What's missing from other IF systems?
Daniel Ellison
is better. I'm just curious about some things. I've been a fan and "user"
of IF for decades; I bought the original Zork just after it came out. In
fact, I bought a floppy drive for my Apple ][ clone specifically to play
Zork. I didn't buy Apple DOS until much later.
I've meant for years to get into creating IF because I've always found the
idea of describing a world in code then walking around in it endlessly
fascinating. I've read this group on and off for years. I know there are a
few dominant systems: Inform, Hugo, TADS, et al. I've even read that Inform
DM everyone's going on about these days. I've seen other systems come in
(e.g. wolf's PAWS) and be met with a resounding shrug.
I'd like to know what's missing in these systems that don't make it?
(Sorry, wolf). Is Inform the most widely used because it was first? I've
found a few other systems that have quite a bit of potential. There's one
called Python Universe Builder (PUB) that allows you to create works of IF
in Python. Yes, I know that's what PAWS does, but there's a different
intent with PAWS: it's meant to enable the non-programmer to use Python with
little knowledge of its internal doings. PUB is different. It provides the
parser, many IF-type objects and a verb system, but doesn't attempt to hide
the fact that you're programming.
Why wouldn't such a system catch on? What's missing? It seems to provide
the most flexibility. As wolf has pointed out on numerous occasions Python
has been ported to almost as many systems as the Z-Code interpreter. I'd
like to settle on one system - one that's very expandable - but I don't want
to expend great effort only to find out there's a very good reason people
stick with <insert favourite major IF system here>. A Python-based system
is attractive to me mainly because I AM a programmer by trade (well, I've
been Peter-Principled into managing, actually) and enjoy the ability to
expand the system in any way I see fit.
Before this gets overly long I'll stop here and send it off. I'll clarify
if need be. Thanks for any input people may have!
Daniel Ellison
Toronto, Ontario
dan...@syrinx.net
L. Ross Raszewski
>
>I'd like to know what's missing in these systems that don't make it?
>(Sorry, wolf). Is Inform the most widely used because it was first? I've
>found a few other systems that have quite a bit of potential. There's one
>called Python Universe Builder (PUB) that allows you to create works of IF
>in Python. Yes, I know that's what PAWS does, but there's a different
>intent with PAWS: it's meant to enable the non-programmer to use Python with
>little knowledge of its internal doings. PUB is different. It provides the
>parser, many IF-type objects and a verb system, but doesn't attempt to hide
>the fact that you're programming.
Well, Inform certainly isn't the first; there were adventure creation
systems available as far back as the 8-bit days. Among the modern
computers, AGT predates inform by a lot, and enjoyed success as the
"dominant" system for several years. TADS also predates inform (though
every time I read the history, I get a different answer as to "by how
much")
The main reason that the "big three" persist as the "big three" is
that the overwhelming majority of the other systems, well, suck.
The "unpopular" systems generally tend to cluster into one of two
categories:
1. Systems which emphasize minimal programming skill
These systems are generally offered (and usually quite arrogantly) as
being able to "fix" the fact that authoring a work of IF requires
substantial programming skill. They tend to be ultra-high-level
languages which rely on an exceptionally savvy interpreter to do most
of the work.
THe problem with a language that abstracts away the "nitty gritty" is
that it has to be *exceptionally* good at it. Writing a high level
language that doesn't suck is much harder than writign a middle-level
language that doesn't suck, simply because you have to allow, in your
high-level code, for the same range of functionality. An "ideal"
hig-level language can do everything a "lower level" language can. In
a lower level language, functionality is available -- you can do
everything the underlying system is capable of (though hard things are
hard). In a higher level language, the language has to provide the
same functionality. For example, in Inform, if all else fails, I can
get any effect the Z_machine is capable of by diving straight down to
assembly. You cant' do that in a high-level language, so the author of
the system has to make sure that "all else" *doesn't* fail.
But as far as I know, this has never happened. WHen you write an IF
game in a "ubereasy" language, what you're restricted to is, in
essence, writign games which are, at some level, just arrangements of
the (very small) set of things which the *system author anticipated* that you
would do.
AGT is notorious for this. No matter how much effort you put in, every
AGT game is, to some extent, restricted to having the general feel of
Zork. A killed NPC always disappears into greasy black smoke, for
instance. All NPCs are either "prevent you from leaving the room until
you kill it" or "follow you around like a puppy". All locks are of
the "find key. unlock lock with key" variety, etc.
These languages generally are designed so that it is only possible (or
only worthwhile -- you can work around the restrictions of AGT that i
mentioned above by special-casing everything, but then the language
stops being "easier than" anythign at all)to write *one* very complex
game -- the game the system author envisioned the system writing --
and a number of fairly primitive games.
Category 2 of the "unpopular" systems are more complex systems. But
they are usually, though not invariably, "one-trick ponies". These
systems are usually offered up (again, very arrogantly) as a brilliant
new way to "fix" some specific thing that's considered "broken" in the
extant systems.
