Splolsky's problem is that his software is even more pedestrian, boring to work on, and useless than Microsoft's data synchronization products that he's making fun of. I know there are FogBugz users out there but I haven't met a single one in person. Besides... a commercial bug-tracker in 2008? Give me a break.
Their Co-Pilot product is pretty interesting project to work on, but it's not their bread-n-butter. FogBugz is. And why would one even need top-notch programmers to write a bug tracker?
You attract the best talent by interesting projects and existing great talent that you've attracted previously. That's a chicken-and-egg problem, unless founders are smart hackers themselves, this is why it's so important.
The "interesting" parts for writing FogBugz, is working with Wasabi, their homegrown compiler that spits out classic ASP and PHP. That's an interesting challenge, right?? /sarcasm
To their defense, FogBugz is more than a simple bug tracker. Predictive scheduling is a novel and useful feature that separates them from pretty much every other free and commercial bug tracker I've used.
FogBugz has an intelligent design that makes it easy to get issues and customer interaction into the tracker so that it gets in and stays there rather than being scrawled on sticky notes and gone over verbally between only two people in someone's office.
It has a good spam filter that does n-category email classification like JGC's POPFile.
It has data entry interfaces that combine the best of an unorganized PIM like 37Signals' Backpack with an organized issue tracker like Trac. (Personal experience: Backpack is great to start out with for just about anything, but does not scale over time.)
But, these are more about good design than turbogenius brilliance, and good design does not require rocket science, it requires perspective and taste.
I really think people are spinning Spolsky's comment about Redmond and Google hiring up all the good people way out of proportion. It was just a throwaway comment at the end of the post; it wasn't the thesis.
No, not a throwaway comment: the entire last paragraph summarizes his rant against MS and Google spending money on salaries and perks "because these companies are like cancers, driven to grow at all cost, even though they can't think of a single useful thing to build for us, but they need another 3000-4000 comp sci grads next week."
Those companies are profitable cancers and Splotsky is annoyed because he can't complete.
I just never thought I'd ever see a comparison between Microsoft's visionary products and 1955 salisbury steak.
Paraphrasing, I think Ray Ozzie (Groove) points out in "Founders At Work", that Microsoft is a horizontal company; they want to deliver the platform which every one will use to build their vertical apps.
Their Co-Pilot product is pretty interesting project to work on, but it's not their bread-n-butter. FogBugz is. And why would one even need top-notch programmers to write a bug tracker?
You attract the best talent by interesting projects and existing great talent that you've attracted previously. That's a chicken-and-egg problem, unless founders are smart hackers themselves, this is why it's so important.