x^y key executes y^x on HP Prime prototype emulator

[N. Jackson]

I was a bit surprised to see in the images of the HP Prime that it has an x^y button instead of y^x.

I've read (for example at http://www.hpmuseum.org/hp35.htm) that the HP35 had and x^y key rather than y^x so that the user could calculate common antilogs because they didn't have enough keys to dedicate one to 10^x.

That explanation isn't entirely satisfactory since x ENTER 10 SWAP y^x is only one more keystroke than x ENTER 10 x^y.

In any case the HP Prime, with 53 keys (if you count the cursor control "rocker wheel" as four keys) isn't suffering a key shortage crisis (despite only having one shift key), so of course it has a common antilog key (although it is a shifted key), so the explanation the choice of x^y over y^x cannot be the same as the putative explanation of the HP-35's x^y key.

I found a copy of HP's early (June 13th) prototype emulator of the Prime lying around on the web, so naturally I ran it to have a look around. (By the way, aside from the bugs inevitable in prototype software, it runs just fine on Wine on Linux -- better than emu48 does (for me at least).)

It turns out (at least on this early emulator) that the x^y key actually executes y^x! So 3 ENTER 2 x^y yields 9 not 8. Is it just me, or does this seem illogical?

Regards,
Neil.

[didier.s...@gmail.com]

Neil,
I gess it is because in RPN mode y^x is logical and in algebraic mode x^y is logical. I hadn't even noticed it when using the emulator !
What do you think of that ?
Didier.

[Joe Horn]

> It turns out (at least on this early emulator) that the x^y key actually executes y^x! So 3 ENTER 2 x^y yields 9 not 8. Is it just me, or does this seem illogical?

Just thinking out loud here...

You are making a false assumption, namely, that the first input is always called "Y", and the second input is always called "X". But that's not true in Prime, and in fact hasn't been true for a long time for most of HP's calculators. Stack levels are only named X, Y, Z and T in 4-level RPN, which all the old HP's had. But in RPL, since there are an indefinite number of stack levels, they are numbered, not named. And in algebraic logic, you can call anything whatever you want. For example, have you noticed the Nth root button? (Shift x^y). Not Xth root; Nth root. Why? Because that's what everybody else calls it, and that's how it's taught.

Therefore, calling the power key x^y, y^x, or just [^] (as some calculators do) makes no difference, since the name no longer represents the order of the inputs. You'll notice that almost all algebraic calculators out there call it x^y, not y^x, and nobody complains that they do it backwards. Even the entire HP 38/39/40 series has called it x^y since 1995.

Bottom line (formerly known as X): I'm rather sure that they chose x^y for Prime for the same reason that everybody else does: That's how it's taught. That's a powerfully compelling reason, when the calculator is primarily intended for education. On the other hand, there is no reason whatsoever to call it y^x, other than human inertia.

-Joe-
Disclaimers: I don't work for HP, or anybody else. The above was my opinion when I wrote it, but might no longer be my opinion by time you read it.

[BartdB]

On Monday, August 5, 2013 11:40:12 PM UTC+1, N. Jackson wrote:

>
>
> It turns out (at least on this early emulator) that the x^y key actually executes y^x! So 3 ENTER 2 x^y yields 9 not 8. Is it just me, or does this seem illogical?
>
>
>
> Regards,
>
> Neil.

The x^y & y^x on old HP's was related to input order because they used a stack with registers called X, Y, etc.

The Prime is first an algebraic calculator. On all algebraics, it has been convention that x^y means (first number)^(second number), and logically x=first number & y=second number.

Even on the 50G, y^x doesn't make sense but was kept because "people were used to it", but its stack levels has numbers and not X, Y etc.

Note that the "first" forerunner of the 50G, the 28C, has the power operator simply as "^" (shifted function above the multiply key).

[TW]

> -Joe-

Just to add on to what Joe said, I am the one ultimately responsible for the exact key text and positioning since I delivered the final artwork for the printing (Prime, 10bII+, 39gII, 30b).

