[MUD-Dev] Text Parsing

Albert thecheezeman at earthlink.net
Fri Jun 4 14:27:40 CEST 1999


At 01:57 PM 6/3/99 -0700, you wrote:
>At 07:56 AM 6/3/99 -0500, you wrote:
>>On Thu, 3 Jun 1999, Caliban Tiresias Darklock wrote:
>>> On 10:02 PM 6/2/99 -0700, I personally witnessed Matthew Mihaly jumping up
>>> to say:
>>> >
>>> >Well, that makes sense, except that in my experience, unless you provide
>>> >such a thing flawlessly, the users complain. The lack of consistency
would
>>> >probably make it far more maddening than having a parser that is less NLP
>>> >oriented. I think that is what Caliban was pointing out.
>>> 
>>> Yeah, it was, but I didn't say it quite so clearly. That happens when I
get
>>> on subjects I'm passionate about. ;)
>>
>>Of course, if you take this idea to its logical extreme, why build a mud
>>at all?  Unless you can make it a perfect mud, you're just going to get
>>complaints from users...
>
>The idea is to minimize complaints. Why institute a system, such as a
>parser that is mostly consistent with some glaring inconsistencies, that
>will just cause more complaints?
>--matt

Yes, complaints will inevitably occur, and in the volumes. But I believe that
the simple of answer that implementing a full scale NLP while at the same
time mainitaining high efficiency on a computer with limited resources simply
isn't plausible. If the NLP route is not the only option give to players,
they'll
probably end up grumbling for a week or two and continue on their merry
way, making use of what they do have. Thereafter complaints would only
arise from newbies.




_______________________________________________
MUD-Dev maillist  -  MUD-Dev at kanga.nu
http://www.kanga.nu/lists/listinfo/mud-dev




More information about the mud-dev-archive mailing list