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

Re: Re: Re: error with Sum and Infinity

• To: mathgroup at smc.vnet.net
• Subject: [mg102488] Re: [mg102409] Re: [mg102387] Re: error with Sum and Infinity
• From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
• Date: Tue, 11 Aug 2009 04:03:45 -0400 (EDT)
• References: <h5bk64$hlm$1@smc.vnet.net> <200908070932.FAA15211@smc.vnet.net> <29EAD82D-4D06-4036-923E-97DD7B2EDC46@mimuw.edu.pl> <4A7C6176.9060002@cs.berkeley.edu> <200908080838.EAA01301@smc.vnet.net> <op.uyfiuix8tgfoz2@bobbys-imac.local> <4031E1F9-B399-4387-A2E7-FB9CBA19F824@mimuw.edu.pl> <op.uygog8nytgfoz2@bobbys-imac.local>

Even if they don't, (almost) nobody will notice any difference.  
Presumably that is why they are undocumented.
Andrzej

On 11 Aug 2009, at 04:30, DrMajorBob wrote:

> If these options (and numerous other features) are not documented,  
> does WRI have any commitment to leaving them alone?
>
> If not, aren't we building our houses on sand?
>
> Bobby
>
> On Mon, 10 Aug 2009 01:49:39 -0500, Andrzej Kozlowski <akoz at mimuw.edu.pl 
> > wrote:
>
>> Yes, its undocumented. So is everything else (I think) you see when  
>> you evaluate
>>
>> SystemOptions[]
>>
>> On the other hand, SystemOptions and SetSystemOptions is documented.
>>
>> I don't know if SystemOptions should be documented or not. I  
>> explained my reasons in an off-list discussion with a certain  
>> mythical scandinavian creature which from time to time visits this  
>> list, but as I do not feel like repeating it all here I will only  
>> say that while I would personally like to see more documentation I  
>> recognize that it would take quite a lot of time and effort to  
>> produce it and most of it would probably be found useful by less  
>> people than it would take to write it.
>>
>> Andrzej
>>
>>
>> On 10 Aug 2009, at 13:31, DrMajorBob wrote:
>>
>>> "SymbolicSumThreshold" is interesting... but undocumented. We  
>>> cannot search for it with ?, and we cannot search for it in  
>>> DocumentationCenter.
>>>
>>> Even now that I've seen your statement "SymbolicSumThreshold" /.  
>>> SystemOptions[], I have no proof it means what you suggest it means.
>>>
>>> Not that I won't take your word for it, mind you... but that's not  
>>> documentation.
>>>
>>> Bobby
>>>
>>> On Sat, 08 Aug 2009 03:38:55 -0500, Andrzej Kozlowski <akoz at mimuw.edu.pl 
>>> > wrote:
>>>
>>>>
>>>> On 8 Aug 2009, at 02:16, Richard Fateman wrote:
>>>>
>>>>> Andrzej Kozlowski wrote:
>>>>>>
>>>>>> On 7 Aug 2009, at 18:32, Richard Fateman wrote:
>>>>>>
>>>>>>> Well, see Elton TeKolste's remarkable post.
>>>>>>> I doubt that the "feature" he illustrates would be known even by
>>>>>>> most
>>>>>>> experienced users.
>>>>>>
>>>>>>
>>>>>> Well, you are wrong. For example, check this post (about the
>>>>>> fastest way of adding up a billion numbers)
>>>>>>
>>>>>> http://forums.wolfram.com/mathgroup/archive/2007/Mar/ 
>>>>>> msg00565.html
>>>>>>
>>>>>> and the rest of the discussion.
>>>>>
>>>>> I think you flatter yourself and mathgroup to think that "most
>>>>> experienced users" will have read your post from March, 2007.
>>>>
>>>> Are you saying that there are "experienced users" who have not read
>>>> all my posts? Impossible!
>>>>
>>>> However, should such a strange phenomenon really exist, he or she  
>>>> can
>>>> always evaluate
>>>>
>>>> "SymbolicSumThreshold" /. SystemOptions[]
>>>> 1000000
>>>>
>>>> which, I think, deals with the rest of your post.
>>>>
>>>> Andrzej
>>>>
>>>>
>>>>
>>>>> I just checked the documentation and I see no notice of that magic
>>>>> number  (though maybe it is there somewhere and I missed it??).
>>>>>
>>>>>
>>>>> Is there a way of finding that magic number without  
>>>>> experimentation
>>>>> or "insider" knowledge?  Would "most experienced users" know that?
>>>>>
>>>>> If it is not in the documentation, the rules of the game mean that
>>>>> WRI is free to change that magic number, or eliminate it, without
>>>>> notice, so any experienced user would be loathe to take  
>>>>> advantage of
>>>>> it for fear that any program utilizing it would cease to work in a
>>>>> new version.
>>>>>
>>>>> RJF
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>> --DrMajorBob at bigfoot.com
>>
>
>
>
> -- 
> DrMajorBob at bigfoot.com

• References:
  • Re: error with Sum and Infinity
    • From: Richard Fateman <fateman@cs.berkeley.edu>
  • Re: Re: error with Sum and Infinity
    • From: Andrzej Kozlowski <akoz@mimuw.edu.pl>