That doesn't mean I am ultimately the decider on what all the keys do - that goes through many revisions based on feedback and adjustments as we develop a calculator and learn what should go where to best support its operation, but I *have* been the one with the most critical eye on things like consistent capitalization, fonts, x/y centering and so on.

I can't claim color choice though unfortunately... :-(

x^y was used in previous HP calculators as was said and doesn't have anything to do with the old XYZT stack. It is just a pure power operator. As has been discovered, it calculates just like on the 50g. True, it could have just been labeled ^ but personally I think that is kind of ugly and less clear to both the math experienced and the math inept.

One consideration that has not been discussed though is that y^x would not fit well on the key without messing up the vertical alignment. The 50g doesn't have "classic HP style" sloped fronts on the keys - nor does it try to fit everything onto the same key. With the sloped front key, you can't print over the boundary there for many reasons. The "tail" of the y would require either reducing the y height, shifting the whole assembly upwards, reducing the size or something like that in order to avoid encroaching into the sloped front of the key.

Feel free to ask any keyboard related questions and I'll try to answer them if I can.


<this next part is just written in general and not directed at anyone>

Top items that people might be wondering about:

1. Large enter key positioning? One of the goals we tried to accomplish with prime was to make it clear this is an HP machine. Hence we tried to incorporate classic and strong design elements into the machine in modern ways. We were not going to miss having a double wide enter for many reasons (or traditionally shaped keys). What usually happens is there is just *1 more* key you want to get a certain function exposed...and someone suggests splitting that single key that takes two spaces and the problem is resolved. This definitely did happen when we struggled with the combined parenthesis vs split, and shifted EEX vs STO and similar.
--Why on the right? We took a lot of design cues from what the "average user" expects. Nearly all keyboards, virtual or hard, have the enter to the right side. Same logic positioned the BKSP/DEL above the enter to right, as opposed to the left or down near the enter.
--What about the ESC key? Well, that one we really did want to have on the left upper right corner. That breaks the tight grouping of the apps/symb/plot/numeric thought which is key to operating the calculator. Could not figure out a way to do it that did not feel wrong in operation.

Combined parenthesis instead of split? Two reasons here - we needed an unshifted EEX since the calc is targeted at engineering students and professional use as well and not just students, and the primary form of input is through a 2d equation editor like interface when using parenthesis. Even in RPN mode when you type an algebraic object it acts more like the EQW on the 50g and formats things 2 dimensionally. It can be mildly annoying if you have to insert a parenthesis early in an equation, but not incredibly so. Also matches the behavior of the { }, [ ] and " " objects.

Why no second shift? This is difficult to explain, but really boils down to several things. 1 - that third shift forces you to begin printing on the background around the key and this *reaaaaly* makes the unit start to look very difficult to use and cluttered. Prime looks great and having text on the brushed metal would be greatly impacted by that. 2 - we are thinking long term here, not just for this unit. At some point tablets/software may take over completely in this space and cramming in huge numbers of functions would make it more difficult to have consistency across platforms in a number of ways. 3 - A properly designed unit can expose functionality in ways that make it able to have more functions closer at hand then a whole shift plane ever could. True, you can pull some functions out, but we just didn't feel the benefit would outweigh the downsides.

Why doesn't <insert_key> match the exact same position as <unit_x>? Basically, the primary interface for a calculator is the keyboard. Even though prime has a touchscreen and it works great, we never try to force the user to use it if not desired. Because of the keyboard being first, calculators always get designed to support how the designers envision or *see* it being used in testing. The keyboard is there to support the use of the device. If we have plans to move things in certain directions, sometimes things get included to try and move usage in that pattern. We do try to keep things as consistent as possible, but sometimes it just doesn't fit perfectly/cleanly. An example of this is the main block of keys on Prime matching exactly the 50g. You have the number keys, decimal and space in the same spot. APPS is in the same spot, eex is nearly in the same spot. BKSP is in the same spot. From the 39 series you have the sin,cos,tan,ln,log block in the same order. vars,math,xttn,frac are the same. Some adjustments to bring in new designe elements always shift things, but it is quite close.

