Language choices?......

10 views
Skip to first unread message

Sam Powell

unread,
Oct 27, 1998, 3:00:00 AM10/27/98
to
If you are an author what most influences your choice in language?
Is it merely the style of the language relative to our coding preferences or
do things such as the history/popularity of the language also sway your
decision?

p.s - I am not asking this aimlesly

Doeadeer3

unread,
Oct 27, 1998, 3:00:00 AM10/27/98
to

In article <7151rq$9jn$8...@newnews.global.net.uk>, "Sam Powell"
<sp...@globalnet.co.uk> writes:

I use Inform.

1. Style - Something I am somewhat familiar with (c-like).

2. Popularity - I did pay attention when I first checked out raif (I lurked
about a year or two, by that I mean I just downloaded some threads now and
then) to what languages were currently being supported. By the author and by
the people in raif.
Based on that, the choice was between Tads or Inform, because they had the most
posts about programming q's and I felt that when I delurked I would want to
have help from other programmers. (Although I stupidly wrote Stuck without
asking any q's until it was almost done, because I wanted to learn Inform on my
own.) I also felt I would look dumb if I asked dumb programming q's (because it
took me awhile to learn the basics). I should have just asked, but initially a
lot of the programmers here really intimidated me (with their knowledge and
sometimes their attitude).

3. Power - Inform has an *open* library (changeable). The Inform language
itself is also flexible (like c). When I first looked the manual and sample
games over, I realized Inform had some limitations, but that a lot of those
could be "gotten around" just by doing one's own additional coding. (Graham's
samples in the DM and sample games made this obvious.) This was important to
me. Special circumstances (that libraries with built-in routines won't/don't
cover) always arise. (I had tried an automatic adventure builder years ago for
my old Apple IIc and it was totally inadequate, I didn't want to duplicate that
frustrating experience.)

4. Z Machine - I liked the fact that Inform used the z-machine and that the
parser, while not identical to the old Infocom games, was very, very similar.
My 15-20 year desire to write games stemmed from playing the old Infocom games
that many years ago. So the fact that Inform used the z-machine was a nice
touch.

Doe :-)


Doe doea...@aol.com (formerly known as FemaleDeer)
****************************************************************************
"In all matters of opinion, our adversaries are insane." Mark Twain

ne...@norwich.edu

unread,
Oct 27, 1998, 3:00:00 AM10/27/98
to
In article <19981027142128...@ngol02.aol.com>,

doea...@aol.com (Doeadeer3) wrote:
>
> In article <7151rq$9jn$8...@newnews.global.net.uk>, "Sam Powell"
> <sp...@globalnet.co.uk> writes:
>
> >If you are an author what most influences your choice in language?
> >Is it merely the style of the language relative to our coding preferences or
> >do things such as the history/popularity of the language also sway your
> >decision?
> >
> >p.s - I am not asking this aimlesly
>
> I use Inform.

I do too. The reason is pretty lame though.

I downloaded Inform after finding XYZZY Magazine from the Infocom site. There
was an interesting interview with somebody who had written a game I never
heard of called, "Curses" in there. He had also written a compiler for
interactive fiction. I downloaded it and tried it, and found it far too
complicated.

The Inform port of Adventure said it came from another port of Adventure
written in another IF writing tool called TADS.

I scrounged up TADS, and tried it. It looked much easier (and much more like
C), and came with a simple example program and tutorial to try, "Gold Skull".
I fiddled around with it for a while, and then I thought it would make more
sense if the Gold Skull showed up in the room description (since it is on a
pedestal in plain site). I couldn't figure out how to do it. I think I found
out last week that that this was a bug in older versions of the TADS program.

At the time, TADS cost $40. Inform was free. So, I decided to embark on the
arduous (I thought) task of learning Inform. I'm still doing that.

I never heard of any other IF systems until I got access to newsgroups, which
was a few years after I started using Inform.

> 1. Style - Something I am somewhat familiar with (c-like).

Inform may look like superficially like C, but using it is almost nothing like
coding C -- and thank god for that.

> 3. Power - Inform has an *open* library (changeable). The Inform language
> itself is also flexible (like c). When I first looked the manual and sample
> games over, I realized Inform had some limitations, but that a lot of those
> could be "gotten around" just by doing one's own additional coding. (Graham's
> samples in the DM and sample games made this obvious.) This was important to
> me. Special circumstances (that libraries with built-in routines won't/don't
> cover) always arise. (I had tried an automatic adventure builder years ago for
> my old Apple IIc and it was totally inadequate, I didn't want to duplicate
> that
> frustrating experience.)

TADS is pretty open as well. Especially now.

> 4. Z Machine - I liked the fact that Inform used the z-machine and that the
> parser, while not identical to the old Infocom games, was very, very similar.
> My 15-20 year desire to write games stemmed from playing the old Infocom games
> that many years ago. So the fact that Inform used the z-machine was a nice
> touch.

I wasn't swayed by this at all. I liked TADS better because a convenient
interpreter came right along with the package. I had no conceived notions
about the Z-machine, and hadn't heard of it. If TADS had been free at the
time, I would have chosen TADS.

--
Neil Cerutti
ne...@norwich.edu

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own

Wildman, the Cuberstalker

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
On Tue, 27 Oct 1998 17:03:41 -0000, Sam Powell <sp...@globalnet.co.uk> wrote:
>If you are an author what most influences your choice in language?
>Is it merely the style of the language relative to our coding preferences or
>do things such as the history/popularity of the language also sway your
>decision?
>
>p.s - I am not asking this aimlesly
>
>

I use TADS. I would say that my decision to do so was entirely arbitrary.
I've downloaded several different authoring systems over the years, and just
stuck with what I felt most comfortable with at the time. For a long time,
it was AGT. Before that, it was nothing - I wrote my games in C and, before
I learned C, in BASIC. Perhaps I'll switch to Inform somewhere down the
line, but for now TADS is what I like.

--
Wildman, the Cuberstalker
"Frank Virga is a spammer, a con man, a liar, an idiot, and a thief." --
The Cuberstalkers
Fight spam - http://www.cauce.org/
DO NOT SPAM THIS ADDRESS

Lelah Conrad

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
On Tue, 27 Oct 1998 17:03:41 -0000, "Sam Powell"
<sp...@globalnet.co.uk> wrote:

>If you are an author what most influences your choice in language?

Though I started with TADS, I switched to ALAN. The clean, clear
look of the language was more to my taste than Inform or TADS, and
makes it possible to focus on writing the code itself, which is a good
thing if you are new to this. ALAN has some fun aspects -- there is
something delightful about scheduling "events" (a special ALAN
construct).

A major drawback is that there are not a lot of games or sample code
for ALAN, and you're definitely on your own when it comes to
libraries. You have to be willing to write up your own stuff from
scratch.

Steven Griffiths has written a marvelous tutorial for ALAN, which you
can get from the archive. It is in my opinion the best teaching tool
right now for any of the languages. At least for beginners.

Lelah

Darin Johnson

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
Just useless 2 cents worth; but I prefer TADS.

1) It feels like more a real high level language than Inform. The
object model feels powerful, there aren't a lot of special
cases (ie, the Inform attributes always made me wince). Inform
has good pragmatic reasons for its design, but I tend to be somewhat
idealistic instead of pragmatic with programming. TADS' goal was
to create an adventure programming system; Inform's goal was to
create a language targetted to an existing out-of-date and cramped
runtime system (gosh, I just noticed, honestly, that this sounds a
bit like Windows).

2) The library was nice, and I know/knew it better than Inform.
Inform still had wobbly legs when I started looking at this
stuff.

3) A novel object-oriented way of doing parsing. It may be somewhat
annoying to some people, but it struck a chord with me. (although
the reliance of whether something printed or didn't print instead of
using normal booleans is pretty gross)

4) A nice printed manual with good examples, etc.

5) Definately NOT popularity. I gravitate towards the less
popular things quite often. Using a language just because
others do seems too much like keeping up with fashion to suit me :-)
*However*, the popularity of the runtime system does matter,
as users need to be able to get a version they can use.

--
Darin Johnson
da...@usa.net.delete_me

David Burke

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
"Sam Powell" <sp...@globalnet.co.uk> wrote:

>If you are an author what most influences your choice in language?

>Is it merely the style of the language relative to our coding preferences or
>do things such as the history/popularity of the language also sway your
>decision?
>
>p.s - I am not asking this aimlesly
>

I was looking for software for my PalmPilot last year. I found
WinFrotz, and from there I found Inform. I downloaded it and the DM.I
read the manual, which I think is one of the best written computer
manuals I have come across.

I like the language, its flexibility, and the support. I have asked a
couple of questions on the newsgroup and received detailed and
accurate answers within the day. I wish the languages I use at work
were as well supported :-)

So in my case it was a random choice, but I do not have the time to
study all the alternatives and everything I have wanted to do, the
language has given me a way of doing.


