OOP, you got it all wrong...

[Chris Boss]

Michael;

I downloaded your IDE and you have a long way to go as far
as I can tell. I applaud your efforts though.

I noticed one strange bug in your IDE. When I ran it one time it
worked OK. After I exited the program and then tried to run it
again, it displayed, but did not respond to the mouse or keyboard.

I had to CTRL-ALT-DEL to get the Rask manager and then terminate
the app that way. I am using WIN95 OSR2, if that helps. Maybe it
is a WIN95 thing !

Now, back to your IDE.

There are two major tasks you working on with your concept of the
IDE :

(1) The Code Editor with a built in Parser which will translate
a higher level language (your OOP commands) into native PB code.

(2) A Visual Designer.

Both are major tasks !

I suggest working on the Code Editor first to get it up to speed
and be able to write code manually (like you can with DDT). Get
that part working first and then add Visual Design second.

If the Code Editor works correctly and its parser works as intended
then you have a sound foundation to work on.

Adding Visual Design is a real "big step" and you want to make sure
your code generation and parser are "very solid" first before
moving on.



------------------

[Mike Joseph]

Michael,
I dont have the wrong idea. As i said before, i think what
you're doing is fine - specifically with regards to those
aspects of the IDE which will ease development of GUI
components by generating API & PB code for the user.

With regards to the OO aids which will allow users to write in
a simulated OO environement, thats fine, I'm just hoping that
PowerBASIC version X or ObjectBASIC version 1.0 wont need it.
As i said originally, this is not an acceptable method of
"implementing" OOP. Please take care to notice that my reasons
for stating this here in this thread are derived from Steve's
response that "...PowerBASIC staff kep(sic) a good eye on what
happens in the forum." This statement from my perspective, was
in the context that the folks at PB may not need to implement
OO in any future version of PB or that your method for implementing
it might be a good suggestion to the folks at PB on how to
implement it. And i find both paths unacceptable.

To answer your very last question, there's nothing wrong with it.
I just hope that a future version of PowerBASIC will be OO and
therefore wont require any external utilities simulate this
functionality in the language. Whats wrong with that?

Finally, if i could respond to your assertion that PB would
turn into VB and that the price would increase, etc. I must
say that such an assertion is a fix looking for a problem.
Ive said this a billion times over and i'll say it again
- Who's to say that the we cant have seperate PB's and ObjectBASICS?
- Who's to say that PowerBASIC cant sell multiple versions
of PB... a Standard version, a Pro version which includes
an IDE and costs more, a Enterprise version which includes
other stuff.
- On the issue of price, you are speculating. If i
were to join in on the speculation, I would say that 90% of
the users on this board would pay Visual Basic Pro prices
for a version of PB that was OO, had a IDE similar to
VB's or Delphi's. Furthermore, I would speculate that PB's
sales and popularity would increase dramatically as a result.
But im just speculating.
- Why do you assume at all that PowerBASIC could ever become
a VB anyway? Isnt it obvious enough already that the
staff at PB would never let that happen? People on this board
who make this or similar claims are just not worth arguing with.
They are just doomsayers and more often then not, they are
just using the concept as cannon fodder rather than a
realistic basis for debate.
(And I dont mean you are one of these people Michael.) Still,
i urge you to define what it means to "become VB" and then
to ask yourself if you think that any of those defining
attributes would ever be allowed to happen with PB? Or if
solutions couldnt be found to deal with such problems. In
effect, are you a doomsayer or not? Im surprised at the
amount of "cant do" attitudes which exist on this programming
related BBS.
- The over-riding sense i get from certain people on this board
is that PB should be kept from going mainstream because
mainstream = bad. Mainstream = bloat.

Im going to keep the faith (at least for the near future) that
PB has their eyes on a bigger prize and that users of PB will
start to find more references to the language on the web and that
PB as a viable language for internal corporate development will
become more recognized.

And maybe you are privledged to more information that the rest of
us, but previous posts from Lance indicated that PB was working on
an IDE. Arent you worried that your IDE will be obsolete
before its even finished? Same thing with the OO stuff. I've read
several older posts which indicate that ultimately, PowerBASIC
wants to have a rich IDE and a compiler that supports true
OO constructs. Do you know that the next version wont already
have OO support? Surely you must or you wouldnt be risking so
much time and energy by embarking on what will surely be a long
and difficult development period.

