What Google Can Learn from Microsoft About Operating Systems

 
 
By Jeff Cogswell  |  Posted 2009-07-08
 
 
 

What Google Can Learn from Microsoft About Operating Systems


We just learned that Google has decided to enter the operating system business. Microsoft, as of this writing, has not commented. I can only wonder, however, what thoughts are going through the heads of the people at Microsoft. (I imagine there's a certain amount of laughter going on.)

In the software business, one problem that we run into again and again is that of a company trying to branch out and do something they're not in the business of doing. I see this often in the software world, where a company that is not in the software business decides to home-brew their own software. I used to work for a company that provided software to the telecom industry. The big telecom companies (you know their names) would hire us to create their software. The telecom companies were certainly capable of pumping money into one of their own groups to build the software. But the reason they didn't is they weren't in the software business; they were in the telecom business.

A close relative of mine works in health care, and a company she worked for decided to have their IT people build their own patient management software. It was a disaster. The person-hired to manage the network-who was single-handedly doing about 90 percent of the coding, was in over his head, and the management knew nothing about how to manage a software project. (The managers knew health care. That's what their business was-not software.)

Google knows search. And they've branched into other areas related to the Web, such as online office software. But do they know the OS market? Are they planning on hiring people from Microsoft, Red Hat and others who are heavily experienced in the OS market to oversee the operation?

The OS business is highly competitive. Microsoft has gone to great lengths to get Windows on computers across the planet, and they have huge agreements already in place with the PC manufacturers. Will the hardware manufacturers be willing to just dump Windows or Linux and go with Google's OS? The manufacturers will want a solid business model in place before attempting such a switch. The last thing they want to do is gamble and distribute a huge volume of PCs only to find the people walking into the stores going right to the Windows machines. Talk about crash and burn.

Besides, people don't want to change. A new OS might be cool and great, but people already know their current OS and are resistant to change. Where's my tried-and-true Microsoft Word? There's also going to be a learning curve. Google might think otherwise because the OS is clearly going to be simple, with the browser being its main GUI. But there's actually a very real concern here: People know the taskbar in Windows and the start menu and the icons on the desktop and how to get to the programs they need. Can they quickly and easily make the change? Linus Torvalds has chastised us all for assuming our users are dumb. But while his concerns are certainly valid, there is a line there. Users who are forced to use a computer but aren't "computer people" per se aren't going to easily just pick up something brand new and switch over night (like switching from Windwos to Linux, Mr. Torvalds?).

What About a Command Prompt and Other Questions


 

Google tells us the OS won't be a windowing system and that the Chrome Browser will be the GUI. As a power user, this gives me concern. I hope that they would include various file management tools. And what about a command prompt? I may be an old holdout, but I do a lot of my work inside the command prompt. I want to be able to go into a directory and manually edit and modify the files myself. But if the operating system isn't a windowing operating system and the browser is a GUI, how do I open my favorite text editor (nedit in Linux, scite in Windows)?

And what about development tools? One reason Microsoft is incredibly successful is because it creates free development tools. Now I would be willing to bet the development tools for Chrome OS will be free (based on open-source GNU). But Microsoft has gone to great lengths to make its development tools incredibly accessible to developers everywhere. I would bet that a version of Eclipse will be available, but then we're back to the other issue about this not being a windowing platform. Will the tools have to somehow run inside the browser? While that's an intriguing possibility, it does give me great concern about its feasibility (and whether such a tool even exists yet).

Now elsewhere I've pointed out that the majority of our work today is done within the browser. That may be true at home, but not at work. Obviously, corporations are going to stick to Windows and Linux for their operating systems of choice. I would bet that Google has already considered this, but then we have an issue of people using a different OS at home. (And what about people working from home? Will there be a VPN client for Chrome OS?)

Next, what about drivers? I have some pretty great gadgets that came with USB cables and Windows (and sometimes Linux) drivers so I can plug the gadgets into my computer. Will the Chrome OS support such devices? Perhaps it will if, at heart, it is really just something similar to Ubuntu, which already has most of the drivers you would need. But if it's truly a scaled-down, lightweight OS, then it likely won't have the drivers. The gadget manufacturers will have to create the device drivers. How long will that take, if ever?

Creating office applications that run in the Web browser is one thing. Creating a full-fledged operating system is another thing altogether. Can they really do it? How much thought have they really put into this? Personally, I'm incredibly skeptical that, from both a technical and business perspective, they can make it happen. It sounds more like a cool Google Labs project but not necessarily a viable business model.

I remember the time a co-worker of mine at a previous job suggested we skip purchasing Oracle and just "write our own relational database management system." He added, "I mean really: How hard can it be?" How hard indeed. If it were that simple, we'd see far more competitors to Oracle than we see right now.

Remember Microsoft Bob? Remember Microsoft WinFS? Remember Microsoft Cairo? And many of us don't remember when Microsoft tried to get into the toy industry (outside of their own business) with its failed ActiMates product. There were times Microsoft ventured out of its main line of business, and, for various reasons, had to throw in the towel.

Google can certainly learn a great deal from Microsoft.

Rocket Fuel