Re: Why do so many people love Teas Instruments?

On Aug 23, 2:01 pm, "John H Meyers" <jhmey...@xxxxxxxxxxxxxx> wrote:

Who forced you to do it "the hard way" when specific "easy ways"
(local variables, use of formulas) were built in?

Even with local variables the syntax is not as clean to read as it is
in TI Basic. The language of the loops is also less clean than BASIC.
Just compare examples given in the TI manual and HP manual to see the
difference between two languages. There is no doubt, that TI
programming is a lot more to a little more readable depending if HP
user uses mixture of algebraic and local variables, or pure RPN
manipulation of the stack.

You caused your own problem, like someone buying a car with Cruise control
but refusing to enable it, then saying how tiring is this brand of car,
because you constantly have to watch and adjust the speed :)

It is not really a problem for me. I use this style because I like
short typing and quick programming even if it is at the expense of
readibility. In fact I can argue, that this enables HP to be a macro
capable calculator, where TI does not really has such functionality.
What I mean by taht is that if I need to have let's say 10 repetable
short operations such as SQRT(a/b+c) to calculate for various values
of a,b and c I'm more than likely assume that a,b,c will be on stack
in that order and quickly write:
<< ROT ROT / + SQRT >> store it, reuse it and finally get rid of the
macro when I'm done.
On TI I wouldn't think about writing such tool for 10 or 20 temporary
operations. First Basic would require more keystrokes, second, it does
not have anything remotely silmilar to stack so all had to be done
either by storing data in variables or through the function
parameters. I think in this case the critical amount of repetable
computations that justify writing such function is higher than it is
on HP due to the overhead of typing it and then amount of typing when
used per single usage.

But if you compare my macro listed above with TI function synatx:

Funtcion example(a,b,c)
return sqrt(a/b+c)
End

(I'm not sure if this is proper syntax, I did not used TI for couple
of years now, so I don't remember exactly how TI Basic function
returns value).

Clearly TI function is way more readable than HP macro.

My case is probably the extreme of how not to use USER RPL if somebody
plan to write large program that require maintenance but I prefer
quick and dirty programming. With usage of local variables and
algebraic the readibility of HP programs does improve. In the given
example probably the "most" user friendly version would look something
like:

<< -> a,b,c << 'SQRT(a/b+c)' ->NUM >> >>

This certainly is much more readable than RPN version but I would
still claim, that it is not as good as Basic. And the preformance (if
used in long loops) suffers (althoguh not that much).

So why don't you take your own advice, and use the self-describing
features of named [local] variables, formulas, etc?

I don't like to mix algebraic and RPN and loose performance.

Maybe you should write in ML, then you'll *really* have
something even more cryptic to complain about :)

Not worth it. I don't use calculator for heavy number crunching
anymore. For such problems I revert to PC. My HP is more and more just
a macro machine where macro is written on the spot to attack some
temporary, simple, repetable problem. Larger simple problems end up on
the Excel spreadsheet with Visual Basic if special functionality non
existent natively is required. Really large problems are either
attacked by specialized software if available or using custom C#
program on PC. This is unfortunate, but it is a reality. I think
convergence of PC tools (many CAD and CAE software now reads or
integrates itself with Excel and have front ends programable in some
variation of C or .NET) will eventually challenge the existence of
advaced calculators.

Oh, is % in MTH REAL by mistake?

Why mistake ? My point was, that % is not a widely recognized unique
mathematical operator but rater a construct of the calculator
manufacturers to help business. That is why it has different
implementations between different manufacturers. They simply do not
have a standard to guide common implementation.

I'm willing to challenge you that if you have two numbers on the
stack:

100
120

and you press % key there are two possible answers:

120 (as implemented in HP50g)

but also 83.333333

as what is the precentage of 100 in 120 :-)

It all depends, how manufacturer implements [%] functionality in
relation to the stack. HP have chosen one way, somebody else could
chose it differently. So what if it is illogical. There is no standard
that forces you to implement it "properly".

2+3*4=

In this case it should always produce 14. On some older calculators
that was not the case but I don't consider them neither algebraic nr
RPN. Any modern algebraic calculator should intepret this statement
correctly.

"Four-bangers" are supposed to get 20 -- they don't have more registers
(or a stack) to save up as yet unused operands (and operators).

Yes, but I don't consider them "algebraic". They do not adhere to
algebraic rules at all and the user is usually warn in the manual that
the proper order of operators is responsibility of the user.

But note that just plain "+,=" or "-,+" is unpredictable
between different manufacturer; it uses a "saved constant,"
the behavior of which varies arbitrarily between brands
(and chip suppliers). Figure out how "Sharp logic"
gets a Fibonacci sequence! ("it's not a kludge, it's a feature!" :)

Again, you are using extensions to speed up calculations. This has
nothing to do with algebra. Proper algebra requires operands to be
separated by the + or - operator. This example is not a fair
comparison. If you do + with single element on HP stack you won't get
anything but a system error so HP is even worse :-)

The "bottom line" of all this talk is that calculators which
"look algebraic" vary all over the map as to how they work,
while you can walk up to most any RPN calculator that you've
never seen before, and already know what it does.