This one thing, the system does very well. But the people who offer up
these languages are almost invariably fully aware of the complexity of
writing a fully operational IF language. The languages they produce
have parsers nowhere near as powerful as those of Inform and
TADS. Critical considerations are totally unaddressed, because it
didn't occur to the system author to address them.
Writing a fully-functional IF-language is **hard**, so when someone
claims to have done it, he's met with skepticism. Usually, the
skepticism is well-founded.
For a new IF-language to really take hold, it has to:
1. Be fully-functional. Which is hard. The number of non-big-three
systems which aren't seriously crippled can probably be counted on one
hand.
2. Offer something that the extant systems don't. A new system "just
for the sake of variety" isn't going to attract current users away
from their system of choice, and new users aren't liable to pick a
system without a veteran following.
3. Have a few good games out. Really good games. Had Curses and Jigsaw
not been so popular, Inform would, very likely, still be considered a
fledgelign language, probably not as powerful as TADS. The IF
community usually demands "proof of concept", because writing a game
is a massive undertaking. You're not liable to invest the time in a
system unless you've already seen that it can do something comparable
to what you want. (Consider the lack of use of Inform's z6 format --
no one uses it because no one has used it. Few people are going to
invest time in v6 because, lacking extant v6 games to look at, they
can't be sure that a v6 game they produce will be up to snuff, or even
be playable (since interpreter development lags behind game release.))
Kevin Bracey
lrasz...@loyola.edu (L. Ross Raszewski) wrote:
> (Consider the lack of use of Inform's z6 format --
> no one uses it because no one has used it. Few people are going to
> invest time in v6 because, lacking extant v6 games to look at, they
> can't be sure that a v6 game they produce will be up to snuff, or even
> be playable (since interpreter development lags behind game release.))
Except that's not really true - there are four major Infocom games that
use V6, and they are playable on a number of interpreters. Someone tackling
a new V6 project would probably prefer to use Blorb, but Blorb support is far
from widespread - they'd probably have to produce dual Blorb/classic Infocom
versions of the graphics (much as Jigsaw came in dual V5/V8 versions). But
there are plenty of people (myself included) who would happily assist
technically in such an endeavour.
I'm sure the main reason there are so few graphical games is that someone has
to actually produce the graphics - a somewhat different skill from either
writing or programming.
--
Kevin Bracey, Principal Software Engineer
Pace Micro Technology plc Tel: +44 (0) 1223 518566
645 Newmarket Road Fax: +44 (0) 1223 518526
Cambridge, CB5 8PB, United Kingdom WWW: http://www.pace.co.uk/
Jonadab the Unsightly One
> Is Inform the most widely used because it was first?
AGT was the most widely used because it was the first to be
readily available for a PC. It's almost dead now, because
of its limitations.
TADS was subsequently the most widely used, because it
was (and is) much more powerful than AGT.
Inform originally gained an advantage over TADS in the
widely-used category because it was free (TADS is now
free as well, but it was shareware originally) and because
it uses the z-machine (which was considered a major selling
point) and because the DM was (and is) particularly good.
After TADS became free and introduced HTADS it experienced
growth; if it hasn't overtaken Inform, that's probably in
large part because of the DM and z-machine factors, and
maybe some residual momentum.
Hugo has never managed to displace TADS and Inform as
the most-used system at least partly because of momentum:
lots of people use TADS and Inform; if new users select
those systems, they're sure to be able to get lots of
help on raif. And it took the release of Guilty Bastards
to overcome the "nobody uses Hugo" momentum enough just
to get it third place; before that it was listed with
AGT and Alan and all manner of even less-known systems.
Someone else used the term "proof of concept" for this
phenomenon and cited Curses for Inform, and I agree.
New systems need a major killer game to get them off
the ground; otherwise they tend to wallow in obscurity.
I personally got interested in Inform after playing
Curses, which my college roommate downloaded (in a ZIP
archive along with a DOS port of the ITF terp -- yeah,
some of us remember ITF) from a collection of games on
a big popular ftp site (I forget which one) circa 1995.
- jonadab
Alex Watson
> > (Consider the lack of use of Inform's z6 format --
> > no one uses it because no one has used it. Few people are going to
> > invest time in v6 because, lacking extant v6 games to look at, they
> > can't be sure that a v6 game they produce will be up to snuff, or even
> > be playable (since interpreter development lags behind game release.))
>
> Except that's not really true - there are four major Infocom games that
> use V6, and they are playable on a number of interpreters.
I believe the point was that there's no source code to look at from Infocom
games, so if someone wants to produce a v6 game, they have to work at it
from scratch, and using Zcode v6 for this is very hard. There is, of
course, the issue of creating graphics, but if this is so important, how
come there are several graphical TADS games, and possibly a few Glulx ones
in the works?
--
Alex Watson
http://www.watson1999-69.freeserve.co.uk/froup/
Replies to me[AT]watson1999-69.freeserve.co.uk
"Programmer - A red-eyed, mumbling mammal capable of conversing with
inanimate objects."
David Betz
was never used much because I didn't do a very good job of documenting it or
supplying a decent sample game. It also used Lisp syntax (like ZIL used by
Infocom) and some people don't like Lisp.
