ISSUE #44 - April 30, 2006

      ___.               .___              _             ___.
     /  _|               |   \            / \           / ._|
     \  \                | o_/           |   |          | |_.
     .\  \               | |             | o |          | | |
The  |___/ociety for the |_|romotion of  |_|_|dventure  \___|ames.

                         ISSUE #44

        Edited by Jimmy Maher (maher SP@G
                       April 30, 2006

           SPAG Website:

SPAG #44 is copyright (c) 2006 by Jimmy Maher.
Authors of reviews and articles retain the rights to their contributions.

All email addresses are spamblocked -- replace the name of our magazine
with the traditional 'at' sign.

ARTICLES IN THIS ISSUE ----------------------------------------------------

The SPAG Interview: Graham Nelson and Emily Short on Inform 7

REVIEWS IN THIS ISSUE -----------------------------------------------------

Akari's Story
All Hope Abandon
The Baron
Damnatio Memoriae
Dracula: The First Night
The Potter and the Mould
The Warlord, the Princess, and the Bulldog

The Baron


From the department of the obvious: This issue is late.  For that, I heartily 
apologize, both to SPAG's readership and (especially) to those who sent in 
content and have been patiently waiting for it to appear.  Now, however, I will 
haul out my excuses.  The first two weeks of delay were entirely due to my own 
personal situation.  I was in the midst of completing my university coursework 
for the semester, and just did not have the time to do an adequete job with 
SPAG.  The next week of delay was to allow the Spring Thing competition to end, 
and for me to print Mike Snyder and Dan Shiovitz's eloquent analysises of the 
games that were entered there.  Finally, I had the opportunity to interview 
Graham Nelson and Emily Short about the forthcoming (and potentially 
revolutionary) Inform 7 for the price of one more week's delay, and I just could 
not pass it up.  The good news, though, is that we have another very substantial 
issue here, thanks once again to all of you, featuring commentary on some fine 
examples of present-day IF.

I have been thinking a fair amount lately about the current state of the art in 
IF.  If we date the beginning of the modern era from the beginning of the IF 
Competition in 1995, we are now far enough into same that larger trends might 
start to assert themselves.  Modern IF, in short, has a history of its own now 
apart from the larger story of IF as a whole.  Prompted partially by some recent 
(and occasionally heated) discussion on the newsgroups, I find myself wanting to 
understand those trends.  Whether my analysis will be correct in the end only 
more time and the better perspective it brings will tell, but I am going to 
briefly describe the way I see IF's recent evolution.

I think we might call the first six or seven years of modern IF, from 1995 to 
2002 or so, the era of experimentation.  It was marked by an explosion of games 
that seemed determined to sort out just what this medium could do and be now 
that it was freed from the commercial constraint that "IF" must be a synonym for 
"(text) adventure game."  That First Competition included C.E. Forman's Mystery 
Science Theatre 3000-ization of Detective, Garath Rees' one-room experiment The 
Magic Toyshop, and Neil Demause's bizarre self-referential Undo.  As time wore 
on, such experiments in form continued unabated.  I am thinking of works like 
Aisle, The Space Under the Window, Shrapnel, and many others here.  Even the 
much-loved Photopia could fall into this category.  (Can we remove all of the 
gamelike elements from IF and still make it compelling?)  Galatea fits as well.  
(Just how believable of an NPC can we manage with our current tools, and can we 
build an interesting work around nothing but interaction with her?)  Of course, 
this was not ALL that was going on.  Plenty of traditional text adventures were 
being written as well.  Still, I think I see a trend here, and possibly a 
dominent trend.

The works I have just mentioned, and many similar ones I did not, were all 
important, indeed essential, to IF's artistic growth.  They were, however, also 
somewhat limited in their artistic success by their very natures as formal 
experiments.  I do not think that Neil K. Guy felt compelled to tell the story 
of an average fellow in a grocery store, and decided that a one-move work of IF 
was the best way to do that.  Similarly, and perhaps more controversially, I do 
not really see Emily Short as writing Galatea with the primary intent of 
illuminating the title character for her players.  I think she wanted to 
experiment with conversation in IF, and chose Galatea after making that decision 
as a clever and interesting subject.  (Photopia, on the other hand, I will leave 
alone as the exception that proves the rule.)

I had a creative writing teacher once who gave me some advice I have never 
forgotten.  She told me that the form of a work should arise organically from 
the nature of what the artist is trying to express.  This remark was made in 
response to the trend nowdays in fiction to break the rules seemingly for the 
sake of it through the employment of fragmentation, deliberate grammatical 
errors, colors and odd typefaces, etc.  Such things, my teacher told me, should 
only be used if the artist cannot get her message across without them.  
Otherwise, they merely draw the reader's attention to the surface of the work 
and away from its real heart.  They are, in short, mere showboating by the 
author, who seems to be crying, "Look at me!  Look how very clever and 
postmodern I am!"  One is often left to suspect that the heart of the work is in 
fact empty, and thus all its author has to work with is its surface textures.

I don't mean the preceding to sound as harsh as it possibly might in the context 
of this discussion about IF.  I don't see the authors of any of the works 
mentioned above as being guilty of showboating for its own sake in the way that 
irritates me with authors of postmodern fiction.  What I do see, though, is 
these authors working out through experimentation and trial and error just what 
the tools of IF really are, and what can and cannot be done with them.  Think of 
Renaissance artists discovering the rules of perspective through a series of 
sketches.  The IF works I have been discussing are works of real quality, 
interest, and historical importance, but they are also limited by the very 
motivation behind their creation.  Their authors did not say, "I have this 
amazing story -- (or, depending on the nature of the work, amazing game) -- to 
communicate, and this form is the best way to do so."  It rather seems to me 
that the form came first, and the details of the story/game followed.  This does 
not make the works any less clever or important, but it does give them a certain 
coldness or even, to use a word batted rather recklessly about the newsgroups 
recently, pretension.  As was also mentioned, though, pretension is not always a 
bad thing, and is often necessary for the advancement of an artform.