While true the Prime keyboard isn't an exact match for the 50g and does pull more from the 39 series, there is not a single key position on there that has not been thoroughly thought out and adjusted at some point. We even went through some early revisions with radically different keyboard blocks (such as numbers off center, or near the screen, or split on both sides). Everything was investigated.


Most issues people have with the keyboard come from a lack of familiarity with any of the 39 series machines and lack of understanding about the "apps" or "aplet" concept since they are only 50g users. That is understandable, but just give it a little bit of time using the machine of software before you claim no thought was put into it. I've had quite a few people tell me that they initially though Prime was crap and they'd never like it, but after using it for a while it really grew on them and they appreciate some of the really great things in there. You're welcome to still hold that opinion of "this is crap", but I do think you'll miss out on a lot of other great things! :-)

The other complaint I've seen is those that just say "it is just a nspire clone". That is soooooo far from the truth. Honestly, in my opinion the nspire did so much wrong that you could come up with a better design for a *handheld* calculator by just doing the opposite of everything they did there. Some similarity exists on the surface, but the style of operation and, more importantly, the ease of use on Prime is so far ahead it isn't even close.

That being said, there are some items on the keyboard that share similarity through convergent evolution. A great example of this is having the directional pad in the middle of the unit. Nobody was more surprised then me when the first 3d printed mockups were made and the dpad near the edge felt out of place and difficult to reach! Basically, because the unit is so thin the best place for it really does end up being the middle when you hold it. It was very surprising.


TW

[TW]

> Why no second shift?

Forgot to include here that having a user defined keyboard lets you adjust things as desired. Those that appreciate having more or certain functions exposed can happily adjust as desired.

TW

[Leviatan]

A TW se le ha ocurrido que:

> Just to add on to what Joe said, I am the one ultimately responsible for the
> exact key text and positioning since I delivered the final artwork for the
> printing (Prime, 10bII+, 39gII, 30b).

Wow, thanks for all the information; it's a privillege to have you
arround here.

I keep using the 48G that I bought when doing me degree and don't think
I'll need a new graphic anymore, but is great to know that there are
nice guys at HP researching in calculators. Maybe when my kids start
high schools in some years...

In fact most of the desk work I do could be done with a simple office
calculator, but having a 48 on the desk is cool, and non-engineers
never try to borrow it!

--
Un saludo,
Alberto

[nljli...@gmail.com]

Joe Horn wrote:
> You are making a false assumption, namely, that the first input
> is always called "Y", and the second input is always called "X".

> Even the entire HP 38/39/40 series has called it x^y since 1995.

Joe,

Thanks for those thoughts. Yup. You're right, my "thinking" about whether it "ought" to be y^x or x^y was influenced by my past calculator experience, and I haven't really been paying much attention to calculators lately and am largely ignorant of the HP 38/39/40 series.

As a long aside, my arc with HP calculators began with the HP 28S. Two factors contributed to my making a rather odd use of it: I was studying pure mathematics and I didn't own a real computer. So most of my time was spent on the calculator programming one thing or another -- I suspect I spent more time on my 28S each day then than I spend on a real computer each day now! But I don't ever recall using the 28S as a calculator. I was still using my Casio fx-3600P from high school for actually calculating anything.

I soon needed more space than the 28S provided and I pre-ordered (and collected the day it was released) a 48SX. That was my calculator until about eight years ago, when I picked up a second-hand 48GX. There was no good reason for that, except the price was good and I'd always wanted one. With its turquoise and purple lettering, in my opinion it is the best-looking calculator ever made. (But I'm sure many are of the opposite view!)

Shortly after I got the 48GX I was studying in the physical sciences and was doing a lot of work with manipulating equations and taking symbolic derivatives and integrals etc. and the 48GX was exasperatingly slow. Of course I could have used Sage or Mathematica, but I was used to working on the calculator, so I bought an HP 50g.

