Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Mike Gunderloy writes in an article for SearchVB.Com about what he believes is more of a perception problem by developers regarding VS 2005 rather than an actual lower quality in the product.
 
"Every single version of Windows, Office, and Visual Studio (and any other product of similar size, for that matter) has shipped with bugs of the magnitude that people are reporting in VS2005. So why are we suddenly so hyper-aware of the VS2005 bugs? The difference is not in product quality, but in two other areas. First, the developers at Microsoft were incredibly open about the development process in this product cycle. For the first time multitudes of people got to see the realities of bugs being triaged to meet a ship schedule. I've seen some online expressions of shock that a bug-fix could be postponed to the next release of the product, but in fact this has always happened. If you don't stop fixing bugs and ship software at some point, you don't stay in the software business for long.

Second, the blogosphere makes it incredibly easy to find and publicize the people who are finding the bugs; previously you had to be in product support to hear those conversations..."
 
I started writing a reply and midway realized I agree with mike rather than disagree, but personally I arrive at a different conclusion than him. that conclusion is that the products were always this buggy, we're just now getting used to ask for better quality, using some new channels like blogs.
 
"First, the developers at Microsoft were incredibly open about the development process in this product cycle"
From my experience, the amount of opacity in a project toward the customer is inversely related to the amount of "shock" and "surprise" the customer experiences. The more the customer knows the things you've been facing in the project (assume you had a customer on site everyday), they know what you're going through, and they know the overall pace of the project and features that need to be cut. They might dislike that some features or bugs are left in, but that's different than expecting that the features that actually stay in the product work correctly. If anything, the amount of opacity on the side of Microsoft has helped to soften the blow for the "customer". Try to imagine what would happen if one day, out of the blue, you'd have gotten the final version of VS 2005 as-is. You wouldn't know what was cut, or why. You'd just see what's there and see that some of is is not working as well as you expect it to.  You remember to yourself that the last time you got a product from Microsoft (VS 2003) it was pretty darn stable, albeit had several lacking features, but the things that worked, worked well. You're much more likely to be shocked and annoyed at something that you feel you had no control over rather than something you feel you were part of, which is what Microsoft was striving for with all of its community effort.
 
   Now, just for the sake of argument, let's imagine the first point by Mike is true. The amount of exposure is directly related to the amount of annoyance. How many actual developers took the time to check which features were being cut, changed or anything else? less than 1% is my bet. Less than 100 blogs do direct coverage of this, and less than 50 that I know of actually participated in the product feedback from Microsoft.
The only thing that would make the general developer community believe that MS screwed up big time is the 2nd argument by Mike: that the widely read blogs basically set the public opinion against Microsoft's products. This is one thing I can't argue with. The media and the power of one are different than they were 2-3 years ago.
But these people usually know what they are talking about, and that usually means their complaints are based on some experience in he field. In other words, the complaints from the widely read blogs are usually justified, while complaints from the less read blogs are,well, less read. so even if they are not justified, they are eliminated by the power of "strongest survives" evolution.  What you get is that the blogs that really tell you what's going on are being read, and the real problems surface. They just get more echo.
 
As an analogy, Mike could have been saying "The government is no more corrupt than it used to be, there is just better coverage of it in the papers". Taking this analogy back to where we came from, you could then say "Microsoft's products are no more buggier than they were before, there is just better coverage" which would mean that they were always this buggy, we just were more quiet as consumers.
My assumption has to be that as developers, and as a community that is more involved than ever in the process Microsoft is using to ship products, we care more, and we demand more. We want better quality than we used to. We want to be heard because we feel more involved. It's like a T-Shirt I saw somewhere. "I yell because I care".
If you ask me, that's a much better situation than it used to be. It puts me and MS in the same barrel where the relationship affects both sides for better and for worse.
That, my friends, is the new "techno-democracy" as I see it.
Published Wednesday, November 23, 2005 11:32 PM by RoyOsherove
Filed under:

Comments

Wednesday, November 23, 2005 5:32 PM by IM

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

I find it hard to judge the stability of VS2005, since Ive been mostly working with WinFX and have Resharper 2.0 EAP installed. The verdict: it's pretty unstable, unsurprisingly :-)
Wednesday, November 23, 2005 5:48 PM by Sam

# sheesh

Ah well, I love VS 2005, I use it everyday, and find it very stable - there are some things it won't do, and sure there are things crashing it, and it can be better.
But still, it is the best IDE I ever used, and I don't wanna switch back, or have waited longer - it's far more than good enough for me.