I believe I see a change in the games being produced by the community in the 
last few years.  The formal experiments have decreased considerably in number.  
Increasing in number, though, are works that internalize the fruits of that 
experimentation, as well as the considerable body of design theory the community 
has built up over the years, in the service of their artistic intent, rather 
than the other way around.  Take Vespers, winner of last fall's Competition, for 
example.  It tells a story that I personally find as emotionally compelling as 
Photopia, yet it manages to also do something Photopia did not: to put the 
"interactive" into its "interactive fiction."  The narrative and the crossword 
are in perfect balance.

For an artist to create great work, she must first understand her tools.  Just 
possibly, we in the IF community are finally reaching that stage.  It may be 
that in ten years whoever is editing SPAG will point to the last few years as 
the first rumblings of the artistic maturity of IF. 

IF NEWS -------------------------------------------------------------------

Wei-ju Wu has developed a new Java Z-Machine interpreter.  Everyone who is still 
using ZPlet to imbed IF games in their web pages should definitely take a look, 
as Wu's program is already at version 0.89 much more attractive and capable.  
Also, Q.P. Liu has written a Z-Machine interpreter entirely in JavaScript (!).  
It's a bit slow, but may be worth a look.

IF's equivalent of the Oscars, for the best games released during the previous 
year, were recently announced.  My, what a good year Jason Devlin had.  
Congratulation to all of the winners!  Visit the XYZZY News website for a 
complete transcript of the ceremony from the IF MUD.
Best game: Vespers, by Jason Devlin
Best writing: Vespers, by Jason Devlin
Best story: Beyond, by Mondi Confinanti (Roberto Grassi, Paolo Lucchesi, and 
Alessandro Peretti)
Best setting: Vespers, by Jason Devlin
Best puzzles: Distress, by Mike Snyder
Best NPCs: Vespers, by Jason Devlin
Best individual puzzle: Following the murderer in Beyond, by Mondi Confinanti 
(Roberto Grassi, Paolo Lucchesi, and Alessandro Peretti)
Best individual NPC: The storyteller in Whom the Telling Changed (by Aaron A. 
Best individual PC: Wendy Little In Tough Beans, by Sara Dee
Best use of medium: Mystery House Possessed, by Emily Short

The Spring Thing competition has just wrapped up as I am writing this.  Thanks 
to Greg Boettcher for administering it for a third year, and to Mike Snyder and 
Dan Shiovitz for their timely reviews of the games to be found in this very 
issue.  All four games are eminently worth playing, but this is after all a 
competition, and so the results were as follows:
1. The Baron by Victor Gijsbers
2. The Potter and the Mould by Robert Street
3. Pantomine by Robb Sherwin
4. The Warlord, the Princess, and the Bulldog by David Whyld

Once again you folks have come through for me, helping out with reviews of older 
games to clear much of my backlog.  There are always more being released, 
however, so your work and mine never really end.  Would we really want it any 
other way?

1.  Eragon
2.  A Spot of Bother
3.  1893: A World's Fair Mystery
4.  Finding Martin
5.  Swineback Ridge
6.  Mystery House Taken Over games (any, some, or all!)
7.  Remaining IF Comp 2005 Games
8.  Dracula: The Arrival
9.  2k Competition 2004 Games
10. IntroComp 2005 Games

THE SPAG INTERVIEW---------------------------------------------------------

We may someday look back upon April 30, 2006, as a hugely significant day in the 
history of IF.  If you haven't heard yet, Graham Nelson has today released the 
latest version of his Inform IF development language.  Actually, thinking of 
Inform 7 as just an update to Inform 6 does not begin to do it justice.  It is 
in fact a major paradigm shift, not just for Inform but for IF development in 
general.  Inform 7 allows its user to create IF not through a conventional 
programming language but through natural language descriptions of the 
environment to be simulated.  The result is a tool that both Graham and his 
closest partner in Inform 7's development, Emily Short, feel is both more 
powerful and flexible and much easier to use than Inform 6, and that has the 
potential to expand IF's authorship community immensely.  The presentation side 
of things has also not been neglected.  Inform 7 games are developed within a 
slick and professional IDE (available for Windows and Mac OSX), and can be 
packaged for release with cover art that is automatically displayed at run-time 
on interpreters equipped with the ability to do so.  I encourage every one of 
SPAG's readers to visit the Inform 7 website at to 
see for yourself.  I think you will come away excited and impressed.

Graham and Emily were kind of enough to take time during the hectic final days 
of Inform 7's development to talk with SPAG.  Their interview follows.

  SPAG: We all know what Emily has been up to recently, namely authoring a 
  string of interesting and important games.  You, Graham, had however retired 
  from the limelight completely.  I had assumed that you had lost interest in IF 
  and moved on to other things.  Now that I realize that is not the case, I 
  wonder just how long you have been working on Inform 7. Could you briefly 
  describe its development history for us?