"Jonadab the Unsightly One" <jon...@bright.net> wrote in message
news:3b8e60ad...@news.bright.net...
Jon Ingold
Infocom
> games, so if someone wants to produce a v6 game, they have to work at
it
> from scratch, and using Zcode v6 for this is very hard.
True enough... I was looking into porting "My Angel" over to v6 so I
could make it a little easier to read, and put the status line at the
bottom, but even with the manual open I couldn't get it to do anything I
needed.
On the subject of which .. does anyone know how to tell how big a V6
"unit" is in pixels? It seems that since the font height is only
calculable in pixels, and the window size in units, there's a problem
with trying to make a two-line-big window, which is what I need. Any
help?
Jon
jpe...@jczorkmid.net
> In message <torcos5...@corp.supernews.com>
> lrasz...@loyola.edu (L. Ross Raszewski) wrote:
>> (Consider the lack of use of Inform's z6 format --
>> no one uses it because no one has used it. Few people are going to
>> invest time in v6 because, lacking extant v6 games to look at, they
>> can't be sure that a v6 game they produce will be up to snuff, or even
>> be playable (since interpreter development lags behind game release.))
> Except that's not really true - there are four major Infocom games that
> use V6, and they are playable on a number of interpreters. Someone tackling
> a new V6 project would probably prefer to use Blorb, but Blorb support is far
> from widespread - they'd probably have to produce dual Blorb/classic Infocom
> versions of the graphics (much as Jigsaw came in dual V5/V8 versions). But
> there are plenty of people (myself included) who would happily assist
> technically in such an endeavour.
I've been unable to create graphics in 'classic' format. If there was
a tool to do this then I'd have a V6 game out by now (I'm not saying
if such a took popped up tomorrow I'd have one out next week).
Jay
--
Jason C Penney (jpe...@jczorkmid.net) Xarton Dragon -=<UDIC>=-
<http://www.jczorkmid.net>
"Time and tide melts the snow man." --The Doctor
jpe...@jczorkmid.net
>> I believe the point was that there's no source code to look at from
> Infocom
>> games, so if someone wants to produce a v6 game, they have to work at
> it
>> from scratch, and using Zcode v6 for this is very hard.
> True enough... I was looking into porting "My Angel" over to v6 so I
> could make it a little easier to read, and put the status line at the
> bottom, but even with the manual open I couldn't get it to do anything I
> needed.
If you want a hand with using V6Lib let me know. I'm more than happy to help.
> On the subject of which .. does anyone know how to tell how big a V6
> "unit" is in pixels? It seems that since the font height is only
> calculable in pixels, and the window size in units, there's a problem
> with trying to make a two-line-big window, which is what I need. Any
> help?
AFAIK on every existing interpreter one pixel = one unit. This could
change with scaling and Blorb... or not. I thought the font height was
only calculatable in units?
Stuart Allen
> I'd like to know what's missing in these systems that don't make it?
I think the crux of the matter is that given Inform and TADS work so
well and are both so well supported, nobody needs a new language. Even
as the author of an unused IF language, I don't find this hard to
accept. I started writing JACL before I had any idea that Inform and
TADS even existed. Had I stumbled across them earlier I probably
wouldn't have written JACL at all, which in some ways would have been
good, in other ways quite a personal loss.
I can understand the way you feel about wanting a language that can be
expanded in any way you desire. Although JACL is quite a simple
language, as its author I have the luxury of adding any feature I like
whenever I have the slightest need for it. And as it is an interpreted
language, this is very easy to do.
As far as I know, JACL is the only IF language that runs on a web
server and returns its output in the form of complete HTML pages. This
allows anything from multi-field forms and JavaScript to video and
sound to be used. But no one other than myself has ever used it. I
think at the end of the day the likely way of things is that other
systems will come and go, and their main contribution will be to
stimulate ideas that will later be integrated into the popular
languages such as Inform and TADS. And as much as hate the fact some
days, this is probably the best way for things to be.
Regards,
Stuart
http://jacl.animats.net
L. Ross Raszewski
<jpe...@jczorkmid.net> wrote:
>
>AFAIK on every existing interpreter one pixel = one unit. This could
>change with scaling and Blorb... or not. I thought the font height was
>only calculatable in units?
Sorta. WOuld be good to clarify that if the interpreter is not
graphics-supportin' (like Frotz in its -d1 display mode), it might
well have characters as screen units.
John Colagioia
> I've been unable to create graphics in 'classic' format. If there was
> a tool to do this then I'd have a V6 game out by now (I'm not saying
> if such a took popped up tomorrow I'd have one out next week).
Ditto. I might not release an actual graphical game, but I'd certainly play around
with the graphics if I could create them.
I actually do periodically try to walk through "pix2gif" to determine the file
format, but I somehow always get called away when I get to the actual graphics,
then manage to lose the work I had done. I think I've identified the file headers
at least three times in the last five years...The most recent time, I even had code
to construct the "MG1" file, assuming I had the graphics available.
Eventually...
John W. Kennedy
> As far as I know, JACL is the only IF language that runs on a web
> server and returns its output in the form of complete HTML pages.