But no matter -- lest we forget, all HP graphing calculators
are fully equipped to handle algebraics, exactly as they should,
and using RPN is an option that one can take or leave.

But so it TI89. There is no ambiguity if you have to enter algebraics
exactly and properly without "+,=" tricks. You enter formula and
calculator does proper handling of the operators as defined by math
rules.

[re his daughter as "casual" user]
Casual generally outnumbers "professional" in every field.
That is why casual user drives what is mass market
and professional drives niche market.

So being a professional engineer, which market do you buy your tools from?

There are only few manufacturers of automotive assembly line tools :-)
You can't buy those in the public store.

Is there anything wrong (or even other than inevitable)
about different products having values
that appeal to different audiences?

No, absolutely nothing wrong with it.

Do you want HP to go for the lowest-level, widest possible market?

To the contrary. I would like HP to stop wasting time on developing
algebraic / history parallel to RPN screen mode altogether in vain
attempt to catch up with TI market and put resources into fine tunning
and further development of RPN. Possibly they could save money to
advertise their product between those who can appreciate RPN and use
it in their professional life. I admire TI for not trying to be all
functionality for all people and not attempting to implement parallel
RPN to lure scientists and engineers away from HP :-)

What, if anything, should differentiate an HP calculator from a TI?

If we talking high end, such as HP50 I think, HP should concentrate on
being faster and more powerfull algebraically than TI. Should not
waste time on implementing parallel algebraic interface. They have
limited resources and should concentrate those in areas that are clear
HP strengths. In my opinion RPN is it's main (and probably only)
reason it is still in existence. It used to be algebra, but frankly I
consider TI89 algebra to be on the same level as HP these days.
Algebraic HP is a lost cause unless it can outperform significantly TI
in algebra and speed for the same price. I don't see that path to be
technically realistic in the current calculator market.

Why not just suggest to TI whatever improvements you want,
and forget about HP?

Because TI is a market leader and has no reason to invest large sums
of money trying to attract small minority of disgruntled users :-)

In case of my daughter the last straw was actually
that she needed to install something later that was not
in the initial Linux package and she wanted to do it herself.
You can do it without any help on Windows in 99% of cases using simple GUI
installer.

If the author has so provided.

In 99% of Windows software it is the case. It has been many years
since I saw last Windows program that required mystic commands from
command line to install. The simplest executables these days might in
the worst case require user to graphically create shortcut to the
"exe" and move it to "Start menu" folder in Explorer so they can start
it from the programs menu.

Just like the free calc software found onwww.hpcalc.org;
some authors have made it simple and foolproof, while others' stuff
only works if you can guess that it has to be installed in a variable
with just a certain exact name, in CAPS, that even the instructions
may not have mentioned -- all this is about the quality (or laziness)
of the author, rather than about the quality of the system the stuff is made for.

You are biased. Just look on any Linux discussion group. Almost ever
problem that newwbies are having with Linux is answered there by the
advice to:

1) Make sure that you have installed packages:

gimtsmimt version 2007.10.57.24
packsrack version 111.49454

2) Make sure you have proper library

librarysmiblary vesrion 232344.34343

Go to command line and execute something like:

blahblah -c#1/25 !@#$%^ grep shit - root smoot #25 bigbang .\user/
superduper\ -password <tralalala>

3) And then you should be able to have whatever functionalty you
wanted.

Even more comic are advices regarding making even simple hardware to
function under Linux.

Sorry, but this is ridiculous. In most cases those "experts" do not
bother to explain what is even the meaning of the cryptic commnad line
switches or the explanation is even more convoluted than the command
itself. There libraries and packages suggested very often have
different latest version when you look on the internet and a lot of
advices still end up being wrong. When I was playing with Linux, many
problems were not solved by those type of internet "expert" advices.

If you go download a Firefox or Opera browser, however, or Thunderbird email,
is it harder to install on Linux than on Windows?

I don't know. I gave up on Linux. They lost me couple of years ago and
I'm not intending to go back in the near future. Unless Windows
becomes pay per use type service, Linux is a lost cause for me due to
the lack of polish and lack of software that could match what I have
for Windows. Contrary to popular believe Linux is not "free". It cost
typical Windows user significant amount of time and effort to
repartition hard drive, dual boot install Linux, test it and
eventually get rid of it. Very often such unfortunate user ends up
fresh installing Widnows after Linux disaster is over. It takes
several days if not weeks of free time to recover from this
experience. Such unfortunate person will think twice about giving
Linux second chance.

Microsoft makes an author's life pretty difficult, erecting a high barrier
to entry -- it's like a "poll tax" that once kept "commoners"
out of the voting booth, and it "improves" the environment
much the same way that a high "cover charge"
keeps the less affluent (and well-dressed) out of the pricier places :)

I'm not sure what you mean by that ? How come they make life
difficult ? Anybody can write programs that run under Windows and
there is no MS fee associated with this. You can get even their
development tools such as Visual C++ Express completely free. Where is
the difficult part ?

The second culprit is vendors who won't bother to enter the market;
why spend so much, when the big returns come from Windows customers?
That's a function only of Windows being bigger, market-share wise,
not better.