--David Burke

Wanderer in the Fourth Dimension
www.btinternet.com/~david.burke/drwho2.html

David Glasser

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
Darin Johnson <da...@cn1.connectnet.com> wrote:

> Just useless 2 cents worth; but I prefer TADS.
>
> 1) It feels like more a real high level language than Inform. The
> object model feels powerful, there aren't a lot of special
> cases (ie, the Inform attributes always made me wince). Inform
> has good pragmatic reasons for its design, but I tend to be somewhat
> idealistic instead of pragmatic with programming. TADS' goal was
> to create an adventure programming system; Inform's goal was to
> create a language targetted to an existing out-of-date and cramped
> runtime system (gosh, I just noticed, honestly, that this sounds a
> bit like Windows).

Ah, this is exactly what I've been trying to say for a while.

> 2) The library was nice, and I know/knew it better than Inform.
> Inform still had wobbly legs when I started looking at this
> stuff.

Right now, the libraries are comparable. They both have their high
points and low points.

> 3) A novel object-oriented way of doing parsing. It may be somewhat
> annoying to some people, but it struck a chord with me. (although
> the reliance of whether something printed or didn't print instead of
> using normal booleans is pretty gross)

Are you sure you aren't me?

> 4) A nice printed manual with good examples, etc.

Well, they aren't printed anymore, but the manual is great. But so is
the Inform DM.

> 5) Definately NOT popularity. I gravitate towards the less
> popular things quite often. Using a language just because
> others do seems too much like keeping up with fashion to suit me :-)
> *However*, the popularity of the runtime system does matter,
> as users need to be able to get a version they can use.

Darnit, get out of my head!

My current Work Not In Progress is in Inform. However, if I ever start
my Really Great Game "Sidekick", it may well be in TADS. I've
"released" games in both languages, and could probably relearn TADS
easily.

--David Glasser
gla...@NOSPAMuscom.com | http://onramp.uscom.com/~glasser
DGlasser @ ifMUD : fovea.retina.net:4000 (webpage fovea.retina.net:4001)
Sadie Hawkins, official band of David Glasser: http://sadie.retina.net
"We take our icons very seriously in this class."

Jon Petersen

unread,
Oct 28, 1998, 3:00:00 AM10/28/98
to
David Glasser wrote:
>
> My current Work Not In Progress is in Inform. However, if I ever start
> my Really Great Game "Sidekick", it may well be in TADS. I've
> "released" games in both languages, and could probably relearn TADS
> easily.

David:

You wouldn't happen to know if your ol' pal Ed Dallas is finishing up
"Bad Guys", do you? I'd be thrilled.

(Seriously, I would!)

Jon

Rene van 't Veen

unread,
Oct 29, 1998, 3:00:00 AM10/29/98
to
Here is a newbie's 0.02c

I was facing the same questions about two months ago or so,
so I tried (still am) both Inform and TADS.

My criteria:

Popularity -
help from raif when you get stuck.
Cost -
it should cost me little to nothing, I do this for fun.
Availability -
can I run it on my machine, can I get it without a lot
of hassle, can I publish it for a sufficiently wide audience,
i.e.. do they have to work before they can play my game.
Open/extendable architecture - with respect to
game/story development that is.
General purpose programming language under the hood -
I tend to prefer OO, multiple inheritance and list processing
Types of story -
I tend to prefer setting the scene and creating
environments and let the player make his own story,
rather than presenting a (possibly non-linear) series of
puzzles/story progressions.
Efficiency -
How easy is it to code up an interesting world?
Quality of the parser -
how good is the parser in understanding
common constructs, can it be extended.
Engine limits -
are there obvious limitations which might prove cumbersome
quickly
Manual -
A well written manual with a good tutorial
section, a comprehensive reference section and some write-ups on
details is a must.
Debugger -
a good source level debugger saves so much time
and trouble.

On the whole I currently favor TADS, but on the basis of the
criteria above it is a close tie, with Inform winning in
most places. I think the debugger, the underlying programming
language and the engine limits clinched the deal for TADS.


----
"Would you tell me, please, which way I ought to go from here?"
"That depends a good deal on where you want to get to," said the Cat
---
Rene van 't Veen - remove the 'nospam's' from the reply-to to email me.


Al Staff

unread,
Oct 29, 1998, 3:00:00 AM10/29/98
to
>If you are an author what most influences your choice in language?
>Is it merely the style of the language relative to our coding preferences or
>do things such as the history/popularity of the language also sway your
>decision?


I use BASIC.
1. I'd rather use something I already knew instead of learning a new language.
2. Who wants to learn some cryptic C type of language anyway?
3. If I WAS going to spend time learning a C type of language to make a text
adventure, I may as well just learn C. At least it would have been useful for
other types of programming as well.
4. In my opinion, the idea that it's easier with the IF creating programs
because the menus and windows are already done for you doesn't hold up because
I think it's easier to program the entire game in a modern BASIC including menu
routines and windows rather than only the heart of the game in a cryptic
language I don't have any experience with.


Al Staffieri Jr.

AlS...@aol.com
http://members.aol.com/AlStaff/index.html

Schep

unread,
Oct 29, 1998, 3:00:00 AM10/29/98
to
> Inform may look like superficially like C, but using it is almost nothing like
> coding C -- and thank god for that.

I have to disagree with that one, Inform the _language_ is very much
like C++. It's the library that makes it easy to program the games.
About 90% of my programming is using library defined constants or else
simple flow control commands/logical operators (if while or...), which
are similar in most languages.

--
Schep -----------------------------------------------------
Reply by email should be schepler at pilot dot msu dot edu.
The other automatic one is an anti-spam device.
Nothing against the meat. (yes, spam contains meat)

David Glasser

unread,
Oct 29, 1998, 3:00:00 AM10/29/98
to
Jon Petersen <en...@ucla.edu> wrote:

Really? I'd have to tell him about it. He might want to take the
suggestion.

Of course, non-cousin Russell might not be thrilled to learn I'd make a
game with similar theme (at least, I think) to his WIP.

John Francis

unread,
Oct 30, 1998, 3:00:00 AM10/30/98
to
In article <19981029171113...@ng91.aol.com>,

You seem to have missed the entire point of Inform or TADS. They don't
do "menus and windows" for you - in fact their windowing support is, at
best, rudimentary.

One thing they do offer you is an advanced (and thoroughly tested) parser.
(Here I'm including the associated libraries as part of the entire system).
Don't underestimate the amount of work necessary to implement a parser that
is anywhere near as good as either of these. Even if you use nothing else
from the libraries, the parser alone is worth the time spent learning the
language.

For a text-based adventure game, the parser is what the user sees most.
A poor parser (or just one that is too limited in ability) will get in
the way, spoiling the total effect of even the best-designed game.

It all depends what you want to do: spend time designing and writing
the actual game, or spend far more time re-implementing the minutiae of
text-based command parsing. Unless you feel sure that your parser will
provide a significant step forward in game interfaces, why bother?
--
John Francis jfra...@sgi.com Silicon Graphics, Inc.
(650)933-8295 2011 N. Shoreline Blvd. MS 43U-991
(650)933-4692 (Fax) Mountain View, CA 94043-1389
Hello. My name is Darth Vader. I am your father. Prepare to die.

satan...@my-dejanews.com

unread,
Oct 30, 1998, 3:00:00 AM10/30/98
to
While working on my own TADS IF game, I've put together an editor for working
on TADS Files - TFE (Tads File Editor). Since I appreciate the games that
others have done, I thought I'd share my editor with other TADS authors.

Some of the Features: - Bookmarking of up to 10 positions (mark a spot & can
jump back to it). - When exiting TFE it saves all of your open files, next
time you start TFE it reloads those files. - Compile & run from within the
editor - Syntax highlighting. - Works in Win95/Win98/WinNT

For now, you can download TFE from you can download it from
http://members.tripod.com/~SatansMutt/index.html

Any feedback, positive or negative, is welcome. Not sure if I'll do anything
with it as far as updates, probably but I don't guarentee squat ;) Pretty
much what you see is what you get, if I add more it's cause I thought it'd be
usefull. Hey, since it's free I'm my most valuable customer so I gotta
keep me happy right? *grin*

later

TenthStone

unread,
Oct 31, 1998, 3:00:00 AM10/31/98
to
jfra...@dungeon.engr.sgi.com (John Francis) caused this to appear in our
collective minds on 30 Oct 1998 00:30:20 GMT:

>For a text-based adventure game, the parser is what the user sees most.
>A poor parser (or just one that is too limited in ability) will get in
>the way, spoiling the total effect of even the best-designed game.

The reverse is true as well: a good parser can make a poor game seem
much better; q.v. Forman's MST 3000 Presents Detective, which is
infinitely easier to play than the actual Detective because of the Inform
libraries.

-----------

The imperturbable TenthStone
tenth...@hotmail.com mcc...@erols.com mcc...@gsgis.k12.va.us

jgoe...@my-dejanews.com

unread,
Oct 31, 1998, 3:00:00 AM10/31/98
to
In article <7151rq$9jn$8...@newnews.global.net.uk>,

"Sam Powell" <sp...@globalnet.co.uk> wrote:
> If you are an author what most influences your choice in language?