And honestly, some of the 'bug'-complaints are plain stupid "if I do this and that the IDE crashes" - so don't do it, or can't you learn?
Or people complaining the scrollbars in the IDE don't use the windows theme - I'd strip these folks of the title 'developer' and name them 'user' then.

As Douglas Adams wrote: nothing spreads as fast as bad news, and blogs make this even faster.
Wednesday, November 23, 2005 7:58 PM by Paschal

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Roy all developers using already VS 2003 expect the quality from VS 2005 because they know what they have already. Nothing like 2002 where everything .Net was new. It's a great version but its somehow frustrating that in name of progress some rules of compatibility were broken.
And regarding the Product Feedback Center, yes it's a great idea, but Microsoft is the one who decide what to consider a bug, which is a bit sad attitude.
Wednesday, November 23, 2005 8:22 PM by Bryan

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

I personally don't care about the bugs. What irks me is that it's going to take 2 to 3 years before the next release and/or a service pack where Microsoft fixes them. Unfortunately, the new release will introduce new bugs and the cycle will repeat.

I don't expect perfect software, but I do expect quality software. Microsoft just can't seem to break the cycle of releasing crap. Until I see regular and routine bug fixes for their development products, they are not moving in the right direction.

Bryan
Wednesday, November 23, 2005 8:37 PM by karl

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

I'm using Resharper AEP too, so I try not to be too quick to judge it, but it is a serious pain.

I spent 10 minutes trying to figure out why VS.Net was reporting invalid errors, only to rebuild and get a compile-time error of:

"The application domain in which the thread was running has been unloaded"

finally, it built ok after this. Maybe it's resharper...
Thursday, November 24, 2005 3:32 AM by Ward

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

I have used vs2005 since beta1 almost every day for a medium sized winforms solution (about 15 projects). The overall quality is acceptable, but the design time of the winforms is much too sensitive to user coding mistakes.

For example, if i misbehave in a default constructor of a Form, vs.net shows an quite cryptic error that it cannot render the form in design time (good) but than completly crashes. Also using inherited forms is a very sad experience.

And in the process of failing, some old temp assemblies are left behind that you need to manually clean up otherwise changes to your form code are never picked up (basic clean solution doesn't help).

The funny thing is, in the last ctp/rc seemed actually more stable to me than the RTM.

A last note about Resharper 2.0 EAP: it's nice for small solutions, but for large solutions the startup parsing it's just too slow when you have multiple vs.net 2005 crashes a day. And stability is not the strongest point of the current build (210)

My 2 cents


Thursday, November 24, 2005 5:04 AM by Frans Bouma

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

"And honestly, some of the 'bug'-complaints are plain stupid "if I do this and that the IDE crashes" - so don't do it, or can't you learn? "
You're the one who's producing stupid complaints. If I type in code into an editor I don't want it to crash. It's just plain text. So you're saying now that I shouldn't type in code into an editor because it might be crashing the IDE? What are you, a manager?
Thursday, November 24, 2005 5:09 AM by Frans Bouma

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Oh, and if you're in a large project and refactoring it from pure .NET 1.x code to .NET 2.0 code, and you have a lot of errors (because not everything is refactored yet), say over a 100, it takes a long delay before it actually shows up the errors after the compile is long done, long delay as in: 5-15 seconds minimum. On 2003 this wasn't the case. I'm not sure what it's doing, but it's weird.
Thursday, November 24, 2005 3:16 PM by Sam Gentile

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Just so you're clear in case you though that was me Frans, that was a different Sam (from Germany). I have no opinion on this matter without further research.
Thursday, November 24, 2005 3:48 PM by Simon Stewart

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Roy, I agree with you in principle that the VS quality thing might come down to the change in the dev world since the last release, but my first experience with the RTM (http://www.brokenkeyboards.com/brokenkeyboards/blogs.aspx?blogid=188) wasn't pretty.
And, after wailing bitterly to PSS about the handful of issues we have had to skirt around each week with VS 2003, I was really hoping for Whidbey to be the solution to my problems.

