Re: LPMOO Questions

• Date: Thu, 30 Nov 1995 10:18:15 PST
• From:
  mpauker@peganet.com (Matt Pauker)
• Cc: moo-cows@parc.xerox.com
• Content-Type: text/plain; charset="us-ascii"

At 2:24 PM 11/30/95, Jason Schlauch (LPCR) wrote:
>On Thu, 30 Nov 1995, Brandon Gillespie wrote:
>
>> On Thu, 30 Nov 1995, Nate Massey wrote:
>> > > Binary DBs (Which LPMOO uses) are much larger because of their nature.
>> >
>> > What of their nature makes them bigger?  I think that LPMOO should do a
>> > good job of making smaller and better binary databases, even though it
>> > does not.  If someone can explain to me why they are so much bigger I'd
>> > appreciate it.
>>
>> They don't bother to pack data on disk or in mem, and their
>> representations of data structures and the like are more bloated than
>> those which MOO has (actually, its more probable that its bloated simply
>> because its emulating MOO).
>
>Correct me if my novice programming knowledge is wrong, BUT:
>Binary representation of data will always be larger than ASCII because
>the binary is a straight dump of what was in RAM to disk.  As a result
>all things are expanded.  Dumping to text forces a pack of the data
>because it is stored in ASCII instead of larger sized binary information.

I don't claim to be an expert on such things (Far from it, in fact), but
this seems pretty correct to me.  For example, in ASCII form, to represent
the letter 'A', it's simply an 'A'.  In binary, though, it's 00200001.  (If
that's wrong, which it probably is, sorry...I know that adds up to 65,
which is the ASCII value for 'A', but I'm not sure if there's extra
information to denote a character.)  00200001 is 7 'characters' longer than
'A', which could account for the size of binary files.

If this is all wrong, just disregard :)

Matt Pauker

• Re: LPMOO Questions
• From: Kipp the Kidd <mcmudge@mailbox.syr.edu>

• Prev: Re: LPMOO and binary db sillyness
• Next: Re: LPMOO Questions
• Index: MOO-cows
• Thread: MOO-cows