I first encountered AGT via David Malmberg's ZORK parody PORK, which had a
great big advertisement built into the game. I managed to locate AGT 1.0 on
a friend's CD-ROM, and messed with it for about 16 months until another
friend downloaded v1.35 for me from the 'net, and eventually I found the
Master's Edition on AOL. Once I got a *real* ISP (30 days later), I
enventually located GMD, and the rest is history.


> Is it merely the style of the language relative to our coding preferences or
> do things such as the history/popularity of the language also sway your
> decision?

I eventually switched from AGT to Inform because of the greater number of
platforms supported by the interpreters. Hence, a larger audience for my
games.

Gome

--
The Gome Page: Downbelow Station
http://netnow.micron.net/~jgoemmer/

JamesG

unread,
Oct 31, 1998, 3:00:00 AM10/31/98
to
Sam Powell wrote:
>
> If you are an author what most influences your choice in language?
> Is it merely the style of the language relative to our coding preferences or
> do things such as the history/popularity of the language also sway your
> decision?

As a new IF author (hey, I've abandoned one game already, I must be
getting *somewhere*...) I've been using TADS. I'd read a few of the
"Which Language?" discussions and from that it seemed clear that TADS
and Inform where the "top two" languages, I found the TADS page, with a
nice clear manual, and it seemed to me to be fairly logical and
intuitive. I had no real experience with the classic Infocom games (but
I finally got hold of the "Masterpieces" CD last Thursday <g>) and so
didn't even get as far as looking at the Inform page (nothing like an
informed decision <g> (no pun intended)).

Having played about with TADS for a while I like it, it seems fairly
solid, I like the Object Orientated style (as an aside I had no
experience with OO programming prior to TADS, a little Basic on an Acorn
and programable calculator (you learn to write tight code on a machine
with less than 2k of memory)) and as said, the TADS manual and assorted
docs, Ditch Day etc. are good to learn from.

JamesG,
who *will* finish a game a some point...
************************************************************************
* Official RASSM Organiser. Will design starships for food. *
* Another original SF short story uploaded *
* (-o-) http://www.geocities.com/Area51/Corridor/2843 <*> *
* "If we knew what it was we were doing, it would not *
* be called research, would it?" Albert Einstein *
************************************************************************


Ben

unread,
Nov 1, 1998, 3:00:00 AM11/1/98
to
In article <363B9C1C...@ukc.ac.uk>, JamesG <ja...@ukc.ac.uk> wrote:

If you don't mind Jay, I'm just going to use your post and change the
TADSes to Informs because it works fine for my needs and communicates my
feelings.. :)

> As a new IF author (hey, I've abandoned one game already, I must be

> getting *somewhere*...) I've been using Inform. I'd read a few of the


> "Which Language?" discussions and from that it seemed clear that TADS

> and Inform where the "top two" languages, I found the Inform page, with a


> nice clear manual, and it seemed to me to be fairly logical and
> intuitive. I

..
> didn't even get as far as looking at the TADS page (nothing like an


> informed decision <g> (no pun intended)).
>

> Having played about with Inform for a while I like it, it seems fairly
> solid, I like the Object Orientated style. And as said, the Inform
manual and > assorted docs, Park etc. are good to learn from.
>


- Ben

> who *will* finish a game a some point...

--
bhi...@san.rr.com
http://members.tripod.com/~tunnels/

Carl D Cravens

unread,
Nov 3, 1998, 3:00:00 AM11/3/98
to
On Wed, 28 Oct 1998, Darin Johnson wrote:

> Just useless 2 cents worth; but I prefer TADS.

I do, too. TADS was built from scratch, while Inform is restricted to
more or less archaic versions of the Z-Machine.

> 5) Definately NOT popularity. I gravitate towards the less
> popular things quite often. Using a language just because
> others do seems too much like keeping up with fashion to suit me :-)
> *However*, the popularity of the runtime system does matter,
> as users need to be able to get a version they can use.

This is a problem for me. If the end user isn't interested in your
game, no matter how superior the *complier*, because he doesn't want
to be bothered to use more than one *interpreter*, you might as well
have not written the game at all.

How many readers of this forum have Inform interpreters but don't
bother with a TADS interpreter because there are plenty of Inform
games already? How many Inform vs TADS games are there in the
competition... 18 to 4. If user popularity follows programmer
popularity, TADS just isn't cutting it. (Even though it's now free as
well... I'm wondering if the "mystique" of writing games to use the
same virtual machine as the original Infocom games has much to do with
that.)

I've been out of this arena for awhile, but am looking at trying my
hand at it again. (Got writer's block on the last story... should
have started something else in the meantime.) But I'm seriously
considering changing compilers just because I'm afraid that my TADS
games won't see as much play as they would if they were Z-Machine
games. And I *paid* for TADS, have the nice, paper manual, and
several hundred lines of TADS code and TADS experience and
understanding.

I wouldn't mind switching so much if I didn't feel that the Z-Machine
itself was a limitation. I suppose I shouldn't... Curses was
certainly larger than anything I'll probably ever complete. And I
feel like I'm losing some features I like... Inform doesn't do dynamic
object creation yet, does it?

I prefer TADS... but is my audience smaller than the Inform audience?

--
Carl D Cravens
carl.c...@lsil.com


Stephen van Egmond

unread,
Nov 3, 1998, 3:00:00 AM11/3/98
to
In article <Pine.GHP.4.05.9811031617210.7324-100000@tiger>,

Carl D Cravens <ccra...@lsil.com> wrote:
>I wouldn't mind switching so much if I didn't feel that the Z-Machine
>itself was a limitation. I suppose I shouldn't... Curses was
>certainly larger than anything I'll probably ever complete. And I
>feel like I'm losing some features I like... Inform doesn't do dynamic
>object creation yet, does it?
>
>I prefer TADS... but is my audience smaller than the Inform audience?

I think in terms of raw numbers of people who can play your game, yes.
Z-Machine interpreters are more widely ported than TADS's. As the person
who does the BeOS port, I must say it never really occurred to me to try
as the Z-Machine materials strike me as being far more portable.

But I feel bad about it, so here, I'll try to make a BeOS port.

------------

Working from:
tads2src-226.tar.gz
at
http://bang.ml.org/if-archive/if-archiveXprogrammingXtadsXsource.html

Hey, look, it's building with the Linux configuration.

Ok, it gagged an osunixt.c. termcap stuff. I'll rename a few things.

Still building.

tadsr built. tadsc built. I just pulled down 2044.gam. It started. tdb
built. Resized the window and 2044 freaked out, but one "look", and it's
fine.

deep space drifter compiled. It seems to run. inconsequential diffs in
the output (formatting mostly: different line widths).


--------

(shakes head, looks around at the people staring at him)

Okay, there's a BeOS port. Anybody want to test it with the competition
games? I'm too busy on other stuff to do anything more comprehensive than
I've already done.

This took about 20 minutes.

/Steve

--
,,,
(. .)
+--ooO-(_)-Ooo------------ --- -- - - - -
| Stephen van Egmond http://bang.ml.org/

Darin Johnson

unread,
Nov 3, 1998, 3:00:00 AM11/3/98
to
Carl D Cravens <ccra...@lsil.com> writes:

> How many readers of this forum have Inform interpreters but don't
> bother with a TADS interpreter because there are plenty of Inform
> games already?

To tell the truth, I've been reluctant in the past to get a Z-code
interpreter because I was unsure which of them all were any good and
worth the time to spend compiling them :-)

--
Darin Johnson
"You used to be big."
"I am big. It's the pictures that got small."

Irene Callaci

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
On Tue, 3 Nov 1998 16:33:01 -0600, Carl D Cravens <ccra...@lsil.com>
wrote:
>

>How many readers of this forum have Inform interpreters but don't
>bother with a TADS interpreter because there are plenty of Inform
>games already? How many Inform vs TADS games are there in the
>competition... 18 to 4. If user popularity follows programmer
>popularity, TADS just isn't cutting it. (Even though it's now free as
>well... I'm wondering if the "mystique" of writing games to use the
>same virtual machine as the original Infocom games has much to do with
>that.)

One of the biggest pluses Inform holds for me (as a player)
is that I can play Inform games on my PalmPilot--anywhere,
anytime. The new Pilot Frotz is wonderful, and the convenience
of playing i-f on a PalmPilot allows me to play a lot more
games than I'd otherwise have time for. Unfortunately, TADS
games can't be ported to the PalmPilot, so Inform has the
edge for me right now.

irene

irene

Jonadab the Unsightly One

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
mcc...@erols.com (TenthStone) wrote:

> The reverse is true as well: a good parser can make a poor game seem
> much better; q.v. Forman's MST 3000 Presents Detective, which is
> infinitely easier to play than the actual Detective because of the Inform
> libraries.

That, and you don't get bored because it's funny.

Jonadab the Unsightly One

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
gla...@DELETEuscom.com (David Glasser) wrote:

> > TADS' goal was
> > to create an adventure programming system; Inform's goal was to
> > create a language targetted to an existing out-of-date and cramped
> > runtime system (gosh, I just noticed, honestly, that this sounds a
> > bit like Windows).
>
> Ah, this is exactly what I've been trying to say for a while.