No denying. Catch 22 works here. Linux cannot win by being niche and
less polished. This will always relegate it to the secondary place and
keep chicken and egg problem going.

I had Windows since 3.1 version and I'm still waiting for this virus
that will infect my machine :-)

Plug it in here and see ;)

I'm plugged all the time. I have broadband @ home (except I do switch
it off when not used to save power).

A few years ago, we couldn't even begin using any W2K system
on our network, because the instant they were plugged in,
they got infected by "Blaster" or "Welchia," which are among
worms which don't need any attachment to spread,
and for which Windows had no way for any user to block them.

Yes, but it is fixed. Holes are also dicovered in Linux. This just
proves, that all software has bugs.
Just last week or two weeks ago serious hole was discovered in Firefox
as I recall.

We spend a great deal of our time tracking down infected computers
that flood our network with ARPs and other attacks,
and 100% of those turn out to be running Windows.

Windows by its market share is the main target of virus writers.
Nobody really bothers to attack Linux. That would be very academic
exrecise ;-)

Sure, you can improve on some simple MS tools.
These were not meant to be "top of the line" products.

Exactly; that's why we all go around replacing them.

I agree, but similarily text editors, file managers etc that are part
of the Linux distributions are often equally primitive or completely
ommited. This doesn't prove anything. I don't buy Windows to be able
to use it with Notepad or Paint. In fact I found, that the same crowd,
who blames Microsoft for it's primitive extras shipped with Widnows
also is first to charge MS with "monopoly abuse", when Microsoft adds
some more polished "add on" to the OS. How many times MS was dragged
to court and charged with "monopoly abuse" because they elected to add
some better tool to their OS ? How many times I've seen such charge
raised on the Linux discussion group ? But the same Linux users who
blame MS for "monoploy abuse" in different thread laugh a MS primitive
Notepad or Paint. Most of the tools that are added to "linux" packages
are really not part of the Linux but freely available from other
developers. I can argue, that this is exactly the same with Windows
except MS does not package it up front, but you must do it yourself
later. I would even argue, that MS probably would be forced to pay
licence if they would attempt to package others software into Windows
distribution.

Try to improve on MS professional tools
such as Visual Studio or Office.

For which you pay several times as much as for Windows itself?

FUD ! FUD ! FUD ! :-)

MS Office Home and Student version cost about $120 licence per three
computers (cheaper than Vista).
Microsoft Visual Studio Express is completely free of charge
(infinietly cheaper than Vista).

Of course, those who cannot afford even $120 for a software that will
work for years on three computers each worth about $500 can always
install free of charge Open Source toys or use one of those free of
charge online office applications from Google etc.

Of course, we're not talking about your daughter's use any longer,
are we? (not VS for sure, maybe Office?)

I have two daugters. If we would talk about the other one, your
assumption would be very true. She would not be a scientificly
"oriented" person. Bue we are talking about the one that is. She
happen to be a student of Bio-Physics at UofM. Don't uderestimate her,
because she is a woman :-)

Open Office is a nice toy for simple tasks but falls well short
when some more features such as programming are needed.

Programming? Your daughter needs an office suite with programming?

She eventually might. I don't know what her proffesors are going to
use it for. UofM requires students to use MS OFffice. But I would
argue, that even if you don't need programming in Excel now, if you
are in science or engineering you will eventually eiter discover need
for such functionality and potential lack of it will be a factor.
Staying with MS Office is simple a good choice.

It seems to me that you're always arguing both sides at once.

Yes. It seems to be my character.

And why do others blindly bash Linux?

Nice sparring with you again, Jacek :)

You are welcome.


Hope your daughter is happy now;
if not, she can always find peers ready to help her,
and no more need for Dad :)

-[ ]-

JAM

.

Relevant Pages

  • Re: [SLE] Developing a Real Time Data System
    ... The KBasic License is valid for both operating systems: Windows and Linux. ... Modern BASIC language ... chapters learning about the programming language. ...
    (SuSE)
  • Re: Why do so many people love Teas Instruments?
    ... that almost every modern programming language ... What you described is some "improvements" to the algebra, ... different calculators will produce different results depending on how ... She lived with Linux for a few weeks, ...
    (comp.sys.hp48)
  • Re: Segmentation in real mode
    ... Except that Windows does have its "prohibition" on "direct hardware ... what you're saying about 32-bit programming is still ... Linux is often ignored in this context but it's actually arguably the best ... a "portable assembler", so, you know, it's not really got anything ...
    (alt.lang.asm)
  • Re: Why do so many people love Teas Instruments?
    ... That is why casual user drives what is mass market ... in the initial Linux package and she wanted to do it herself. ... You can do it without any help on Windows in 99% of cases using simple GUI ... when some more features such as programming are needed. ...
    (comp.sys.hp48)
  • Re: Flame Bait! Windows vs: The Unices
    ... I use the term "Linux" to refer to a typical ... > saying for a while about the advantages of Windows... ... Database server running on a remote machine. ... so I've decided to discuss /only/ programming (this ...
    (comp.programming)