# Re: compound propositions

From: Bob Badour <bbadour_at_pei.sympatico.ca>
Date: Thu, 18 Mar 2010 18:55:26 -0300
Message-ID: <4ba2a0d9\$0\$12442\$9a566e8b_at_news.aliant.net>

paul c wrote:

```> Bob Badour wrote:
>
```

>> paul c wrote:
>>
>>> Bob Badour wrote:
>>>
>>>> paul c wrote:
>>>>
>>>>> Bob Badour wrote:
>>>>>
>>>>>> paul c wrote:
>>>>>>
>>>>>>> David BL wrote:
>>>>>>> ...
>>>>>>>
>>>>>>>> This boolean valued function can be said to represent a
>>>>>>>> predicate under an interpretation but I'm not sure if that's
>>>>>>>> what you
>>>>>>>> mean. More specifically, what do you mean by "satisfy" when you
>>>>>>>> say
>>>>>>>> relations satisfy predicates?
>>>>>>>> ...
>>>>>
>>>>> ...
>>>>>
>>>>>>> match the variable names apparent in the predicates and the
>>>>>>> attribute types are applicable for whatever manipulations (eg.,
>>>>>>> aggregation) the predicate states.
>>>>>>
>>>>>>
>>>>>> In other words, the extension of a predicate is the set of all
>>>>>> tuples that satisfy the predicate.
>>>>>> ...
>>>>>
>>>>> Yes, but David B asked what 'satisfy' means.
>>>>
>>>> In that case, I suggest you not shy away from equality and boolean
>>>> truth values. Unless you can think of a situation where "satisfy"
>>>> means something other than "predicate evaluates to true".
>>>
>>> I don't know why the fuss about the word 'satisfy'! Admittedly its
>>> casual but some big names use it from time to time. I certainly
>>> wasn't trying to alter anybody's vocabulary but I like it because it
>>> encourages me to distinguish header from value which helps me think
>>> concretely about implementation. I just don't see the usefulness of
>>> repetitious acknowledgement that 'it is always true that there is a
>>> set of featherless bipeds'.
>>
>> I, too, find your use of the word sloppy, because a relation is a set
>> of things that satisfy a predicate. The relation, itself, doesn't
>> satisfy the predicate. Extent or extension is well-defined as a set of
>> instances and describes what a relation is: a set of instances that
>> satisfy a predicate.
>>
>> I am not suggesting you never use the word "satisfy". I'm merely
>> suggesting more careful use and a different approach when asked for a
>> definition.
>> ...
```>
> I didn't give a formal definition and didn't pretend to (my disclaimer
> was snipped).

```

Which is what opens you to questions like David BL's and to criticisms like mine. Received on Thu Mar 18 2010 - 22:55:26 CET

Original text of this message