GN: I tend to get involved publically only when I feel I have something to say, 
I suppose, but my apparent invisibility in recent years is partly an aspect of 
the natural life-cycle of community software. Inform 6 is not gcc, or emacs, or 
even curl, but it is (I hope) a fairly dependable command-line tool. After rapid 
flux in 1993-96, during which Informs 1 to 5 never stood still long enough to be 
debugged, I spent about six months recoding the compiler from first principles - 
this was Inform 6. It then needed stability and a trustworthy infrastructure of 
support and documentation. In 1999-2001, spurred on by the kindness and interest 
of Mike Berlyn and later David Cornelson, I worked mainly on the Designer's 
Manual, notionally the fourth edition (people took to calling this the "DM4") 
but really the first to be a genuine book. Gareth Rees's editing forced me into 
doing the job properly: the best kind of editor is also the author's conscience. 
The DM4 is the scene of a collision between two quite different books - one on 
the history and practice of IF, the other on Inform 6 - but perhaps that is what 
gives the book its cult status, if I can call it that. I have always rather 
liked hand-made books which expose the inner landscape of the writer - Robert 
Harbison's "Eccentric Spaces", say. Sometimes such books put a spurious 
intellectual gloss on individual captivations which leave others cold, or worse, 
embarrassed: and they can be awfully gauche, like "Godel, Escher, Bach", or just 
plain mad, like "The White Goddess", but there's a kind of honesty in them. I 
don't compare myself, but the DM4 belongs to that genre. It continues in print, 
and Dan Sanderson published a third impression just this month - actually, I 
received the first hardback copy today. Buy early for Christmas.

After 2001, Inform 6 stabilised to its published description. The Inform-
maintenance mailing list organises bug fixes - Roger Firth, Cedric Knight, David 
Kinder and other fine people do all the work; I doze benignly on my peacock 
throne, occasionally murmuring when something radical is said, but otherwise 
leaving Inform 6 maintenance to the professionals. A few years ago, Roger asked 
me what would be the next step forward for Inform 6: I think I disappointed him 
a little by saying that I wanted stability. People do sometimes push for, say, 
Java-like exceptions, or other new linguistic features, but I think they would 
add little. While there are minor process improvements to Inform 6 which I think 
would be helpful - more transparent Unicode support, for instance, a little more 
work on Glulx compilation - and while we do need more reliable unit testing, 
Inform 6 is in an essentially final state.

So in 2001-02, I felt that I had pretty well said what I wanted to say about IF. 
Except... Off and on, I worked on a little draft of an IF, which never got much 
further than a few scenes because I kept grinding to a halt. I found using 
Inform at the command line to be fussy, inelegant. To start a project took ages. 
And why was I typing so much verbiage just to establish that one thing was in 
another, and so forth? I formed some hazy thoughts about what would be a better 
interface, but they seemed hopelessly unrealisable.

But over 2002-03, I began again in earnest, and persuaded Andrew Hunter to join 
the project. I had by now written what I thought of as the "Book of Inform", a 
brief description of how Inform should present itself to the user - the name 
being a homage to Jef Raskin, who wrote a similar "Book of Macintosh" at Apple 
twenty years ago. Andrew, who had already written a beautiful Z-machine 
interpreter for Mac OS X - Zoom - agreed to take on the coding of the user 
interface: I would write the actual compiler inside, and the two together would 
be Inform 7. The final interface isn't exactly as once specified - many ideas 
came later and were Andrew's, not mine - but it's pretty close to the original 
"vision", if I can call it that. David Kinder's user interface for Windows came 
later and is in some sense a clone of the Mac OS X one, but they have no code in 
common - his is an independent miracle of programming.

The history of Inform 7 can be divided into five: initial sketches and false 
starts; completion of a superficial, but just about IF-capable first draft (it 
was this that Emily used to write "Mystery House Possessed", whose banner was 
the first public appearance of the words "Inform 7"); going back to the drawing 
board, which meant fairly intensive study of linguistic semantics and even a 
little philosophy of language; realising that the new features added as a result 
of this had enormously expanded the scope of Inform, and developing the 
consequences; and finally, polishing, documenting, finishing up.

I hope it will not be thought that I am the only begetter of Inform 7. Emily 
Short became increasingly closely involved after the first year, and went from 
being an adviser to becoming - as she now is - in practical terms co-regent. 
Andrew Plotkin remained an occasional adviser, so his contribution was more 
modest, but he injected exactly the right radical ideas at the right time. 

Emily, want to talk about what it was like from your point of view? 

ES: I, er, gosh. A blur. I guess I spent roughly a year filing hundreds of 
fiddly bug reports and writing long, annoying email about all the difficult 
features I wanted added; and another year or so writing test code and examples 
and full games. So during the first part I spent a lot of time thinking about 
what the strengths of the system were, and how they could be enhanced -- whether 
there were ways to generalize or abstract the things that it already did, to 
make them more powerful. The second half was turning around and applying those 
things again, and making sure Inform really could handle all the applications I 
envisioned for it. There are still, oh, two dozen or so directions I want to 
develop further, but it's time to let other people have a crack at it too.

  SPAG: Many Internet programs, especially those intended to be used by a   
  community, are written in public, with open source and frequent published 
  drafts. Why is Inform 7 only now making its first public appearance?

GN: Partly my nature, no doubt. TADS 3 has had a much more open process - and 
perhaps a more confident one. I will certainly plead guilty to being a control 
freak. The hard decisions are what to take out, not what to put in, and that 
becomes more difficult when there is a nascent user base. (Also, you need to be 
careful which wish-lists to read: the wish-list you need to pay most attention 
to is the one written by a designer of actual IF, not a well-motivated 

Had Inform 7 been developed in open source, I am fairly sure it would now be an 
elaborated version of the superficial prototype, and that it would be much the 
poorer. And it ought to be remembered that for at least the first year of the 
project, I wasn't at all sure it would ever work - "work" in the sense of being 
capable enough to be useful.

