Matt Messier (Marius of MudOS) denies LpMud Community

[somerville32]

Earlier today, Mat Messier (Marius of MudOS.org) has denied TMI-3 the
rights to continue development of MudOS. The following e-mail was sent
to Messier on June 30th:


"Dear Marius,
My name is Cody Somerville and I am the current director of the Tmi-3
Project. Tmi-3 is an umbrella project that aspires to revive the
LpCommunity through offering services such as IRC Chats;
standardization; a mutual location for research, education, and
certification; a modern and powerful lpc mudlib; and MudOS driver
development. However, to fulfil this mandate, it would be only fair if
I ask you for permission in regards to the development of MudOS as you
are the current maintainer. I understand there has been a dispute
between yourself and an affiliate of mine, Edward. I ask that you
please disregard that incident and allow us to continue the development
of MudOS in good spirit in hopes of reviving something you once
enjoyed. I thank you for your time and consideration.

Thank you,

Cody Somerville"


In response, Marius had the following to say:


"NO. Not a chance. I believe I was quite clear on this."


In a related story earlier this month, there was a dispute between the
developer of MUMS 1.0 (Formally MudOS 1.0) and Messier. To quote
specific statements made earlier (which can be found at www.mudos.org),
Messier said the following:

"In the end, I don't much care about MudOS anymore myself; otherwise, I
would still be actively maintaining it." ... " You do not now--nor will
you ever--have my blessing or permission to release _anything_ under
the MudOS moniker. Am I clear?""

The TMI-3 community said they believe that Marius is denying the
LPCommunity the ability to grow due to a personal issue with members of
their development team. Tmi-3 returned Messier's e-mail, quoting him
and saying:


"However, I am not Edward. And furthermore, TMI-3 is not affiliated
with his version of MudOS. TMI-3 simply stated that it planned to
continue to development of MudOS, how is that morally conflicting for
you? We are a totally
different party and should be treated as such. I hope you reconsider."


Though it is not clear exactly why Marius denies TMI-3's request, some
speculate that it is because Marius feels that TMI-3 tried and/or is
trying to lay claim to "Marius's zealously guarded code", noting
sarcasim. It appears that it is completly a personal attack against
specific members of the TMI-3 community, which is a real shame. TMI-3
states that they never tried to steal or lay claim to other people's
code, they simply stated they planned to continue development of MudOS.
This is a sad day in the LpMud Community when the community does't even
have control anymore. TMI-3 urges all those interested in the
development of the lpmud community and driver development of MudOS to
send an e-mail to Messier urging him to transfer maintainership to
TMI-3.

[Jon A. Lambert]

[quote]
TMI-3 urges all those interested in the development of the lpmud community
and driver development of MudOS to send an e-mail to Messier urging him to
transfer maintainship to TMI-3.
[/quote]

Why? Just who is TMI-3 and what have they done to develop the MudOS driver
that would entice the current maintainer to hand the reigns over? How many
patches have TMI-3 submitted? And when is TMI-3 going to explain the deal
with the GPL license? Who is Edison, Edward, Robert Fuller? Is not Robert
Fuller an administrator of TMI-3 and was asked not to use the MudOS name?

What is MUM and just what sort of silly stupidity is this?

[quote]
This mudlib is distributed under the terms of the General Public License,
which is included in this directory as the file Copyleft. Read that file
for details. Succintly, you may modify this source code in any way you
like, and you may freely redistribute it, subject to three major
restrictions.
First, you may not charge money for the software. Second, any modifications
you make must be clearly indicated in the source code. Third, you must
include this license agreement and the original copyright notice in any
distributions you make. The Copyleft has other provisions and you should
read it carefully before distributing this code. In the event of a conflict
between this notice and the Copyleft, the provisions on the Copyleft are the
one which apply.
The restrictions of LPMUD also apply to this release. They are more
restrictive than the provisions of the General Public License in certain
regards, and you are bound by both in your use of this software.
[/quote]

*boggles*

The GPL license forbids applying restrictions..
[quote]
6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the original
licensor to copy, distribute or modify the Program subject to these
terms and conditions. [b] You may not impose any further restrictions on
the
recipients' exercise of the rights granted herein.[/b]
[/quote]

Yet another mud and mudlib release with an invalid and completely unusable
license. And yet you have not clarified the TMI-3 flirtation with the GPL
despite being asked by several people.