Oh great, i see that ive been rambling again...
-Mike

------------------

[Michael Purkhardt]

To: Chris Boss

Chris, I ran it on my 95, 98 and 2000 without any problems!
But again, I didn't create 30 forms with all the controls on
each one. If that's what you did as a test, let me know so that
I can correct the problem. Thanks

As for your suggestion, I agree. Perhaps a OO translator first
would be the way to go. But it'll have to be different than the
one of the IDE because the IDE would 'hide' the codes needed to
create forms and controls. But again, we could add new OO code
to it, say...

Form1 = Form.Create
Form1.Caption = "Test form"
Button1 = Form1.Control.Create(Button)
Button1.Caption = "Cancel"
Form1.Show

...and the IDE would use the above when generating the code before
the OO tranlator use PB compiler to make the EXE.

It is a major task, but will be worth it, I'm sure.

-Michael-

[Michael Purkhardt]

To: Mike Joseph

Mike, perhaps I didn't understand your point of view, but you are
right about PB making an IDE, why not. But I disagree with you on
a third party IDE. Not because the IDE comes from PB that means
it's the best. It's better having PB work on a more solid compiler
than to give up optimization for an IDE. But again, if they would
have an army of programmers, they can do it all. Sure, the normal
PBDLL and a PBIDE from PB Inc would be fine. But do they have the
time? The man power? When was last time we've seen an update? Once
a year right? I would hate to see PBDLL v7.0 to be nothing more
than PBDLL v6.0 with an IDE. I would rather see v7.0 have more
functions, subs, better tricks on the compiler, perhaps more API
calls in a PB format, not so primitive, you name it. Just like
the implementation of TCP/UDP and DDT. PB could then work with
third parties just like Microsoft, Adobe and the other big guys do.
Saves time and money for all, and the more heads you have, the
better it is. If PB becomes "monopoly" on their products, they
won't go far. They have to work with third parties. Why? Marketing
is one thing. Look, Photoshop for "WINDOWS", Corel for "WINDOWS"
PBSedit for "POWERBASIC", EZGUI for "POWERBASIC", PBSEdit for
"POWERBASIC". Do you get the picture now?
PB can NOT be alone and do it all. They'll go nowhere.
They need others with web sites selling product for PB. Then the
world will start asking, "what is PB?", "Wow, never thought it was
better than VB", "Hey, did you get PB?", "I've seen this...for PB",
"Damn, so many things available for PB", and bla bla bla...

-Michael-

[This message has been edited by Michael Purkhardt (edited October 02, 2000).]

[Steve Hutchesson]

Michael,

I think the response shows that you are on a winning idea, over an extended
period many programmers who support this forum have indicated that they would
like an easier environment to write their PB code in yet very few are willing
to make the sacrifice involved in terms of compiled output size and speed by
using some of the other products on the market.

PowerBASIC have made it clear for a long time that they support after market
vendors who extend the PowerBASIC product and that they actively take notice
of what their customers require from whatever source the information comes
from, direct contact, forum discussion, phone or whatever else.

Your approach is a particularly good one in that it addresses both performance
and ease of use without compromising either. Empowering basic programmers is
a lot of what appeals to PowerBASIC users and the willingness of the staff to
pick up demand and ideas is part of the reason why the product has continued
to get better.

I think I add my voice to many in encouraging you to forge ahead with this
development and don't be put off by some of the suggestions, I seriously doubt
that PowerBASIC will ever compromise their compiler power to follow the same
path that other environments have taken.

Regards,

[email protected]

------------------

[Michael Purkhardt]

Steve,

I think your point makes perfect sense.

Thanks
Michael

------------------

[Mike Joseph]

Steve, your last sentence sums up much of what i said in my last
post. But I noticed that you too seem to be indicating that you
know the next version of PB will not have OO support. I could
not otherwise see how you could in good conscious encourage
someone to head down a path (particularly Michael considering
the recent loss of his business) developing an app whos utility
to others may be diminished with the next major release of PB.