Actually, the Z machine can do this. (If no-one else has such an
interpreter, at least I do.)
--
John W. Kennedy
(Working from my laptop)
Jon Ingold
to help.
I've now roughly got it to work, though it looks a lot worse on the
whole than the z5/z8 version did because I've lost the boxes, the
italics, and I don't know how to do any decent effects in v6. (This
isn't using V6lib, I've been doing it from scratch with the manual, but
I'll have a look at it now I've realised it exists). However, it only
seems to run properly under DosFrotz - Nitfol, Jzip, WinFrotz; none of
them are displaying even halfway correctly.
I don't find this very encouraging for the format really. Glulx it is..
if I can work that!
Jon
jpe...@jczorkmid.net
>> If you want a hand with using V6Lib let me know. I'm more than happy
> to help.
> I've now roughly got it to work, though it looks a lot worse on the
> whole than the z5/z8 version did because I've lost the boxes, the
> italics, and I don't know how to do any decent effects in v6. (This
> isn't using V6lib, I've been doing it from scratch with the manual, but
> I'll have a look at it now I've realised it exists). However, it only
> seems to run properly under DosFrotz - Nitfol, Jzip, WinFrotz; none of
> them are displaying even halfway correctly.
One of the things I've tried to do with V6Lib is try to make games
compiled with it work the same (or close to it) with as many
interpreters as I could. This includes Infocom's own V6 terps, unix
frotz, dos frotz and winfrotz. Doing this by hand was no fun... so I
tried to hide it.
V6Lib also fixes boxes and menus.
Stuart Allen
> Stuart Allen wrote:
> > As far as I know, JACL is the only IF language that runs on a web
> > server and returns its output in the form of complete HTML pages.
>
> Actually, the Z machine can do this. (If no-one else has such an
> interpreter, at least I do.)
Although the above sentence clearly doesn't say it, I was actually
thinking more that I haven't seen another engine that allows both HTML
output and the processing of a multi-field form. In a JACL program you
can define mappings between URL parameters and variables. For example:
parameter STATUS player(status) 1 40
With a parameter like this defined, if a URL reads:
http://mysite/cgi-bin/mygame.jacl?STATUS=30&command=examine+chest&user_id=fred
player(status) will be set to the value 30 before the command 'examine
chest' is processed. It is possible to define as many of these
parameters as you like. The '1 40' bit is an optional feature for
doing bounds checking.
Regards,
Stuart
jacl.animats.net
Jon Ingold
> compiled with it work the same (or close to it) with as many
> interpreters as I could. This includes Infocom's own V6 terps, unix
> frotz, dos frotz and winfrotz. Doing this by hand was no fun... so I
> tried to hide it.
>
> V6Lib also fixes boxes and menus.
Consider it downloaded...
Jon
Jonadab the Unsightly One
> Actually, AdvSys predates AGT or GAGS and was even available for CP/M.
Ah. This must have been before computers entered my existence.
(I got my start on PC-DOS 3.3)
> was never used much because I didn't do a very good job of documenting it or
> supplying a decent sample game. It also used Lisp syntax (like ZIL used by
> Infocom) and some people don't like Lisp.
I actually rather like lisp syntax. Except for the extra backslashes
needed in regexps; that can get to be a pain. Though I have a theory
that better syntax highlighting could improve this.
- jonadab
James Taylor
John W. Kennedy <jwk...@attglobal.net> wrote:
>
> Stuart Allen wrote:
> > As far as I know, JACL is the only IF language that runs on a web
> > server and returns its output in the form of complete HTML pages.
>
> Actually, the Z machine can do this. (If no-one else has such an
> interpreter, at least I do.)
Really? That's interesting.
I'd like to try putting z-code games on my website.
Where would I get such a terp?
Will it run on Linux through Apache?
--
James Taylor <james (at) oakseed demon co uk>
Based in Southam, Cheltenham, UK.
PGP key available ID: 3FBE1BF9
Fingerprint: F19D803624ED6FE8 370045159F66FD02
Sean T Barrett
>Actually, AdvSys predates AGT or GAGS and was even available for CP/M. It
>was never used much because I didn't do a very good job of documenting it or
>supplying a decent sample game. It also used Lisp syntax (like ZIL used by
>Infocom) and some people don't like Lisp.
AdvSys was cool enough that I worked pretty hard on developing a
game with it; I seem to recall rewriting a bunch of the world
model though (maybe container stuff, I was doing vehicles and
some wacky robot stuff). But then I ran into the size limit for
the PC DOS version, and I didn't know that source was available,
so I ended up reimplementing it from scratch with 2 or 3x the size
limit (it partitioned the data into several sections, so I gave
each one its own 64K chunk)... I forget why I never finished the
game at that point, either because I hit the new limit, or because
I got sucked into MUDs...
I seem to recall noticing some reference to AdvSys in Return to
Zork, possibly a string somewhere in the executable, as if some
version of it was being used there?
SeanB
John Colagioia
>In article <3B8FA036...@attglobal.net>,
>John W. Kennedy <jwk...@attglobal.net> wrote:
>> Stuart Allen wrote:
>> > As far as I know, JACL is the only IF language that runs on a web
>> > server and returns its output in the form of complete HTML pages.