So why should anyone bug the MudOS maintainer to entrust development to
people who cannot/wont even address licensing of it correctly in order to
PROTECT the people who wrote MudOS?

--
J. Lambert

[larnen]

Post by Jon A. Lambert
Yet another mud and mudlib release with an invalid and completely unusable
license. And yet you have not clarified the TMI-3 flirtation with the GPL
despite being asked by several people.
So why should anyone bug the MudOS maintainer to entrust development to
people who cannot/wont even address licensing of it correctly in order to
PROTECT the people who wrote MudOS?

As a rather old hand on these newsgroups, with no affiliation to TMI-3,
MUM or whatever and as someone who has run a mud for over 11 years now,
I would like to ask a couple of questions.

While handing over MudOS to people who have yet to justify their ability
to continue its development in a sensible fashion is prudent, it does
seem to miss the wider point that MudOS is, effectively, dead. I would
like nothing more than to see MudOS spring back to life, and while I
would rather that it be done in a structured way, handed on to someone
who has proven their credentials, we have to consider the following:

1) No such person has stood forward in years.
2) MudOS has been effectively dead for years.
3) The current maintainer has stated publically he intends no further
development.

I can't help but wonder if we are fighting hard to maintain the sanctity
of a corpse.

I would say that the following critera need to be met by any prospective
developer:

1) That development continue from the latest public release, not a spin
off from an earlier point.
2) That the prospective developers demonstrate a proven track record of
C development.
3) That they have signficant experience using MudOS.
4) That they understand and abide by the existing (albeit inconvinient)
license.

Other than that I dont think it much matters who does it. Any
resurrection is better than a protracted death. I suspect the only
reason that there are any MudOS muds left in existance is because
conversion to LDMud or DGD would be too painful. If someone comes along
with enthusiasm to kick start the project we should give them a chance
to prove their credentials, not mock and dismiss them out of hand.

Larnen

(13 years mud development, 12 years mud head admin of large, active mud,
14 years C programming and networks, in case anyone wondered ;)

[Lars Duening]

Post by larnen
I can't help but wonder if we are fighting hard to maintain the sanctity
of a corpse.

Maybe, but in the open source world respect towards the previous
maintainers is pretty much the only reward those maintainers get. I have
the impression that this whole boondoggle is less about MudOS itself,
and more about the "Gimme!" attitude of certain people.

[larnen]

Post by Lars Duening

Post by larnen
I can't help but wonder if we are fighting hard to maintain the sanctity
of a corpse.

Maybe, but in the open source world respect towards the previous
maintainers is pretty much the only reward those maintainers get. I have
the impression that this whole boondoggle is less about MudOS itself,
and more about the "Gimme!" attitude of certain people.

Absolutely, and again, my own experiences with running something for
over a decade with only peoples appreciation of what I do as reward
reinforce this.

I was broadening the subject a little - putting TMI-3 and MUM aside, how
do we best help MudOS as a project. It has a lot of excellent work gone
into it, but do we best applaud that by stagnating it, or by developing
it further, even if not in quite the way or style we would have chosen
ourselves?

Should we, as a community, have input on what happens next, or should
the current maintainer retain total control? (Thats not a loaded
question btw - Marius has more right to dictate what happens next than
any other individual other than Beek etc).

Perhaps, if Marius is just looking for the right person to hand it on
to, we should set some minimum requirements in terms of technical
ability, demonstrated project delivery etc, and then require that person
to be nominated by a percentage of the people who are old hands/active
MudOS mud maintainers/contributers etc etc

Really Id like to take the opportunity of the recent unpleasantness to
try and look forward as to how we avoid similar spats while also
ensuring the continued development of MudOS. Maybe MudOS is dead and any
development work should be integration patches for other drivers. But
whatever is decided the community should put forward its opinions and
concerns first and have a chance to comment on, if not necessarily
dictate, the path development takes.

Thanks,
Larnen

[John Viega]

Larnen,

I think it's worthwhile for the MudOS community to look for a good
person to be the active maintainer. I suggest taking it to the MudOS
list, since a lot of those people don't check here.

John

Post by larnen

Post by Lars Duening

Post by larnen
I can't help but wonder if we are fighting hard to maintain the sanctity
of a corpse.

Maybe, but in the open source world respect towards the previous
maintainers is pretty much the only reward those maintainers get. I have
the impression that this whole boondoggle is less about MudOS itself,
and more about the "Gimme!" attitude of certain people.