I do not quite buy the argument put by Eric Raymond in "The Cathedral and the 
Bazaar", that the bustling, self-organised world of the bazaar gets things built 
better and faster than the secretive clergy. I suspect he chooses a cathedral to 
rebut the famous passage in Brooks's "The Mythical Man-Month", which points to 
the aesthetic triumph of those cathedrals whose masons stuck to the original 
design conception, rather than adding incoherent but fashionable motifs in each 
generation. But I think Brooks is right more often than Raymond. (Actually, we 
need both kinds of software development: we need both Linux and Mac OS X, the 
classically open bazaar and classically secretive cathedral - they keep each 
other honest.)

In particular Raymond's argument seems to me applicable only to programs whose 
basic operation is clearly understood by all parties in advance. The bazaar gave 
us Linux, but Linux was essentially a duplicate of an already very well 
established mode d'emploi, as we might say. To be fair to Raymond, though he 
says some silly things about academia and is somewhat in the thrall of the Ayn 
Rand and Robert Heinlein school of society - I'll bet he read "Farmer in the 
Sky" as a kid, and thought it just dandy - let me note that his actual writings 
are more pragmatic and less ideological than the 20-point headlines printed 
above them. I think Raymond accepts that all projects go through both cathedral 
and bazaar phases.

Besides, these two cultures often overlap. Chartres cathedral used to have an 
actual bazaar in it, in the middle ages - you could buy vegetables, or even 
livestock, and it must have been mayhem sometimes: but that was all right, 
because the occasional runaway piglet was never going to be able to knock over 
the columns holding up the walls. Well, if Inform is a cathedral, its explicit 
support for extensions is the equivalent of inviting the townsfolk in to set up 
their stalls.

To give a perhaps ridiculous example - only I don't think it is ridiculous - of 
what I mean by holistic design, over the last week we have been worrying about 
the colour of two of the icons used in the Scene index. This tiny corner of the 
user interface is based on the metaphor of traffic lights - the icons are green, 
amber, red. But was the middle colour too yellow? (We looked at photographs of 
actual British and American traffic lights at this point.) And was a round shape 
quite right, or would it wrongly suggest affinities with other round elements in 
the user interface? We didn't spend long on this, but we went through five 
iterations of icon designs, and it was time worth spending. I think my ideal for 
software may not be as ambitious as a cathedral, but it is good to try to build 
something like a Charles Rennie Mackintosh house: one in which the furniture, 
the decor and even the cutlery are part of the line of the whole.

The question, of course, is whether anyone would feel able to sprawl out and 
relax in a Charles Rennie Mackintosh house. We shall see.

  SPAG: Until now, there have been two paradigms for the development of IF.
  There is the programming model, which brings great power and flexibility but 
  can be difficult to learn, especially for the non-technical.  And there is the 
  GUI-based model, which is quick and easy to learn but sacrifices power for 
  ease of use.  With Inform 7's natural language model, you have introduced a 
  third way.  What led you to the idea of using natural language for IF 

GN: It's funny: I used to think that conventional styles of programming "bring 
great power and flexibility", but now I'm not so sure they do, in this 
particular arena - though I certainly accept that they are way ahead of point 
and click models.

I think it is significant that the impetus for natural language came out of a 
desire for better semantics, not better syntax. Many systems which adopt 
English-like scripting languages do so in a way which might be called 
"procedural programming plus syntactic sugar", and the result is seldom very 
edifying, except to produce readable configuration files now and then.

The answer is that this came out of a tiny piece of unfinished business from the 
DM4. One of the last chapters to be written in the DM4 was a formal semantic 
description of the Inform world model - semantic in the sense of linguistics, 
not computing: there were sections on how time, space, and so forth were 
conceptualised. I would read this - mostly lucid, mostly precise - and then look 
at the quite horrible expression of the same ideas in the Inform 6 library, 
which is written in Inform 6 code. And I would think: why can't Inform simply 
read the world-model chapter of the DM4, and use that as its library?

ES: I confess that when Graham originally wrote to me to say he was working on a 
natural language version of Inform, my reaction was, er, extreme skepticism. I 
didn't see how such a system could be anything but infuriating to program in; 
the natural language bit seemed to be at best a way for the novice to sketch out 
some rooms and objects, but I imagined the real work would have to be done at 
the I6 layer, and that anyone half-competent at I6 would probably find they 
preferred just staying at that level. So I wrote back what I hope was a polite 
letter, expressing an interest in seeing what he came up with, but hinting that 
trading Inform's native power for this sort of convenience would be a serious 
mistake. I also had some concerns about the fact that this kind of language 
would be less accessible to non-English speakers.

At some point, though, I started to realize that this natural language business 
had deeper implications than I had previously appreciated, to do with making the 
parser and the world model and the output all work together properly, and that 
it was adding power to the system, as well as readability; which is the point at 
which I became really committed to working on the project through to the end.

  SPAG: I understand that Inform 7 is not really a compiler, but rather a pre-
  processor that generates Inform 6 code, which is then in turn compiled by the 
  Inform 6 compiler.  Was there any reasoning behind this choice beyond 
  technical expediency?

GN: Well, again, I hope it will not be thought that Inform 7 "is" the compiler. 
It is the compiler and the interface and the documentation, all merging one into 
another from the user's perspective. (The compiler is actually called NI, for 
"natural language Inform", but this name is not outwardly visible.) As for "pre
-processor", perhaps I might observe that the Inform 7 compiler source is much 
larger than the source for Inform 6.

It is true that Inform 7 code-generates to Inform 6, rather than to native Z-
code machine language. This has obvious benefits in terms of the proven 
reliability of Inform 6 as a code generator, and also for cross-platform 
compilation to the Glulx virtual machine (an eventual goal, though one that 
isn't yet supported). I think of Inform 6 as a rather low-level language, but 
one that is securely implemented: I'm naturally keen to leverage all of the 
programmer-hours which have gone into making it work. It also contains a solid 
kernel of run-time code - notably the parser.

