Patch That Browser

I don't use Internet Explorer even though it's still installed on my computer (this thing is dang impossible to take off!). I switched years ago to Firefox before IE 7 came out with tabs, and because Firefox is arguably the safer browser. Plus, it's just cooler -- like an iPod versus a Zune.

But was that the right choice? Well, according to a new report, it was dead-on. IE is more vulnerable.

One problem is that despite the well-publicized Patch Tuesdays and automated tools like Windows Update, IE users patch their browsers less than other users. The report also argues that it takes Microsoft too long to find a flaw and write a patch.

What makes the results just a wee bit suspect? The research was done by IBM and Google. Hmm.

What's your favorite browser and why? Tell us all by writing [email protected].

Bill Gates retired last week, though he remains Microsoft's chairman. Some chairmen have a soft touch, are more figurehead than figure. I expect Bill will be different, that he'll err on the side of being a strong rather than a weak chairman.

Over the years, I've been asked many times what Microsoft would be like without Gates. My theory was that Microsoft wasn't so much focused on a single product as it was on building an integrated system, like a quilt. Each piece didn't have to be better or as good as the competition's; it just had to fit better.

I also believed that Microsoft was a bit like the Pentagon: It had plans for nearly every contingency. My conclusion? Gates could leave and Microsoft would be fine for the next five years. It just has to keep making more quilt pieces.

To some degree, that's still true. Desktop OSes and Office suites are still rich clients, and servers by and large are big hunks of IT-managed software. But the world is changing, and services are taking hold. Microsoft has to be agile and make faster and, sometimes, radical decisions.

So Microsoft will be different without Gates full-time. Steve Ballmer will be more free to be Steve (that'll be fun to watch!), and it'll be put-up-or-shut-up time for Ray Ozzie. Speaking of whom, if you want the real inside dope on Ozzie's challenge, read Redmond Editor Ed Scannell's recent cover story, "Cloud Man," here.

Do you care about cloud computing? Share your insight by writing [email protected].

Defending Nick Hogan
Yesterday, I started an item about Bill Gates by saying, "Unless you were living in a bio-dome or were in a Nick Hogan-induced coma, you must have heard that Bill Gates retired last week."

Redmond Report reader Chris took offense and had this to say:

"Bad taste, Mr. Barney. I could see that coming from a young staff writer, but not from an editor in chief."

I wrote Chris back to point out that Nick Hogan isn't worth our admiration. Here's that message:


Take a look at these two links: 1 and 2. Nick Hogan essentially killed his friend, then he and his dad bad-mouthed the victim and plotted to make money from the accident.

Hulk Hogan showed disrespect to the court by not removing his do-rag even when testifying. Nick Hogan is a dirt bag."

Agree? Disagree? Sound off by writing me at [email protected].

Mailbag: Hyper-V More Than Hype?, More
On Monday, Doug posted a letter from reader Mark who was less than enthusiastic about Hyper-V. Here are some more of your thoughts on the Microsoft hypervisor:

Sounds like he has a axe to grind. Hyper-V demonstrates a much more mature software product than the pre 1.0 that the reviewer exclaims. I have found Hyper-V to be an extremely stable and flexible product with a full Windows 2008 install or with a Server Core, the latter being much more locked down and hacker-safe.

In the next year or two, this product will have "world-class" written all over it.

The issue at hand is that everyone (including Microsoft) is comparing this product to VMware ESX. Just because they strip the Start menu out of the OS to make Server Core, doesn't mean this is a bare-bones hypervisor. It's merely an evolution of Virtual Server allowing deeper penetration into VMware-entrenched territory. Now with the use of clustering, you can perform cold migrations and have the ability to use a ridiculous amount of processors in a VM.

There are other more glaring problems than killing a host through the parent partition -- like not being able to over-commit the level of RAM or iSCSI-only support. Hyper-V is a good ESX 1.0 competitor but no one will take it seriously until you get those two limitations straightened out, along with Live Migration.

I have been using Hyper-V for about two months with few problems except for the following:

  • NO support for wireless adapters. Yes, you may argue that there is no place for wireless in a server environment, but what about us developers who emulate the complete client system on our laptops? I have been told that the reason is that Microsoft can not clone the MAC address of the wireless -- but it worked with Virtual PC!
  • Still no USB support.

As to running out of resources on the root/parent machine, that has never been a problem. I have configured it as core and not used the core machine for anything (other than Hyper-V). Bottom line, I think it is a good product and will become even better when Exchange 2007 is certified to run on Hyper-V.

VMware Enterprise has many very cool features that Hyper-V lacks, but the reality is that Hyper-V is version 1 and really cheap to buy and use! For a test environment, it works great and is easy to use and set up. Heck, VMware now gives some of its products away for free in reaction to/anticipation of Microsoft's entry to the virtualization market.

The writing is on the wall for VMware. By the time Hyper-V is in rev 3 or 4, it will be able to support enterprise virtualization very well. Why pay 5K a processor for VMware Enterprise when you can get it much cheaper in Windows? Why hire a VMware OS expert when anyone that can admin Windows can admin Hyper-V?

I'm planning to try Hyper-V as a small computer science experiment so that I can run Vista and XP in parallel. Toward this end, I built a new machine. I've installed Vista Ultimate on this machine while waiting for the Hyper-V RTM and discovered that it does a good job of supporting the hardware the Vista analyzer said wouldn't be supported (Epson Perfection 1650 scanner and ATI HDTV Wonder). But even so, I want to continue with my experiment to see what happens and possibly write up my experience for the benefit of others. This is going to be a budget project; I got Vista Ultimate and a one-year Windows 2008 trial from Microsoft for free. Now my wait for Hyper-V is over and I'm ready to start.

But where to start? This is the purpose of this e-mail. Do I have to build a new system from scratch, starting with Win08, reinstall Vista, etc.? Or can I "import" an existing installation of Vista and install a new version of XP? And what do I do with the downloadable images Microsoft provides? Should I use a core Win08 60-day trial and just extend it three more times as Microsoft suggests? Then can I update that install with my one-year trial product key? Can I extend my one-year trial three times? Does virtualization share peripheral hardware (i.e., NICs) or do I really need to have two machines worth of hardware on one motherboard? It doesn't seem likely that two OSes can share the same MAC/IP address, I must admit, but that would be quite a problem if I wanted to run 12 OSes as "astute reader Mark" desires to do.

Keep the info coming. I'm indeed about as giddy as Mark suggests but I'm not going to let his comments spoil my buzz. I can see from his comments why an actual server administrator might be a bit less than over-enthused by Hyper-V 1.0. If I can get just XP Pro and Vista Ultimate to coexist, I'll be one happy camper.

And after Doug's announcement that he's once again manning the Redmond Report column full-time, Gordon wanted to get just one thing straight:

Not THIS purple dinosaur?!

Join the fray! Leave a comment below or send an e-mail to [email protected].

About the Author

Doug Barney is editor in chief of Redmond magazine and the VP, editorial director of Redmond Media Group.


comments powered by Disqus