>> Actually, the Z machine can do this. (If no-one else has such an
>> interpreter, at least I do.)
>Really? That's interesting.
>I'd like to try putting z-code games on my website.
>Where would I get such a terp?
>Will it run on Linux through Apache?
I'd be rather interested, myself, mostly to satisfy my own curiousity.
David Betz
news:GJ0My...@world.std.com...
> I seem to recall noticing some reference to AdvSys in Return to
> Zork, possibly a string somewhere in the executable, as if some
> version of it was being used there?
Yes, Activision used a variant of AdvSys for a few games. This was back when
they wanted to do development on PCs and ZIL only ran on Unix machines or
the PDP-10. I'm sure they would have used Inform had it existed then.
John W. Kennedy
> Really? That's interesting.
> I'd like to try putting z-code games on my website.
> Where would I get such a terp?
> Will it run on Linux through Apache?
It's a Java servlet, so it needs Jakarta or something of the sort (which
certainly runs on Linux through Apache). I wrote it mainly for WAP,
hoping for the Nokia contract, but it also does HTML. It only supports
V1-V3 -- higher levels of the Z machine really aren't compatible with
HTML, or HTTP, for that matter (and, of course, are grossly incompatible
with WAP).
At the moment, the code is a little unstable (I was halfway through a
bit of redesign when it became clear I wasn't going to be selling the
thing), and I only have a dummy mechanism for handling user ID's and
passwords (a Java .properties file -- similar to a Windows .ini file or
a Unix .rc file). I could probably pull it together in a few weeks.
Magnus Olsson
Jonadab the Unsightly One <jon...@bright.net> wrote:
>Someone else used the term "proof of concept" for this
>phenomenon and cited Curses for Inform, and I agree.
>New systems need a major killer game to get them off
>the ground; otherwise they tend to wallow in obscurity.
>I personally got interested in Inform after playing
>Curses
There's also another important aspect of this: I think that any IF
writing system needs to be put through the agony of having at least
one major game written in it. This will not only detect a lot of bugs
and misfeatures, it will also, above all, detect *missing* features.
I'm thinking particularly of the world model here - it's easy on the
paper to design an IF library, but I think it takes a real game
to make the world model powerful enough. This is the equivalent of
playtesting a game, or letting real users test a consumer product.
I think part of Inform's appeal is that its development was so
intimately tied to that of a popular game, Curses. IIRC Graham's
original goal was simply to write a compiler that could generate
Z-code. If he'd stopped at that, and just been content to release one
of the early versions of Inform, Inform would probably have been a
curiosity today.
--
Magnus Olsson (m...@df.lth.se, m...@pobox.com)
------ http://www.pobox.com/~mol ------
Magnus Olsson
This is extremely interesting. Did they actually use AdvSys for
graphical games? Got any more details?
Magnus Olsson
Really? That's amazing - and, in a sense, a great waste that so much
energy was spent writing mediocre AGT games when it could have been
spent writing great AdvSys games.
>It was never
>used much because I didn't do a very good job of documenting it or
>supplying a decent sample game.
It's been almost ten years since I last looked at AdvSys (when I
was betatesting "The Sound of One Hand Clapping"), but I seem to
recall that the state of the documentation was such that you
more or less had to know Lisp, and that it, unlike, say, the Inform
DM, didn't explain much about the basic concepts of IF programming.
I first read about AdvSys in Byte in 1986, IIRC, and I immediately
realized that this was what and IF writing system *should* look
like - a general, object-oriented programming language with some
special constructs for adventures. At the time, I was busy writing
"Dunjin" (in Pascal) and I already knew Lisp, so I had no problems
with the docs.
However, I must confess that I didn't get very far writing IF in
AdvSys. Partly, I suppose, because "Dunjin" was too far advanced for
me to switch languages, but also because I found AdvSys a bit lacking
- the world model was a bit primitive, especially the way NPC orders
were handled, and I missed some features in the language itself (IIRC,
I couldn't get it to respond to an incomplete order like "give bone"
with "To whom do you want to give the bone?" because I couldn't access
the verb string from AdvSys). I tried to hack the compiler to provide
the feature I wanted, but didn't quite succeed, and got tired of the
exercise.
I think you're correct in that the documentation was one of the
reasons for AdvSys' lack of success, but I think the major thing that
was missing was a largish example game. As I wrote elsewhere in this
thread, such a game would not only have been a showcase for the
system's abilities, but it would also have been an enormous help for
you (or whoever wrote the game) to improve the parser and the world
model.
>It also used Lisp syntax (like ZIL used by
>Infocom) and some people don't like Lisp.
I think that in this case, the big problem was that most of the
prospective users didn't know enough Lisp even to get started.
Anyway, even though AdvSys didn't catch on, you made at least me
an enormous favour, both by pointing out how IF *should* be done,
and by showing that it *could* be done that way.
John Colagioia
So...am I completely out of it, or does this mean that Activision used
AdvSys to build the world model in "Return to Zork"? Obviously there would
have to be significant enhancements to the known runtime engine to make it
work at that level, but I find that a fascinating detail in the historical
scheme of things.
