Gotta chime in...
Massimo Valle wrote:
The main problem, as usual, is the fact that every new release fixes
some bugs and introduce new ones. Sometimes, I suspect, fixing a bug
expose the new bugs, sometimes the new bugs are related to new features
and some other times the new features break the existing ones.
This is a cat biting his tail.
But this is always a problem... with everything. Every new iteration of
a "thing" is better and worse than the previous for various people.
I currently have 4 releases of REALbasic on my hard disk and I have to
choose for every project which of them is better for that particular
project. Perhaps I should say which of them is less worst... Sadly, I
often end to use RB 5.5.5 which, IMHO, is the most stable and productive
release ever.
I'm having the opposite experience. I went back and opened a project in
5.5.5 recently... and it annoyed me. Sure, I liked it when I was using
it some 2 years ago now. But I missed the new features, the language
improvements... and the IDE of 2k5/2k6.
Sure, there have been a few of the 2k5/2k6 releases that I haven't used
much due to various bugs (such as 2k6R3 due to the Linux cpu usage bug)
but R4 runs each and every one of my projects without fail. No crashes.
No speed problems.
The fact RS respected the rapid release schedule it's a matter of
marketing, too. They simply release *what they have* at the predefined
date (more or less).
If by "more or less" you mean "they took into consideration what they
felt were reasonable goals for that particular release schedule and
planned their work accordingly and, where necessary, moved some planned
features to other releases"... then yes. I agree.
Which is exactly what every good developer does.
The rapid release model could be a good choice, starting from a solid
base, but not with RB 2006. At least RS should stop to add new features
and focus the efforts in fixing bug. I'd really appreciate a release
without [New] tags in the Release Notes, and I'm sure many other people
too.
It seems to me that RB2k6 is a more "solid" base to build from than the
5.x series. It is made with RB itself, has a newer compiler. Plus...
it's what we have now. So that's what is built upon.
Another hot topic, at least for me, is the fact we are not only
customers of RS, but full time beta-testers too. Testing and bug
reporting should be something volunteer (and it is) but the actual
situation turned to be "if you want a bug fixed, then file a bug
report". Ok, I understand that if the guys at RS are not informed of an
existing bug, they can't fix it. But why we (users) find bugs and RS
doesn't?
Well. There's a ton of beta testers and users. Beta testers and users
will always find bugs the in house developers/testers will not. It is
this way in every single project. Always. For ever and ever.
Too often, reported bugs are indeed, stupid bugs. Everyone
which extensively use RB can recognize them. But the fact is, sadly, RS
doesn't take seriously bug reports. Just for example I filed a bug
report regarding the fact that when something goes wrong in the 2006r4
IDE and show the "An error occurred...blah blah" window, it also include
the exceptions stack. This often cause the dialog box to go off-screen
and prevent me to press the "Continue" button. The RS evaluation for
this bug was "please provide a sample project".
This makes me think if they read the bug reports or not... How is
supposed I can provide a sample project that programatically crash the IDE?
Were you working with a completely blank project?
So we are now some sort of venture capital for RS.
We do buy update plans to have the ability to partecipate to beta
testing and support the REALbasic development. Hoping to see, a day, the
result of this effort.
Geez. I really don't see it that way. I've been making my living by
(primarily) developing software with RB for a few years now. Some of
the work I've done would simply not have been possible with other tools
(certainly not in a reasonable/profitable time-frame). RB's price has
been very worth while for me.
Sure, there are new features I want. And, you bet... I've got a handful
of bugs that really annoy the snot out of me. But I have the same
troubles with the other tools that I use from time to time (VisualStudio
2k5 and MonoDevelop). RBs' benefits outweigh its problems for my work.
And that's the key with any toolset.
The RAD myth is gone together with other nice words like "productivity",
"ease of use", "really working"... and my money for the next update plan.
Why do I get the feeling like you're upset? :)
Seriously man, I understand your frustration. I just don't have the
same problem here.
-Bryan
_______________________________________________
Unsubscribe or switch delivery mode:
<http://www.realsoftware.com/support/listmanager/>
Search the archives of this list here:
<http://support.realsoftware.com/listarchives/lists.html>
