Hi Dennis,

since I retired from psychology, I work hard not to hypothesize about others' motivations, since I'm not being paid to do so. <s> I have to admit if I were in the position many others are, I would likely feel as you describe, but I can't speak for others, and every one is an individual.

As for the win-win: I couldn't agree more. I was told by an MS PM 3 years ago, I think, that it would be at least this (VS2010) or next (VS2012 from what I read) version of the tools before .Net approximated the development and language facilities of VFP. And it was just in the last month that a codeplex project was created with a lightweight data persistence framework that I know I can convert to use our existing metadata (sans the VFP code which would have to be converted), and be able to run against any backend using our existing tools.

My observation, and that of others, about complex systems is that it take 10 years to mature. This applies to complex jobs and skills, but I think applies to complex products as well. .Net is coming up on 10 years, and it's starting to get there. When I read the 1.0 books (I can remember the afternoon, where I was sitting, what kind of day it was, when the first 3 arrived and I pored over them), I knew (with a sinking feeling) that it was going to be a long haul before it matured. From my own experience, it then takes about 5 to 7 years more, beyond initial maturity, to fulfill the capabilities of the system. If that holds, .Net will be "there" in 6 to 8 more years. And when it gets there, there will be dynamic languages that directly access data. I'm willing to bet any and all a pizza on that prediction. <s>