Absolutely, and again, my own experiences with running something for
over a decade with only peoples appreciation of what I do as reward
reinforce this.
I was broadening the subject a little - putting TMI-3 and MUM aside, how
do we best help MudOS as a project. It has a lot of excellent work gone
into it, but do we best applaud that by stagnating it, or by developing
it further, even if not in quite the way or style we would have chosen
ourselves?
Should we, as a community, have input on what happens next, or should
the current maintainer retain total control? (Thats not a loaded
question btw - Marius has more right to dictate what happens next than
any other individual other than Beek etc).
Perhaps, if Marius is just looking for the right person to hand it on
to, we should set some minimum requirements in terms of technical
ability, demonstrated project delivery etc, and then require that person
to be nominated by a percentage of the people who are old hands/active
MudOS mud maintainers/contributers etc etc
Really Id like to take the opportunity of the recent unpleasantness to
try and look forward as to how we avoid similar spats while also
ensuring the continued development of MudOS. Maybe MudOS is dead and any
development work should be integration patches for other drivers. But
whatever is decided the community should put forward its opinions and
concerns first and have a chance to comment on, if not necessarily
dictate, the path development takes.
Thanks,
Larnen

[somerville32]

I just want to say I'm sorry. I've been very arrogant and ignorant and
I completly understand where you guys are coming from. I will take your
advice and hopefuly I can fix this mess I've made. Even though alot of
us, obvisously, haven't been on the best of terms, your help and advice
has been priceless and so I thank you all for putting up with me.

~Me

[Lars Duening]

Post by larnen
Should we, as a community, have input on what happens next, or should
the current maintainer retain total control? (Thats not a loaded
question btw - Marius has more right to dictate what happens next than
any other individual other than Beek etc).

I don't think that there's an either-or in here.

Of course the community's control is limited as they aren't then ones
working on the driver - the maintainer is. On the other hand the
maintainer's control is limited as well: if the community doesn't like
his changes, his driver won't be used.

Post by larnen
Perhaps, if Marius is just looking for the right person to hand it on
to, we should set some minimum requirements in terms of technical
ability, demonstrated project delivery etc, and then require that person
to be nominated by a percentage of the people who are old hands/active
MudOS mud maintainers/contributers etc etc

If the MudOS community does it this way, I think Marius could be
convinced to give up ownership - even though it means that the driver
will take a new direction.

However, all this hinges on the one problem of having a person willing
to take up the development, knowing that that is a multi-year
commitment. Without that, everything else is just a daydream.

[John Viega]

Your original mail comes off as if you're trying to get Edison the
right to call MUM MudOS. You should be taking his response in that
light. Note that you do not make it clear whether you want to take
maintainance of MudOS as it lives today, or if you are offering to call
some other code base MudOS, which is derived from a version of the code
base that is about 10 years out of date. It's not fair to any of the
current users of MudOS to have so many of their features taken out of
the main development line. Plus, you haven't shown that you have any
of the real experience with C, with MudOS, or so on.

Also, you should stop right away trying to imply anything bad about
Marius. He's been perfectly willing to turn over the reigns of MudOS
to someone talented, motivated, responsible and mature. I'm sure those
qualities are debatable, but a lot of people don't find Edison's
actions to be those things. And ultimately, it is Marius' opinion who
counts. If you want to read Marius' judgement of Edison based on
those properties as a personal attack, then go right ahead. Or, if you
think the personal attack was against you, remember that none of us
know jack or shit about you beyond your newsgroup posts from the last
couple of months.

And, whether or not you are affiliated with Edison, opinions on you are
about even. It's amazing that you would continue to use the name
TMI-3. I mean, if you're going to post contents of private emails to
usenet, let's talk about the mail you sent Leto asking for his
permission to use the TMI name. He said approximately that he wished
you wouldn't, didn't think it was a good idea, but wasn't going to do
anything to actively stop you, more or less because he sees the whole
genre of LPMud as dead. I've made it clear that I don't think you
should be doing it, in part because of the over-entitled way in which
you go about doing things. The average person would say, "the name
isn't that important to what we're trying to accomplish... let's try to
respect the opinions of those people", and change the name. Now, you
just look like people who are just trying to leverage whatever little
recognition value the names "TMI" and "MudOS" have for your own
personal ego trip without doing anything to earn the recognition.

Considering the circumstances, you should find it no surprise that he
said "no". Plus, see the other posts in this thread so far, which make
good points... particularly with respect to you guys and licensing.

