Well, hasn't this been a hum-dinger-ringer of a past week? That Vista 2007 post is up above 430 comments so far.
(This is one of those navel-gazing posts, so be forewarned - after a fire-burner of a week, I'm trying to bore things down here.)
Last week started with me beginning to collect positive buzz around Microsoft via the Mix06 conference (and write notes about dev vs. pm vs. test). It's all lost now, but I remember seeing appreciative quotes like "...this seems like a very different Microsoft," and how people were beginning to get that the next generation of Microsofties were busy turning Microsoft into a positive, transparent, honest we're working with you company. We have lots of sins to amend for and lots of contrition to work through before we can be back in balance. Mix06 seemed to be a nice, delightful skip along that path. We are working on it.
Then, as if to torpedo ourselves, the bad news and reorganization erupts and the Mix06 attention meter drops to nil. As if scanning the sky amidst the destruction of their city, Microsoft watchers fretted, "Where's Scoble?"
Could we have at least waited a week to break the news and built up a wee-bit of positive goodness from Mix06 and the Office developer's conference?
Truth be told, at this date I'd much rather start writing about how Microsoft is turning itself around versus pounding out diatribes raging against the machine regarding how we need to start obvious enforcement of accountability. This accountability thing is a core value to me that translates into a better company. A number of commenters have pointed out that slipping Vista due to the non-business components not being ready was the right thing to do: shipping something shoddy early is an awful sin compared to shipping later and high quality. Yeah, okay, "duhs" all around there, but that's a small insight at the end of a long running slo-mo train-wreck.
One small reflection on accountability in this comment, in the midst of fixing automation-generated issues:
[...] I’m willing to be accountable for my mistakes, but first I want to see some GMs held accountable for theirs. I’ve made lots of mistakes in my career, and I’ve been accountable by making up for them with even more successes. But now every good thing I do is craptulized by someone farther up the management chain. We are working hard, but can’t make a difference because we don’t have any coordination or direction. The managers who should provide that are MIA. Off buying villas in Italy, I guess. So Mr. Just-Suck-It-Up, what do you propose I do? Stage a dev coup and tell my PM team that I’m calling the shots now, so they can forget about those last few DCRs? That should look good on my September review, considering my boss the GM used to be the GPM. Or maybe I should march into Amitabh’s office and tell him I’m firing all his flying-monkeys and bulk resolving their PREFIX bugs. Should I go over to building 9 and tell the Shell team to dump Glass and just go back to the XP shell in the name of shipping?
[...] Second of all it's contrary to reality. While I was working on Vista - and I don't think you ever have - there was no shortage of dedication on the part of anyone I knew of, at least those at the bottom of the org chart. Devs, testers, maybe even PMs, putting in the grueling hours and present on weekends and not just at crunch time.
The real problem comes from those higher up the chart, the ones who keep tossing obstacles into our paths. The ones who release LH versions that wouldn't even complete installation, the ones who turned the checkin process into a nightmare of needless steps each exhibiting random failures, the ones who insisted that their favorite hobbyhorse had to be part of the process.
Regarding accountability: SPSA. What's that? Well, first there's this:
All us partners were awarded our humungous SPSA grants 8/2003. They vest this August. For some reason I thought they were going to vest a little later, closer to the november original date.
For me, I collect 68,000 shares on 8/29 so I hope the slip hype blows over quickly. I'll take my $1.8m this August, then get pumped and help push this bad boy out the door!
November would have been pushing it for me anyway cause my house in tuscany is supposed to be done late october and we were planning on spending a month there once its ready.
-a distinguished partner
p.s. - go ask your vp if you think I am being a bs/troll. this is real. the spsa program is huge awards tied to company performance, BUT does anyone honestly think that bill/steve have the balls to say that since our performance has been shit that the multiplier is 0? See ya in tuscany!
Now that's just a nipple twister for my soul. A follow-up that doesn't twist so much:
The 68k spsa number is a reasonable number, and the vesting date of 8/29 is accurate. Trust me, I am a senior partner (L80). The numbers are in line with reality and the award amount is based on company performance metric goals established 3 years ago. I also agree with the poster that its unlikely that the award multiplier will be 0, BUT I also don't expect it to be 150% (the allowable max).
The sad truth is that comp plans like spsa are designed for the long term and are designed to reward people at the time they are granted. Its hard to look forward and its impossible to recall a previously granted award.
If LisaB wants to rebuild trust between the front-line team members and our executive leadership, how about making everyone's rewards transparent. As I dole out a couple of percent raises here and a few hundred stock awards there, it becomes impossibly difficult to explain the money-grab happening at the top of our corporation when results do not match the awards. Profit sharing should benefit every full-time employee.
Going back to getting Vista out, one good comment (well, good other than the M. Anti-Mini start) has the following observations:
[...] I am one of the several components owners who consulted with Brian on this latest Vista stuff. And to be honest, Vista is nothing so different from the complex beasts that Brian has nutured to fruition in the past. Of course if you have been here for only five or so years you'd think Vista is the only challenge MS has ever faced. But I digress..
