Path: dp-news.maxwell.syr.edu!spool.maxwell.syr.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!wns13feed!worldnet.att.net!24.30.200.11!news-east.rr.com!news.rr.com!news-server.columbus.rr.com!tornado.ohiordc.rr.com.POSTED!53ab2750!not-for-mail
From: J M Davitt <jdavitt@aeneas.net>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.2) Gecko/20040803
X-Accept-Language: en-us, en
MIME-Version: 1.0
Newsgroups: comp.databases.theory
Subject: Re: Why all the max length constraints?
References: <1148773922.267863.170930@j55g2000cwa.googlegroups.com>   <1148775197.195932.299730@y43g2000cwc.googlegroups.com> <1148780864.560153.47550@g10g2000cwb.googlegroups.com>
In-Reply-To: <1148780864.560153.47550@g10g2000cwb.googlegroups.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Lines: 108
Message-ID: <9X7eg.49923$P2.38771@tornado.ohiordc.rr.com>
Date: Sun, 28 May 2006 02:30:29 GMT
NNTP-Posting-Host: 24.123.195.58
X-Complaints-To: abuse@rr.com
X-Trace: tornado.ohiordc.rr.com 1148783429 24.123.195.58 (Sat, 27 May 2006 22:30:29 EDT)
NNTP-Posting-Date: Sat, 27 May 2006 22:30:29 EDT
Organization: Road Runner High Speed Online http://www.rr.com
Xref: dp-news.maxwell.syr.edu comp.databases.theory:40108

dawn wrote:
> Marshall wrote:
> 
>>dawn wrote:
>>
>>>[why all the length constraints?]
>>
>>You asked a very general question, so I'm going to respond in
>>very general terms.
>>
>>Length constraints are a physical reality.
> 
> 
> Yes, the computer needs to allocate space somehow.  Understood.  I
> don't tell it when I start a new Word document how much space to use,
> but I might get a message that I'm out of space and not be able to save
> it.
> 
> 
>>Your computer is finite,
>>and it can only take so much data before choking.
> 
> 
> Thanks, Marshall, but with explanations like that I'm worried that you
> are buying into the Fox news spin on my intellect too ;-)
> 
> 
>>And there are
>>times when software is buggy and will write data forever, filling
>>up your disk or dbms or whatever.
> 
> 
> Most assuredly
> 
> 
>>It is best if the entire system
>>not die under those circumstances.
> 
> 
> I knew I would learn something new from you ;-)
> 
> 
>>(There are also times where
>>such bad behavior is deliberate, as in denial-of-service attacks.)
>>So no matter how it appears, there are real limits somewhere.
> 
> 
> Yes.
> 
> 
>>Note that Java strings are conventionally referred to as variable
>>length, but they are in fact a maximum of 2^31 characters long.
> 
> 
> I am only referring to logical variaibility, so I think of Java String
> as logically having a similar spec to a Pick attribute.  Are you
> wishing that you had to put a max length on every declaration of a
> variable of type String?  I'm not.
> 
> 
>>Realistically, if you have a field for state/province, and someone
>>enters 6 megabytes of character data, something has gone wrong
>>somewhere.
> 
> 
> Yes, and there should be no data entry widget that permits entry of
> more than 2 characters for a state code, for example, more likely
> selecting from a drop-down.
> 
> 
>>Better to trap it around character 255 than to
>>let it just run along sucking up resources.
> 
> 
> I do want the software product (as a whole) to limit anything which has
> a conceptual limit.
> 
> 
>>There is nothing specific to RM here.
> 
> 
> In the other response I gave, I asked questions related to how those
> dbms systems where attributes have variable length seem to also be the
> ones that do not have the unordered requirement of the RM.

Apples and oranges.

   Also, might
> set processing have anything to do with it?  In other words, are you
> sure that it is just a coincidence that all existing implementations
> (flawed though they may be) of the RM work extensively with length
> constraints

This is certainly not the case.  The concept of maximum length fields
has nothing to do with the relational model.  Stop looking for
something that doesn't exist.

  while many of those (not all, for sure) that do not
> implement the RM (whether OO, XML, MV, MUMPS?...) are more inclined
> toward variable lengths.

There are, I"m sure, tens of thousands of products that "do not
implement the RM" in which field lengths are either fixed or
limited to a maximum.

> 
> --dawn
>