Post by somerville32
Earlier today, Mat Messier (Marius of MudOS.org) has denied TMI-3 the
rights to continue development of MudOS. The following e-mail was sent
"Dear Marius,
My name is Cody Somerville and I am the current director of the Tmi-3
Project. Tmi-3 is an umbrella project that aspires to revive the
LpCommunity through offering services such as IRC Chats;
standardization; a mutual location for research, education, and
certification; a modern and powerful lpc mudlib; and MudOS driver
development. However, to fulfil this mandate, it would be only fair if
I ask you for permission in regards to the development of MudOS as you
are the current maintainer. I understand there has been a dispute
between yourself and an affiliate of mine, Edward. I ask that you
please disregard that incident and allow us to continue the development
of MudOS in good spirit in hopes of reviving something you once
enjoyed. I thank you for your time and consideration.
Thank you,
Cody Somerville"
"NO. Not a chance. I believe I was quite clear on this."
In a related story earlier this month, there was a dispute between the
developer of MUMS 1.0 (Formally MudOS 1.0) and Messier. To quote
specific statements made earlier (which can be found at www.mudos.org),
"In the end, I don't much care about MudOS anymore myself; otherwise, I
would still be actively maintaining it." ... " You do not now--nor will
you ever--have my blessing or permission to release _anything_ under
the MudOS moniker. Am I clear?""
The TMI-3 community said they believe that Marius is denying the
LPCommunity the ability to grow due to a personal issue with members of
their development team. Tmi-3 returned Messier's e-mail, quoting him
"However, I am not Edward. And furthermore, TMI-3 is not affiliated
with his version of MudOS. TMI-3 simply stated that it planned to
continue to development of MudOS, how is that morally conflicting for
you? We are a totally
different party and should be treated as such. I hope you reconsider."
Though it is not clear exactly why Marius denies TMI-3's request, some
speculate that it is because Marius feels that TMI-3 tried and/or is
trying to lay claim to "Marius's zealously guarded code", noting
sarcasim. It appears that it is completly a personal attack against
specific members of the TMI-3 community, which is a real shame. TMI-3
states that they never tried to steal or lay claim to other people's
code, they simply stated they planned to continue development of MudOS.
This is a sad day in the LpMud Community when the community does't even
have control anymore. TMI-3 urges all those interested in the
development of the lpmud community and driver development of MudOS to
send an e-mail to Messier urging him to transfer maintainership to
TMI-3.

[Lars Duening]

Post by somerville32
Earlier today, Mat Messier (Marius of MudOS.org) has denied TMI-3 the
rights to continue development of MudOS.

Misunderstanding aside (Marius only refuses to hand over the name
'MudOS' - he can't prevent anybody from forking the project): Big
friggin deal.

Seriously - why this insistence on using 'TMI' and 'MudOS' as names? Ok,
you get the recognition effect, but only among the old hands, and...
well, don't you guys have any creativity? or ego?

If I built something great, I'd want to make sure that it bears a name
_I_ chose, not somebody else's.

Post by somerville32
This is a sad day in the LpMud Community when the community does't even
have control anymore.

Speak for yourself.

The 'LPMud Community' you are talking about is just the subset
interested in the TMI-3 revival - it's not even the whole MudOS
usership, and definitely not the LPMud community as a whole.

Don't get me wrong - I applaud your enthusiasm in reviving the spirit of
TMI and MudOS, but right now you're hung up on trivialities.

Write some great code, build a great mud, and you'll find out that it
won't matter if it's called 'MudOS' or 'Walter'.

[John Viega]

Well said. Mr. Somerville, you're perfectly welcome to fork MudOS as
long as you don't call it that. No one will complain for a second.

Post by Lars Duening

Post by somerville32
Earlier today, Mat Messier (Marius of MudOS.org) has denied TMI-3 the
rights to continue development of MudOS.

Misunderstanding aside (Marius only refuses to hand over the name
'MudOS' - he can't prevent anybody from forking the project): Big
friggin deal.
Seriously - why this insistence on using 'TMI' and 'MudOS' as names? Ok,
you get the recognition effect, but only among the old hands, and...
well, don't you guys have any creativity? or ego?
If I built something great, I'd want to make sure that it bears a name
_I_ chose, not somebody else's.

Post by somerville32
This is a sad day in the LpMud Community when the community does't even
have control anymore.

