This blog has moved http://www.sharplogic.com/blogs/ed

How do we know that it has caused PR and morale damage? Where do you get that? Not a single person from either team has contacted me. No one from Waggener Edstrom has contacted me. Are you making that up? Please explain further who is complaining to you. And, explain why you are being used by these people. Especially since you aren't a Microsoft employee anymore.

If Newsweek or someone else is calling the PR team today as a response to what I wrote, the PR team should be celebrating.

That is an invitation to start a conversation about where the Windows Media team is going.

Also, are Microsoft employees really that thin skinned?

If they are, they might want to quit and go work for Apple.

If I were on the Windows Media team I'd be far more worried about all the white headphones I see on campus.

I'd be far more worried about Time Magazine mentioning Steve Jobs in "person of the year."

I'd be far more worried about iTunes sales outstripping MSN Music's sales.

I'd be far more worried about time to market. About image. About customer satisfaction numbers (which, you have intimate knowledge of).

I'd be far more worried that investors are looking at Apple's stock price quadrupling while ours has gone up a few percent.

I'd be far more worried about Newsweek putting the iPod on the cover of the magazine.

I'd be far more worried about Apple stores opening with lines around the block waiting for the ribbon to be cut (and cheering when they are open).

That's what I'd be worried about. If they are worried about what a lowly blogger said about them, maybe they are worrying about the wrong thing.

One last thing. So far, none of my employee goals are to blog or to get more traffic to my blog. So, I doubt that either of those things will help my review score in June.

My personal experience is that only shaming MS publically actually gets a response. And then the response is generally half-hearted at that.

As evidence I point to the MSDE 2000 Merge Module diabolicle and the lame-assed solution they came up with to finally address my bitching and moaning in every public forum that I could find.

It's only now with SQL Server 2005 (Express) that we're getting somewhere, and it still ignores the most basic problem with the way they want you to install the thing. (i.e. that it should be transparent to the user and on unistall or failure, nothing should be left behind from the installer... these criteria are completely ignored STILL)

Talking nicely to MS rarely actually gets you anywhere beyond getting bug fixes sent to you. You guys really tend to ignore people and what makes sense if it didn't come from inside. (and it sounds like even if it does come from inside you ignore it.)

Case and point the Combo Box control in .net (any version) doesn't support Name/Value pairs. Thousands of people complained in the original VS.net 2002 beta, it was in VB6 and you pulled it out of Vs.net 2002. People are still complaining in VS.net 2005 betas and you STILL haven't figured out, that the only way to use a combo control 99% of the time is with a Name/Value pair and that this functionality is not optional, and having to do it by hand in code is not acceptable. (yes, you can do it with databinding but you can't build a static Name/value pair list without writting it in code)

Oh ya, and then there is the DateTimePicker that is utterly useless that you keep ignoring even though their are 1000s of bug reports on it... that one sets a new record. VB 3 was broken with the datetimepicker and it's still broken 12-13 years later! You'd think that in that time someone would have figured out that allowing a null entry in a date time picker is required when dealing with databases... especially when you have 1000s of people telling you so....

Yes, this is a Microsoft wide problem. Even now with your blogs, you love to talk, you rarely like to listen. Yes, it has gotten better with VS.net 2005 but you keep giving us what you think we need, instead of listening carefully and giving us what we are telling you specifically we really need.

You guys need to stop with the needs analysis that happen inhouse and start using your own products more (yes, there are NO major MS applications written with WinForms in .net yet) and anything you have to write a custom control for the mimic basic functionality of windows, you stop, and write a stanard control or improve the one that is in .net already. If you can't build Word lock stock and barrel without any custom controls (and Outlook) then you haven't done your job. Once you've done that, then start asking all of your customers (not just the huge paying ones) what is really needed and really listen carefully. I mean have the developers listen and make sure that they know that they aren't writting software for themselves, they are writting software for the users of MS products, so they're opinion on the subject doesn't really matter at all. You don't think that it's a big deal that the date time picker doesn't work in any database driven situation? Do damn bad. Your users do, thus you make it work your opinion be damned.

BTW. Most of this is a flaw in how you do your beta cycles.

Here's how you do it now.

1. Do a needs a analysis.
2. Impliment (buggy, but that's fine)
3. Solicit feedback from a very small handfull of people (I'd be fine if I was one of them I suspect :) )
3. Add other stuff, take other stuff out (object spaces).
4. Refine like mad. (i.e. bug fix)
5. Release beta 2.
6. Refine some more. (i.e. more bug fixes)
7. Release RC1.

Notice the problem here?