John Colagioia
>In article <WKpk7.7115$zj5.1...@typhoon.ne.mediaone.net>,
>David Betz <db...@xlisper.mv.com> wrote:
>>"Sean T Barrett" <buz...@world.std.com> wrote in message
>>news:GJ0My...@world.std.com...
>>> I seem to recall noticing some reference to AdvSys in Return to
>>> Zork, possibly a string somewhere in the executable, as if some
>>> version of it was being used there?
>>Yes, Activision used a variant of AdvSys for a few games. This was back
when
>>they wanted to do development on PCs and ZIL only ran on Unix machines or
>>the PDP-10. I'm sure they would have used Inform had it existed then.
>This is extremely interesting. Did they actually use AdvSys for
>graphical games? Got any more details?
Heh...Ignore my post, since Magnus pretty much summed mine up in a much more
readable way. Must be all the darn textbooks I've been reading the last few
days getting to me...
David Betz
for a PC port of Manhole. I think they ended up using it for something in
the Zork series but I'm not sure which game. They had me remove the parser
and make some improvements to the runtime to allow virtual objects and more
efficient storage utilization. I've got the notes around somewhere for my
improvements.
"John Colagioia" <JCola...@csi.com> wrote in message
news:3b93c...@excalibur.gbmtech.net...
wo...@one.net
Hi Magnus,
>I think you're correct in that the documentation was one of the
>reasons for AdvSys' lack of success, but I think the major thing that
>was missing was a largish example game. As I wrote elsewhere in this
>thread, such a game would not only have been a showcase for the
>system's abilities, but it would also have been an enormous help for
>you (or whoever wrote the game) to improve the parser and the world
>model.
If anyone cares I wrote a manual (for ADVSYS version 1.2?) years ago.
If anybody wants it they can email me: wo...@one.net. Be warned, it
doesn't wax poetic about the Land of Infinite Sadistic Parentheses...
:)
Respectfully,
Wolf
"The world is my home, it's just that some rooms are draftier than
others". -- Wolf
Magnus Olsson
>If anyone cares I wrote a manual (for ADVSYS version 1.2?) years ago.
>If anybody wants it they can email me: wo...@one.net. Be warned, it
>doesn't wax poetic about the Land of Infinite Sadistic Parentheses...
Why not upload it to the IF archive? If it were sitting next to the
AdvSys distribution, people wouldn't have to look around for it.
Branko Collin
>>If anyone cares I wrote a manual (for ADVSYS version 1.2?) years ago.
>>If anybody wants it they can email me: wo...@one.net. Be warned, it
>>doesn't wax poetic about the Land of Infinite Sadistic Parentheses...
>
>Why not upload it to the IF archive? If it were sitting next to the
>AdvSys distribution, people wouldn't have to look around for it.
If you decide to indeed upload it, please make sure your manual is
accompagnied by a decent distribution license.
--
Real Men Don't Need Anaesthetics
wo...@one.net
Hi Branko,
>>>If anyone cares I wrote a manual (for ADVSYS version 1.2?) years ago.
>>>If anybody wants it they can email me: wo...@one.net. Be warned, it
>>>doesn't wax poetic about the Land of Infinite Sadistic Parentheses...
>>
>>Why not upload it to the IF archive? If it were sitting next to the
>>AdvSys distribution, people wouldn't have to look around for it.
>
>If you decide to indeed upload it, please make sure your manual is
>accompagnied by a decent distribution license.
Actually I never thought about it, it isn't a widely used language
these days and the manual was written in the days when IF was still a
potentially paying market. (20 years ago?) :)
David Betz has a copy, if he thinks it worthy to upload I have no
objection. ADVSYS is his creation after all and I wouldn't want to
step on anybody's toes...
By the way the manual and accompanying sample code (Quest, the *EARLY*
(incomplete) version :)) has been released to the public domain. So
there's no license to worry about.
David Betz
> these days and the manual was written in the days when IF was still a
> potentially paying market. (20 years ago?) :)
Not a widely used language? That's an understatement. Is anyone at all still
using it? I'd be surprised. The languages of today have so much more to
offer.
> David Betz has a copy, if he thinks it worthy to upload I have no
> objection. ADVSYS is his creation after all and I wouldn't want to
> step on anybody's toes...
Don't worry about stepping on my toes. If I ever start working on AdvSys
again, I'll be sure to include your manual in any new package I release.
Thanks very much for writing it.
Branko Collin
PD is good enough! :-)
The reason why I was asking is because often people do not think about
something as trivial as copyrights when making the product. They
figure that if they upload something themselves, that that counts as
implicit permission to distribute the work. Copyright is getting big
business, though, and for instance a German lawyer could have sued GMD
(if the IF Archive would still have been hosted there) for copyright
infringement on your behalf without you knowing about or agreeing to
it. This is what happened recently to for instance the author of
Killustrator, who got sued by Adobe without Adobe knowing about it or
even agreeing to it.
