Viktor Szakáts escribió:
Pritpal Bedi escribió:
Hi ...
This is the reason I floated the idea of extending Xbase++ framework
which has not been accepted by the group by now.
Well on this I disagree with Viktor. I dont need backward compatibility with 
xbase++, so extending claes and/or creating new ones is ok with me.

I don't know where the misconception comes from about my POV on this, but I'd like to stop it now.


Hi Viktor !

Well first of all my english is not very good so I could be misunderstood. I've said I was in disagree with you when refering backward compatibility among xbase++ and harbour. For me xbase++ is more like a "philosofy of work" than a rigid set of rules. If compatibiliy risks the viability of the qt GUI we need, then I think we can and we should break it. I insist: It's only my personaly POV.

This is exactly what I wrote:
---
"I don't know, but please document it, we have a long history in Harbour to document extensions, so we should IMO do this also for hbxbp. We don't have any interest in locking in users into Harbour + hbxbp, IOW we should allow them to move to Xbase++, or create apps which work on both platforms.

The easiest and recommended way to self-document it is by using a simple macro, f.e. HBXBP_EXTENSION. Which is #defined by default."
---

Agreed !

Which seems quite clear, but to clarify further, it means that while I can't decide whether we _should_ extend (I trust Pritpal and actual Xbase++/hbqt users on this), but I know we _may_ extend it, if we stick to a few rules: we extend it wisely and document all extensions in code. These rules should be similar to what we've been using when extending Clipper, maybe not so strict.

IMO it's very important to do extensions along a documented procedure, otherwise there is no guarantee whatsoever that Xbase++ compatibility can be kept, or even tested. Which means we cut off Harbour + hbxbp as an upgrade path for Xbase++ users and make the transition a one way step, plus we can't guarantee any sort of parallelism through the transition process (I mean app can't be build for both platforms at the same time).

Also agreed !

If make Xbase++ users life more difficult to move to Harbour, it means Harbour userbase can't grow the way it otherwise would. Which isn't very good.


I don´t think hords of xbase users will come here since they are so Windows headed. They like things like OLE, skins, owner drawn parts, GUI fancy objects, all kind of non multiplatform M$ toys. Also the false feeling of support you get when you pay for something. Anyway the whole idea and consistency of the xbase++ GUI framework is a good model to follow. It's yet to prove if it's easy to integtrate tools like qtcreator and qt development style in this model or we'll have to invent a whole new development philosofy derived from the zillion c++ examples they have on their page.

[ about a few technical aspects of extending wisely and in documented way, pls see my recent msg. ]

Yes I've read it and I trust your judjement and the vision of quality you have imprinted into Harbour project. This post was only a very personal POV ( I insist ) and has no intention to harm anyone.

My best regards and continue with your superb managing of the project.

Angel


_______________________________________________
Harbour mailing list (attachment size limit: 40KB)
[email protected]
http://lists.harbour-project.org/mailman/listinfo/harbour

Reply via email to