I'd suggest that it probably isn't worth it to spend a significant amount of time to support an audience that small. To determine if this is true, the company should track conversions from those using the IE browsers. If they have an average or higher than average number of conversions, one could argue that it is worth the time spent.
Otherwise, you are probably wasting significant engineering and design resources to solve problems for laggards who probably won't give you any money.
This article is obviously written for beginners, so I won't be too hard on it for lacking comprehensive detail, but I will make a few recommendations.
In the reset section, showing the * reset is okay, but I think it would be good to mention a few of the more popular/established/supported ones - like the Eric Meyer reset or normalize.css, and link to those. Beginners can obviously google "css reset", but it may be hard for them to discern which results are good and which are bad. Showing a few that are popular or commonly used will help them move forward on this point.
For the vendor prefix area, I would also recommend including a paragraph about checking sites like caniuse.com to see if the CSS property that you plan on using is even supported in browsers like IE8 or IE9.
Since the article is for beginners, too much detail within the article may be overwhelming for them, but links to resources they can use to expand their knowledge on the points you mention would be nice, as it would allow them to build on top of what you've written at their own pace.
One good thing that happened were auto-updating browsers. You don't see these articles anymore about supporting old versions of firefox. Microsoft has to get onto the auto-update bandwagon, and we can finally stop bending code backwards.