Speak for yourself.
The 'LPMud Community' you are talking about is just the subset
interested in the TMI-3 revival - it's not even the whole MudOS
usership, and definitely not the LPMud community as a whole.
Don't get me wrong - I applaud your enthusiasm in reviving the spirit of
TMI and MudOS, but right now you're hung up on trivialities.
Write some great code, build a great mud, and you'll find out that it
won't matter if it's called 'MudOS' or 'Walter'.
--
Lars Duening; lars at bearnip dot com

[Alan Schwartz]

Post by somerville32
Earlier today, Mat Messier (Marius of MudOS.org) has denied TMI-3 the
rights to continue development of MudOS. The following e-mail was sent

It's been pointed out, but I'll say it again.

Your right to develop the source code of the MudOS server is whatever
the MudOS license says it is. That license may also specify what
you're allowed to call a forked release, or it may not.

If it doesn't, you should indeed ask the developers of MudOS if they
prefer that you include the name "MudOS" in the name of the forked release
(they might want you to, to show its lineage), or if they prefer you
don't (they might not want you to, to distinguish the two codebases).
It would then be polite and customary to follow their wishes.
Although nothing legally requires you to if the license doesn't speak to
this issue and the name isn't trademarked, only a jerk would do
otherwise.

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
***@M*U*S*H (mush.pennmush.org 4201) | Alan Schwartz
| ***@pennmush.org
***@DuneMUSH, and Javelin elsewhere | PennMUSH Server Maintainer
=-------------------------------------------------------------------------=
PennMUSH God's Guide: http://www.pennmush.org/~alansz/guide.html
PennMUSH Source: http://download.pennmush.org/Source
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

[Trimaster]

I remember a time when there were several forked drivers of lpc capable
drivers and no one gave orcs butt care about the licenses.

However, I personally see no reason to not have a bunch of forked
releases again. Having them will encourage (maybe) people to make a
better driver for all the lpc mudlibs out there.

As far as you using TMI and MUDOS as names, I personally don't care and
I'm lumped in with the average person, just change the name of the mud
and the mud driver/lib forking. Considering most people do like credit
and it is enevitable that even after the forking someone will notice
the similarities you should mention at least in a history file
somewhere where the initial forking branched from or whatever.

Personally I have never heard of TMI-3 as a mudlib and I have to wonder
do you in fact actually train people to code on other muds? (I am
guessing not) If you do not, then I wouldn't call it TMI. Or even
research new ways of doing things, I ponder how many different ways you
can calculate THAC0 or if you have a truely sentient npc in your
codebase.

With that asside, let's consider what has happened and is happening,
much faster than your thread indicates; but first my creds eh? -- a
seasoned mud gamer since 92 and a Mud admin since 93 with various roles
in between including the current of mudlib consultant -- So what has
been happening. I am sure Marius and Beek are aware of it. I am sure
even Mr. Viega is aware of it. Of course the only real stink being made
is by the TMI-3 (Can't you change that name? I don't like giving you
that type of recognition, I'm betting there are no TMI or TMI-2 staff
on your staff). And I am sure our mudos friends don't like it when
others fork or spoon. heh.

So, everyone knows what I am going to say, but I will not name names,
however, I do know of several forked drivers, actively maintained
forked drivers at that. And do you want to know when that forking
began, I'll tell you, when they all decided the MudOS branch was dead.
Which brings us to the other point. Keyword: branch. MudOS simply put
is a branch. I personally was against it, grew to like it and almost
love. It's like heroine -- you only have to shoot up once then you are
hooked. MudOS is as close as I'll get to heroine. When was I against
it. I was against it around 1993 to 1994 when it first started show
it's little sapling head. Our mud Highlands was reaching maturity, We
had a driver staff and a mudlib staff. We were actively forking (see)
LPmud. We were on that tsunami wave between 2.4.5 and 3.1.2 driver/libs
(I say that because there existed drivers that could do one or the
other and both, Amylaar's is pretty popular for doing both). MudOS
occured at a time when standard LPmud was losing it's steam, beginning
to wither and show yet more signs of age and code bloat (ie, features
without the optimizations). Being the elder fart that I was
(Admin/Coder), I couldn't stand MudOS. However, that all changed for
the better. The funny thing is, I can't stand for what MudOS stands for
now, I can stand what it used to stand for however.

