[Date Index] [Thread Index] [Author Index]

Re: Re: Inequality not documented in 6.0

• To: mathgroup at smc.vnet.net
• Subject: [mg90081] Re: [mg90065] Re: Inequality not documented in 6.0
• From: Jack L Goldberg 1 <jackgold at umich.edu>
• Date: Sat, 28 Jun 2008 05:52:26 -0400 (EDT)
• References: <200806271020.GAA24405@smc.vnet.net>

I beg to disagree.  One place where a beginner could run into trouble 
is pattern matching. I did and I did so precisely on this issue.  
Moreover, I fully disagree with the judgment that the documentation is 
adequate. In my opinion it is far from adequate, it is basically 
non-existent in the sense that most people using Mathematica would 
accept as "adequate" documentation. This is my opinion of course, but 
I want to add again that I ran into this issue and puzzled over it for 
far too long.


Quoting Bill Rowe <readnews at sbcglobal.net>:

> On 6/26/08 at 5:34 PM, drmajorbob at att.net (DrMajorBob) wrote:
>
>>> I suppose you aren't counting the description returned by
>>> evaluating,
>
>>> ?Inequaltity
>
>> The result of that is:
>
>> "Inequality represents a sequence of relational statements."
>
>> Personally, I do NOT count that as documentation,
>
> =46ine. I don't care to argue this specific point
>
>> There's not even a "More" button.
>
> Right. And I believe there is a reason for this fact.
>
>> How would one type ?Inequality in the first place, if one didn't
>> already know it existed, with precisely that name?
>
> Better question. Why would you want to do that except as an
> attempt to learn about what is returned by
>
> FullForm[x>y>=z>a]
>
> If you think about the documentation for Greater, GreaterEqual
> etc., it makes sense for these to have heads identical to the
> symbol and a separate page since they can take an arbitrary
> number of arguments, i.e., x>y or x>y>z or x>y>z>a etc.
>
> But this scheme makes no allowance for something like x>y>=z>a.
> Clearly, something more is needed to allow for this possibility.
> But I think it is equally clear the sequence of inequalities is
> far more natural to enter as x>y>=z>a than Inequality[x > y, y
>  >= z, z > a]. Using FullForm it is easy to verify both are the
> same. So, given the far more natural entry possible, what need
> is there for more documentation than what is returned by ?Inequality
>
>> And how does a newbie know that?
>
> I would not expect a newbie to know that. In fact, I would not
> be surprised many experienced Mathematica users would think to
> type in ?Inequality without first having seen Inequality in an
> expression they were examining by using FullForm. But I do think
> the documentation that does exist is adequate in this particular
> case for the reasons given above.
>
> I am not suggesting I think Mathematica documentation is
> complete, clear and cannot be improved. I do think this is not
> an example of an instance where the documentation is incomplete,
> unclear or needs improvement.
>
>
>
>

• References:
  • Re: Inequality not documented in 6.0
    • From: Bill Rowe <readnews@sbcglobal.net>