  SPAG: Emily, you stated on the newsgroups that Inform 7 actually allows you to
  do things easily that were difficult or impossible in Inform 6.  I assume it 
  must have advantages and capabilities that go beyond being more readable and 
  easier to use than Inform 6.  Now that the cat is fully out of the bag, as it 
  were, could you elaborate a bit about what those are?

ES: This is a pretty large question, so I will stick with a couple of major 
points I'm not mentioning elsewhere - one to do with natural language, one to do 
with abstractions, and one to do with Inform as an application - though in fact 
there are others.

Re. natural language: One of the things one is always doing in traditional IF 
coding is trying to get the parsing, the screen output, and the world model to 
line up: make sure all the words that are ever used to describe something are 
also valid for parsing it, make sure that important conditions of objects are 
always described to the player, that sort of thing. And it is even more work 
dealing with temporary conditions. Inform 6 does a bit with understanding "lit" 
and "unlit" to distinguish objects, for instance, but it's not easy for the 
author to add to that system. Along the same lines, I6 automatically describes 
things as lit/open/etc., but it takes considerably more effort to turn off that 
automatic description or get the game to narrate some other aspect of the world 
model entirely. 

With Inform 7 this was something we had an opportunity to improve, and the 
natural language component made it easier to bring those three aspects - input, 
model, output - in line with one another. Adding a concept like "color" or 
"weight" to an I7 game doesn't affect the model world alone; it also means that 
Inform 7 knows how to print and parse those words or units. This offers a lot of 
help in implementing objects that change in complex ways (like liquids, or 
burning objects), and in letting the player specify precise interactions (like 
"cut two inches off the twelve-inch string").

Second, the language allows the author to define (and then refer to) new kinds 
of relationship between things. The classic I6 world model knows about some 
basic relationships, like the relationship between a container and the contained 
object, but it is not always possible to expand upon that as richly as one might 
like. In Inform 7 it is possible to make up entirely new kinds of relation, and 
then test for those relations, change them easily, and run path-finding routines 
through them. So I could, if I wanted, add to my model world the concept of one 
object being under another, and from then on refer to "the things which are 
under the chair", for instance. But I can also use these relations to say things 
like "this conversation topic relates to these other three conversation topics": 
now Inform has an idea of relations between abstract ideas, which was never 
present in the built-in world model, but I can bring to bear on that relation 
various abstract tools. I can write a grammar line that tells the parser only to 
understand topics related to the current one, for instance; I can change the 
interrelation of topics midway through the game; I can write NPCs who seek a 
path to a specific conversational goal, just as though they were seeking through 
the rooms of the map. In "Bronze", I used relations to describe which objects 
are required to solve the puzzles presented by other objects, which means that 
the adaptive help system is working on a (primitive, but adequate) model of 
puzzle structures, and I didn't have to think of all the possible 
interdependencies myself. "Damnatio Memoriae" runs through all the kinds of 
relation that Inform recognizes, partly as a test of the system, but also 
because this was a very trim and tidy way to express the in-game style of magic.

Finally, Inform 7 includes a number of features that make it easier to develop a 
solid game. Some of this is reflected in the indices: you can check out your map 
and your world layout easily, and see whether Inform agrees with you about the 
way you've set things up. Less immediately flashy, but hugely valuable, are the 
pieces that keep track of previous playthroughs and let you automatically verify 
an entire walkthrough or a single portion of the game. "Bronze" was less painful 
to beta-test than just about any game I've written in Inform 6, though it does 
some challenging things, because it was developed simultaneously with the 
mechanisms for automatic testing; as I put new pieces in place, I was able to 
verify the outcome more thoroughly and efficiently than before.

  SPAG: Many aspects of Inform 7 are forthrightly booklike.  I speak not only of
  the natural language model itself, but also the appearance of the IDE as an 
  open book with two facing pages, and the inclusion of "cover art" with Inform 
  7 story files. Is this indicative of a desire to drive Inform 7 toward a more 
  literary model, and if possible to attract more literary people (as opposed to 
  technophiles) to IF?

GN: Again, we use these computer-programming terms! I suppose the user interface 
for Inform is indeed an "IDE", but somehow that isn't how I picture it. The 
question I have tried to ask is: what is the natural representation, the natural 
expression? How do we think about something, and how can we make the user's 
engagement with it unconscious, rather than being a deliberative act of 

I do want to make Inform accessible to a wider community. The manual says that 
it is for "computer programmers intrigued by writing, and writers intrigued by 
computer programming", but truthfully, I'd like to see IF tools - not just 
design systems, but also iTunes-like browsers and interpreters - which open up 
IF to that huge creative community of people who write blogs, and design their 
own websites, often startlingly well. IF will never be for everyone, but I would 
like it to be on the table as a viable form of artistic expression.

On cover art, I think people have been wary in the past because of a suspicion 
of illustrated IF - bad partly because only graphic artists can make it work, 
but also perhaps partly because it is associated with the collapse of the glory 
days of commercial IF: it is seen as a slippery slope, a deviation from purity. 
I'm sympathetic to that, and wouldn't, for instance, be much impressed to find 
drawings every twenty pages in a new novel by Ian McEwan. On the other hand, I 
do not expect a new McEwan book to have a uniformly beige cover.

ES: You might be impressed to find a vintage children's novel with the original 
plates, though, or a manuscript with illustrated initials, or a textbook with 
historical map inside the front cover. There are plenty of uses for illustration 
in IF other than trying (and failing) to emulate a graphical adventure game. 
There have been a handful of IF games in recent years that do this really right, 
and I would like it to be easier. I'd like to be able to write a game that had a 
frontispiece illustration before each major plot scene, for instance. Probably 
one drawn by someone other than me, but that's beside the point. It's possible 
to find collaborators for these things.