I do remember tho, I was like WTF! WTH is MudOS! Because they didn't
take on an old name like LuG as a moronic example for Lars uber Game,
that would've been so wrong and I respect them now for doing that, not
then, but definately now. And they do in their history file indicate
who they branched from. Also we know from all the past posting on here
that they too didn't really follow all the licensing that much.
However, we know from the history license that Lars wasn't anal either.
Ofcourse, they didn't start out with LuG either which might have made
him to be anal.

Damn if the fault doesn't lay both ways here.

Anyways, you guys need to change your mud name, it will be way easier
for your small group to get over it and fork a new driver that supports
all the libs with a different name than it is for the rest of the
community trying to afford you recognition you don't deserve. It's all
about the recognition nothing more. Because if it wasn't you would have
changed your name to "The Mud Corporation" and your driver to "Rivet"
perhaps that as indication of you making something will build upon what
is. Who knows. Just change the names and get on with it, and stop
asking for blessing from which none will come.

Now where is that driver of yours, I'm wanting to check out something
new.

[Ed]

Post by Trimaster
Now where is that driver of yours, I'm wanting to check out something
new.

The driver was to be MUM. MUM is Unofficial MudOS. This is the
MudOS-derived gamedriver that has been running the Astaria game for the
last decade. (I called it a MudOS branch until certain people had cows.
Now I call it MudOS-derived.)

My plan was to port TMI-3 lib to MUM this summer. Unfortunately, I
accidentally got a summer consulting gig which pays infinitely more than
porting TMI-3 lib.

However, I did test and bundle TMI-2 lib with MUM. The download is found
here:

ftp://mum.dyns.cx/tmi2-1.1.1+MUM-1.0.3.tar.gz

-Ed.

[John Viega]

(I called it a MudOS branch until certain people had cows. Now I call it MudOS-derived.)

It *is* a MudOS branch, from a really old version of MudOS. Call it
that all you like. You were calling your branch MudOS, which is what
upset people.

[Ed]

Post by John Viega

(I called it a MudOS branch until certain people had cows. Now I call it MudOS-derived.)

It *is* a MudOS branch, from a really old version of MudOS. Call it

And how many times are you going to beat the "really old version of MudOS"
drum? How is this different than the "official" BeekOS MudOS? After all,
it branched from the same "really old version of MudOS."

Post by John Viega
that all you like. You were calling your branch MudOS, which is what
upset people.

It IS MudOS. I am not arrogant enough to think that I've changed it so
much to justify giving it a new name. Yet, your camp insisted I rename
it, so I gladly did. I consider that permission to call it what I want
and a privilege to not have to associate it with the so-called "current"
version.

-Ed.

[John Viega]

Post by Ed

Post by John Viega

(I called it a MudOS branch until certain people had cows. Now I call
it MudOS-derived.)

It *is* a MudOS branch, from a really old version of MudOS. Call it

And how many times are you going to beat the "really old version of MudOS"
drum? How is this different than the "official" BeekOS MudOS? After all,
it branched from the same "really old version of MudOS."

There was no "BeekOS MudOS". There was BeekOS and MudOS for a time
until he was given control of the driver, at which point he merged the
two together. He was not such an asshole as to try to claim he was
releasing MudOS without differentiating the name.

Post by Ed

Post by John Viega
that all you like. You were calling your branch MudOS, which is what
upset people.

It IS MudOS. I am not arrogant enough to think that I've changed it so
much to justify giving it a new name. Yet, your camp insisted I rename
it, so I gladly did. I consider that permission to call it what I want
and a privilege to not have to associate it with the so-called "current"
version.

It is not MudOS, it is a branch of the same code base. You don't have
the situational awareness and tact of the many, many people who have
created their own MudOS branches, and their own branches of other
people's software. Get over it.

And as far as I'm concerned, slam MudOS all you like. I've personally
been of the opinion for about 10 years that MudOS is no more than a
non-robust, poorly featured dynamic programming language, with very few
exceptions (such as the parsing infrastructure). Most dynamic languages
are better choices for a mud than LPC "drivers". So your opinion of it
is probably not much different than mine. But then again, my opinion
of MUM isn't going to be any higher.

[Ed]

Post by John Viega
It is not MudOS, it is a branch of the same code base. You don't have
the situational awareness and tact of the many, many people who have
created their own MudOS branches, and their own branches of other
people's software. Get over it.

Get over what? I'm not the one foaming at the mouth. I suggest you worry
about yourself.