(The reason some scrupulous lawyers in Germany have reverted to this
tactic is because it makes good money. One cease and desist letter
nets you 2000 DM without trouble. So a search of Google for brand
names is an hour well spent for them.)
This is just an example, though. A license is always better for the
user than no license at all. Of course, it is the IF Archive's own
choice to accept works that are not accompagnied by a license, so in
the indeed I am just appealing to authors' sense of being a nice
guy/gal to the end-user.
Magnus Olsson
Branko Collin <col...@xs4all.nl> wrote:
>Copyright is getting big
>business, though, and for instance a German lawyer could have sued GMD
>(if the IF Archive would still have been hosted there) for copyright
>infringement on your behalf without you knowing about or agreeing to
>it.
Does German law really allow this? Under Swedish law, a third party
in general can't sue for damages; only the injured party can sue.
Or do you mean that the lawyers act first, and then tell the copyright
owner "I've found somebody who's infringing your copyright; if you
pay me $X and sign these papers I'll sue them for you"?
BTW, I doubt very much that any lawyer would want to sue for copyright
infringement of a manual for a freeware compiler - there's no money in
it (except for statutory damages, but they're not likely to be very big
in a case like this).
Branko Collin
>>Copyright is getting big
>>business, though, and for instance a German lawyer could have sued GMD
>>(if the IF Archive would still have been hosted there) for copyright
>>infringement on your behalf without you knowing about or agreeing to
>>it.
>
>Does German law really allow this? Under Swedish law, a third party
>in general can't sue for damages; only the injured party can sue.
>
>Or do you mean that the lawyers act first, and then tell the copyright
>owner "I've found somebody who's infringing your copyright; if you
>pay me $X and sign these papers I'll sue them for you"?
Yes, I mean the latter. Actually, when the lawyer sends the letter,
the sueing has yet to start.
>BTW, I doubt very much that any lawyer would want to sue for copyright
>infringement of a manual for a freeware compiler - there's no money in
>it (except for statutory damages, but they're not likely to be very big
>in a case like this).
Well, here's the trick as I understand it. It is just a gap in the
law, but judges have little choice but to accept it.
The lawyer can send a letter on behalf of anyone to anyone. It does
not really matter how nonsensical the case is, because the lawyer is
essentially allowed to retrieve the costs from the person receiving
the letter. I seem to remember that if you fight back, you can get the
other party to bear the costs of the letter (assuming you win), but
the costs of having it out before the court is usually more than what
the lawyer is charging. The other sick bit is that the lawyer who
started is also found out that the law is unclear to what the costs
are. He figures, and this is where the 2000 DM come from, that for
every letter he sends out, he can charge the initial costs of finding
out how the law applies here. (The guys name is Gravenreuth, IIRC).
This is of course deeply immoral (some people would call it
blackmail), but it is not illegal. This started being issue something
of two or three years ago. Germans are fighting this practice tooth
and nail (i.e. you should be able to find more reliable sources than
me through Google), but AFAIK the law has not changed yet.
This is all from memory, so if somebody wants to correct me, please go
ahead.
Branko Collin
>The other sick bit is that the lawyer who started is also found out
^^
this
>that the law is unclear to what the costs are.
--
Branko Collin
>col...@xs4all.nl (Branko Collin), je schreef:
>
>>The other sick bit is that the lawyer who started is also found out
> ^^
> this
aargh!
"the lawyer who started this also found out"
^^^^
(sorry)
Michael Brazier
>The lawyer can send a letter on behalf of anyone to anyone. It does
>not really matter how nonsensical the case is, because the lawyer is
>essentially allowed to retrieve the costs from the person receiving
>the letter. I seem to remember that if you fight back, you can get the
>other party to bear the costs of the letter (assuming you win), but
>the costs of having it out before the court is usually more than what
>the lawyer is charging. The other sick bit is that the lawyer who
>started is also found out that the law is unclear to what the costs
>are. He figures, and this is where the 2000 DM come from, that for
>every letter he sends out, he can charge the initial costs of finding
>out how the law applies here. (The guys name is Gravenreuth, IIRC).
What happens if the letter's recipient forwards the letter to the
supposedly injured party, offering to settle out of court? Seems to me
that Gravenreuth is then out of luck.
--
Michael Brazier But what are all these gaieties to me
Whose thoughts are full of indices and surds?
X^2 + 7X + 53 = 11/3
-- Lewis Carroll
wo...@one.net
>>By the way the manual and accompanying sample code (Quest, the *EARLY*
>>(incomplete) version :)) has been released to the public domain. So
>>there's no license to worry about.
>
>PD is good enough! :-)
Ok, I uploaded ADVSYS_MAN.ZIP to ftp://ftp.ifarchive.org/incoming/ if
anyone wants it. :)
Alexander Deubelbeiss
Michael Brazier schrieb in Nachricht <3b99510a...@News.CIS.DFN.DE>...>>[German "Abmahnung" weirdness]
>
>What happens if the letter's recipient forwards the letter to the
>supposedly injured party, offering to settle out of court? Seems to me
>that Gravenreuth is then out of luck.