But anyway, you were saying about cover art...

GN: I think people have a sophisticated understanding of the level of artistic 
statement being made by a cover picture: that it is not of the book, yet sets 
the tone for it: that it may come and go with new editions, and indeed may give 
an old text a new lease of life. I think it's good to employ the sophisticated 
understandings we already have, rather than miss out on a potentially 
interesting form of expression.

There's also a practical reason for cover art: most of today's IF tools have 
their origins in the age of FTP, when cover images would have been meaningless. 
Today we must engage with the Web, and the world of blogs - a world of 
promiscuous quotation not only of text but also of the image.

Lastly, it's not an accident that Inform has a large suite of features for 
putting finishing touches to works, and that one whole chapter of the manual is 
about "Publishing". That isn't entirely a matter of providing a whole solution, 
like the teaspoons in my imaginary Charles Rennie Mackintosh house. It is also 
one of several ways in which Inform tries to manipu - excuse me, to encourage 
the user to produce something finished, and tidy, rather than makeshift.

ES: I think "seduce" is the word you're looking for.

  SPAG: Are any plans afoot to offer the Inform 7 manual in hardcopy form, as
  was done with the Inform 6 Designer's Guide?

GN: Yes, but not yet: there are sure to be changes, and perhaps major changes, 
as Inform 7 meets its users. What we would really like to publish - I say "we" 
because while I wrote the main text, Emily wrote almost all of the examples - 
would be an elegant, almost coffee-table book. One whose design doesn't evoke 
computing books of the drab Addison-Wesley style, but is more like a lifestyle 
cookery book, or one of the more aesthetically pleasing Photoshop guides. I 
certainly want something that would comfortably lie flat open. Digital printing 
has come a long way since the DM4's first impression in 2001, but our 
aspirations are probably a bit ambitious for now. Still, we might just be able 
to do it, if we printed the first edition by subscription, as David Cornelson 
did with the DM4: let's see.

Actually, "the" manual is really two different volumes interleaved: the main 
text, and the examples, which are organised into a recipe book aiming to show a 
comprehensive range of techniques. From one viewpoint the examples footnote the 
main text, from the other it's vice versa. On screen, one can have it both ways, 
as the reader prefers: for the printed manual, we would have to choose.

  SPAG: You mentioned in your notes on this project that you are attempting to
  construct a rules-based rather than object-based model for IF development. Can 
  you briefly explain what you mean by this?

GN: This is Andrew Plotkin's idea more than my own. As you say, I've gone into 
this in some detail in the formal paper about the project, but broadly I would 
say that Inform 6's biggest semantic failing is that it ties all events and 
behaviours to individual objects (well, or classes, to be fair: but then you 
don't have easy access to the class Object, to put it mildly). Suppose one 
possible event is Alice eating a cake and shrinking. Is that a behaviour of 
Alice, or the cake? Why must we choose? Source code for IF games like Scott 
Adams's "Adventureland", which is stuffed full of combinations of objects doing 
something or other, becomes impossible to read in any linear (i.e., plot-
respecting) way. Worse, if it's a change to the standard rules of realism (to do 
with what containers people can go inside, say) then it may be impossible to 
implement at all, without rewriting chunks of the Inform 6 library. There is 
constant pressure for said library to call more "entry points", to provide more 
hooks to hang customised code on, etc. The system grows steadily more complex 
and arcane - and it's never enough.

ES: For instance, with "City of Secrets", I had to replace the library's 
handling of movement verbs to allow for the player to use a >GO BACK command. In 
Inform 7 the same job is accomplished with a couple lines that add an extra step 
to the Go action, without disrupting any of the things that are already supposed 
to happen.

GN: Rules allow us to treat the world model and the actual game as interleaving. 
There is no dichotomy between the world model (fixed and with universal 
validity) and the objects in play (free to be altered but in ways that have only 
very specific validity).

ES: I have found that the rule-based aspect means I tend to rough out a whole 
system fairly efficiently (for conversation, or some simulationist purpose or 
other) and get a structure working fast. There's still a lot of refinement to do 
after that point, but it's comparatively easy to build new features into the 
framework without doing a lot of rewriting. 

GN: Also, we can write the source in whatever order we think best. So once the 
framework is in place, it's easy to tidy up - to rearrange under suitable 
headings, say - without affecting its functionality. I reorganised "Reliques of 
Tolti-Aph" two or three times until I was happy with its chapter and section 
breakdown. I actually feel that this freedom considerably adds to the expressive 
power of the language even though it makes no difference to the outcome - it 
enables me to have the source text read back in a way which reflects how I think 
about the problem.

  SPAG: The two of you recently released three demonstration games as a taster
  for Inform 7.  Some quite sophisticated effects were to be found therein that 
  would have required considerable programming skill, and even in a couple of 
  cases recourse to Z-Machine assembly code, to achieve.  Were these games 
  written entirely in Inform 7's natural language?

GN: "Reliques of Tolti-Aph" - a frivolous piece of IF, whose real purpose is to 
test whether Inform could handle highly algorithmic puzzles - is written 
entirely in natural language, about 33,000 words of it. The text can be browsed 
at the Inform website.

ES: There is almost no direct use of Inform 6 in mine either; all the screen 
effects and status line elements rely on extensions included with I7. (The 
extensions do drop back into Z-machine assembly code, but the user of the 
extension doesn't have to worry about that.)

GN: It may also be worth mentioning that all three of these "worked examples", 
as we call them, were written very quickly - "ROTA" in the spare evenings of 
less than a fortnight, for instance, plus a little time for play-testing.

