Board index » delphi » Re: Borland snubbing out the hobbyist programmers with Delphi 2005

2004-10-17 03:04:07 PM
delphi68
"David Clegg" <XXXX@XXXXX.COM>writes
development environment *will* function without outside connection or I will
find another. (and I think I speak for a number of others).
However, we were talking about some conversion help, improvements to the
Delphi Help system (remember, you can not even find MsgBox() clues), popup
help in the editor environment, etc, etc. These sound like things that
could be managed within Diamondback... but somebody has to understand the
issues and do it.
Dan
 

"Dan Barclay" <XXXX@XXXXX.COM>writes:
The four letter word begining with F originally had a
non-{*word*201} agricultural meaning.
Now, I do not even dare to spell the word out in this forum since
the message will probably be deleted :-)
 

"ozbear" <XXXX@XXXXX.COM>writes
That dose *not* mean making Delphi over in VB's image. I have posted earlier
about the complete and absolute lack of clues for VB(Visual Basic) knowledgeable people.
Using Delphi Help, try converting the following code:
Public Sub HelloDelphi()
MsgBox "Hello Delphi"
End Sub
new keywords.
Having said that, new keywords (provided they are created in the spirit of
the core language) are a GoodThing. The *bad* thing is to change existing
keywords or behavior. In fact, that is precisely why you will find VB(Visual Basic) users
interested in Delphi. They will *not* want Delphi to change in ways that
break existing code because that is the precise mindset they're trying to get
away from.
They (me anyway) will be interested in keeping some of the useful features
of VB, but that doesn't mean they need be implemented in the same way.
I'll give you a good example: When I am working on a new I dea I like being
able to code new fragments without having my thoughts distracted by having
to go to the top to declare variables. VB(Visual Basic) handles this by allowing us to
use undeclared variables with explicit data types (then declare them later).
It appears Diamondback would give me this same "feature" by using automatic
declaration insertion... same end result, I can code pretty much without
distraction. It would also be cool if I could define the data type using
some clue in the variable name so the automatic declaration would be
unambiguous. Poof, problem solved with *no* change in Delphi using a
feature that can be made optional to the user.
Now, what if Diamondback would take MyInt% and change it to MyInt (as
integer) and insert the declaration? You wouldn't even notice... VB(Visual Basic) folks
would.
Dan
 

Dan Barclay writes:
companies where the software is their flagship product which continues
to grow. eg if MS would otherwise collapse if Windows was required for
them to live, would someone be brave enough to label it as "legacy"?
--
Dave Nottage [TeamB]
Newsgroup guidelines: www.borland.com/newsgroups/guide.html
 

Dan Barclay writes:
--
Nick Hodges -- TeamB
Lemanix Corporation -- www.lemanix.com
Read my Blog -- www.lemanix.com/nick
 

"Dan Barclay" <XXXX@XXXXX.COM>writes:
hope there's a strategy of some sort in place to take advantage of this.
 

John Kaster (Borland) writes:
www.sdmagazine.com/documents/s=9411/sdm0411a/sdm0411a.html
(you have to sign in with a CMP Media account, but signup is free and
doesn't require a subscription IIRC)
Quoth the article--
"The most common languages or platforms? Oracle or other database, at
67%, Java at 64%, C++ at 63%, Perl or other scripting languages at 53%,
.NET at 50%, C at 48%, J2EE at 40%, JavaBeans at 35%, C# at 34% and
SOAP at 30%. At the opposite end of the spectrum, legacy technologies
Cobol and Delphi have fallen steadily in popularity over the past five
years: Their 2000 standings were 28% and 65%, respectively; they stand
today at 20% and 8%. CORBA/COM/middleware has also dropped from 2000's
36% to 2004's 18%. For the first time, Java 2 Mobile Edition (J2ME)
made the list at 7%."
According to the article Delphi dropped from 65% to 8% (see quote).
Seems like a mighty sharp drop... far larger than even I would have ever
imagined.
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
qc.borland.com/wc/wc.exe/details
 

Dave Nottage [TeamB] writes:
"*Methodology*
The seventh annual salary survey was prepared by the editors of
Software Development. CMP Media's Information Week, along with Hewitt
Associates LLC, a global management consulting firm that regularly
conducts professional compensation and benefits studies, helped
redesign the questionnaire in 2000. San Diego, California-based CIC
Research Inc. collected and tabulated the data.
An e-mail invitation asking 49,000 software developers to fill out a
Web-based survey was sent on July 9, 2004; the data collection period
lasted nine weeks. After removing students, unemployed, consultants and
part-time employees from the pool and cleaning the data, a total of
4,518 records remained."
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
qc.borland.com/wc/wc.exe/details
 

Will DeWitt Jr. writes:
--
Rudy Velthuis [TeamB]
"Believe those who are seeking the truth. Doubt those who find it."
-- Andr?Gide
 

Rudy Velthuis [TeamB] writes:
says they sent out "an e-mail invitation asking 49,000 software
developers to fill out a Web-based survey," but it doesn't mention if
any measures were in place to prevent ballot stuffing, or, where the
49,000 e-mail addresses were acquired (did they use every willing
subscriber to MSDN Magazine-- if so, seeing a Borland product slip that
much would hardly be surprising).
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
qc.borland.com/wc/wc.exe/details
 

Eric Grange writes:
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
qc.borland.com/wc/wc.exe/details
 

Nick Hodges [TeamB] writes:
reserved word because, heaven forbid, somebody might have been using it
for a variable name or function name.
If someone installs a new version of Delphi, compiles their code and
notes compiler errors about reserved words, it should be obvious to
everyone but the most inexperienced that this means they need to change
the variables name. With the new refactoring support in Delphi 2005,
this should be really /really/ easy to deal with, don't you think?
Seems obvious to me.
Will
--
Want a 64-bit Delphi compiler for AMD64 / IA-32e? Vote here--
qc.borland.com/wc/wc.exe/details
 

Will DeWitt Jr. writes:
--
Rudy Velthuis [TeamB]
"Life is like a box of chocolates." -- Forest Gump
 

Will DeWitt Jr. writes:
Borland have *always* tried to keep the proliferation of new keywords to a
minimum, not just to avoid breaking existing code, but to keep the language
reasonable clean. If they had favoured new keywords every time a new feature
was added, the language would be an awful mess by now.
The general rule is if a feature can be added without new keywords and can
be made to fit reasonably in keeping with the language style then that is
favoured. Only on failing such tests are new keywords added. In it's
history, *many* new keywords have had to be added anyway, but to do so
indiscriminately is hardly good language design.
for <item>in <list>is every bit as readable, understandable, and
compatible with the language as foreach would be but without requiring a new
keyword, thus a very reasonable choice. Complaining that it should be the
same as C# or VB# has as little weight as does arguing that Delphi should
trade-in begin and end for { and } - many have asked for that too but it
won't - *and shouldn't* happen.
--
Wayne Niddery - Logic Fundamentals, Inc. (www.logicfundamentals.com)
RADBooks: www.logicfundamentals.com/RADBooks.html
"True peace is not the absence of tension, but the presence of
justice." - Martin Luther King, Jr.
 

"Will DeWitt Jr." <XXXX@XXXXX.COM>writes
existing code, except to require name changes for user symbols. That is
*not* breaking code. If you'd ever had your code broken you'd clearly
understand the difference.
Code *breaks* when you change behavior existing keywords or tinker with
default behavior. Details available on request.
Dan