before executing, an applet class loader that determines if and when classes can be loaded across the internet and the security manager which restricts ways in which the applet can interface with your computer. Newer versions of java allow for signed code to step outside the “sandbox” and run on your system, but the security model of java has been discussed in more depth, elsewhere, than I intend to go here. The point is, the structure upon which .NET has been based upon has been built with security in mind, not the latest, whizbang features.
As for older software, I’m sure .NET will initially support older formats. How else would you get your existing Excel and Word document into the newer .NET version. If Microsoft can pull this off, it must be a wet dream come true for them. No longer will you purchase Office, you will subscribe to Office. Switching the software over to a subscription model will dramatically increase annual profits by virtually eliminating production costs of installation CD’s, its assorted packaging and shipping costs. The pricing model may appear attractive to previuosly hesitant consumers as they may only pay twenty dollars a month for the subscription as compared to the hundreds of dollars they would’ve payed for the boxed set of disks. This could potentially bring in more customers for Microsoft and more firmly entrench the Office suite than it has already become.
That’s all of the good aspects of .NET, now let’s look at some of the more likely, less pleasant aspects of its implementation.
In order to use any of the Office applications, once the migration to .NET is complete, the end user will be dependant upon an internet connection. You’ll need a pretty fast one too as you’ll be downloading entire applications every time you want to use them if they’re not locally cached, or every time they’re updated. Considering Microsoft last security patch, it could take some time. This won’t be an issue for coporate LANs or consumers with broadband, as there might be some initial latency but, that should disapper once the application’s cached. Consider for a moment, however, that the majority of the United States is on slower modem connections, not to mention users overseas.
Even though .NET is based on Java, it is still produced by Microsoft and, therefore, more prone to gaping, security holes you could drive a tailer-truck through. For instance, Microsoft claims that hooks are there for almost any programming language, but how many .NET applications can you find that aren’t built with Redmonds proprietary programming languages (C#, Visual [basic, C++])? And we all know how useful, and proliferate Microsoft’s Visual Basic can be.
Also consider that many corporations and consumers aren’t running the latest and greatest hardware. Many consumers use hardware until it dies and many corporations buy used or surplus machines for simple document editing and spreadsheet use. Java and, subsequently, .NET are very resource hungry languages. Having to upgrade system software to take advantage of .NET will relegate existing, useful hardware to the scrapheap earlier than expected. Even if companies are able to cut back IT staff with a more secure product and in-place upgrades, and even if the subscription model will be cheaper than traditional licensing, the cost of upgrading all of the machines to the latest and greatest so they can essentially act as dummy terminals may give some businesses pause.
Also, daily “fixes” could be pushed onto subscribers that prevent interoperability with other software. What if you decide to take the plunge and convert your company over to a .NET implementation of Office for a year? What if, after the first year, you decide .NET is not for you and you want to switch back? How many documents do you think will gracefully degrade back to their original formats? It’s already been reported that Microsoft version of XML that it uses as it’s file format with the new Word is nearly impossible to parse and convert to a format that can be used with any software, even it’s older versions.
Finally, nobody used this technology when SUN introduced it, what’s changed? Nothing, as far as I can tell. Aside from all of the benefits that .NET could possibly introduce, it’s taking a lot on faith that Microsoft won’t use this technology to force you into their upgrade cycle and keep you there if you ever want to get your data back out. Call me cynical, but putting my faith in Microsoft to keep my companies data where I can get to it without feeling extorted just isn’t going to happen in this reality. .NET is a good idea for Microsoft, a bad idea for companies and consumers. In the end, I think it’s safe to say that, the time has come for business and consumers to consider migrating their office/productivity suites to other vendors.