First, you only ask a couple of thousand at most what they think.
Second, by the time that the general public that uses your software the most gets a chance to comment on the functionality, the functionality is largely locked down (i.e. only when 1 million people or more complain do you do anything to change (i.e. C# Edit and Continue) and even then you try very hard to ignore it (DateTimePicker, Combo Control))

There is no general feedback cycle for the general populace.

You treat beta tests as bug fix sessions with free testers. What the beta test is actually usefull for is to figure out what you're missing in real world scenarios. It is very UNUSEFULL (see www.joelonsoftware.com for lots of proof of this) to use the general public as bug testers. They suck at it. Programmers are even worse than normal people at bug testing. They not only suck at it, but are really lazy about it.

So if you adjusted how you went about your test cycles, you'd have a chance. And the key here is that if it's a gap, and there is a reason why you'd want to do it, even if there is another way that is 10x harder and requires way more maintance (Name/Value pair combos) you still fix it and make it work intelligently.

The entire .net/vb component market has been built up as a result of MS being completely incapable of creating components in .net that actually work the way users will use them.

This is why Apple has those lines around the corner btw. WMP is a perfect example of a passible product that meats the needs analysis document that was made when it started developement, and can be used if you want, but users either just don't care or detest it.

Why does Apple do so well and write software that people love to use? (to the point of stupidity I might add)

Because they have UI people (Steve mainly I suspect) that know how to develop excellent interfaces. You guys STINK at it. YOu're great at little things like 3d buttons and the concept of a wizard etc. but you suck at the big picture UI design stuff.

There are two ways to solve this:

1. Find another Steve Jobs (i.e. you could hire me, and yes I really am that arrogent, but it's because I could do a BETTER job of UI design than steve and do so on a regular basis)
2. Use your customers and listen to them. Not just your huge ones that couldn't get their asses together long enough to ask the real users of your product in the organization what is really wrong, I mean the developers, the users the people on the street that are using it, why they like apple and what is missing. Lots of them will be stupid and won't know, but every once and a while you'll get a gem that will make your product fantastic. It's the little things that automatically happen because there is nothing else that you could possibly want to do, that make software insanely great. (This is trivial, but WMP actually does this now, so credit where credit is due: I.e. if I use WMP 10 and right click on a song and click add and then add to burn list, it automatically brings up your burn list on the right. Now that automatically part is great. The crappy part is how damn hard it is to get to the burn list (or even knowning that WMP actually will burn CDs) without knowning it's there. Try teaching a 70 year old how to burn a mix cd for their grand kids sometime... it's frighening... they get out the pen and paper and just write it all down because it's WAY too complex.)

This is the problem at MS and Scoble in his non-politically correct way (not that I'm much better most of the time) is telling you this. He's searching for ways to get you to this point but instead you're killing him for the way he's saying it instead of realizing that WHAT he's saying is absolutely the problem at MS. That he says it directly and sarcastically should tell you MS's failing. People only get that way when they've been beating their heads against a wall for a long time and they're fed up, but still love who/what ever they're trying to fix so dearly that they're still willing to try. Thus the fault is with the people not listening to what he's saying, not him for getting fed-up.

Microsoft listened to users on the Pocket PC. They didn't DO everything we wanted, but they did a couple of big things that made a big difference. The really ironic one is that Microsoft now does Graffiti better than Palm.

So they *do* listen, sometimes. You have to catch the right people at the right time.

Of course, they seem to have left PPC to float while they try and convince people a laptop with a stylus really deserves a special name. You win some, you lose some...

Sorry Peter, but I was one of those guys that kept asking what the point was with the PPC. Phones can do Appointment books and contact managers, and if you need something more, get a laptop... I love the concept of Tablet PCs. I hope that one day they are on every kid's desk in every class room in North America, with all of their text books in there and all of their work done on them.

Take that to the business world and you see my point. But the pen SUCKS for input on a computer unless you really don't know how to type, and then, go learn. I can type 5x faster than I can write, no matter how sloppily I do it. What the pen is great for is doing math, drawing diagrams, etc. Smokes a keyboard. Using both on the same device is the way of the future, but a little 2x3 screen isn't the way it's going to be done.

Toshiba has that part right, now the OS needs to evolve with the understanding that the pen is not a hand writting input device, it is not a replacement for a keyboard. It is a replacement for a mouse, and it is a completely new way of inputting sketches, diagrams and math forumlas. This is the key failing of the Tablet PC as it stands right now, the wrong assumption was made. I'm not saying that the Toshiba flip top is perfect, I think the perfect design would be more like a RIM with a keyboard that flys down when you need it and snaps back up when you don't, but it's definately the right approach.