To me, it all comes down to managing expectations. If I make a client wait for a month or more before handing out updates, I'm not going to be popular. However, if I *try* and get hotfixes out often and make sure that the client doesn't feel that they are paying for it, then I'll stay in business.

If I told the client to wait for the next version, I'm sure they would be critical no matter what when it finally come out. Understandably too.

Microsoft's handling of the client's (our) expectations wrt Visual Studio has been poor. And it's made me a little meshuga to say the least.

Ward: "if i misbehave in a default constructor of a Form, vs.net shows an quite cryptic error that it cannot render the form in design time (good) but than completly crashes. Also using inherited forms is a very sad experience."
This is scary as most of our work (and IDE bugs) are in the Winforms space. Hopefully the infamous 'disappearing usercontrols' bug is fixed.

And, I've been finding speed issues with .NET 2.0 when compared to .NET 1.1. This, and the poor support, is definitely reason enough for me to not move my team onto .NET 2.0 until SP1.
Saturday, November 26, 2005 2:55 AM by Sam

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

"You're the one who's producing stupid complaints. If I type in code into an editor I don't want it to crash. It's just plain text. So you're saying now that I shouldn't type in code into an editor because it might be crashing the IDE? What are you, a manager?"

There are only a very few very special cases of source entered that can crash the IDE. So learn them, and don't type them. Programming in VS 2005 with source highlighting and intellisense is very obviously not 'only entering text'. If you really only want to enter text, use notepad.

VS 2005 is a tool, and a tool for a professional who should be knowing what he is doing, who should be able to learn what this tool is good for, and what not to do with it, and act accordingly.
Tuesday, June 13, 2006 12:18 PM by Will

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

In the final release of Visual Studio, the MonthCalendar Windows control has a terrible visible flaw and I wonder how that control made it passed testing.

To see the flaw:

Drop the control on a form and run.
Select a range of dates within a month.
The display within the control is messed up.
Change the month.
The display continues to be messed up.

Also, I'm disappointed that snaplines and grid don't work together.  They must be used independently.

I do like VS.NET as a tool.
Monday, July 03, 2006 11:33 AM by Chris

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

Yes Microsoft has screwed up and has been screwing up Visual Studio since VS6 The reason is simple, there has not been a patches or bug fixes for a compiler since VS6 SP6! Yes if you compare the release of VS6 to VS2002-2003 or 2005 you can say they had similar issues. However the thing that irritates us all is that you don't make that comparison. You compare VS200X to the VS6 SP6 release which had been cleaned up. For some reason supporting the development community in this way doesn't seem important to Microsoft. I realize that there have been many new features; they have all come at the expense of stability and predictability. VS200x releases have had the most heinous of bugs and for some reason they weren't worth patching. For example: VS2003 C++ project references wouldn't consistently build code correctly. So yes, you learn and go back to the old Force using technique. However, this is only after months of annoying intermittent problems with multiple developers and few weeks reporting the bug to microsoft support. All in all probably tens of thousands of dollars wasted.. only to be told wait for VS2005 and MsBuild.. hmm lets see that was late 2003. I'm sorry.. VS6 SP6 was very solid when it came to developing 32bit windows applications. At the very least it was predictable and you knew you're devil. I was hopping that VS2005 was going to clean things up a bit. Sorry, I've already reported a bug in the VS2005 Managed C++ compiler that is preventing the transition of 500,000 lines of C++ and C# code from VS2003 to VS2005 And for C++/CLI... Hmm.. you have to ask.. wouldn't it have been nice to stabilize the C++ environment before making radical changes like those involved with C++/CLI. The only reason VS2005 isn't screwed up is because there isn't a reasonable competitor! Otherwise we would be using someone else's product like SomethingElsePlusPlusCLI.NET So, to summarize the Screw up with VS2005 is this: NO PATCHES! Microsoft: We need faster fixes to development tools so we can respond to our customers in a more predictable fashion. Christopher
Wednesday, July 19, 2006 3:28 AM by Ravikiran Malladi

# re: Has Microsoft screwed up with VS 2005, or is it a perception problem? Some thoughts.

do a clean project and rebuild, instead of build.