No, no, it's really not like that at all. Windows was designed to
turn a PC into a Macintosh, which is obviously misguided since
a Macintosh makes a much better Macintosh than a PC does
any day, and if you want a Mac you should just buy one
and leave the PCs to us character-prompt people. That
has almost no resemblance to Inform's goal of targeting the
ancient and venerable z-machine.

I'm going to go wrap myself firmly up in asbestos now...

> > 5) Definately NOT popularity. I gravitate towards the less
> > popular things quite often.

You mean like IF? I think most of us here do.
How do you think I end up defending things like
DOS and BASIC? It's sure not by taking popularity
polls, that's certain.

> > Using a language just because
> > others do seems too much like keeping up with fashion to suit me :-)

Sounds like java.


Jonadab the Unsightly One

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to

What influenced my decision? Mainly, the first language I ever ran
into that was specifically designed for IF was Inform. It came with
a great Designer's Manual (for all practical purposes a tutorial,
except that it goes a lot further than most tutorials), and it was
very easy to learn. In no time it was my second-best computer
language. I never tried another IF language, and I don't know if I
ever will. There isn't anything Inform can't do. (The VM is a bit
limited, yes, but I'm hoping somebody's going to fix that...)

Actually, TADS is on my list of languages to try out, but Perl is
higher on the list, and so is elisp, so it'll be a while before I get
to TADS, most likely. Nothing against TADS, mind. I just found
Inform first, and I liked it too well to go looking for anything else
to take its place.


Doeadeer3

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to

In article <Pine.GHP.4.05.9811031617210.7324-100000@tiger>, Carl D Cravens
<ccra...@lsil.com> writes:

>I wouldn't mind switching so much if I didn't feel that the Z-Machine
>itself was a limitation. I suppose I shouldn't... Curses was
>certainly larger than anything I'll probably ever complete. And I
>feel like I'm losing some features I like... Inform doesn't do dynamic
>object creation yet, does it?

Yes, it now can. You have to specify the object ahead of time and the max
number of objects you will want, but you can use create().

>I prefer TADS... but is my audience smaller than the Inform audience?

Probably. I tend to like Inform's default responses to behavior, on the whole,
"Violence isn't the solution to this one." (although I change some of them).
Tads responses sometimes (maybe because of lack of familiarity), throw me off.

To me it would seem very ironic if Tads started disappearing NOW, just when
Mike Roberts is back to support it and when Graham Nelson hasn't been around to
support Inform (not saying Graham won't be back).

Also when probably relatively soon a new VM for Inform may be developed that is
NOT the Z-machine.

I would suggest the Tads interpreter be ported to more platforms (if it can
be). I would also suggest (because of what I said above and I know that I am
not the only one) that some of the Tads standard library responses be changed.
Actually, I haven't played a Tads game in quite awhile, but I think "take" was
one of the ones that was confusing (when the object is scenery and not
takeable).

Andrew Plotkin

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
Doeadeer3 (doea...@aol.com) wrote:

> >I prefer TADS... but is my audience smaller than the Inform audience?

> Probably. I tend to like Inform's default responses to behavior, on the whole,
> "Violence isn't the solution to this one." (although I change some of them).
> Tads responses sometimes (maybe because of lack of familiarity), throw me off.

So we should distinguish between people that *can't* play TADS games, and
people that *don't* (or tend not to) because of the many small differences
between TADS and Inform (and Hugo, and every system) -- everyone is going
to have some trivial preferences about some of those things.

> I would suggest the Tads interpreter be ported to more platforms (if it can
> be).

What's missing, besides the Pilot and Newton?

My hackerly opinion is that porting TADS to the Pilot is impossible as the
source stands. TADS keeps tremendous amounts of data on the stack, and the
Pilot stack is an inflexible 4K maximum. (Under PalmOS 3.0. Future Palm
devices may change this, I suppose, although I don't much expect it.)

With intensive source-code surgery to change to dynamic allocation
everywhere, it may be possible. Very large job.

> I would also suggest (because of what I said above and I know that I am
> not the only one) that some of the Tads standard library responses be
> changed.

Tricky -- I know there are TADS messages that confuse *me* sometimes, but
then I'm used to the Inform library. (I'm surprised when "read" and
"examine" *aren't* synonyms, just because they *are* in Inform!) So you
have to be careful that you don't cause as much trouble for people used to
TADS as you fix for people who aren't.

--Z

--

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."

Jonadab the Unsightly One

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
Carl D Cravens <ccra...@lsil.com> wrote:

> This is a problem for me. If the end user isn't interested in your
> game, no matter how superior the *complier*, because he doesn't want
> to be bothered to use more than one *interpreter*, you might as well
> have not written the game at all.

Neither Inform nor TADS is that low on players.

> How many readers of this forum have Inform interpreters but don't
> bother with a TADS interpreter because there are plenty of Inform
> games already?

1. I had a z-machine for a couple of years before I got a TADS
runtime, but you may attribute that first of all to the fact
that InfoTaskForce was distributed together with Curses
in a ZIP file my college roommate found on some DOS
games site. There was also info on Inform, which I
subsequently found and downloaded, but I didn't know
about TADS until I found XYZZYnews much later.

2. I'm rather slow in the playing department, being able to
count on my fingers the number of IF works I've completed,
so my to-play list is large. It includes one AGT game,
two or three TADS games, and perhaps a dozen
Inform games, as well as several of the works I got
in LTOI and LTOI2.

> How many Inform vs TADS games are there in the
> competition... 18 to 4. If user popularity follows programmer
> popularity, TADS just isn't cutting it. (Even though it's now free as
> well... I'm wondering if the "mystique" of writing games to use the
> same virtual machine as the original Infocom games has much to do with
> that.)

Mac users IIRC have available to them a single interpreter capable
of handling both formats. I think Windows '95 users also have a
package (not a single interpreter, but it comes in a single
autoinstaller) that does both -- namely, Adventure Blaster.
IMO more of this sort of thing would benefit the less popular
language (in this case TADS) because more players could play
the games without downloading something additional.
Just a thought.

> I've been out of this arena for awhile, but am looking at trying my
> hand at it again. (Got writer's block on the last story... should
> have started something else in the meantime.) But I'm seriously
> considering changing compilers just because I'm afraid that my TADS
> games won't see as much play as they would if they were Z-Machine
> games. And I *paid* for TADS, have the nice, paper manual, and
> several hundred lines of TADS code and TADS experience and
> understanding.

Inform as a language is easy to learn.

> I wouldn't mind switching so much if I didn't feel that the Z-Machine
> itself was a limitation.

That's the most serious drawback. It's the memory limitation you're
looking at, no?

Actually, in another way the Z-Machine is actually more flexible
than the TADS runtime -- that is, it can do things like Space
Invaders and Tetris. But that's only important if you want to
put something like that in your game; similarly, the memory
limits are only significant if you want to write a large game.

> I suppose I shouldn't... Curses was
> certainly larger than anything I'll probably ever complete. And I
> feel like I'm losing some features I like... Inform doesn't do dynamic
> object creation yet, does it?

Hmmm...

I don't think it does as *such*, although I remember seeing something
in the Balances example that sure *looked* like it. Maybe Inform 6
manages something Inform 5 didn't, or maybe I misread the example.

> I prefer TADS... but is my audience smaller than the Inform audience?

I think the question is how *much* smaller, and whether it's
significant enough to change. I don't know how to answer.


- jonadab

Roger Carbol

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
Doeadeer3 wrote:

> To me it would seem very ironic if Tads started disappearing NOW, just when
> Mike Roberts is back to support it and when Graham Nelson hasn't been around to
> support Inform (not saying Graham won't be back).


I'm of the opinion that HTML-TADS will change all that, and TADS
will experience a surge in audience growth.


I'm trying to create the game to prove it, of course.

.. Roger Carbol .. r...@shaw.wave.ca .. not forgotten

Dylan O'Donnell

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
erky...@netcom.com (Andrew Plotkin) writes:

> Doeadeer3 (doea...@aol.com) wrote:
>
> > I would suggest the Tads interpreter be ported to more platforms (if it can
> > be).
>
> What's missing, besides the Pilot and Newton?

The Psion Series 5 is another palmtop which has a Zterp but not a TADS.
(That's the reason that I've played all the Inform comp games, but none
of the others yet - was easy to take on holiday with me.)

How easy it'd be, I don't know, though. The S5 is reasonably high-spec
as palmtops go, so it _should_ be feasible...

--
: Dylan O'Donnell : "Now is not this ridiculous, and is not :
: Demon Internet Ltd : this preposterous? / A thorough-paced :
: Southend slave deck : absurdity -- explain it if you can." :
: http://www.fysh.org/~psmith/ : -- W.S. Gilbert, "Patience" :

Roger Burton West

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
In article <erkyrathF...@netcom.com>,

Andrew Plotkin <erky...@netcom.com> wrote:
>Doeadeer3 (doea...@aol.com) wrote:
>
>> I would suggest the Tads interpreter be ported to more platforms (if it can
>> be).
>
>What's missing, besides the Pilot and Newton?