I've always been highly ambivalent about the 50g. It does most of what I need it for, and it does it fast enough to be useful. In that, for me, it is by far the best calculator available. But it's a toss-up whether it's more ugly on the outside or the inside, with it's mulligan stew of functionality all added on higgledy piggledy without any apparent design or planning. When one goes back and reads the manuals for the 48SX or even the 28S, which are beautiful reflections of coherent design, the 50g is almost enough to make one weep. (I'm not really surprised that no one ever wrote real manuals for it. Who would want to?) And the feel of the 50g is so inferior to the excellence of the 48SX and 48GX that every time I use the calculator I feel a little sad about the demise of Hewlett Packard from a once-great engineering company to what I tend to think of as little more than a marketing company.

And yet... along comes the HP Prime out of left field. Something completely new. (Although apparently not so new if one had followed the the HP 38/39/40 series at all, which I did not, other than to glance as the specs as each calculator came out and see that it was insufficient to meet my requirements.) A single glance at the Prime hints at the excellence that one used to expect from Hewlett Packard. Deeper investigation has not disappointed me yet, and indeed, has given me that "new toy" feeling that I haven't felt in a very long time! It remains to be seen if this machine will meet my requirements for a calculator, but if it does I will be very happy for it to nudge my 50g into retirement, and to once again have a calculator that I'm not embarrassed to be seen using!

Regards,
Neil.

[nljli...@gmail.com]

> TW

Tim,

Thank you for sharing your comments about some of the design decisions that went into the keyboard layout for the Prime. I for one find that sort of thing fascinating. (And thank you for all the other information and comments you've provided about the calculator in other threads in this group.)

> Feel free to ask any keyboard related questions and I'll try to
> answer them if I can.

I was curious about the philosophy of the escape key. Coming from a 48 perspective I naturally use the ON key (in it's role as ATTN or CANCEL) as an escape key. When I was playing with the emulator, I found that sometimes ON would get me out of things but sometimes I needed to use Esc. I haven't yet finished reading the draft manual that was lying around with the emulator when I found it, but I was wondering if there is a way of thinking about it that would lead me to using the right key the first time every time. Or is it as simple as always using Esc?

Regards,
Neil.

[TW]

> I was curious about the philosophy of the escape key. Coming from a 48 perspective I naturally use the ON key

The difference here really stems from the way the underlying OS is designed. In Prime, you have the idea of view switching. A view is a base location that immediately takes your inputs and saves them. An example of this is the SYMB view, where on typing an expression the value is immediately saved. There is not "OK" to accept what has been done, or a "cancel" to revert the change. The home settings screen is another. If you compare that to MODES on the 50g, you will see that any change you make is immediate, there is no OK/CANCEL type operation, and in all it is much cleaner and consistent overall. Thing how on the 50g MODES you can open another page with CAS and they get nested together. Which OK/CANCEL goes where? What if you cancel the first page after pressing OK on the second?

Another important thing here is that a view can immediately switch and leave to another screen. If you are in home settings, you could press APPS, CAS, PLOT and so on and immediately switch and be there. On the 50g, everything really is stacked up on top of the stack screen, and you must exit first back to the stack before moving to something else. Thus in this case having a CANCEL operation exit makes sense. Everything is basically a modal interface.

Esc will always work to back you out 1 level. Cancel (on) behaves more like a real cancel - it will stop input, or cancel a modal dialog- but usually does not exit a location. Note that there are some things in Prime which are like the 50g in that they pop up a dialog on top of what you are doing. Something like the HELP screen, or a choose box is a natural fit for a cancel just closing it.

I am sure there are places that are screwed up and we'll need to find them and make the esc/cancel behaviors match the user perspective even better, but I hope I could expain at least my thought process behind things.

One thing that is clear about the 50g is that a certain variety of users never understand the way that ON exits, and why that makes sense there. I've personally seen people that have used the calculator for over a year and never new how to exit a screen and that ON doubles as a cancel... :-(

TW

[bertran...@gmail.com]

Le jeudi 8 août 2013 19:45:27 UTC+2, TW a écrit :
> [..]
> Esc will always work to back you out 1 level. Cancel (on) behaves more like a real cancel - it will stop input, or cancel a modal dialog- but usually does not exit a location. Note that there are some things in Prime [...]
> TW


I found that to be pretty true everywhere and I think it's good :)
However, on the program editor, Esc brings me back to the home screen (instead of the previous screen, being the program list).
Any chance of having that "fixed" ?

