Did Remi get it even further wrong?

I don't know why Kevin is going on about Tech-Ed Barcelona except to boil VFP developers' blood but his post referring back to Tim Anderson's post about new projects illustrates some fairly interesting points.

Remi Caron's point about FoxPro forms never looking quite right on Vista is the equivalent of saying a VB 6 form won't look quite right on Vista but then, my thought on the whole matter is that a Javascript application also won't look right on Vista.

And my own attempts at building desktop applications in Visual Studio all suffer from the same "god, it looks awful" approach as well until I go through and design around various UI guidelines. If you want to design for the new interface, you'll want to use the new tools. So do you use WinForms or WPF? The argument goes around and around. When MS comes out with Vienna or whatever is next and introduces a whole new round of UX (user experience) changes, their toolset will have to be revamped again too.

Maybe the real rule of thumb should be : don't expect any application tool to give you a Microsoft designed user interface (a la Office) without some severe tweaking.

At least, Tim was bang on with "so I am not sure that the arguments were fully persuasive. FoxPro applications will trundle on for a while yet. It is also interesting that C# 3.0 is only now getting integrated database query, which has been a feature of xBase from its earliest days"

His last comment about "FoxPro coming to the end of its life as a product under active development" has to be taken in context. "as a product under active Microsoft development" would likely be more accurate.

But wait, there's something else to be said here. For years now, Microsoft has promoted n-tier design as the ideal way of building applications. There was DNA, COM, COM+, Web Services, and on it goes. And as Toni Feltman pointed out in her Infopath article in Advisor and at Southwest, if you have built your FoxPro application with "redesign" in mind, you are also not likely to be using FoxPro as your user interface (experience) tool - but as your business layer and possibly your data layer.

If the future of user interface design is in tools like Sparkle and WPF (which is XML based), then quite honestly, I would want the best string manipulation tool around to really create awesome interfaces. And Ken Levy showed how you can use VFP to do this back as early as 2005 Devcon.

This isn't to say that if someone said "I want to build a next generation product", the first choice would be Visual FoxPro but it also wouldn't necessarily be Visual Studio, which is what gets touted at these conferences. Remember - Vista isn't built exclusively on DotNet nor is Office. The first response should be "to do what?" and then you take it from there. In many cases, developers may recommend other tools to get the job done.

For years, the FoxPro community used to complain that Microsoft was not doing enough to promote it from within. Ken, Yag and others tried their best - but as this session shows, the issue may not simply have been within Microsoft (although they are always the easiest target to point to). It's the spreading of misinformation from whoever decides to spread it.

It's hard to really say if Remi's intention was to spread mis-information. It's hard because Tim's comments may have been from his own realizations or they may have been from what Remi was saying.

And if you read the comments on Tim's post, they all bear this out. From Snorkel's comment
"Win32 applications are going to be viable for a long long time no matter what MS says. Hell, .net is just a layer on top of win32."
to Dave Crozier's note "Please don’t knock the product without being aware of its capabilities and knowing what you are talking about."

Don't shoot the messenger, in this case, Tim - and I'm sure Remi has some great skills in building database desktop applications, much gleaned from his experience in FoxPro back from 2000 - but today's Visual FoxPro is much different than that from 2-3 years ago.

Comments

Remi GJ Caron said…
Well well either my English is really bad or i used the wrong words. It might also be that having a different look at things is not appreciated. Without getting to defensive I shared my view on the matter and will repeat that over and over again because it is what I believe. There was more said in the session I simpy reply to the items “touched” in this post here.

VFP =! Bad
I never said VFP was bad and never the tool of choice in any given scenario. I simply said that it is not very likely to be the first choice like it was for many years (also in my shop). I’ve been doing FoxPro since foxbase and still use the latest VFP 9 in cases where it has its use.

UI stuff
I said VFP apps will run on Vista, but it is highly unlikely that you will be able to use a sparkle created UI with VFP. In other words I do not see the form designer being replaced or adapt those capabilities. The rendering engine would have to change dramatically and I (personally) do not see MS do that. At best I can see a UI use a VFP written business layer object.

Business objects
If there is a VFP one and it works (Com+) use it! Don’t fix if it isn’t broken many times though people didn’t write their code (with change in mind) that way. And when you do need to separate BL from ui from scratch VFP is one of your options. A business layer is usually a set of functions without any UI so if you have to build one why choose a language that can only be invoked with the aid of a wrapper (extra layer) and still brings on the DLL hell over a more modern one without at least the last two mentioned issues.

VFP end of life
Yes I did say that I and I believe that even a blind man can see that. The market for VFP jobs is drying up staffing new people willing to learn these skills is hard (if doable at all). Due to the marketing effort of MS towards .Net yes. But a realistic look at the product can’t hurt either. Its basic ideas were dreamt up over 2 decades ago and it has been enriched over the years with Com capabilities (not all the way though), Web service support (SOAP), OO (better than VB 6) bUt still not all the way. VFP is a procedural language with OO capabilities with a very fast database engine and there is nothing wrong with that.

For today’s demanding business world that is not enough and there are other tools out there that enable you to get into mobile development and tablet pc applications (to name just two). I never said that you need to abandon VFP altogether I simply said that the adding a extra tool to the belt is probably a very good idea. I also said that there is an alternative to .Net when it comes to adding tools to the belt. But I happened to choose .Net for my own reasons.
Anonymous said…
Let's not forget that VFP has been "at the end of its life" since VB4.0 was the greatest thing since sliced bread.

If there is anything useful to be taken here it is that disenchanted ex-users are not the ones who decide when a product is dead- or when it is alive for that matter.

Popular Posts