Psion 5. ARM architecture, flat memory model, but no freeware compiler
for it. Should be doable, I'd have thought. Of the comp98 games I've
played so far, at least half have been on PsiFrotz.

Roger
--
Roger Burton West
Frontline Administrator, Demon Internet Ltd
Home: ro...@firedrake.demon.co.uk
Web: http://www.firedrake.demon.co.uk

David Glasser

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
Jonadab the Unsightly One <jon...@zerospam.com> wrote:

> Carl D Cravens <ccra...@lsil.com> wrote:
>
> > This is a problem for me. If the end user isn't interested in your
> > game, no matter how superior the *complier*, because he doesn't want
> > to be bothered to use more than one *interpreter*, you might as well
> > have not written the game at all.
>
> Neither Inform nor TADS is that low on players.

Right.

> > How many readers of this forum have Inform interpreters but don't
> > bother with a TADS interpreter because there are plenty of Inform
> > games already?
>
> 1. I had a z-machine for a couple of years before I got a TADS
> runtime, but you may attribute that first of all to the fact
> that InfoTaskForce was distributed together with Curses
> in a ZIP file my college roommate found on some DOS
> games site. There was also info on Inform, which I
> subsequently found and downloaded, but I didn't know
> about TADS until I found XYZZYnews much later.

This has nothing to do with anything. This was ancient history.

> 2. I'm rather slow in the playing department, being able to
> count on my fingers the number of IF works I've completed,
> so my to-play list is large. It includes one AGT game,
> two or three TADS games, and perhaps a dozen
> Inform games, as well as several of the works I got
> in LTOI and LTOI2.

I miss your point.

> > How many Inform vs TADS games are there in the
> > competition... 18 to 4.

Personal hypothesis: HTML TADS came out quite recently, and so many
people are in the middle of an HTML TADS game they didn't have time to
finish for the competition. Just wait until next year.

Plus, quantity != quality. I've played two (that's half) of the TADS
games from this competition so far, and they were both excellent.
Though I've also played two excellent Inform games from the competition,
that was out of about seven.

> > If user popularity follows programmer
> > popularity, TADS just isn't cutting it.

Sure it is. Competition != every programmer.

> >(Even though it's now free as
> > well... I'm wondering if the "mystique" of writing games to use the
> > same virtual machine as the original Infocom games has much to do with
> > that.)

I know some people are influenced, but I think that's a rather silly
idea. I mean, I doubt that those people would prefer ZIL on MIT's 70s
machines to Inform on an G3 PowerBook.

> Mac users IIRC have available to them a single interpreter capable
> of handling both formats.

Nope, but Zarf has made two similar acting interpreters.
Mmm...orthogonality.

> I think Windows '95 users also have a
> package (not a single interpreter, but it comes in a single
> autoinstaller) that does both -- namely, Adventure Blaster.
> IMO more of this sort of thing would benefit the less popular
> language (in this case TADS) because more players could play
> the games without downloading something additional.
> Just a thought.

YM "would benefit the language that just happened to have a few less
entries in a competition and has a few less questions from newbies about
its bugs (possibly because it is less buggy) (in this case TADS)". HTH.
HAND.

> > I've been out of this arena for awhile, but am looking at trying my
> > hand at it again. (Got writer's block on the last story... should
> > have started something else in the meantime.) But I'm seriously
> > considering changing compilers just because I'm afraid that my TADS
> > games won't see as much play as they would if they were Z-Machine
> > games. And I *paid* for TADS, have the nice, paper manual, and
> > several hundred lines of TADS code and TADS experience and
> > understanding.

Stick with TADS. Learn Inform if you have free time, but stick with
TADS until you do. If you decide to learn Inform, you have twice as
much freedom in writing your games. But they are about equally good.

#include <stdhugo+alanplug>

> Inform as a language is easy to learn.

As is TADS.

<ANECDOTE type="personal">
When I first started to learn Inform (as a TADS programmer), I kept
thinking the same thing over and over: "What in the world was Graham
Nelson thinking when he implemented foo this way? A much better way
would be like in TADS! This is the dumbest thing I've ever seen!"

Now that I've done more work in Inform, I understand it is a different
beast (and I do think beast is a good word for both languages) than
TADS. I still think that about half of its features work a hell of a
lot better in TADS.

Attributes for, say, openness? Why not an isopen property? One could
have code that says "I'm open in such-and-such a case". A built-in
object tree? Again, I can't make a routine to return location, relying
on tons of "move" commands. (Though TADS' approach, requiring indirect
routines instead of object.location = foo can be annoying.) Doing
x.isquux where isquux hasn't been defined for object x crashes the
program? Say *what*? Three separate library files that must be
included at the perfect place in your code, or else everything dies?
There are *how many* different ways to specific defaults for properties?
The property declaration, ChangeDefault, editing something in the
library, something I've forgotten, and of course forgetting to and
having your game crash?

Of course, anyone who wasn't accustomed to the object-oriented
free-flowing bliss of TADS wouldn't mind. In TADS, my main complaint
was that
noun = 'foo' 'bar' 'baz'
wasn't very orthogonal to all other property definitions (unlike, say:
noun = ['foo' 'bar' 'baz']
) But if I was to make a list of such stylistic anti-OO-and-freeform
complaints in Inform, I'd go on for pages and pages!
</ANECDOTE>

That said, Inform is a very nice language once you get used to it.

> > I wouldn't mind switching so much if I didn't feel that the Z-Machine
> > itself was a limitation.

It's only if you are crazy. Like that silly Irene, overloading the
ZMachine weekly.

> That's the most serious drawback. It's the memory limitation you're
> looking at, no?
>
> Actually, in another way the Z-Machine is actually more flexible
> than the TADS runtime -- that is, it can do things like Space
> Invaders and Tetris. But that's only important if you want to
> put something like that in your game; similarly, the memory
> limits are only significant if you want to write a large game.

I'd guess that Space Invaders/Tetris are now possible (or at least more
possible than before) under HTML TADS.

I mean, HTML TADS is damn flexible. Spoilers for Arrival, rot13ed (do
NOT read if you haven't played and rated it!):
Gung jrocntr. V ybir vg. Vg ehyrf. V arire jbhyq unir rkcrpgrq vg,
ohg...jbj.

> > I suppose I shouldn't... Curses was
> > certainly larger than anything I'll probably ever complete. And I
> > feel like I'm losing some features I like... Inform doesn't do dynamic
> > object creation yet, does it?
>
> Hmmm...
>
> I don't think it does as *such*, although I remember seeing something
> in the Balances example that sure *looked* like it. Maybe Inform 6
> manages something Inform 5 didn't, or maybe I misread the example.

It does something that pretends to be dynamic object thingy.

> > I prefer TADS... but is my audience smaller than the Inform audience?

NO.

OK, maybe there are three people who would play Inform and not TADS.
The person who only plays IF on a PalmPilot (but most who do also use a
computer); the person who uses a computer (such as, IIRC, the Acorn)
that doesn't have a TADS (and TADS is quite portable to any computer
with enough memory); and the lunatic who believes that TADS is the spawn
of Satan and has scripts set up to automatically delete any .gam on
his/her computer and bows down to an altar of Graham Nelson every night
(with voodoo dolls of MJR burning somewhere off to the side).

The first group is limited by the fact that even the new PalmFrotz
doesn't do ZCode as well as a computer with, say, a bigger display. The
second group is rather small, limited to a few types of computers that
could easily get a port, and old 80s tiny computers like Apple II's (and
what is anyone doing on the Internet with one of them, anyway? Not that
the Apple II's didn't rule...). The third group is also very small.
Plus, you wouldn't want them playing your game anyway.

> I think the question is how *much* smaller, and whether it's
> significant enough to change. I don't know how to answer.

OK, here's David Glasser's Official Which-is-better-TADS-or-Inform
Answer:
--------------------------------------------------------------------
TADS:
HTML TADS is AFAIK the best implementation of graphics and sound in IF
yet. (Well, there's Hugo, but I've never played it, as there isn't even
a crashes-once-a-minute Mac port of it.)

It's not buggy, really, and now that MJR is back, he fixes everything
with his mad-skills Implementor magic. There's only one real bug that I
know of (Which do you mean, the TADS or the TADS?).

It is so much more orthagonal than Inform. *Everything* is an object
(well, you can also make standalone functions, but that's not common,
and there are a few extra declarations that are mainly used by the
library). And, in general, you can put things wherever you want,
without worrying about order. Oh, and 'replace' and 'modify' rule.

Inform:
Slightly (very slightly) more ports than TADS. Of course, this only
means it is harder to make any substantial updates.

Graphics/sound standard, and bigger memory model, are in the works.
However, I hate to be a pessimist, but I predict that we won't see
either frequently used until at least 2001 or so. At any rate, TADS is
the king of graphics and sound.

More complaints^H^H^H^H^H^H^H^H^H^Hconfusion^H^H^H^H^H^H^H^H^Hcomments
on raif. The fact that most Inform threads consist of "Q: Why can't I
do X? A: Because the Library/compiler is seriously screwed up." is
mostly ignored by people who say "Inform has more threads on raif, it
must be better!"

