June 09, 2008
Seats at the TableOne theme that emerged at the Forum last week was groups outside of Development wanting to get a "seat at the table." This is a conceptual table, but I suppose if you wanted to picture a real table, it's the table in a conference room where a triage meeting was taking place, trying to decide which bug fixes to accept late in a product cycle. Or perhaps a meeting about adding a feature after the coding phase has begun.
Program Management unquestionably has a seat at the table; if there was a battle to achieve this, it happened before I got to Microsoft. Test did not have a seat a decade ago, but by dint of long hard work they have achieved it. In Proudly Serving I wrote that at Microsoft the 1980s were the "era of the developer" and the early 1990s were the "era of the program manager" (in terms of which group had the most power in determining what features and bug fixes were accepted). I also claimed that the "era of the tester" began around 1995, but that "Unfortunately, not too many people are aware that Microsoft is currently in the era of the tester—in particular, the testers themselves don’t realize how important they are." But I think, just a decade or so late, the era of the tester may finally be upon us. I was in a Forum session last week, in a discussion among people from different disciplines, and a tester explained something with the phrase "Testers Rule"; this was not the first harbinger of the possibility that test may be where it's at at Microsoft these days. I myself have toyed with the notion that if I go back to a product group, test may be a more rewarding place to pursue my interests.
Anyway, so you have Dev, PM and Test at the table, and others want in. In particular there were 3 Forum sessions I attended in which Project Management, User Experience (UX), and Operations discussed how to get a seat at the table.
(Interestingly, I heard somebody discussing how Apple does it, and they said they (metaphorically) have a table with four seats, with Project Management and Design (aka UX) occupying two of them, Marketing in one, and Engineering (meaning Dev/Test/etc--the people who currently fill all the chairs at Microsoft) confined to the single remaining one. There was nothing about Operations, but I assume for a web service product, Apple would pull up a fifth chair for them, rather than stuff them into the Engineering chair).
The general tone of the seat-table sessions was "We need to prove our worth to be respected", which I think is a healthy change away from the "We just need to work harder and harder and do our jobs better and better" angle that I used to hear, but it's still unfortunate that Dev can't just recognize the value of these disciplines, which has been shown over and over outside of Microsoft (and inside of Microsoft, if anybody cared to look beyond their own group). The table-sitters generally still have a "throw it over the wall" attitude towards the other disciplines (this was how Dev was vis-a-vis Test a decade ago; now it may be that Dev is grabbing the arms and Test is grabbing the legs).
All three organizations have the advantage that Test has softened up Dev a bit, but they also face a challenge that Test did not. Although Test had to fight to be listened to, in the end when you show a developer a bug they will admit that it is a mistake that should be fixed. UX, on the other hand, could point out a fit-and-finish error that confuses a user, and Dev may never come around to admitting that it is worth fixing late in the game; most bug bars (meaning the level of seriousness that a bug has to hit before it will be accepted) only talk about data loss and crashes, plus geopolitical, legal, accessibility, and all that. Project Management and Operations fall in-between UX and Test on the scale of how much empathy Dev would have for their problems. On the other hand, the UX Excellence team is a peer of the Dev Excellence team, whereas the Ops Excellence team is organizationally just a bit further away, and there is no Project Management excellence team (Project Management is currently a neglected subset of Program Management). I'm not sure how much the relative position of a discipline's EE avatars affects what happens in the product groups; we'll have to wait to see how this all shakes out.
Posted by AdamBa at June 9, 2008 09:19 PM
I think you hit the nail on the head here. In a head-to-head engineering contest I'd never bet against Microsoft. Grounds-up innovation is different matter though. Windows Mobile is just never going to have the same appeal as the iPhone.
Apple understands at a deep organizational level that it's about the user experience not the software or the hardware or the box it all comes in. That's why applications aren't allowed to run in the background in the iPhone (where they will drain battery, CPU cycles and memory for little/no gain). It's why they will spend big money on a beautiful high resolution 3.5 inch screen that you can't take your eyes off of.
Google shows evidence of similar thinking. Without them who would have thought to do such a clean interface to an advertising-supported search engine?
Once you've made the conceptual leap to the fact that UX is at the core of everything then the only people who warrant a seat at the table are those who worry about the system as a whole. When your organization is used to creating products that are both hardware and software based it's going to value those people who can think about the system as a whole. When all you ship is software you are, by and large, going to value software people over everybody else (software dev and test are two sides of the same coin, the only interesting thing from a system point of view is the delivery of working software, how that is achieved is lower layer problem).
I'm not surprised that in Apple the people whose opinions matter are project management, design, marketing and engineering as dividing representation into those groups naturally encourages system-level thinking.
Posted by: Andrew at June 10, 2008 02:47 PM
My UX with Vista was dreadful and I paid someone perfectly good money to downgrade it to XP (which I had to buy besides). What was wrong? I couldn't do this, couldn't do that, in one case couldn't even find the setup program on the CD. And I could not get the find program to find anything. I know bit more now, but all the real information was well hidden inside a useless help program that did not make up in glitz what it lacked in functionality.
I am about to buy a new laptop at the last minute before MS stops selling XP. All the while, the bad UX of Vista has become a cliche. What is it with MS, are they trying to go down the drain?
Posted by: Marble Chair at June 10, 2008 06:17 PM
I think this depends a lot on the group you work in. My day to day experience is that PM, Test and Dev Mgmt take themselves seriously enough to have discussions without Dev at the table.
PM: Such-and-such behavior is strategically important.
Dev Mgmt: Sounds good, we agree.
Dev: This is braindead and will not work with users because of X.
Dev Mgmt: PM owns the vision and usability, you just schedule and implement
Test: Oh, and please schedule your checkins so we can plan the testing a month in advance
Test: This feature is awful
PM: New information we learned indicates that against all expectations, blah doesn't perform with end users
Dev Mgmt: Really, Dev, this is your fault. You had some objections but you didn't fight for them strongly enough.
Posted by: at June 12, 2008 07:36 AM
Ahem well yes, when I say "Dev" I mean dev, dev lead, dev management...the entire discipline, not just front-line developers. What you describe is more of a general "bad manager" issue which could happen in any company, no matter who sits at what table (and yes, I'm sure it happens at Microsoft too).
Posted by: Adam Barr at June 12, 2008 11:46 AM
For your vista UX experience, o noble paterfamilias, in fact UX is the team that should be able to point out, "You know, those privilege elevations dialogues really aren't accomplishing what you think they are" and avoiding a UI that only functions well for people who work on the Windows team. So I would view Vista as partway along a good path, not heading down a bad one, and UX should be the guide.
Posted by: Adam Barr at June 12, 2008 11:53 AM
Oh noble prodigal son, you have missed my main point. Although it is certainly interesting to know how Vista got out the door with that unusable UI, the more interesting question is how MS decided to pull the plug on XP despite its successor having become a laughing stock and a cliche for bad UX.
Posted by: Marble Chair at June 12, 2008 04:33 PM
Decisions like when to stop supporting XP are business decisions, outside the realm of engineering entirely, and I have no idea how they are made.
Posted by: Adam Barr at June 12, 2008 09:25 PM