[John Viega]

Post by Ed

Post by John Viega
It is not MudOS, it is a branch of the same code base. You don't have
the situational awareness and tact of the many, many people who have
created their own MudOS branches, and their own branches of other
people's software. Get over it.

Get over what? I'm not the one foaming at the mouth. I suggest you worry
about yourself.

I've got nothing to worry about... You've clearly got a chip on your
shoulder over the whole issue. It may be Matt's actions, but I
actually assume it's a 12 year old chip against me. You clearly feel a
need to take a little dig every time you post on anything, no matter
how misguided (thus, this thread coming back to life).

I may be happy to spend a few minutes each day responding to you, but
it's not actually because I'm hung up on it. You seem to have a
long-standing grudge, but personally, I long, long since moved on from
whatever friction was between us at TMI-2. I wish you nothing but the
best of luck, with your game driver, with your mud, and whatever else
you choose to do in life. I've moved way on to other things, and I
hope you have too.

The only reason I've been posting (beyond the late night diversion) is
because I still have some residual emotional attachment to some of the
people and to the work. I personally think, in the context of this
whole MudOS thing, it wasn't a good idea to call the thing MudOS. Your
explanation is certainly different from how it looked. The perception
led people to be extremely upset at first, myself included. Once you
changed your software's name, everyone was fine with your release,
etc., and I think I personally tried to make that clear.

Pretty much the only thing that keeps me posting is you taking your
digs on people I respect (some friends, some not), and maybe the
occasional thing I perceive as a factual error. It's not actually
anything personal against you (I don't think there was ever a point
where I was intentionally trolling). In the threads I haven't tried to
attack you personally, or make anything I would perceive as a dig (I am
sure I have failed for that occasionally, such as in saying "get over
it" and for that I apologize). I have tried to be respectful to you,
wishing you luck once you changed the name, and so on. I don't think
I've been trying to flame you or disrespect you, but I definitely see
that you *feel* I am. There are certainly one or two places where
I've exhibited just a tiny bit of frustration with you, as you
continually refuse to acknowledge other people's feelings as reasonable
feelings and act like it's an indellible truth that you are in the
right. But it doesn't really get to me, cause I think I understand the
reason why you act like that, and your opinion of me (or Hollebeek, or
Messier, or ...) doesn't matter to me. This is, in many ways, a
reasonably fun trip down memory lane for me, maybe a bit of fun, and
not much more. I'm not trying to be a huge thorn in your side, or
anything like that.

Let me be clear that I do not hold any grudge against you. As for the
TMI quibbling that I suspect is the cause for the continual hostility,
I don't even remember most of the details. I don't have a vested
intrest in muds anymore, haven't for most of a decade. I've moved on,
and I hope you have too. Of course, you're free to hate me, I'm not
trying to change that, and I'll sleep well at night either way. I'm
just trying to let you know that I don't hate you. In fact, I hope
you're happy and successful, personally and professionally.

[Lars Duening]

Post by John Viega
And as far as I'm concerned, slam MudOS all you like. I've personally
been of the opinion for about 10 years that MudOS is no more than a
non-robust, poorly featured dynamic programming language, with very few
exceptions (such as the parsing infrastructure). Most dynamic languages
are better choices for a mud than LPC "drivers".

Now I'm curious: which languages, and why?

[John Viega]

Lars,

If I were to build a mud today, I might look at BOO, because it has
many of the advantages of languages like Python, plus static type
safety (inferenced), which I value for projects as large as a mud.
Python would be another reasonable choice for me. It's not type safe,
but it's got tremendous libraries and is pretty easy to read and write.
Off the top of my head, I'm not 100% sure that the security model we
used to use (which is very Java-like) could be implemented in Python,
but that wouldn't be a concern for me anyway. Of course, this is all
theoretical, as I'm doing nothing of the sort :) And using a different
language may not be the right choice for other people... it's just what
I would do (i.e., I wouldn't use LPC for a new mud).

The mud-specific functionality in LPC drivers can, in general, be
easily duplicated. For example, under a bizzare set of circumstances
that aren't worth explaining, I ended up building a much more powerful
parsing infrastructure than the MudOS system in about 6 hours, in
Python (mainly because it was going to be too much work to rip out the
parser code and bind it to Python... this was for a little one-off that
was in use for a weekend). Then there were lots of little things about
LPC that, to me, were more kludged than anything (and annoying), such
as the object model, the type system and the poor exception handling
capabilities. Of course, you could improve on those things (and people
have in some cases), but other languages already have adequate
solutions here, so I would use them, since the rest of the
infrastructure is so easy to provide (and there was a big effort to
expel a lot of it from the driver anyway, in at least a couple of
drivers).