Yes the original plan was to get this puppy out August ending to ensure the two classes of OEMs (Direct and Indirect, e.g. Dell and Circuit City) had a level playing field for the year end Vista fueled hardware & software sales. Vista is being delivered with a bottoms up project management method where each component's (and there are hundreds) schedule make up the master. So the Brian review showed a few components (I can count with the fingers on one hand) needing a few more weeks to wrap. Because of the nature of these components (input vs output) there is no other alternative other than adding these extra weeks (half in Beta 2, 0.25% in RC1 etc).
Now Brian has humongous titanium balls - Go check his history from Exchange 5.5 to Win2K etc. He has his rules that cannot be overriden and some of these are:
1) If it ain't done, I a'int serving it. Period.
2) A customer will soon forgive you for shipping good stuff later than you promised but will never forgive you for shipping bad stuff on time.
The added time was not even much, prolly 6weeks (or 8 worst case) and will serve mainly those few components. Over 90% of the Windows components will be Vista complete in June and then cruise along the project ending dance to August as originally planned. However the added time pushed overall RTM to mid-October.
Regarding that, a number of people followed up saying, "Wow, that's exactly the kind of perspective I wanted to hear from our leadership." Exactly. Not that bungled spin put out by Allchin. And where is Ballmer? Where is Gates?
Regarding Mr. Sinofsky, this comment shines forth with silver lining for those folks looking forward to working on Vienna:
If you take a look through the Office org (including Client, Server, Shared), count how many PUMs, Directors, or GMs there are. You can probably do it on one hand (maybe two). That's not an accident. Sinofsky is not a big fan of the PUM model because it can lead to fiefdoms (see: Windows org). He's more about letting smaller, flatter product teams of Dev/PM/Test do the right thing and be hard core about carving out a vision and sticking to it. And that means not doing every feature you want in a single release and adding 1000 DCRs after you're code complete.
And with some fiefdom busting, you can easily rebalance people around according to needs versus spending time breaking through empire building and useless features.
(And if you want to piss off someone from Office, here's something I witnessed: poke them about how Office 2007 is slipping just as much as Vista. Expect to be told how they are just fine, thank you, and on target to be completely done way before the end of the year. It seems the people ready focus around shipping Office is our OEM customer-people and making their pre-install lives easier to deal with two major releases at the same time vs. back to back.)
And what have I learned?
I'm encouraged by the mention of agility in Kevin Johnson's memo:
3. Agility: Lay the foundation for accelerating our pace of innovation, including focusing on ways to improve clarity of decision making, drive greater accountability, and reduce layers in the organization so we can move faster. It also means utilizing existing expertise within the division to embrace services -- and rapid release cycles that services can enable -- to all aspects of our business. Our software + service approach and the expertise we have built in MSN can support innovation agility as we enable the Live era.
Reducing layers, accountability, moving faster, and agility are all pretty much one of the Pillars of Mini. I know we can do it, it just has to be something that's both valued and something that the leadership, from top to the very front-line, are all compensated for succeeding at as part of their commitments. Otherwise, it's just lip service and a nice slide to pontificate in-front of a crowd of rolling eyeballs.
I've discovered that if you link to relevant topics currently on-top of tech.memeorandum you can achieve perhaps more attention than you really want... even being the top story for a small bit. From a technical + sociological + news gathering + attention economy point of view, it's fascinating. It's like letting a steely ball loose inside of a complex pinball machine. Sometimes it's one boink! and the ball goes straight down between your flippers. Other times, it bouncing around and careening through the bumpers and traps all on it's own, even bouncing into the land of Slashdot.
Speaking of Slashdot: I've once again had to flip comment moderation on. I let a lot of the anti-Microsoft comments pile up for a while, but then it just got offensively silly. You know, recently I was talking with a medical doctor. He said how much he was impressed by Microsoft Corporation and what high esteem he held for the company, along with the community values that lead to lots of charitable giving by the employees and the company. He was bemused and confused to have discovered recently that there was this "Evil Empire" point of view by some people. Most people in the world view Microsoft with polite indifference. The incoming comments by the Any But Microsoft crowd certainly serve as a small reminder of the lightening rod of hatred we still are for a techy subculture. The one thing it makes me ponder is, "Where are the Microsoft zealots? The ones who only want to run Microsoft technology and wouldn't dare suffer a Mac or Linux?"
What's your reaction to that question?
I look forward to flipping the moderation off again. Perhaps this post will be boring enough to warrant that. Plus, I'm going to have to take a break because I need to read some books and enjoy the sun and the wind against my face (and mended knee) vs. lurking in the dark with my laptop and 92 new morning comments to scan through (I know, rub your fingers together to play me a nice, mournful violin tune, but how about you imagining what moderating Slashdot-based comments must be like. Take a moment. Okay, now go to your happy place.).
Oh, and that 60% rewrite brouhaha? On the face of such a comment: what?!? No one would ever believe that!
Then it shot off like wildfire on a dry West Texas plain. Maybe they meant one particular small component of Vista was requiring a 60% rewrite, but the whole OS? There's a lesson here, though: technical people are going to pick up on wild stuff like this and start propagating it and turning it into real news quickly, where it dries and becomes fact. There are going to be plenty more examples of this in the future, especially around releases of Vista and Office 2007. Microsoft (and other companies worried about misinformation) needs to have an fire-brigade that can disperse in minutes blog postings regarding the facts, linking to the offending posts, to stop the propagation as soon as possible. Don't rely on one guy. Blogging is not his day job.
Updated: fixed two small typos that irked me.