>
"you have infringed on someone's trademark, please pay the TM holder
2000.- as a compensation" but "you have infringed on someone's
trademark, please pay me 2000.- for pointing this out to you."
Branko Collin
[licenses]
>The reason why I was asking is because often people do not think about
>something as trivial as copyrights when making the product. They
>figure that if they upload something themselves, that that counts as
>implicit permission to distribute the work. Copyright is getting big
>business, though, and for instance a German lawyer could have sued GMD
>(if the IF Archive would still have been hosted there) for copyright
>infringement on your behalf without you knowing about or agreeing to
>it. This is what happened recently to for instance the author of
>Killustrator, who got sued by Adobe without Adobe knowing about it or
>even agreeing to it.
Man, I must have been sleeping. The German cases I referred to are
trademark cases, not copyright cases.
Sorry,
Branko Collin
>Michael Brazier schrieb in Nachricht <3b99510a...@News.CIS.DFN.DE>...
>>On Fri, 07 Sep 2001 15:05:35 GMT, col...@xs4all.nl (Branko Collin) wrote:
>>
>>>[German "Abmahnung" weirdness]
>>
>>What happens if the letter's recipient forwards the letter to the
>>supposedly injured party, offering to settle out of court? Seems to me
>>that Gravenreuth is then out of luck.
If they settle out of court, Gravenreuth is out of luck. This is
basically what happened in the Killustrator. By that time, however,
the programmer of Killustrator had already changed the name of the
program and his employer had hired a lawyer to deal with this
situation. So the Abmahnung was not entirely without consequences.
>As I understand Branko's explanation, what the lawyer is saying isn't
>"you have infringed on someone's trademark, please pay the TM holder
>2000.- as a compensation" but "you have infringed on someone's
>trademark, please pay me 2000.- for pointing this out to you."
Well, that is what it more or less amounts to.
Here's a short story about the Adobe case:
http://www.theregister.co.uk/content/4/20431.html
Anyhow, this all revolves around trademarks and has little to do with
copyrights (although the lesson may be that if you do not protect
yourself, someone may find a way to take advantage of that).
What I am saying though, is that you please include a license with
anything you upload to the IF Archive.
Magnus Olsson
>Hi Branko,
>
>>>By the way the manual and accompanying sample code (Quest, the *EARLY*
>>>(incomplete) version :)) has been released to the public domain. So
>>>there's no license to worry about.
>>
>>PD is good enough! :-)
>
>Ok, I uploaded ADVSYS_MAN.ZIP to ftp://ftp.ifarchive.org/incoming/ if
>anyone wants it. :)
I just had a look at it, and my first reaction was confusion, because
it contains documentation for TADS and for QUEST.
But it's not that TADS, and not that QUEST :-). This earlier
TADS is based on AdvSys, and QUEST is a game, with source code.
Perhaps the Quest game should be uploaded separately to the
games/source direcotry? Or has that already been done?
wo...@one.net
Hi Magnus,
>I just had a look at it, and my first reaction was confusion, because
>it contains documentation for TADS and for QUEST.
>
>But it's not that TADS, and not that QUEST :-). This earlier
>TADS is based on AdvSys, and QUEST is a game, with source code.
TADS stands for Text Adventure Development System, so it seemed
natural at the time. Remember this was what, 1985?
>Perhaps the Quest game should be uploaded separately to the
>games/source directory? Or has that already been done?
I really don't care to, whever the IF archive decides to put it is
fine with me. This version of Quest has a commpleted level 2 and
several combat sequences, unlike the PAWS version. After all, it's not
like a lot of people still use ADVSYS...
David Betz
news:31dtptcjdjiqvs4ff...@4ax.com...
> After all, it's not like a lot of people still use ADVSYS...
What do you mean people don't use AdvSys???? *I* still use AdvSys. Well, I
keep *intending* to use it anyway. Maybe some day...
:-)
Magnus Olsson
>>I just had a look at it, and my first reaction was confusion, because
>>it contains documentation for TADS and for QUEST.
>>
>>But it's not that TADS, and not that QUEST :-). This earlier
>>TADS is based on AdvSys, and QUEST is a game, with source code.
>
>TADS stands for Text Adventure Development System, so it seemed
>natural at the time. Remember this was what, 1985?
Of course. It's perhaps not so much of a coincidence that we got two
systems called TADS.
Magnus Olsson
>In article <31dtptcjdjiqvs4ff...@4ax.com>, <wo...@one.net> wrote:
>>>I just had a look at it, and my first reaction was confusion, because
>>>it contains documentation for TADS and for QUEST.
>>>
>>>But it's not that TADS, and not that QUEST :-). This earlier
>>>TADS is based on AdvSys, and QUEST is a game, with source code.
>>
>>TADS stands for Text Adventure Development System, so it seemed
>>natural at the time. Remember this was what, 1985?
>
>Of course. It's perhaps not so much of a coincidence that we got two
>systems called TADS.
Oops, that didn't come out the way I intended it. It *is* of course
a coincidence (I'm not accusing anyone of plagiarism here); what I
meant was that as a coincidence, it wasn't a very strange one.
John Colagioia
Ditto. I did re-download it, at the very least.