Similar arguments could be made about "Only 4 TADS games in the
competition, nanny nanny boo boo".

Hugo, ALAN, etc.:
I haven't as much as played a Hugo game (no interpreter for Macs...grr),
but I've heard Good Things about it (and Kent Tessman is one cool guy).
I've had limited experience with ALAN, and am turned off a bit by the
poor Mac interface (though I couldn't do any better) and the fact that
it doesn't accept many "standard" IF commands.
--------------------------------------------------------------------

Contrary to the implications in DoeaDeer's statement, the Inform default
messages have no binding tie to Inform. It would take me maybe 10
minutes to make TADS' messages equal to those currently in Inform.

Go with what your mind tells you to do.

Wow, this is a long article.

--

David Glasser

unread,
Nov 4, 1998, 3:00:00 AM11/4/98
to
Jonadab the Unsightly One <jon...@zerospam.com> wrote:

> What influenced my decision? Mainly, the first language I ever ran
> into that was specifically designed for IF was Inform.

In the end, the first language tends to be the favorite.

> It came with
> a great Designer's Manual (for all practical purposes a tutorial,
> except that it goes a lot further than most tutorials), and it was
> very easy to learn. In no time it was my second-best computer
> language. I never tried another IF language, and I don't know if I
> ever will. There isn't anything Inform can't do. (The VM is a bit
> limited, yes, but I'm hoping somebody's going to fix that...)

Yep, meguesses.

> Actually, TADS is on my list of languages to try out, but Perl is
> higher on the list, and so is elisp, so it'll be a while before I get
> to TADS, most likely. Nothing against TADS, mind. I just found
> Inform first, and I liked it too well to go looking for anything else
> to take its place.

Hmm. Perl and elisp aren't much for IF (though I'm sure one could write
a Perl IF library with a tiny bit less effort than writing, say, TADS).
Of course, if you want to learn a language for purposes other than IF,
Perl is great, and I've heard Many Good Things about lisp.

--David Glasser, who just got the Camel Book! Yay!

Lelah Conrad

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

Alternate languages and alternate runtimes stretch your imagination!

Lelah :)

member of the ALAN cabal

and familiar to some degree with the non-widely supported human
languages of Sanskrit, Portuguese, and Classical Greek

Geoff Bailey

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

In article <1dhzctz.1gh...@usol-209-186-16-183.uscom.com>,
David Glasser <gla...@DELETEuscom.com> wrote:
>
> TADS:

[ ... ]

> It's not buggy, really, and now that MJR is back, he fixes everything
> with his mad-skills Implementor magic. There's only one real bug that I
> know of (Which do you mean, the TADS or the TADS?).

I haven't tried to program in either TADS or inform (and quite frankly, I'm
not likely to in the foreseeable future), but there was one thing that struck
me about the TADS entries in this year's competition. At least two of them
(I can't remember whether the other two were also like this or not) would
lead to dialogues like the following:

There is a red box and a blue lamp here.
> Get red box
There appear to be extra words after that command.
> Get red
Taken.

[ Going from memory, so that message might not be quite correct. ] This was
really annoying, as often I would have to leave out the nouns to get it to
accept the sentence at all. Is this a problem with TADS or the way such
objects were implemented? If the latter, can I suggest that the manual make
specific mention in some suitably hard to miss way about this possibility
and how to avoid it?

Thanks,
Geoff.

-------------------------------------------------------------------------------
Geoff Bailey (Fred the Wonder Worm) | Programmer by trade --
ft...@cs.usyd.edu.au | Gameplayer by vocation.
-------------------------------------------------------------------------------


Jonadab the Unsightly One

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
gla...@DELETEuscom.com (David Glasser) wrote:

> Jonadab the Unsightly One <jon...@zerospam.com> wrote:
> > Actually, TADS is on my list of languages to try out, but Perl is
> > higher on the list, and so is elisp, so it'll be a while before I get
> > to TADS, most likely. Nothing against TADS, mind. I just found
> > Inform first, and I liked it too well to go looking for anything else
> > to take its place.
>
> Hmm. Perl and elisp aren't much for IF (though I'm sure one could write
> a Perl IF library with a tiny bit less effort than writing, say, TADS).
> Of course, if you want to learn a language for purposes other than IF,
> Perl is great, and I've heard Many Good Things about lisp.

I didn't mean for IF. Perl is on my list because a lot of people
write one-liners in it, so it can't be all bad ;-)

elisp I want to learn so I can adjust emacs to emulate the One
True Text Editor, UED, which, unfortunately, is only available
for DOS, and I have way too many commas in that sentence.

That, and lisp has a great history.

- jonadab

Sam Carmean

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
There are a lot of good games written in both languages, so I'm a member
of both audiences (like many others here, I imagine). One thing I like
about Inform is that it produces much smaller game files. Not as
important now as it was when the Z-machine was formulated, but
aesthetically pleasing nonetheless...I can fit 10,000 games on my hard
drive instead of just 5,000! Seriously, I *am* curious to know the
reasons for the size difference. (There's got to be more to it than the
Z-machine's baroque text-encoding.)

--
sa...@primenet.com

Jonadab the Unsightly One

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
gla...@DELETEuscom.com (David Glasser) wrote:

> > 2. I'm rather slow in the playing department, being able to
> > count on my fingers the number of IF works I've completed,
> > so my to-play list is large. It includes one AGT game,
> > two or three TADS games, and perhaps a dozen
> > Inform games, as well as several of the works I got
> > in LTOI and LTOI2.
>
> I miss your point.

My point is that if I played a lot of IF I'd play a lot of TADS games
(as well as a lot of Inform games).

> Personal hypothesis: HTML TADS came out quite recently, and so many
> people are in the middle of an HTML TADS game they didn't have time to
> finish for the competition. Just wait until next year.

Not entirely unlikely.

> Plus, quantity != quality.

Of course not. Unless you're MicroSoft.

> > Mac users IIRC have available to them a single interpreter capable
> > of handling both formats.
>
> Nope, but Zarf has made two similar acting interpreters.
> Mmm...orthogonality.

Oh. I was thinking... nevermind. I was thinking of something else.

> Stick with TADS. Learn Inform if you have free time, but stick with
> TADS until you do. If you decide to learn Inform, you have twice as
> much freedom in writing your games. But they are about equally good.

Yeah, I'd say stick with your first language for primary development,
and learn the other for fun.

> #include <stdhugo+alanplug>

Yes, yes. And Quest. Et Cetera.

> > Inform as a language is easy to learn.
>
> As is TADS.

Good. Then I'll have an easy time WIGART.

> Now that I've done more work in Inform, I understand it is a different
> beast (and I do think beast is a good word for both languages) than
> TADS. I still think that about half of its features work a hell of a
> lot better in TADS.

Heh. And half of the features of C++ work better in BASIC,
because it's what I know better. What you're used to *really*
matters. So stick with what you know, provided it's good
at doing what you want to do. Now, if you wanted to do
something that TADS isn't good at handling, then it would
be time to look into alternatives. Similarly, if I was
programming in BASIC and needed to write a program that
could allocate 5 MB of RAM, it'd be time to switch languages.
But if I'm just snarfing in a text file and stripping out the HTML
tags and spitting out a tagless text file, it's not necessary.
(I am *not* saying TADS is like BASIC and Inform is like C. I
wouldn't say that. (I like BASIC too well to say that ;-) I'm
saying that TADS and Inform are different -- and that BASIC
and C are different. Go ahead. Disagree. Flame me.)

> Attributes for, say, openness? Why not an isopen property? One could
> have code that says "I'm open in such-and-such a case".

Aaargh! So much *work* that would be.

> A built-in
> object tree? Again, I can't make a routine to return location, relying
> on tons of "move" commands.

A routine return an object's location?
I'm not sure I want to know about that.

> x.isquux where isquux hasn't been defined for object x crashes the
> program? Say *what*?

Huh? No. Testing for an attribute never crashes, because every
attribute is defined for every object. Now, if x isn't really an
object, then you have a problem.

> Three separate library files that must be
> included at the perfect place in your code, or else everything dies?

Okay, that one's a little weird.

> There are *how many* different ways to specific defaults for properties?

As many ways as you want. Each with its own advantages and
disadvantages. It's a matter of programmer style!

> > > I wouldn't mind switching so much if I didn't feel that the Z-Machine
> > > itself was a limitation.
>
> It's only if you are crazy. Like that silly Irene, overloading the
> ZMachine weekly.

Hey, how can you write a quality interactive novel in 256K?
Short story, sure. But longer works are out in the cold!

Maybe I'll go into interactive free verse poetry...

> I'd guess that Space Invaders/Tetris are now possible (or at least more
> possible than before) under HTML TADS.

I imagine HTML TADS opens itself up to all sorts of things,
but I was under the impression it hadn't been ported to
very many platforms yet... Give me a few months to get
used to its existence...