And Steve, I think your attitude is lousy. You want a PowerBASIC
compiler that is fast, produces compact code, which is procedureal,
has an inline assembler, and which supports static libraries in
the future. Thats fine. But many many other people
want _some_ version of PB that is truely OO. You wont even
concede that an OO version of PB doesnt necessarily have to
adversely impact any of these existing attributes. You have
also never said you would support seperate versions (one OO and
one for procedural) so I can only assume you are against that
idea as well. Why?

If it just boils down to you not wanting PB to divert resources
from their primary compiler, then I'll be glad to hear it. I
happen to think that you're in the minority and that there's
not a big future for PB in the direction you want them to go (or
should i say, the lack of any direction you want them to go). PB
could stop development of PB6 (cept for bug fixes) and as far as
i can tell, you'd be just as content since you already feel
PB is the best compiler in the land. So you and yours are
already happy, its time for PB to broaden their appeal.

Michael, I agree that 3rd party support is necessary for _any_
compiler. Im hardly disputing this. I just think most people
dont think of full blown IDE's when they hear "3rd party."
Either way, I just cant get over the idea that you would risk
developing something that could be made obsolete. I guess both
of you, having been working with PB for alot longer, have a
better idea of what is in store for all of us.

Enuf of this thread for me. These things frustrate me... not
because of the back and forth nature, but because I start to
believe more and more that OO in PB is a pipe dream. I cant help
shake the notion that your (Steve) arguments are pre-emptive
in nature. If it turns out to be true, that will
be the end of one thing and the start of another for me.

-Mike J

p.s. Michael, this is absolutely none of my businesses, so you're
obviously free to ignore the question but, were you insured?

------------------


[This message has been edited by Mike Joseph (edited October 02, 2000).]

[Michael Purkhardt]

To: Mike Joseph

Mike, I do agree with you, but the fact is, PB do not need any
beta testers as of yet. So the next release of PB is a long way
down the road. So I'll take the chance. Also, from talking with
the PB guys (also Bob) they are VERY strong on "not another VB"
when I was mentioning OO to them last year. They argue to the
max and would NOT understand any of suggestions. So again, I'll
take the chance.

I think PB should leave the IDE/OO stuff to third parties, and
when PB is fully ready to have it embedded, they can join venture
or even buy the rights of the IDE/OO of their choice. This way,
all the bugs and ideas would be all done for them, and the major
release of PBOO (hey, not bad of a product name!) would be something
everyone wants.

I did have insurance Mike, but because I had a "trustee" taking
care of business and of my house, they can't do a damn thing
about it and I have to take it to civil court. They sold everything,
and are long gone. So trying to sue someone with no money is a
waste of time. All I have left are the source code of all my apps
and the design of my web site. That's it, nothing else. No clothes,
no furnitures, no books, no documents, no computers, nothing. They
totally emptied out my house and my office. And I mean, totally.
They knew I had death in the family and that I would be gone for
about a month, so they took advantage of it. My "trustee" was a
friend for 10 years. Go figure this out, what money can do!

-Michael-

------------------

[Gregery D Engle]

Micheal,

I've been reading the discussions on this thread and I like your idea's very much. I can't wait for something to try out

I could never imagine what your going through right now with everything being stolen. Let alone a friend stealing it. I hope you get through this ok.



------------------
-Greg

[Steve Hutchesson]

Mike,

I will try and background why I hold a range of views about the direction
of PowerBASIC. In late 97 I discovered PowerBASIC version 5 after brain-
picking my local ISP who also happens to be an agent for PowerBASIC in
Australia. I have written in basic for a long time as my ISP has so it did
not take that big a "hard sell" for me to buy my first copy of PowerBASIC.

PowerBASIC was a much smaller company then but it has continued to expand
on the basis of its compiler performance and that has been its hallmark,
small fast binary code in a viable dialect of basic that addressed a big
hole left by other vendors. When version 6 was introduced, it added support
for TCP/IP, DDT technology and across a broad range of functions, it got
faster through intensive internal code optimisation.

When you have worked in detail in a particular language for a couple of
years, you start to get a reasonable idea of how it hangs together and the
broad philosophy behind it, "war on bloatware" is a slogan from PowerBASIC
I agree with, thats one of the reasons I write in it, it has not been
compromised in performance terms and on the basis of past performance, I
doubt that it will in the future.

