Dawn, I have nothing against you, but it does bother me when people
casually throw out uncommon (at least to CDP) terms with no supporting
explanation or examples.

Duck typing is a term that I don't ever recall being used in CDP
before. Maybe it has, but it would certainly be quite rare. As such is
the case, it is reasonable to assume that many in CDP might be
unfamiliar with it and how it applies to MV programming.

Your own post mentioned duck typing with QM and Caché. So, I asked for
elaboration on QM because I am familiar with it. Also, QM has been
rising in popularity among CDPers, so it seemed more reasonable than
asking about Caché.

In your reply to my request, instead of providing an explanation or
example of duck typing, you chose a defensive stance in which you
declared that you're "comfortable" using the term as you do.

Well, yay! Such a "hand wavy" answer didn't help anybody, and as I
said, just led me to believe that you didn't really know how duck
typing applied to MV or possibly didn't really understand the term at
all.

Of course, you had to follow-up my explanation by taking on the role of
the victim in what I guess you imagine to be my evil plot to disparage
you and become the all-knowing master of CDP. - btw, that Master of CDP
position is already taken by a Nebulous character. LOL.. Just kidding,
T. ...Sort of.

It's unfortunate that you took my initial, simple request personally.
I believe that both my request and assessment, based on the information
(or lack of) that you provided, were fair.

--
Kevin Powick

On Jul 19, 4:53*pm, Excalibur21 <pgmcmur... (AT) gmail (DOT) com> wrote:>
Hmmm I'd like to see timing tests for this theory.
It's certainly not true for R83 or R91 that this behaviour exists i.e.
"both string and numeric flags set".

Rather what occurs is the numeric Zero is moved into the variable
Foo. If you examine the run-time space you see a numeric zero (if you
can read bytes) not a string.

However if you say Foo = "0" then the string 0 is put into variable
Foo, in which case you should really do a Foo+0 before a numeric
comparison but few people are that rigorous.

Will

Correct, a variable's descriptor either contains a short string,
a long string, or a number (integer). There's no string flag,
there's no numeric flag; there's only a descriptor.

--
frosty

On 2011-07-20 13:14:20 -0400, frosty <frostyj (AT) bogus (DOT) invalid> said:

KA-BOOM!
--
Kevin Powick

On Jul 20, 8:32*am, Kevin Powick <nos... (AT) spamless (DOT) com> wrote:
OK, good, because I was "feeling" some negative energy from you Kevin.
I wasn't sure what I did to trip your buttons some time ago, and we
all know that not everyone likes everyone, so I just figured you were
a detractor. Given this statement, I'll put you back in my "maybe he's
can be helpful" bucket.

I did not know whether it was a known term or not among cdp readers,
but putting "loose typing" in parens seems like it would have
mitigated that issue. The fact that people can google would mitigate
any such issues too.

I have used this phrase for maybe a decade, at least as long as I have
written Java. Now that there are two MV BASIC alternatives that
include OO, I figure it is a useful term. Additionally, instead of
copping to "procedural" I would rather have us say that like
JavaScript, MV BASIC employs duck typing. As I mentioned I feel free
to use this term for non-OO languages too. Basically for any language
in which typing is associated with the value rather than the variable,
I think this term can apply. I do not like the phrase "scripting
language" but it is more hip than "procedural language" so in my use
of the terms and in my opinion, I might say that MV BASIC is a duck-
typed scripting language, similar to JavaScript. [Also very different
but I will not write "lambda" anywhere lest you want me to write a
treatise on closures here].

I have not used QM OO, but, yes, I did assume that it would have duck
typing in the stricter sense of the word too.

I "felt" as if I had just mentioned something about painting using an
eggshell finish and now was being asked to defend just how much I knew
about eggshell and other paint finishes. It wasn't important to the
context.

From previous posts, I thought this was a continuation of you thinking
I knew nothing of anything. Given that I am comfortable with people
thinking I know nothing (someone once told me never to wear red if I
wanted to look feminine and recessive), I would rather step back and
say "yes, bro, I'm clueless" than to spew out my understanding of a
term that anyone can look up simply to prove that I am occasionally
knowledgeable.

Yes, that is the stance I took. This was not the first time I have
felt this vibe from you, so I decided to let you know this time.

Of course.

Yes, I did and if that is not how it was intended, then I was wrong
and I apologize.

Fair and helpful are two different things. Have a great day, Kevin. --
dawn

On Jul 21, 4:16*am, Kevin Powick <nos... (AT) spamless (DOT) com> wrote:
Oh dear me so many who have not read the manual I quote

"In FlashBASIC, variables are assigned a data type at assignment time:
string or numeric. If the value assigned to the variable is a non-
numeric literal, only the string flag is set. If the value assigned is
a numeric literal, both flags are set. If any numeric operation is
performed successfully upon the variable, the string flag is set off
and the numeric flag is set on.

When two variables are compared, their type flags are checked for
compatibility. If both fields have their string flags set, the
computer performs a string compare ("001" # "1"). If both fields have
their numeric flags set and neither has a string flag set, a numeric
compare ("001" = 1) is performed.

It is important to note that this behavior can be controlled by the
use of "+0" to normalize numeric variables and force numeric compares;
while ":' '" concatenates a space, forcing the numeric flag off and
forcing string compares."

I recommend that one uses the map compile option to see the change in
D3 variables since the early days Pick

On the topic of the manual is there a setting in Windows Explorer that
will let a particular file such as the manual run Active X without
having to say yes all the time?
Peter McMurray

On 2011-07-20 20:35:09 -0400, Excalibur21 <pgmcmurray (AT) gmail (DOT) com> said:
Interesting. You are correct that this is a behaviour for FlashBASIC.
I learned something new about FlashBASIC, Thanks.

However, the text you quote is found under the heading "FlashBASIC and
BASIC Differences". Since in a later post you even mentioned that
you've stayed away from FlashBASIC, and there was no indication in your
previous posts that you were referring to FlashBASIC, I can understand
how some of us thought you were speaking about regular BASIC.

I don't know, but why don't you just download the PDF versions of the
manuals? I prefer them in that format. Navigation has been much
improved in recent years. e.g. TOC entries are linked to the actual
content.

BTW, in case you haven't heard of it, the FoxIt PDF reader (free) is
way better, IMO, than Adobe's Acrobat Reader.

--
Kevin Powick

On Jul 21, 10:35*am, Excalibur21 <pgmcmur... (AT) gmail (DOT) com> wrote:

Assume the "explorer" in this instance is internet explorer? If so,
simply setting the soyurce of the manual as a "Trusted Site" under
internet options may do the trick

I wonder if App Inventor will be a casualty of the closure/scaling
back of Google Labs?
http://www.crn.com.au/News/264438,go...y_newslet ter

It's not. I'm in the AI developer forum and that's been a topic of
discussion, with reassurances from Google that AI will still be
available, largely unchanged, like most of their Android properties.

I am, however, feeling like I'm in a state of mourning over the demise
of many of the other Labs projects.

T

Ross Ferris wrote: