Level Extreme platform
Subscription
Corporate profile
Products & Services
Support
Legal
Français
How to answer negative VFP attitude? Help...
Message
 
 
To
24/10/2000 14:02:15
General information
Forum:
Visual FoxPro
Category:
Other
Miscellaneous
Thread ID:
00427554
Message ID:
00433968
Views:
21
>>
I believe this thread would benefit by breaking itself into smaller and non-emotional specific issues.
<<

I hear you on the non-emotional comment. I have never understood why people get so emotionally attached to things like development tools. It is kind of like the emotional attachment one gets to a car. That too I never understood. Cars to me are like Bic lighters. You use them for a while and then toss them away to get a new one.

Some of the comments are interesting like "Why do you insult VFP?" This is an interesting comment given the fact that VFP itself is incapable of feeling anything. I am fond of the phrase "I am not my development tool."

I think most of it is more borne out of fear. The sad reality is that many in the VFP devleopment community have become very "institutionalized" in the way they think of data. Data always has to be expressed in terms of VFP's native format. This is not to say that VFP's native format is necessarily a bad thing. Heck, I use VFP cursors borne from SPT all of the time. VFP Cursors are an integral part of DataClas 2000.

Again, the native mechanism that VFP has for working with data is not a bad thing. However, when it is the only thing you know, the only way you are capable of viewing and interacting with data, that is where the cause of concern is.

The message I have is not for those people that are happy doing the small/local data/1-2 tier applications. Quite frankly, and I don't say this for the purposes of being nasty, but the reality is, I don't care about those applications. For that matter, MS does not care either. It simply is not worth the investment for me. For one thing, VFP, both the langauage and its data model are the perfect tools for those type of applications. I cannot improve on that. For those folks, just ignore what I say as it both does not apply to you, in all liklihood, what I say will royally piss you off.

Rather, my message is for those folks you have a mandate to dive into the C/S arena, to build apps that have to scale now or at sometime in the future. The irony of course is that most, if not all apps have to scale to one degree or another in the future. Example, when was the last time you wrote a single-user application? Even if it was to be used by only one person, you still wrote it with multiple-users in mind; becuase you knew that was going to be the next requirement. Further, significant extra work was not required to program a multi-user application.

Getting back to who the message for - folks that are diving into the C/S arena where large-scale applications are involved in terms of data-size, number of users, and or reach is involved.

The single biggest disservice that MS did for the VFP community was Remote View. Not because of the Remote View per se. But because they never finished the damn thing. The design is incomplete. One example, Why can't I modify the SQL property? Why can't I perform sub-selects? Why isn't the view smart enough to ignore schema changes that now not only break the view, but prevent you from opening the view in the designer. And of course, why don't RV's have the ability to use SP's?

IMO, individuals tasked with large scale C/S development and who use VFP with RV's are at a big disadvantage for two reasons. First, the tool set simply is not optimized for this sort of development. We have seen the arguments with regard to n-tier development, SP's, scaleability. Heck, Mike Feltman himself will tell you and did tell you on the wiki that RV's are the least scaleable alternative out there. This was/is not a secret - and it has not been for quite some time. Tell me, if somebody where pitching you on buying a tool, and they told you it was the least scaleable thing out there, would you buy it?

The second disadvantage for the VFP developer lies in what the rest of the world is doing. From a career standpoint, you need to see what else is going on. How portable are your skills. Given the fact that the concept of RV's only exist in VFP, I would say that those skills are not very portable. Now, one might say, "Yea, but in the end, it is all about SQL, and I know that, so I will be OK." The SQL language is only 1 part of it.

MS believe it or not, completed the design of RV's. The technology however is called OLE-DB and is manifested to you and I in the form of ADO. Working with data objects is very different than working with VFP cursors. And, the concept of SP's can also be a mind-bender as well. MS pulled the rug from under VFP developers and took everything, with the exception of VFP in another direction. The world, being a bunch of lemmings, has followed suit. The question is what does the VFP developer do now?

The deal is, which sandbox do you want to play in today. The bigger and more important question is what sandbox do you want to be prepared to play in tomorrow.

For many, it is not a matter of having to re-learn something. IMO, RV's are not a skill you can build upon. Rather, it will be a matter of un-learning and then learning skills that not only will allow you to use VFP in a scaleable C/S environment, you will be able to use other tools as well...


< JVP >
Previous
Next
Reply
Map
View

Click here to load this message in the networking platform