  SPAG: Two classic pitfalls of IF development are the modeling of liquids and
  the modeling of ropes.  Will Inform 7 offer any help here?

ES: Neither is built into the default world-model, but I did write some example 
code for rope-handling, and several different pieces of code for liquid 
implementation. These range from the very simple (containers that are simply 
full to different levels or not) to a full system of mixture that keeps track of 
the components within and automatically identifies recipes (so that if you've 
poured the correct amounts of orange juice and champagne, say, it will relabel 
the liquid as "mimosa", with appropriate parsing and naming).

GN: Is that what a mimosa is? I do wish people would just say orange juice and 
champagne if -

ES: Yes, well, as we see not everyone wants this level of complexity.

There are a number of other similar things in the examples as well: models for 
different types of conversation, telephones, objects that can be broken into 
component parts, cutting, burning, more sophisticated treatments of lighting, 
NPC goal-seeking, parsing keywords and adverbs, automated help for stuck novice 
players, organization by plot and scene rather than by location, and so on. Some 
of these are very standard simulation problems, and some have more to do with 
overriding the defaults of the world model entirely and making interactive 
fiction that behaves in an utterly different way. As I wrote the examples I 
spent a huge amount of time combing through extension libraries, RAIF posts, the 
IFwiki - anywhere where anyone had posted about wanting to be able to implement 
any hard thing at all - and tried implementing these things at least to a basic 
level in I7.

Sometimes I ran into trouble, and then I pestered Graham for feature additions. 
A few of these involved major reconsiderations of how the model works - non-
player character actions are now handled more or less on a par with player 
actions, so it's possible to make the same rules apply to both. The lack of this 
had always struck me as a major weakness in Inform 6; combined with Inform 7's 
action-handling procedures, it means that goal-seeking behavior can be written 
for NPCs using the same mechanisms that we use to write implicit actions for the 

Similarly, Inform 7 has a more developed mechanism for dealing with output. It 
was an early conversation with Nick Montfort that convinced me we should be 
thinking about descriptions and output as a separate aspect of the system, 
essentially, alongside the parser and the model world; and though there is quite 
a lot more one could do with this, it is now much easier to do things like keep 
track of what the player has been told about, and also to write rich and complex 
scene descriptions that vary depending on what's present and don't look so auto

But quite often I found that Inform 7 did already provide the leverage I needed, 
or would with just a little minor modification, which I find encouraging.

So I am hoping that people will be able to pick and adapt from these whatever 
they find most suited to the game they're writing - and, of course, people who 
develop more elaborated systems can, as always, offer those as extensions. I am 
not at all under the illusion that these examples are the final word in any of 
the areas they cover. But there was a kind of intentional propaganda in writing 
a lot of disparate examples that implied entirely different kinds of IF: I 
wanted to show off some of the true flexibility of the system, and encourage 
authors to think beyond the usual paradigms. We've had a number of interesting 
discussions on the newsgroups over the years about what IF might be like if it 
were less based on geographical rooms, a simple physical model, and so on; here, 
I think, is a tool suited to writing some of those works.

  SPAG: Has your work on natural language processing for the programming side of
  Inform 7 lent any insights that might be useful in improving our in-game 
  parsers, or are we speaking of apples and oranges here?

ES: Hm, well. It's certainly made me more aware of things I wish an in-game 
parser would handle intelligently, some few of which have actually been 
implemented and the rest of which I would like to see later. On the other hand, 
I don't know a lot about the practicalities of this. Inform 7 is able to 
extrapolate a good deal of implied information from the text it is given, and I 
would be interested in something similar in an IF game, but at the same time I 
imagine there might be speed issues.

GN: It was in some ways the other way round. It annoyed me that Inform 6 did not 
parse source code as well as the Inform 6 library parsed commands in play - the 
latter was a little more linguistically aware. But as Emily says, there's more 
that could be done here.

  SPAG: Inform 7 does not currently support the Glulx virtual machine, but I
  understand that plans are in the works to do so. Are there any plans to expand 
  the language to support the display of images and playback of sounds within 
  Inform 7's natural language level?  Will Glulx replace the Z-Machine as the 
  standard Inform 7 virtual machine at some point?

GN: Yes, cautiously yes, and we shall see - I do not intend to force any such 
change of allegiance. Inform 7 makes Z-code which runs fine, though I've 
published a note of what an interpreter could do to have an easier life playing 
Inform 7 games (notably, have a larger stack capacity). But it's easy to hit the 
limits of the Z-machine, certainly: "Reliques of Tolti-Aph" is right up against 
them. Coding in Inform 7 makes it so quick to throw in vaguely nice if 
inessential extras (let's have not just Teddy Roosevelt but also a bison's head 
on his wall, oh, and a stuffed bear, why not, and...), that one hits limits 
earlier. And certain Inform 7 techniques do use up memory - various-to-various 
relations on large kinds, for instance.

ES: I am quite interested in getting the ability to drop in images and sounds, 
myself, so I occasionally raise this topic. So far we've mostly been preoccupied 
with other more basic functionality, but the time will come, I hope.

GN: Oh yes, I do hope eventually to have "Figures", just as in Emily's example 
of a children's book ("Figure 6. The hook-nosed stranger, his hand shaking, 
pointed the revolver straight at Biggles").

KEY TO SCORES AND REVIEWS--------------------------------------------------

Consider the following review header:

TITLE: Cutthroats
AUTHOR: Infocom
EMAIL: ???
DATE: September 1984
PARSER: Infocom Standard
SUPPORTS: Z-code (Infocom/Inform) interpreters
URL: Not available.
VERSION: Release 23

When submitting reviews:  Try to fill in as much of this info as you can.
Authors may not review their own games.

