[MUD-Dev] Re: META: What are you looking for in this list?

[J C Lawrence]

On Thu, 24 Sep 1998 12:18:47 +0200 
Ola Fosheim Grøstad<olag at ifi.uio.no> wrote:

> J C Lawrence wrote:

>> I'm actively considering posting the next monthly MUD-Dev advert
>> (suitably reworded) to rec.games.programming, rec.games.design,
>> and possibly an IF group as well as rgm.announce/admin.

> What about the artificial life groups? (carefully reworded?)

Sure.  I'm not familiar with them.  Care to suggest some, and
applicable rewordings?
 
> But what is the main focus of the list?  Is it to promote new
> designs, to evolve existing designs (polishing the surface) or to
> aid new implementations of existing designs?  I sortof hoped that
> the main focus was to promote new designs, hopefully RADICALLY new
> (Including Caliban's roll-your-own-windows-mud-with-little-pain
> vision).

Yes.  (Translation: I'm not willing to mandate or define a priority
line through the above.  I see them all as valid and interesting
aspects of the same central thrust.  While we don't do a lot of
surface polishing or cross-port assistance, it does happen, and
tends to bi-directionally support the new-design attempts (people
roll new ideas into old bases, new designs incorporate old design
idea, new designs use new ideas etc).)

> If not, then basically my whole curriculum and the extension of it
> would be on topic? 

> I personally fail to see how discussing "Is the
> fixed dimensional array evil?" under the topic "let's call it a
> spellcraft" can possibly be on topic on a mud development list? 

While I'll freely admit that the thread has little relevance to the
Subject Heading (for shame!), as a list moderator I don't see it as
worth my time to police subject header changes beyond encouraging
them.  As for the validity of the "fixed arrays are evil" topic in
and of itself, its definitely borderline, but I count it relevant as 
far too many older MUD server designs suffer critically from poor
decisions in this area and a short discussion of it may assist
such future poor decisions being less frequent.

> Let me be somewhat more specific. Here are some issues which I
> have run into and struggled with on my own in the MUD design
> process:

> - how to represent a portable pixel for maximum efficiency 

> - how to make an abstract API for object local protocols with
> maximum efficiency 

> - how to efficiently cache and update client objects - how to
> handle exceptions (errors) within the engine 

> - how to write a flexible 2D renderer with antialiasing 

> - how to render thousands of spheres 

> - how to generate parameterized images 

> - how to make a simple real time 2D layout algorithm with
> constantly changing constraints 

> - how to organize and write clean c++ code (still working on that
> one)

> Are these issues on topic?  

There is no fixed answer to that question.  I can see versions of
all of the above (encluding the C++ coding format concerns) being
acceptable posts on MUD-Dev tho often in very particular cases.

Perhaps what is missing here is that I do not attempt to define the
topic for MUD-Dev by rule.  I define it by purpose.  Part of the
purpose of MUD-Dev is to assist the MUDding community.  As such,
forays of various lengths into outlieing fields are acceptable.
This does not mean that a long thread on coding formats or naming
styles is suitable for MUD-Dev, but quick question on the area
followed by various external references would be an acceptable
stretch of the topic definition.  Similarly, detailed discussion on
a specific BSP implementation would be off-topic, but a discussion
of the area, its concerns, advantages, difficulties and a few
specifics would be quite welcome, and a short very pointed thread on
specific problems with a specific implementation would also be
acceptable as a brief OT discussion where the knowledge base of the
membership is quickly tapped.

> And what about minimal spanning trees, traveling salesman problem,
> bucket/grid algorithms etc?  There is a sillion issues which can
> be related to MUDs...  I just hope that those with the radically
> new visions don't stop posting to this list if/when it is flooded
> with such issues, because I haven't seen them anywhere else.

<nod>

If and when it becoems a problem I'll handle as I see fit.  I don't
see this as an area which is suited to the blanket application of
rules.  So far what we have seems to work and I'm willing to
continue with it for the nonce.

--
J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...