Thanks,
Adrien

[tcab]

> I found that to be pretty true everywhere and I think it's good :)
>
> However, on the program editor, Esc brings me back to the home screen (instead of the previous screen, being the program list).
>
> Any chance of having that "fixed" ?

Agreed, I often want to 'back out' one level but ESC is so drastic and always takes you to the home screen. So I have to remember where I wanted to back up to, and key the strokes to get me there again.

I think the concept of 'back' is pretty common these days, esp. in browsers - I wonder what the HP Prime design thinking was with regards to this.

Its probably the most annoying thing I have encountered in using the Prime (well, the emulator anyway). It not only affects the program list view, but also the matrix view and other similar views. It also affects the help system (though you can mitigate it by not hitting ESC and hitting Tree instead which gets you 'back' to the tree). Probably affects some other places too. Ironically the 'On' key backs you out of multi level menus, so acts a little like a back button in those situations, but that's about it.

[tcab]

> Agreed, I often want to 'back out' one level but ESC is so drastic and always takes you to the
> home screen. So I have to remember where I wanted to back up to, and key the strokes to
> get me there again.

Actually there is a workaround - kind of. That is, to accept the design decision of the strict, clean way in which ESC always takes you out of what you are doing and back into home (umm... except for backing out of submenus and menus ;) Then instead of hitting ESC from the program editor and expecting to see the list of programs, you hit "SHIFT Program" to achieve the same effect, that is, to see the list of programs.

This also works in e.g. when editing a matrix and wanting to go back to the list of matrixes view. Instead of hitting ESC and expecting to go there, you instead just hit SHIFT Matrix. That will take you out of the matrix editor and straight into the list of matrixes view. Etc.

So its a different mindset - you just tell the calculator where you want to go at all times, rather than relying on some sort of implicit history/back mechanism. Maybe that's easier for some users to understand?

[electr...@gmail.com]

I haven't read all the posts, but x^y seems an extremely unfortunate and illogical choice to me because so many users will be familiar with older HP calculators. If it is intended to have nothing to do with X, Y, Z, T, then why use the letters X and Y at all? Why not just say A^B instead?

[Joe Horn]

> I haven't read all the posts, but x^y seems an extremely unfortunate and illogical choice to me because so many users will be familiar with older HP calculators. If it is intended to have nothing to do with X, Y, Z, T, then why use the letters X and Y at all? Why not just say A^B instead?

Simple answer: That's what almost all *algebraic* calculators call it, regardless of manufacturer. Yes, they COULD have called it "^" or "A^B" or "**" or whatever... but the name which causes the LEAST confusion in an *algebraic* calculator is "X^Y" because that's what everybody else uses.

Nothing will make everybody perfectly happy, except a ham sandwich.

Proof:
Nothing is better than perfect happiness.
A ham sandwich is better than nothing.
Ergo, a ham sandwich is better than perfect happiness, Q.E.D.

-Joe-

[Erwin Ried]

Anything is better than nothing...
Ergo, anything is better than perfect happiness?

[electr...@gmail.com]

We’ll have to agree to disagree on that. I can’t see anyone getting confused to see A^B just because another calculator says X^Y. On the other hand, if other calculators (including previous from the same manufacturer) said B^A, then A^B would not be a good choice. Yes, people can learn new things which are defined the exact polar opposite from their previous definitions, but human factors principles suggest we shouldn’t make them do that if we can avoid it.

That’s just my opinion fwiw (Opinions are like ham sandwiches, or something like that.).

[TW]

> On the other hand, if other calculators (including previous from the same manufacturer)

So I see we are in agreement then since Prime continues the previous series of calculators (38,39/40,39gII) in a much closer way then any other series. ;-P

TW

[electr...@gmail.com]

ok, you got me on technicality. I still say a^b would be more user-friendly.

But if that's the worst I have to complain about, life is pretty good.