I have also been at MASM for over 10 years and it started from the need
to add more grunt to both Microsoft C version 6 and Quick Basic, C6 because
of its lack of facilities, Quick Basic because of its toothless nature. In
PowerBASIC there already is a very good inline assembler so if I need a
capacity that is not available in standard basic and it needs to be faster
than I can construct in general purpose capacity, I can do it with no real
problems.

In the development direction, I have plugged for the things that will give
the language more power again, macros and libraries as this general purpose
low level capacity makes many more things possible including an efficient
OOP capacity for PowerBASIC programers who wan to write code of this type.

I would not hold your breath waiting for PowerBASIC to abandon its compilers
at the current level and try and follow the OOP(s) implementations that are
currently on the market. I suggest that the action is in providing the
capacity for programmers to produce their own implementations.

Lets look at the general problems of current implementations, oversized
binaries, diminished performance, ugly opaque syntax with no access behind
it, unfixable objects, memory size blowouts, code duplication within
different objects in the same binary file. Next comes the visual garbage
generator on the front end and a gaggle of DLLs to support this junk.

If I was the only one who saw it, I could be written off as a dinasaur but
I suggest that if you cannot see the problems, you must be one of the few
people left who has not. I support the programmers who want to be able to
implement OOP instead of OOP(s) with additional performance in PowerBASIC.

I have also made the comment before that PowerBASIC are the best informed
when it comes to the interests of the great silent majority, I simply
don't accept that you are the spokesman for people who do not actively
post views in the forum.

This has been a good forum for the couple of years that I have supported
it and it has been based on the capacity of programmers from many places
being able to toss around ideas, help other members, post bits of code and
spread their particular expertise for others to use if its useful. The
range of expertise floating around here is enormous and I think it is one
of the good things about the forum.

Tapping into that experience is something I will continue to do, with or
without your approval as the benefits outweigh the occasional bit of flack
for doing so. If you think that the implementation of OOP in Microsoft VC6
is the right one for you, you are free to write in that language but if
you attempt to inflict that style of programming om PowerBASIC
programmers, you will probably run into a number of people who see it
differently.

I characterise the evangelistic approach to spreading the current
implementation of OOP(s) as someone saying, "The rest of the world has
cholera, you should catch it too".

Regards,

[email protected]

------------------

[Michael Purkhardt]

Thanks Greg. I have no choice but to get over it. Either that
of suicide! It's extremely hard loosing 14 years of hard work
and accomplishment.

-Michael-

[Michael Purkhardt]

Steve,

I'm not sure what your point is this time! But I do not think
that PB will abandon their compilers at the current level. PB
comes a long way, and will go a long way. As long as they don't
do like VB and Delphi. Think of it this way. The engine of a car.
That's all we care about. Horse Power. The rest of the car, body
looks, ABS etc. are all additions to the great engine. If the
maker of the engine start concentrating on body looks, the
engine will never change, or for that matter, slowly. And that's
what I'm afraid would happen to PB if they start other things.
Like PowerGen for example, was a useless project now that we have
DDT built into PB.

My idea of IDE/OOP for PB is not what anyone think it is. It's
ONLY a translator to PB code, nothing more. No DLLs required.
It would be the same as having the experience programmer to code
APIs in his PB code to handle forms and controls, except that those
API routines will be written for you, by the IDE/OOPS engine.

One thing I would REALLY like PB to do next on their compiler is...
Not to compile a whole INC/BAS file with the EXE. Extract ONLY the
functions, subs, var, type... needed. Forget the rest. Delphi
does that already. This way, we can write many functions into
INC files, as libraries, without worries that all 500 functions
will be compiled into the EXE. That's another thing my IDE/OOPS
would do, IF pb doesn't want to do it. I've told them this a few
years ago, with no response.

