A sweet little poll about the WP 31S layout

[Massimo Gnerucci]

Walter, now you're going to hate me... ;)

I found that ISO 80000-2 states:

Quote:Clause 3 specifies that variables such as x and y, and functions in general (e.g., f(x)) are printed in italic type, while mathematical constants are in Roman (upright) type. Examples given of mathematical (upright) constants are e, π and i. The numbers 1, 2, 3, etc. are also upright.

Time to shuffle some pixels, again? :D

[Jonathan Cameron]

(04-19-2014 09:19 PM)walter b Wrote:  There are some who want to see more lower case letters on the keys. So the new alternatives are

a) or c) now.

What do you think about them?

I'm not opposed, in principle, but I find the lower case 'a' a bit hard to distinguish in this font. I think the capital versions are easier to read (in this font).

-Jonathan

[Thomas Klemm]

(04-19-2014 09:53 PM)Massimo Gnerucci Wrote:  Walter, now you're going to hate me...

I found that ISO 80000-2 states:

Quote:Clause 3 specifies that variables such as x and y, and functions in general (e.g., f(x)) are printed in italic type, while mathematical constants are in Roman (upright) type. Examples given of mathematical (upright) constants are e, π and i. The numbers 1, 2, 3, etc. are also upright.

Time to shuffle some pixels, again?

You should quote the next paragraph as well, not just what's written in Wikipedia:

Quote:3 Variables, functions and operators
Variables such as x, y, etc., and running numbers, such as i in Σ_i x_i are printed in italic (sloping) type.
Parameters, such as a, b, etc., which may be considered as constant in a particular context, are printed in italic (sloping) type. The same applies to functions in general, e.g. f, g.

An explicitly defined function not depending on the context is, however, printed in Roman (upright) type, e.g. sin, exp, ln, Γ. Mathematical constants, the values of which never change, are printed in Roman (upright) type, e.g. e = 2,718 218 8...; π = 3,141 592...; i² = −1. Well-defined operators are also printed in Roman (upright) style, e.g. div, δ in δx and each d in df/dx.

The ISO 80000-2 is behind paywall (CHF 158.-), which is a very nice way to spread a standard.

[Thomas Klemm]

(04-19-2014 05:27 PM)walter b Wrote:  So we'd need ln, lg, sin, cos, tan, asin, acos, atan, hyp, deg, and rad.

We rather need: \(\ln x, \lg x, \sin x, \cos x, \tan x, \arcsin x, \arccos x, \arctan x\)
Or if you prefer: \(\sin^{-1} x, \cos^{-1} x, \tan^{-1} x\)
How are you going to explain that inconsistency to your grandchildren? Using \(\sqrt{x}, x^2, y^x\, 10^x, x!\) but then suddenly forgetting the parameter \(x\) in all the other functions?

[Massimo Gnerucci]

(04-20-2014 01:22 AM)Thomas Klemm Wrote:  You should quote the next paragraph as well, not just what's written in Wikipedia:

For the interested reader: 80000-2

(04-20-2014 01:22 AM)Thomas Klemm Wrote:  The ISO 80000-2 is behind paywall (CHF 158.-), which is a very nice way to spread a standard.

I agree.

[Sanjeev Visvanatha]

Option a), please. Our standard should reflect prior well thought out traditions in our community, IMO.

[walter b]

FYI, if no one of those who wanted lower case letters will speak up, it's going to be option (a) by tomorrow 15:00 UTC. Else, discussion may last a bit longer until a majority vote is found. After all, we know what works for the WP 34S since 2011.

d:-)

[Massimo Gnerucci]

(04-20-2014 02:06 PM)walter b Wrote:  FYI, if no one of those who wanted lower case letters will speak up, it's going to be option (a) by tomorrow 15:00 UTC. Else, discussion may last a bit longer until a majority vote is found. After all, we know what works for the WP 34S since 2011.

d:-)

I really don't care for UPPER/lower case... just don't explain anymore your adversion for LOG vs LG advocating international standards.

[walter b]

(04-20-2014 02:10 PM)Massimo Gnerucci Wrote:  just don't explain anymore your adversion for LOG vs LG advocating international standards.

Even you can admit LG is closer to the standard than LOG is.

[Thomas Klemm]

This poll reminds me of something that a moderator removed.

[walter b]

Thomas,

I'm sure you're feeling bad meanwhile about your thoughtless post. Else please give proper reason why you think (!) it being appropriate. (As you know I can stand quite a good lot of humour but that's beyond my limit. I will delete this post here as soon as you delete the preceding one).

d:-(

[htom trites]

I think that the use of upper case for calculator key labels comes from slide rule scale labels, which were almost always upper case. Few slide rules had the names of the functions. Those that did, usually had them correctly cased in addition to the traditional upper case scale labels, eg. "sin(x) S", Tan-CoT (x) T.

Depending on when and where and disciple you learned logarithms, the standard and computer program and language you usually use, and the publication's standards and editor's preferences, you will "naturally" have different meanings in your mind for log, ln, lg, lb.

lb -- seems to be used only for log 2

lg -- means log (base indeterminate unless specified), or log 10, or log e

ln -- means log (base indeterminate unless specified), or log e, or log 10

log -- means log (base indeterminate unless specified), or log 10, or log e

It would have been convenient if hundreds of years ago it had been settled to always write lgbase(argument) but it wasn't and so we live with the mess.

Lne and Lg10 might be the least confusion propagating (although to my taste they should be ln and log.)