Hank

>Hank,
>
>Perhaps this agressivity could be linked to being forced to jump to another development environment. Not only that but moving to a new platform that causes a lot of deceptions.
>
>Many can't afford to be patient. They have to follow the path traced by major companies that hire them to build the new killer apps.
>
>So in the meantime they have to suffer and adapt as fast as possible to the new paradigm while others can afford to wait and see what will happen with their beloved development environment. Apparently there are still many that can afford to wait and that really get on the nerves of those always forced to adapt to that new cool platform.
>
>So like you I still can afford to wait and hope for the best from enterprises like etec. And if nothing happens with etec then the alternatives will only be better, easier and faster to work with.
>
>I see this as a win-win situation.
>
>
>>Hi Mike,
>>
>>in my partial defense, part of the story you may not know, or perhaps didn't remember. I was hired to first be the tech writer for the RAD book, and then the author of the screen builder book. For whatever reason, the author of the RAD book was unable to proceed, so I was asked to step in, which I did. It was a difficult book -- from scratch I wrote an event-driven, metadata driven framework for 2.5/2.6. That book was late, and by that time, I had no time to write the screen builder book. I have a lot of patience for getting things right, which at times causes others to have great consternation. <s> That's why I spend my time, for the most part, writing stuff that can be used over and over again. At least there's some balance between input and usefulness.
>>
>>And that (being dogged, being interested in the long term) explains why I am doggedly persistent about VFP.Net. The software language world is moving in 2 directions at once: toward dynamic languages (have you heard of IronJS? It will be joining IronPython and IronRuby, and will likely become another first-class .Net citizen); and toward functional languages that can handle parallelism (F# being the best known in the .Net world). Now, if you've been following dynamic languages and functional languages, you are likely to have come across the interesting tidbit that Python, a dynamically typed language, is capable of being a Functional language (http://www.ibm.com/developerworks/library/l-prog.html -- this is not new knowledge). Cool, eh?
>>
>>Well, what does Python have that VFP doesn't have? And the answer is: not much, once you count in Execscript and & and Methods as object (which VFP.Net has, but VFP does not). So, in terms of where software development languages are going, VFP.Net has a chance to hit the sweet spot. That's worth supporting. Perhaps not waiting for: all the infrastructure work I'm doing will equally apply to (my own version of) the Agile Ado.Net Persistence Framework, but worth supporting so long as there is hope. And as my wife will ruefully attest, I am near-to-eternally hopeful. <s>
>>
>>Which brings me around to ORM's, and why they are a mistake. I started looking at them in 1996 (the first well-known one was by a professor from Australia, IIRC), and what I saw was that the abstractions were, well, abstractions, which by definition lose detail. Why would I want to lose detail? Could I afford to lose detail? And so here we are 14 years later, and yesterday I read this, from a DBA guru (http://blogs.msdn.com/buckwoody/archive/2010/02/16/code-that-writes-code-a-good-idea-or-not.aspx), saying that the abstractions don't mesh with what happens at the database level. Ted Neward's (I gather famous) blog post about ORM being the Vietnam of Computer Science (http://blogs.tedneward.com/2006/06/26/The+Vietnam+Of+Computer+Science.aspx) makes the same point (and a lot of others). I don't agree with all his points, but in general, the issue of an impedance mismatch seemed accurate to me in 1996, and still seems accurate. Interestingly, the Entity Framework has been described (in the MSDN introductory article on the EF) as an attempt to alleviate the impedance mismatch between data and objects. Oh well.
>>
>>Now, this bit about the impedance mismatch between ORM and database is pretty obvious, it seems to me. It is the difficulty in writing data apps in a statically-typed language that creates the need for some way to make having data classes in .Net more manageable that it otherwise is (and that, too, is taken right out of the MSDN introductory article). VFP.Net cuts the Gordian Knot, it seems to me, giving the best of both worlds.
>>
>>And when I contrast the problems inherent in ORMs and databases with the ease of working with dynamic metadata (those who have used xCase2VPM know what I mean, but it's hard to convey because it's so far from what developers are used to doing), I shudder at any prospect of being forced down the ORM road. The competitive advantage I provide to those who employ me is based on the ability to program primarily through metadata. Give that up and I'm competing with, well, name some place with $600/mo programmers, not entirely, but to a degree that is uncomfortable, not just because of the economic competition, but even more because I know it can be done better, because I have been doing it for 9 years.
>>
>>The real mystery to me, then, isn't my behavior, which of course I see as incredibly rational <s> -- and seriously, I think I have done my homework. The mystery to me is why other people, whom I don't think I am harming by supporting VFP.Net by my persistence, should be apparently upset, not about me in particular, I understand that, but upset at the persistence of a relatively small group of people who are hanging on for that chance to still do it right, as they see it. One usually sees this kind of negative fervor only when there is a threat, but what's the threat?
>>
>>Anyway, I hope this clears up your mystery. <s> I am naturally dogged, and my loyalty is based on self-interest, in the sense that there is a lot to gain (being in that sweet spot of programming languages as it is evolving), and little to lose (the work I am doing will be used either way).
>>
>>Hank
>>
>>>The mystery to me is why Hank Fay is so doggedly loyal to them. I honestly don't get it.
>>>
>>>Then again, I know Hank has patience. Years ago he was contracted to write a book on the Screen Builder for Hentzenwerke, part of their Pros Talk Fox (I think) series. The manuscript was eons late. Whil finally had enough and told Hank to get lost. He asked me if I wanted to step in, which I didn't want to at the time because I was working on a lucrative project. Steve Sawyer wound up writing the book (and did a better job than I probably would have).
>>>
>>>>Have you read this ? http://groups.google.com/group/vfpnet-compiler-community-support-group/browse_thread/thread/121abf96c598ef01/e9a02cdb2ce49ab8
>>>>
>>>>They were in one of the ut magazines - years ago - when they talked about the vfp compiler.
>>>>It seemed like a waffle to me
>>>>
>>>>
>>>>
>>>>>While I bought their product early, I still find it amusing - especially after reading that thread - that people who don't want to go to .NET because they can't trust Microsoft etc are pinning their hopes on these guys <bg>
>>>>>
>>>>>
>>>>>>>Hi Cecil,
>>>>>>>
>>>>>>>sorry, I haven't been to the UT for a few days. The message was written by someone called JR1. The whole thread is here:
>>>>>>>
>>>>>>>http://groups.google.com/group/vfpnet-compiler-community-support-group/browse_thread/thread/1858112008af647f/f152aa0d54d0f018#f152aa0d54d0f018
>>>>>>>
>>>>>>>(And you responded to him/her!)
>>>>>>>
>>>>>>
>>>>>>Interesting link - to say the least