Another thing I've ask PB is to update their editor. Right now it
****s. We can't always remember functions params, nor the syntax
of vars or types, nor capitalization. VB and Delphi has a simple
"Ctrl-Space" key to show you a list of what's available. For example,
I'd like the editor to show me the params of the function I'm
typing. Also, a pulldown list of the vars, subs, funcs, etc. of
what I'm typing. Send"Ctrl-Space" would come up with a list of
things starting with 'Send' as in SendMessage. Auto capitalization
of words, like if I have a function called IntToStr, I'd like
the editor to write it that way, not as inttostr. It saved so
much time on typing. Beside, if it doesn't capitalize like the
one you've declared, it must be typed wrong. Then you can use the
Crtl-Space key to find out the proper syntax.

Another thing, make the editor 'smart' with functions, subs, declares.
Like delphi, I hate to write code in one long list, having to
scroll to subs and functions, Ctrl-Tab to other windows to find
syntax etc. Right now, the editor only has a dropbox of your
subs and functions. That's it. Way too primitive.

The next version of PB should have a powerfull editor, capable
of helping the programmer with params and syntax. Quick jumps to
other subs/func and things like that. Separate the subs/func from
the rest of the code (as an option) like VB does.

In my opinion, you don't really need a buch of BAS files to make
a big app. The editor should handle all that for you. It should
classify all that. Let me start another thread for suggestion...

-Michael-

[Steve Hutchesson]

Michael Purkhardt,

I was actually responding to Mike Joseph's posting as we have a difference
in our views about listening to the wishes and ideas of PowerBASIC
programmers.

I like your basic ideas and I think you are on a winner with the direction
you are developing. It was just a case of the wrong Mike.

Regards,

[email protected]

------------------

[Michael Purkhardt]

Oh OK Steve. That's why I put "To:" in my replies.

Thanks
Michael

[Lance Edmonds]

Originally posted by Michael Purkhardt:
...but the fact is, PB do not need any
beta testers as of yet. So the next release of PB is a long way
down the road. <...OOP discussions snipped...>

If the following seems like a flame, it is not intentional... However, not responding would be out of character for me. OK, here we go...

Michael, just because PowerBASIC has not publically asked for applicants for our beta teams (in recent times) does not mean we don't already have teams. PowerBASIC has been known to approach people directly too, not just make public offerings. IOW, you cannot possibly know what we are doing today unless you are _personally_ involved.

On this basis, please refrain from making speculations and personal opinions sound like they are based on real facts. Thank you.

Therefore, if/when/how OOP is implemented will remain to be seen...

As we have noted regularly in these forums, we simply cannot divulge what we may or may not have in the pipeline, planned, in development, on test, etc...

Also noteworthy is that the number of PowerBASIC customers that frequent this BBS (and only a modest number of you actually contribute publically) represent only a tiny proportion of our customer base... The volume of opinions that are posted to this BBS account for only a portion of the information we receive... IOW, we receive a great many more suggestions and comments (and compliments!) through avenues other than this BBS.

I hope this makes our position a little clearer on some of these matters. Thanks for your time, and I sincerely appologize if I have upset anyone with my response.

{rant mode off!}


------------------
Lance
PowerBASIC Support
mailto:[email protected][email protected]</A>

[Michael Purkhardt]

ok I've just sent it.

-Michael-

[Peter P Stephensen]

Michael --

Sounds interesting...could you be more specific on your OO-concept?

Regards
Peter


------------------

[Michael Purkhardt]

Lance,

Sorry if you took this personal but, I was debating the subject that
PB might already be working on OOP, so why should I waste my time
making one. My answer was "I'll take the chance" based on th fact
that PB has not yet have the need for beta testers. But that does
not mean PB are not working on a OOP. I only stipulate PB are not,
based on the non-demand for beta testers at this time. That was it.

Other than defending the fact as to why I should continue developing
PBSEdit, I was only trying to explain my idea as to what it is and
what it will do.

Lance, you seem to analyze every single words, and make a big deal
out of something you think I've meant. Not everyone has a perfect
vocabulary, and interpretation varies between individuals.

Regretfully,
Michael

[Michael Purkhardt]

Peter,

I've explained the concept in a previous post. You might want
to go back on page 1 and start from there.

Regards,
Michael

------------------

[Peter P Stephensen]

Michael --

I have read it...are you aiming at simpel VB-classe? No real inheritance?

Regards
Peter



------------------