For example, let's talk about the type system. It wants to be staticly
type checked, but can't statically type across objects. If I am happy
with dynamic typing, then having to add the syntax is a pain in the
neck. If I want static typing, then I'm not getting a reasonable
degree of type safety across objects, and so why bother? If I want
mostly static checking with the ability to occasionally circumvent the
system, I would want to do this in a more intelligent manner than on
object boundaries (only on a rare, as-needed basis). And, finally, if
I do want static typing, I personally would want the language to do as
much of the typing as possible for me (i.e., type inferencing),
particularly if I wanted novice programmers to be able to deal with
this stuff (I know understanding the type system was a huge stumbling
block for novice LPC coders when I was doing this stuff).

I hope I didn't offend those still doing driver work :) I have a lot
of respect for you and LDMud, which I did look at sometime recently,
simply out of curiosity. I know the level of effort involved in
maintaining a programming language. I just think there's an advantage
to having much larger user communities.

John

Post by Lars Duening

Post by John Viega
And as far as I'm concerned, slam MudOS all you like. I've personally
been of the opinion for about 10 years that MudOS is no more than a
non-robust, poorly featured dynamic programming language, with very few
exceptions (such as the parsing infrastructure). Most dynamic languages
are better choices for a mud than LPC "drivers".

Now I'm curious: which languages, and why?
--
Lars Duening; lars at bearnip dot com

[Lars Duening]

Post by John Viega
If I were to build a mud today, I might look at BOO, because it has
many of the advantages of languages like Python, plus static type
safety (inferenced), which I value for projects as large as a mud.

Just took a quick look: interesting!

Post by John Viega
[ valid points snipped ]
I hope I didn't offend those still doing driver work :) I have a lot
of respect for you and LDMud, which I did look at sometime recently,
simply out of curiosity.

Thanks! And no, I didn't take offense - after all I asked for it :-)

I actually agree with pretty much all of your points; and I have long
realized that a lot of things people would like to see in LPC would
transform it into a general-purpose scripting language, of which there
are certainly enough. Given enough time, I could probably come up with a
way to streamline LPC without making it a PythonPerl-Clone, but time is
currently at a premium.

However, I think that LPC has at least one feature over other languages
for writing Muds: the ability to set resource limits (evaluation time,
array size) and execute in a sandbox. I have yet to see any other
language offering a similar feature (maybe I didn't look sharp enough),
and I know that it is a valuable safety net for existing muds.

Post by John Viega
I know the level of effort involved in
maintaining a programming language. I just think there's an advantage
to having much larger user communities.

My user community may not be large, but is very engaged, so it evens
out.

[John Viega]

Lars,

Post by Lars Duening
However, I think that LPC has at least one feature over other languages
for writing Muds: the ability to set resource limits (evaluation time,
array size) and execute in a sandbox. I have yet to see any other
language offering a similar feature (maybe I didn't look sharp enough),
and I know that it is a valuable safety net for existing muds.

Well, many popular high-level languages have some "restricted
execution" facilities that allow you to execute in a sandbox. I never
expected that the Python one worked all that well, just because of the
nature of Python (thus my concern about whether you could implement a
good security model), but I'm pretty sure that you could build a good
restricted execution layer in BOO if one isn't already done. I also
wouldn't be against contributing to Python and moding the language
itself if the functionality were needed.

As for evaluation time, this is easy to do and more useful to do with
threads. It is relatively easy to have your main game thread killed
when it runs for too long. This also allows people to specify their
max eval cost in terms of how long they are willing to let the mud hang
if someone messes up with a while(1), instead of specifying in terms of
number of virtual instructions, where those instructions aren't even
constant time.

As for array size, and so on, I don't see people accidentally making 2
gig arrays too often. That kind of limitation has only ever been a
frustration, in my experience. That is, I don't think there's much
cause to try to prevent accidents, and what's in place doesn't stop
intentional problems, because I can just alloc tons of things that bump
up against the maximum size if I want to have the mud eat up too much
memory. The process itself can clearly be constrained at the OS level,
protecting other processes, and that is probably enough.

I'm glad to hear you still have an enthusiastic and active community!

John