Comments on Craig Bailey on Microsoft: .Net Return on Complexity (ROC)

Craig VS 2005 is here now for sometime now and I a...

2006-10-12T01:23:19.996+10:00

Craig VS 2005 is here now for sometime now and I am sure you should have something to say as followup to your original post now. Please share how do you see to move ahead with .net hype? What kind of strategy you have adopted in your organisation, which I think should be atleast 30 developer company by now? Cheers :-)

Interesting blog. I have a xml dom blog.

2005-10-03T01:30:45.310+10:00

Interesting blog. I have a xml dom blog.

Yabbbbsssss,What a load of hot cocky sh¦t. Now I'v...

2005-07-18T06:55:30.496+10:00

Yabbbbsssss,

What a load of hot cocky sh¦t. Now I've been an "acquaintance" of this Craig chap for a while now and have heard some pretty ludicrous, even outrageous things in the past, but this blog takes the cake. Mr Bailey, you have truly outdone yourself this time.....

Nah, just pulling your leg....good post, no really, I mean it this time.

Here's my take on this issue, call them the Mulky laws of software developement:

First: All things being equal (and that's important to keep in mind here), it is faster to develope a running, deployable application with technology N+1 than it is to develope with technology N (where "N" is the technology at any given point in time and N-1 is the previous version). Craig, think Citect Pages.

So for example, it's faster to do a .Net application than it is to develope an MFC C++ or VB application (which were the previous techologies).

Second: All things are never equal!. The only constant in our industry is change. Customer expectations will continue to rise because they need every edge possible over their competitors. If they currently have X cools features in their product, the next version better have X+5, otherwise their competitors will go X+5 or better, or cheaper and their customers will switch. Customer loyality is a thing of the past.

Complexity: Yep, it's out of control. But what ya gonna do. This is what you can do: You make sure that your team knows what the hell their coding about. Find out how all this new stuff works. Don't believe the marketing hype, don't even believe MSDN (been shafted too many times by the MS documentation it's not funny). You've got to find out yourself and to do that, you've got to have dedicated SW research time. Get a software architect, pay them well, and make them find out how this stuff works and how to use it. I'm not talking about inventing new languages, a better compiler etc etc, just find out how the existing stuff works. Yeah it will cost, but yeah it will pay.

SAP: Don't start me on that.....too late. So far, the only two companies that I've worked in that introduced SAP was a financial and usability disaster. Final costs were in excess of double the original estimates, and took almost twice as long to execute, had much poorer performance, much less functionality and finally, resulted in redundant data entry. If I ever work for another company again that decides to introduce SAP, I'll quit.

Maks: Outsourcing is not the solution, in fact, it could be one of the biggest mistakes a company could make. Why? Domain Knowledge Loss. DKL happens quite simply when a company ABC decides that it can outsource it's development because it will be cheaper for version X and just maybe X+1 of their products. Come version X+2, when foreign currencies and wages have adjusted, the contractors want better deals, or they just move onto other more lucrative and interesting projects, the previous projects were way late due to cultural, language, expectational and time zone differences, company ABC will decide that it's time to either move onto another external company (complete DKL) or start building up it's own internal staff (short term DKL, long term no DKL). This will result in version X+2 also taking way longer (due to knowledge build up) and being more expensive than expected, and you may just find that had you stuck with your original (local) developement team, you would already be up to version X+3, or X+4, where your competitors are right now.

Man of many names.

hi craig ,hmmm base on my observation I found out ...

2005-07-09T09:50:19.416+10:00

hi craig ,

hmmm base on my observation I found out that majority of your problem is the word cost. Since in your article you have mention it a lot of times. The answer to your problem is simple outsource your way to the third world. I can guarantee you that an equal talent your looking for where your paying $80K I bet in a third world you can pay them for as low as $170 a month.

Think about it. Outsourcing is the future.

My thanks to all who commented. Your thoughts are ...

2005-07-08T22:26:52.863+10:00

My thanks to all who commented. Your thoughts are much appreciated. The main issue facing us software managers is how to ensure our businesses not only survive, but grow in the light of these concerns. My aim in voicing them is to generate thought about how we can best respond to them. And for those who may have missed it Andy's post is very interesting reading:
http://weblogs.foxite.com/andykramek/archive/2005/07/06/719.aspx

Hi Craig,You are right, its high time to decide on...

2005-07-08T01:58:18.420+10:00

Hi Craig,
You are right, its high time to decide on which direction we are heading. Just reminds me of a joke

Joke: Potential & Reality
A kid comes home from school with a writing assignment. He asks his father for help. "Dad, can you tell me the difference between potential and reality?"

His father looks up, thoughtfully, and then says, "I'll demonstrate. Go ask your mother if she would sleep with Robert Redford for a million dollars. Then go ask your sister if she would sleep with Brad Pitt for a million dollars. Then come back and tell me what you've learned."

The kid is puzzled, but decides to ask his mother. "Mom, if someone gave you a million dollars, would you sleep with Robert Redford?"

"Don't tell your father, but yes, I would."

He then goes to his sister's room. "Sis, if someone gave you a million dollars, would you sleep with Brad Pitt?"

She replies, "Omigod! Definitely!"

The kid goes back to his father. "Dad, I think I've figured it out. Potentially, we are sitting on two million bucks, but in reality, we are living with two sluts."

Well does it say everything... :)

Aashish

But what can we do? Customers have eagerly lapped ...

2005-07-07T15:36:54.093+10:00

But what can we do? Customers have eagerly lapped up the kool-aid. Plus we have ex-gurus in most tools going round drumming up business for their conversion services by claiming to be honest advisors who "love" existing technology but are smart enough to understand the need to upgrade. Who can compete with that? I don't think it is a P/E ratio issue, I think there is a stampede of lemmings that we must knowingly join if we want to feed the kids. Madness. -- Sanjeev

I agree. MS beleives complexity equals simplicity...

2005-07-07T01:11:12.723+10:00

I agree. MS beleives complexity equals simplicity.

Hey Craig, I agree with your general thesis and yo...

2005-07-07T00:12:24.486+10:00

Hey Craig, I agree with your general thesis and your posting has inspired me to create my own Blog entry on this topic - so I will add a link to this one to my own when I post it. Luck!

Hi CraigI couldn't agree with you more! As the MD ...

2005-07-04T18:44:02.220+10:00

Hi Craig
I couldn't agree with you more! As the MD of a SME programming company, we are moe concerned with producing solutions for our customers than the sexiness of the tools. We confidently use solid RAD products such as Visual FoxPro. I am unsure of the real business benefits of the .NET hype in comparision. If the hype is meant only to promote Microsoft and lock in young programmers,then surely this hype is harmful to the Industry in the long run.

Hi Craig,great thoughts, and they are 100% on targ...

2005-07-03T19:44:34.473+10:00

Hi Craig,

great thoughts, and they are 100% on target. I'm also running a Programming shop with about 20 Visual FoxPro dedicated collegues here in Germany. And everytime a DotNet coder shows up with some "very cool stuff" which he did with "just a bunch of 50 lines of code"; I wonder myself what the heck is wrong with VFP, where I do the same with 5 lines of code... Granted, there are areas where DotNet is better, but unfortunately, for 70% of software which is regularily built for the day to day use, DotNet and its requirements are just overkill...