> OK, maybe there are three people who would play Inform and not TADS.
> The person who only plays IF on a PalmPilot (but most who do also use a
> computer); the person who uses a computer (such as, IIRC, the Acorn)
> that doesn't have a TADS (and TADS is quite portable to any computer
> with enough memory); and the lunatic who believes that TADS is the spawn
> of Satan and has scripts set up to automatically delete any .gam on
> his/her computer and bows down to an altar of Graham Nelson every night
> (with voodoo dolls of MJR burning somewhere off to the side).

You may be exaggerating just a smidgen on that last point.

> The first group is limited by the fact that even the new PalmFrotz
> doesn't do ZCode as well as a computer with, say, a bigger display. The
> second group is rather small, limited to a few types of computers that
> could easily get a port, and old 80s tiny computers like Apple II's (and
> what is anyone doing on the Internet with one of them, anyway? Not that
> the Apple II's didn't rule...). The third group is also very small.
> Plus, you wouldn't want them playing your game anyway.

Sure you would. You'd want them as beta testers, so they can do
kooky things and break your game ;-)

>
> > I think the question is how *much* smaller, and whether it's
> > significant enough to change. I don't know how to answer.
>
> OK, here's David Glasser's Official Which-is-better-TADS-or-Inform
> Answer:

[Basically, he says it's not that much smaller and not
significant enough to change. We could try to verify
this by taking a poll, but we'd get four responses before
the thread went off topic never to return.]


- jonadab

Jonadab the Unsightly One

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
Sam Carmean <sa...@primenet.com> wrote:

> aesthetically pleasing nonetheless...I can fit 10,000 games on my hard
> drive instead of just 5,000! Seriously, I *am* curious to know the
> reasons for the size difference. (There's got to be more to it than the
> Z-machine's baroque text-encoding.)

I'd finger one-bit attributes. They've *got* to save space.

Hmmm.... Maybe the library's more compact?
Just a guess; I don't know.

- jonadab

pke...@mailexcite.com

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
> From: erky...@netcom.com (Andrew Plotkin)

>> I would suggest the Tads interpreter be ported to more platforms (if it can
>> be).
>
>What's missing, besides the Pilot and Newton?

Commodore 64, Commodore 128, Apple IIe, etc. That's one of the probs with
TADS, it has very few restrictions. Sure, this is sometimes a good thing,
but in a portability sense it's a bad thing. This is why Infocom games
were available on such a large number of machines, the interpreter knew
that the game couldn't be larger then a specific size and it knew what the
maximum resources would be.

Patrick

Neil K.

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
0In article <71ri49$4m0$1...@nnrp02.primenet.com>, Sam Carmean
<sa...@primenet.com> wrote:

> There are a lot of good games written in both languages, so I'm a member
> of both audiences (like many others here, I imagine). One thing I like
> about Inform is that it produces much smaller game files. Not as
> important now as it was when the Z-machine was formulated, but

> aesthetically pleasing nonetheless...I can fit 10,000 games on my hard
> drive instead of just 5,000! Seriously, I *am* curious to know the
> reasons for the size difference. (There's got to be more to it than the
> Z-machine's baroque text-encoding.)

I suspect that's a fairly large part of the difference. Remember that
TADS stores 8-bit clean ASCII whereas the Z-machine uses a most "baroque"
5-bit form of encoding. Right there you're going to get a percentage file
size difference. Inform also does a bit of compression by tokenizing
repeating text elements; TADS does not. There are probably a bunch of
other factors involved in the way TADS and the Z-machine encode their
respective virtual machines, but I'd be the last person to know the gory
details.

- Neil K.

--
t e l a computer consulting + design * Vancouver, BC, Canada
web: http://www.tela.bc.ca/tela/ * email: tela @ tela.bc.ca

Neil K.

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
ft...@cs.usyd.edu.au (Fred the Wonder Worm) wrote:

> I haven't tried to program in either TADS or inform (and quite frankly, I'm
> not likely to in the foreseeable future), but there was one thing that struck
> me about the TADS entries in this year's competition. At least two of them
> (I can't remember whether the other two were also like this or not) would
> lead to dialogues like the following:
>
> There is a red box and a blue lamp here.
> > Get red box
> There appear to be extra words after that command.

There is a known bug in the most recent versions of TADS in which typing
something that results in an error message sometimes erroneously brings up
the text of a previous error message. It's possible you got bitten by that
bug. As for disambiguation issues TADS seems somewhat more vulnerable than
Inform to annoying disambiguation prompts (which tree do you mean, the
tree or the tree?) but I don't consider that intrinsically a problem with
TADS, per se - authors have to be careful to avoid coding up such
situations. I don't know about the competition entries you cite, as I
haven't come across those bugs - maybe more details?

T Raymond

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
On Thu, 05 Nov 1998 06:30:04 GMT, jon...@zerospam.com (Jonadab the Unsightly One) wrote:
> gla...@DELETEuscom.com (David Glasser) wrote:
>
> to TADS, most likely. Nothing against TADS, mind. I just found
> Inform first, and I liked it too well to go looking for anything else
> to take its place.

This sounds about like my comments for TADS *G*

> I didn't mean for IF. Perl is on my list because a lot of people
> write one-liners in it, so it can't be all bad ;-)

For proessing text and doing things with it Perl is awesome. Certain nameless online companies (who
barage people with disks) had a nice list of Internet mailing lists and aside of editing the data (done
mainly with Emacs), the entire formatted list was built using Perl scripts.

> elisp I want to learn so I can adjust emacs to emulate the One
> True Text Editor, UED, which, unfortunately, is only available
> for DOS, and I have way too many commas in that sentence.

Hmm, run emacs in C mode and write a key file for inserting handy bits of coding from your favorite
IF-producing language *S*

Tom
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Tom Raymond a...@usa.net
"I hate it when you fumble a rezrov and frotz the cat."
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

HarryH

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
In article <3641453b...@news.bright.net>, jon...@zerospam.com says...

>gla...@DELETEuscom.com (David Glasser) wrote:
>I didn't mean for IF. Perl is on my list because a lot of people
>write one-liners in it, so it can't be all bad ;-)

Learn Perl. As someone who has learned it, I can say that it is great. In
fact, I may write an IF in Perl someday.

>elisp I want to learn so I can adjust emacs to emulate the One
>True Text Editor, UED, which, unfortunately, is only available
>for DOS, and I have way too many commas in that sentence.
>

>That, and lisp has a great history.

I suggest that you try Andrew Plotkin's Lists and Lists. That is definitely
Lisp-like. See if you like that before you dive into Lisp.

-------------------------------------------------------
IFC0.1 --C -P++ --A --r -i++


Stephen Granade

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
On 5 Nov 1998, Geoff Bailey wrote:

> I haven't tried to program in either TADS or inform (and quite frankly, I'm
> not likely to in the foreseeable future), but there was one thing that struck
> me about the TADS entries in this year's competition. At least two of them
> (I can't remember whether the other two were also like this or not) would
> lead to dialogues like the following:
>
> There is a red box and a blue lamp here.
> > Get red box
> There appear to be extra words after that command.

> > Get red
> Taken.
>
> [ Going from memory, so that message might not be quite correct. ] This was
> really annoying, as often I would have to leave out the nouns to get it to
> accept the sentence at all. Is this a problem with TADS or the way such
> objects were implemented?

Were you perchance running WinTADS 1.2.0? My update had a serious bug
whose symptoms were exactly as you describe. If this is the case, dash on
over to GMD and get the 1.2.1 update.

In short, it's not the game authors' fault, it's mine.

Stephen

--
Stephen Granade | Interested in adventure games?
sgra...@phy.duke.edu | Visit Mining Co.'s IF Page
Duke University, Physics Dept | http://interactfiction.miningco.com


Andrew Plotkin

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
Jonadab the Unsightly One (jon...@zerospam.com) wrote:
> Sam Carmean <sa...@primenet.com> wrote:

> > aesthetically pleasing nonetheless...I can fit 10,000 games on my hard
> > drive instead of just 5,000! Seriously, I *am* curious to know the
> > reasons for the size difference. (There's got to be more to it than the
> > Z-machine's baroque text-encoding.)

> I'd finger one-bit attributes. They've *got* to save space.

> Hmmm.... Maybe the library's more compact?
> Just a guess; I don't know.

That easy to check, though; make a one-room no-object game in each system
and compare.

I'd guess the size difference is 30% text compression and 70% differences
in the code encoding. The Z-machine opcode format is very compact, and it
has lots of abbreviations like one-opcode forms for "print and return", or
"if (a == b) return".

Irene Callaci

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
On Wed, 4 Nov 1998 19:22:09 -0500, gla...@DELETEuscom.com (David
Glasser) wrote:
> Carl D Cravens <ccra...@lsil.com> wrote:
>> > I wouldn't mind switching so much if I didn't feel that the Z-Machine
>> > itself was a limitation.

>It's only if you are crazy. Like that silly Irene, overloading the
>ZMachine weekly.

