J Allard Rumored to Be Leaving Microsoft | Hacker News Reader

J Allard Rumored to Be Leaving Microsoft

1up.com

28 pointseplanit21 comments

I think the article does a disservice by focusing on Allard’s work on the Xbox and Zune. The truth is JAllard has been at Microsoft since 1991 and has had some pretty big accomplishments. The man built TCP/IP into Windows for God’s sakes. He’s credited as one of the people who put the web onto Microsoft’s radar (he was the only non-VP involved in the original plans for Internet Explorer and was pushing the web browser internally years before Gates knew what one was). He (with his team) created IIS (Microsoft’s web server for LAMP folks out there). And so on…

I mean, there’s basically a book written about him: http://www.amazon.com/How-Web-Was-Won-Idealists/dp/076790048...

(Good book. Very pro-Microsoft but still entertaining)

So the fact that he’s leaving after 19 years says something very profound about the current state of Microsoft. It says the wheels have really come off the wagon at this point and that they’re pushing away the few good people they have left.

Couldn't the fact that he's leaving after 19 years be because.... he's been there for 19 years?

> The man built TCP/IP into Windows for God’s sakes.

Before Windows Sockets, getting on the net was a technical exercise. You had to configure a packet driver and maybe even NetBIOS. Or you had to use an expensive third-part net library. Netmanage Chameleon I argue that the reason why Microsoft Internet Explorer became a monopoly was because of winsock32.dll.

it's more correct to say that ms has "integrated" tcp/ip into windows.

J was one of the people behind DHCP. His name is on the RFC http://www.faqs.org/rfcs/rfc2131.html

If I'd been involved in any way with Courier, I'd be on my way out of there as well.

I've heard Microsoft called "the biggest waste of some of the industry's best talent." Courier may be the poster child for this, but if it's systemic, that's how they're going to continue to lose people like Allard.

I'm sad to see this happen. He should have been on the way up at Microsoft; he could have changed things for the better if they'd let him.

"the biggest waste of some of the industry's best talent"

I've seen this first-hand. Incredibly talented people losing DAYS of productivity to trivial disagreements, corporate power grabs, and other wholly avoidable problems. If you start noticing it, it is really quite painful to watch. Sadly, most people just let it float on by.

but then there aren't many places where you can basically work whatever you want to on. Take Allard example, he has worked on TCP/IP protocol in Windows, Internet explorer, Xbox, Zune, and now courier.

This is why I prefer startups.

Can you do anything you want at someone else's startup when startups are supposed to be frugal and focus?

At someone else's startup you're often required to do a variety of different things, which provides the sort of intellectual stimulation that extremely differentiated larger companies lack.

Yes, variety of different things that most of the time didn't yield the best solution.

Pardon me for being a bit sarcastic. It is what it is.

To clarify: This is why I'd prefer to start a startup. You might not succeed, but at least you're working on the project you want to work on.

Makes absolutely perfect sense. Working at someone else's startup sucks unless you're given full freedom; which is rarely the case.

You can basically be assigned to any team you want but the leadership generally does not let you do work except in a few teams (CLR, Research, etc.).

There are a few reasons that struck me during my time there - all related to perverse incentives:

1. There are too many PMs. You could fire all the engineers and the PMs would keep holding meetings for a couple of years reshuffling priorities, co-ordinating interactions across teams and "making hard feature cuts". The last thing is especially comical: The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship. In spite of this, I saw some very good people become PMs instead of devs since that's the only way you get any say in the direction of the product. They all end up spending an insane amount of time with powerpoint decks for shallow reviews by upper management.

2. Secondly, the split between test and dev orgs leads to some perverse incentives. As a dev, you are looked upon slightly favorably (though not goaled on) for fixing bugs that you created.

3. The mid-level leadership is not hungry/driven enough. Virtually all the mid-level people running teams have little to prove after they make partner. Its the standard principal agent problem where the agents are too risk averse.

These factors make it hard for people (especially engineers) to feel like they are making a difference.

The dev team would be behind schedule and the PMs would be very proud that they are macho enough to drop certain features in order to ship

Whats wrong with the above?

A large part of your performance as a PM is measured on how much BS you are capable of generating in order to manage perceptions - which is natural since you do not have authority over engineers. This is one instance of that.

Obviously, teams differ so you might have a different experience.

On the other hand, a lot of people are saying that Microsoft should focus more on its core, not keep expanding into so many markets, that it is not winning.

XBox overall is yet to make a return for Microsoft. Sure, over the last year or so it has started to turn a profit, but when you look at all the money MS has invested in this, they are heavily in the red.

Zune is yet to turn a profit.

Why would courier be any different to these?

I fully agree that Micosoft has been neglecting their core competencies lately, but I think they should expand into other markets, only they should be a lot smarter about it. I do not understand ther obsession with the strategy tax and the preservation of the windows monopoly. They seem to have made bets on many of the right things (online gaming, tablets, Internet tv, mobile computing), but except for online gaming they have pretty much been outhustled hy the competition, despite their head start.

While the article suggests that he is leaving due to cancellation of a project that he was involved in, after 19 years anywhere, a person might consider moving on just to see what else is going on in the rest of the world.

Less interesting that he is leaving microsoft, more interesting is where is he going to ?

Reply on news.ycombinator.com