My Turn: Ten Insane Ideas for Microsoft

re: My Turn: Ten Insane Ideas for Microsoft

Darrell — Wed, 27 Jul 2005 15:42:49 GMT

Oh man VS 2002 is VERY painful. Office not so much.

re: My Turn: Ten Insane Ideas for Microsoft

Troy Phillips — Wed, 27 Jul 2005 16:43:20 GMT

This is the most useful of the lists so far.

It is also encouraging to see that Microsofties are aware (and publicly acknowledging that they are aware) of Mini :)

re: My Turn: Ten Insane Ideas for Microsoft

Eric Newton — Wed, 27 Jul 2005 18:44:55 GMT

You make good points. I see Microsoft becoming very agile in the near future with your suggestions and with whats already happened.

With the relatively successful VS Community Collaboration, the product is tight. Biggest benefit: EnC for C#... thats a feature that was directly because of community insistence. I believe a couple of other lukewarm features were dropped because nobody in the community thought they were that grand.

This is a good step because now Microsoft isnt just listening to a select few, but to the developer community that wants to buy good quality software that meets their needs.

re: My Turn: Ten Insane Ideas for Microsoft

jledgard — Wed, 27 Jul 2005 18:58:10 GMT

Ok ok, maybe suggesting you stick with VS 2002 is going a bit too far. I did upgrade all those bits to VS 2003 as soon as we had a beta. :-)

Yeah, there's a lot of us who know about Mini. He's on my blogroll for sure. I don't agree with everything he says and in some ways it sort of like watching a train wreck except more fun.

re: My Turn: Ten Insane Ideas for Microsoft

TAG — Thu, 28 Jul 2005 01:31:29 GMT

Strongly agree on Item 5 !!
MS did a very poor job marketing VS 2005 Betas to REAL customers.
On one side you put ads everythere asking everybody to install betas, on other side you write in EULA/Readme that this software may trash out your Windows installation.
I.e. you promote everybody to trash out their PCs !

So ? What you can do with Item 5 ? Need ideas ?

re: My Turn: Ten Insane Ideas for Microsoft

Norman Diamond — Thu, 28 Jul 2005 05:04:50 GMT

> there are several Microsoft projects I
> can’t even get permission to report bugs to
> their database or view their plans without
> knowing the special handshake

In the real world Microsoft usually allows victims to report bugs if victims pay 4,200 yen in advance. Surely you could do the same. Surely you won't because the idea offends you as much as it does me, but you could.

An exception was eMbedded Visual C++ 3. When the compiler reported an internal error it told me to view a particular Microsoft web page. That page said I could report the bug if I paid US$195 in advance. I didn't report it.

re: My Turn: Ten Insane Ideas for Microsoft

Norman Diamond — Thu, 28 Jul 2005 07:06:42 GMT

Wednesday, July 27, 2005 6:31 PM by TAG

> On one side you put ads everythere asking
> everybody to install betas, on other side
> you write in EULA/Readme that this software
> may trash out your Windows installation.
> I.e. you promote everybody to trash out
> their PCs !

Actually I thought that wasn't a problem. I install betas in virtual machines under Virtual PC 2004.

The July CTP won't install though. I think I've figured out the reason. Recently I installed a checked build of Windows XP SP2 with plans to use it for testing a driver, but in the meantime it was a pristine environment for trying out a new Visual Studio CTP. I think I've figured out that the July CTP depends on Windows Installer 3.1, but contains a version of Windows Installer 3.1 which refuses to install itself onto a checked build, and there's no available version of Windows Installer 3.1 that would agree to install itself onto a checked build. I'd guess the July CTP would be installable onto a retail build.

I'm not in the mood at the moment to create another guest machine just for the purpose of playing with the CTP for a few weeks though. I did exactly that for Beta 1 and Beta 2 though.

Geek Noise 2005-08-01

Geek Noise — Tue, 02 Aug 2005 08:24:28 GMT

re: 6. Merge the dev and test teams on a project

Anonymous Dev — Thu, 11 Aug 2005 18:14:20 GMT

I don't think there's a chance in hell we could successfully merge dev and test. First, our testers have very few technical skills. The half of them who had a bulk title change to SDET because of their level mostly don't meet the criteria (or even come close). My team has two testers (out of ~20) that I would trust with a writable SD enlistment. Less than half even have a RO enlistment after a year of encouragement.

Entrusting that team (some of whom are great testers despite a lack of technical skills) with the responsibility to create our products would be a disaster.

And, finally, this idea ignores the fact that test and dev are different disciplines with different goals. Most testers like testing, not writing code. The good ones have the skills and the degree, but they chose to go into testing rather than development because that's what they want to do! One of my friends was a tester on a team who did exactly this - they retitled all of their testers as SDEs, and he hates it with a passion.

re: My Turn: Ten Insane Ideas for Microsoft

jledgard — Thu, 11 Aug 2005 20:31:27 GMT

AD: I don't think this is a change that could happen overnight or with the current group of employees that most teams have. Something like this would take time to avoid the skill issues.

Regarding disciplines specialties... this is where we might disagree. Show me a dev who can't effectively test code and I'll show you some crappy code.

re: scale the frontline

Dave Cortright — Fri, 12 Aug 2005 09:24:04 GMT

I totally agree about scaling the frontline, but what MSFT needs to do is have frontline support people who are full-time MSFT empolyees with stock options that have a stake in the game. It amazes me that Microsoft fully outsources the only human face the company has to most of its customers.

re: My Turn: Ten Insane Ideas for Microsoft

Anon — Fri, 02 Sep 2005 20:02:32 GMT

"9. Make more bug databases more public"

You really think customers seeing that there are 1000+ bugs is a product is a good thing?