Hey! I resemble that remark! (But now that L. Ross has written
imem.h, my worries in that department are over, at least for the
time being! Yay!)

irene

Doeadeer3

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

In article <1dhzctz.1gh...@usol-209-186-16-183.uscom.com>,
gla...@DELETEuscom.com (David Glasser) writes:

>Contrary to the implications in DoeaDeer's statement, the Inform default
>messages have no binding tie to Inform. It would take me maybe 10
>minutes to make TADS' messages equal to those currently in Inform.

No, I agree. I prefer messages that are close to the old Infocom messages and I
have been meaning to alter some of the Inform library messages to be closer to
that standard, I just haven't gotten around to it yet. (Also there wasn't a
standard response for some/most of them in the Infocom games, they differed
slightly from game to game, so they would have to be random or I would have to
settle on one).

I have altered one, a la, Ken Fair's post to raif about a year ago, "Sorry, I
don't know the word", whatever, "." I appreciated that in the old Infocom games
and I appreciate it now. It quickly lets the player know they are wasting their
time with that particular WORD, that it is not some other problem (untakeable
scenery, sentence constructed wrong, etc.). I can't figure out why more people
don't like it, but that is okay (I presume because they don't want to give away
an object that may not have been revealed yet, but phooey, I think that is a
minor point, if the game doesn't know "green" I am happy to find out.
Inadvertantly it also helps me debug my games, let's me know if I have included
the right synonyms or forgot to give a "dictionary" name to an object.)

Inform's original is "You can't see any such thing.", which in my games it
still does if the object is not in the location.

Off hand I don't know how Tads handles this.

I also replace "That is not something you need to refer to during the course of
this game." with "That is not something you need to bother with." (That's one
of the messages that differed from Infocom game to Infocom game, but I don't
like Graham's substitution, using the word, "game", I feel it breaks mimesis
too much.)

Ditto, I am not sure, off-hand how Tads handles this.

Doeadeer3

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

In article <1dhzctz.1gh...@usol-209-186-16-183.uscom.com>,
gla...@DELETEuscom.com (David Glasser) writes:

>Contrary to the implications in DoeaDeer's statement, the Inform default
>messages have no binding tie to Inform. It would take me maybe 10
>minutes to make TADS' messages equal to those currently in Inform.

No, I agree. I prefer messages that are close to the old Infocom messages and I

Doeadeer3

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

In article <364056...@shaw.wave.ca>, Roger Carbol <r...@shaw.wave.ca>
writes:

>Doeadeer3 wrote:
>
>> To me it would seem very ironic if Tads started disappearing NOW, just when
>> Mike Roberts is back to support it and when Graham Nelson hasn't been
>around to
>> support Inform (not saying Graham won't be back).
>
>I'm of the opinion that HTML-TADS will change all that, and TADS
>will experience a surge in audience growth.

It probably will. Also just the fact that Mr. Roberts is back should make a
difference. I think the rapid growth of Inform was PARTLY due to the fact that
for so long Mr. Nelson was around and Mr. Roberts wasn't. Now, interestingly,
the reverse is true.

It makes an author feel a bit more secure when they know a language is being
currently supported by the author.

Phase

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
jon...@zerospam.com (Jonadab the Unsightly One) writes:
>> with enough memory); and the lunatic who believes that TADS is the spawn
>> of Satan and has scripts set up to automatically delete any .gam on
>> his/her computer and bows down to an altar of Graham Nelson every night
>> (with voodoo dolls of MJR burning somewhere off to the side).
>
>You may be exaggerating just a smidgen on that last point.

Yeah, I personally worship Zarf and Mary Kuhner, but have yet to add
Mr. Nelson to the pantheon. :-)

I went with Inform, btw, just because I saw it first and got the
impression that it was more popular, true or no.

--
PHASEFX @ VM.SC.EDU - http://www.cs.sc.edu/~jason-e
Intelligence is supposed to be our greatest survival trait, yet
it is emotion that motivates us. Never rationalize away love.

Roger Carbol

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
Doeadeer3 wrote:

> I have altered one, a la, Ken Fair's post to raif about a year ago, "Sorry, I
> don't know the word", whatever, "." I appreciated that in the old Infocom games
> and I appreciate it now. It quickly lets the player know they are wasting their
> time with that particular WORD, that it is not some other problem (untakeable
> scenery, sentence constructed wrong, etc.). I can't figure out why more people
> don't like it, but that is okay (I presume because they don't want to give away
> an object that may not have been revealed yet, but phooey, I think that is a
> minor point, if the game doesn't know "green" I am happy to find out.


I mostly dislike it because I'm never sure just to whom the "I"
in the sentence "Sorry, I don't know the word fenestration" is
referring to.

"The word 'fenestration' is unknown" is a bit of an improvement,
but it still implies that it's unknown by *someone*.
"Unrecognized" is perhaps a bit better. "Unimplemented" is better
yet, but then we start getting close to the mimesis cliff again.


.. Roger Carbol .. r...@shaw.wave.ca .. I, zorkbot

Frank Filz

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to

How well does Inform use these optimized opcodes?

Another possible difference: Properties. (also, are TADS integers 32
bits?) I assume TADS objects are implemented more like C structures
whereas Inform uses the Z machine objects which have lists of properties
identified by property number and data.

--
Frank Filz

-----------------------------
Work: mailto:ff...@us.ibm.com
Home: mailto:ff...@mindspring.com

T Raymond

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
On 05 Nov 1998 18:11:57 GMT, doea...@aol.com (Doeadeer3) wrote:
> Inform's original is "You can't see any such thing.", which in my games it
> still does if the object is not in the location.
>
> Off hand I don't know how Tads handles this.

Unless it's been tweaked, I beleive the standard response is
"I don't see any here"

> I also replace "That is not something you need to refer to during the course of
> this game." with "That is not something you need to bother with." (That's one
> of the messages that differed from Infocom game to Infocom game, but I don't
> like Graham's substitution, using the word, "game", I feel it breaks mimesis
> too much.)
>
> Ditto, I am not sure, off-hand how Tads handles this.

I'm not sure in the general sense, but there is a class called decoration which you can
use to add something in like a painting on the wall or such, that the player can look at,
but can't take, etc. The default message for them is
"The <<painting, or whatever>> is not important."

I'm sure somebody with more time using TADS can correct me if I'm wrong. *s*
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Tom Raymond a...@usa.net
"I have everything except the combination... where is it?"
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

David Glasser

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
Andrew Plotkin <erky...@netcom.com> wrote:

> Jonadab the Unsightly One (jon...@zerospam.com) wrote:
> > Hmmm.... Maybe the library's more compact?
> > Just a guess; I don't know.
>
> That easy to check, though; make a one-room no-object game in each system
> and compare.
>
> I'd guess the size difference is 30% text compression and 70% differences
> in the code encoding. The Z-machine opcode format is very compact, and it
> has lots of abbreviations like one-opcode forms for "print and return", or
> "if (a == b) return".

Would this mean that a Z-machine interpreter would have to be bigger,
though?

David Glasser

unread,
Nov 5, 1998, 3:00:00 AM11/5/98
to
Doeadeer3 <doea...@aol.com> wrote:

> In article <1dhzctz.1gh...@usol-209-186-16-183.uscom.com>,
> gla...@DELETEuscom.com (David Glasser) writes:
>

> >Contrary to the implications in DoeaDeer's statement, the Inform default
> >messages have no binding tie to Inform. It would take me maybe 10
> >minutes to make TADS' messages equal to those currently in Inform.
>

> No, I agree. I prefer messages that are close to the old Infocom messages
> and I have been meaning to alter some of the Inform library messages to be
> closer to that standard, I just haven't gotten around to it yet. (Also
> there wasn't a standard response for some/most of them in the Infocom
> games, they differed slightly from game to game, so they would have to be
> random or I would have to settle on one).

I figured you undestood what I meant, but I did want to make it clear
that one can change the messages. (Though it would probably mean
reading through the entire library, as there is no LibraryMessages.
Sigh.)

> I have altered one, a la, Ken Fair's post to raif about a year ago,
> "Sorry, I don't know the word", whatever, "." I appreciated that in the
> old Infocom games and I appreciate it now. It quickly lets the player know

> they are wasting their time with that particular WORD, that it is not some


> other problem (untakeable scenery, sentence constructed wrong, etc.). I
> can't figure out why more people don't like it, but that is okay (I
> presume because they don't want to give away an object that may not have
> been revealed yet, but phooey, I think that is a minor point, if the game

> doesn't know "green" I am happy to find out. Inadvertantly it also helps
> me debug my games, let's me know if I have included the right synonyms or


> forgot to give a "dictionary" name to an object.)
>

> Inform's original is "You can't see any such thing.", which in my games it
> still does if the object is not in the location.
>

> Off hand I don't know how Tads handles this.

The way you like it.

> I also replace "That is not something you need to refer to during the
> course of this game." with "That is not something you need to bother
> with." (That's one of the messages that differed from Infocom game to
> Infocom game, but I don't like Graham's substitution, using the word,
> "game", I feel it breaks mimesis too much.)
>
> Ditto, I am not sure, off-hand how Tads handles this.

I don't think anything like that is built in (though I could be wrong).

David Glasser

unread,
Nov 5, 1998, 3:00:00 AM11/5/98