LOGx (y) or other variations ... please don't do that, as tempting as it is as the shift of x^y.

[walter b]

(04-20-2014 05:24 PM)htom trites Wrote:  Depending on when and where and disciple you learned logarithms, the standard and computer program and language you usually use, and the publication's standards and editor's preferences, you will "naturally" have different meanings in your mind for log, ln, lg, lb.

lb -- seems to be used only for log 2

I concur.

Quote:lg -- means log (base indeterminate unless specified), or log 10, or log e

Never seen that being used for log_e nor for an indeterminate log.

Quote:ln -- means log (base indeterminate unless specified), or log e, or log 10

Only used for log_e according to my experience.

Quote:log -- means log (base indeterminate unless specified), or log 10, or log e

Never seen that being used for log_e.

Quote:Lne and Lg10 might be the least confusion propagating ...

I guess I see what you wanted to tell and agree on ln and lg.

Quote:LOGx (y) or other variations ... please don't do that, as tempting as it is as the shift of x^y.

We already did - on the WP 34S. Please see g-shifted 9 (LOG_x ) reverting y^x.

d:-)

[htom trites]

I've seen "lg" for "log" here and there, and thought it a typo. Recently I've been told by some students that "lg" is the abbreviation of "log" -- and they don't recognize "log" as being the abbreviation of logarithm. Some think "sin" is an abbreviation of "sinusoidal" and that "sine" is not a word at all. Schools these days ...

I should have included "lge", too, as a sometimes synonym of ln e .

Prescriptive vocabulary vs descriptive vocabulary may be good in some things but I think prescriptive works better in math.

LOGx, eh? I think you just sold me on a 34s!

[Thomas Klemm]

(04-20-2014 07:56 PM)walter b Wrote:  

(04-20-2014 05:24 PM)htom trites Wrote:  log -- means log (base indeterminate unless specified), or log 10, or log e

Never seen that being used for log_e.

Definition of Logarithm from MathWorld:

Quote:Note that while logarithm base 10 is denoted \(\log x\) in this work, on calculators, and in elementary algebra and calculus textbooks, mathematicians and advanced mathematics texts uniformly use the notation \(\log x\) to mean \(\ln x\), and therefore use \(\log_{10} x\) to mean the common logarithm. Extreme care is therefore needed when consulting the literature.

Cheers
Thomas

[Jonathan Cameron]

(04-20-2014 03:17 PM)Thomas Klemm Wrote:  This poll reminds me of the Anschluss:

I had hoped this post would go away before now. I reported it as abuse and just wanted to publicly say that I think it is offensive. I hope the site admin removes it soon.

Thanks to Walter for all the work he has done on the WP-34s and the WP-31s.

-Jonathan

[Thomas Klemm]

You started a poll with two variants:
a) uses LG
b) uses LOG

Later you stated:

(04-19-2014 07:54 AM)walter b Wrote:  Be assured, a label LOG for the function lg(x) = log₁₀(x) will not return on our calculators. Period.

Now tell me how is there still a choice?

Cheers
Thomas

[walter b]

(04-21-2014 06:17 AM)Thomas Klemm Wrote:  You started a poll with two variants:
a) uses LG
b) uses LOG

Later you stated:

(04-19-2014 07:54 AM)walter b Wrote:  Be assured, a label LOG for the function lg(x) = log₁₀(x) will not return on our calculators. Period.

Now tell me how is there still a choice?

The poll was about overall Layouts, not about a single letter. Who's literate, read!

[Massimo Gnerucci]

(04-21-2014 06:17 AM)Thomas Klemm Wrote:  Now tell me how is there still a choice?

Cheers
Thomas

I hope this could represent your feelings in a lighter mood... :)

[Manolo Sobrino]

It is Walter's calculator, he wants LG so he gets LG, period. The nice thing about this calc is that it is GPLv3, so anybody can change whatever they want. There's no point in arguing about it. Some banter is fun, but I wouldn't take it too far.

The whole matter is funny. The designer/creator goes for a standard that no one is really using, because that is the very German right thing to do. And then people complain endlessly as if this were a design committee. Well, it is not.

You had the choice to vote for a layout with LOG, just 1/3 of voters did. Walter gave you the chance. I'm genuinely surprised he proposed having this poll: layout b) was just an exercise after all (see the other thread). Either he was really interested in the layout, or it was the easy way of legitimising his choices although he didn't have to! There's nothing bad with it.

I'm more surprised that no one is really thinking about the general idea behind these layouts. There haven't been any just scientific, non programmable HP calcs since the 31E, 32E. And those were 30-key calculators. This one has 37. The other comparable scientific is the 45, but then it has 3 rows of 5 keys and 5 rows of 4 keys.

The 20b/30b has 2 rows of 6 keys and 5 rows of 5 keys (and you can't change the ON position)... great, so now we can make it look like a 42S... Why? You guys haven't been using a just scientific with these settings. This is a new problem for a different use case and you can't see it (for instance, the prominent position of STO/RCL is arbitrary). IMO discussing these options would have been much more interesting than arguing about the O in LOG. It should be LOG10, BTW.

I'm done with this layout, if Walter wants it that way, so be it. I can assign keys, I can print a different one... I can even change other internal choices in mine(*). What's the problem?

(*) I've been thinking about more interesting things that %CH. I'd like to test a key that would give you the product of all the non zero numbers in the stack, or a key that would give you the sum of the squared numbers in the stack, the first one would change the way I use RPN, the second would be a handy macro.