REVIEWS -------------------------------------------------------------------

From: Sara Brookside (jsh11a SP@G

TITLE: Akari's Story
AUTHOR: "Taleweaver"
EMAIL: tralu SP@G
DATE: 2005
SUPPORTS: ADRIFT interpreters
AVAILABILITY: Freeware; Shadowvault archive
VERSION: Release 1

In this game, you play a Japanese teenager on a typical weekend day (in other 
words, no school!)  I would place this piece of IF in the "slice-of-life" genre 
and the game does do a relatively good job of faithfully simulating Akari's 
daily life.  Unfortunately, that simulation can sometimes be a little bit 
pedantic in the sense that it provides little in the way of "escape value."  I 
wasn't particularly caught up in the story or involved in the action, despite 
the fact that the world-modeling implementation was adequate.  Along those same 
lines, Akari's day lacked a sense of urgency or any clear goals.  Accordingly, 
there wasn't very much to command action or to require much of the PC.

The walkthrough reveals that the game ends after a certain number of turns. at 
the end of the day, so to speak.  So, there is no real way to "win" the game, 
although there is a scoring system that awards points for certain actions.  The 
walkthrough also revealed that I tried many of the actions that the author had 
in mind, while there were others that I missed completely and would have never 
thought of had I not read the walkthrough.

One of the most interesting aspects of this piece is that it DOES provide 
insight into another culture (unless, of course, you happen to be a Japanese 
teenager yourself!)  Japanese customs and terminology and even dietary 
preferences are woven into the game, which is quite intriguing!  The game also 
reveals a bit about what is important to modern Japanese youngsters. also neat 
to know.  On the downside, this may have the effect of making the player feel 
more like a spectator than a participant.  It is as if one is observing Akari's 
life, rather than participating in it or living it, which makes the pace of the 
work feel rather slow at times.

The writing is rather sparse, in the sense that room descriptions are relatively 
brief and many nouns are non-examinable.  Still, I didn't note any particularly 
jarring errors in grammar or spelling, which certainly helped make for a 
pleasant reading experience in that regard.  In short, however, I felt much as 
if I was reading an essay by a Japanese teenager about her life, rather than 
playing a game.

There are puzzles in the game and they are reasonably well-crafted, although 
certainly not complicated.  I wish there had been more of a sense of payoff to 
successfully solving the puzzles, however.  Because the problems posed were 
essentially of the routine, day-to-day variety, and there was very little 
urgency, it didn't seem to matter much whether I solved the puzzles or not.  The 
only real impact for doing so was the point value added to my score for 
performing certain actions.

The characters in the game were largely undeveloped, except for the PC.  All of 
the NPCs felt rather static and cardboard to me, almost as if they were objects 
rather than characters.   Conversation is minimal, except if you happen to guess 
the few things that the author has allowed you to "ask [character] about," but 
this is not an uncommon problem by any means.

As for plot and story, both were a little thin.  Without a compelling goal to 
spur action, the experience was much more like an exploration than an 
interactive narrative.  Game play progressed smoothly, though, with little 
evidence of "bugginess."  There was an occurrence or two of "guess the verb," 
but I found those issues to be relatively easily solved and certainly not game-
stoppers.  A reading of the walkthrough definitely revealed several cases of 
"read the author's mind" and in each case, I had failed to do so.

In conclusion, this game could be much improved by augmenting descriptions to 
add atmosphere and capture the attention of the player, as well as implementing 
more variety and innovation in the tasks of the PC to make for a more compelling 
story line.

Overall rating:  ** out of ***** for faithful simulation, fair puzzles, and 
cross-cultural value.


From: Paul Lee (bainespal SP@G

NAME: All Hope Abandon
AUTHOR: Eric Eve
EMAIL: eric.eve SP@G
DATE: May 2005
SUPPORTS: TADS 3 interpreters
AVAILABILITY: Freeware; IF Archive

Only a well-done game can successfully combine such elements as a rather 
detailed plot, puzzles of all sorts, a theologically dense theme, and Greek 
alphabet characters. "All Hope Abandon," though the prospect seems dizzying in 
retrospect, does all of the above and somehow manages to seem somewhat natural. 
This feeling of being natural is not inherent; this is especially apparent early 
on when the player is quickly thrown into the deep end at which point the game 
world seems like a disgruntled fairy tale. Thankfully, this "fairy tale" feeling 
soon dissipates into the more interesting scheme of the game.

The player is cast as Dr William Fisher, a New Testament scholar, who at the 
opening of the game is found listening to the lecture of the deranged Professor 
Wortschlachter. Earlier at breakfast, William had met a pretty blond woman who 
somehow also ends up trapped in Limbo land (though the reason this happens is 
never adequately explained -- I think on purpose). Apart from adding a little 
romance to the story, she serves to symbolize the theme of hope throughout the 

The text is not overly complex or "purple," but is sufficient and usually easy 
to read. Sometimes the writing even has a slight pinch of wit, such as in the 
introductory text. The game is very detailed -- even unnecessary objects are 
carefully described. A "THINK" verb is implement which allows the player to 
THINK ABOUT certain topics, and I found the responses to cover mostly everything 
that could be thought about.

The detail in prose and design is complimented by many puzzles. These puzzles 
come in a range of difficulty, generally becoming more difficult as the game 
progresses. The solutions to quite a few of these at least partially require the 
player to scrutinize every single detail of an important object. This sounds 
tedious, but it works alright because one gets used to the high level of 
implementation and also because the puzzles all at least require some level of 
critical thinking besides just examining. The game is structured so that it is 
quite easy to finish the game without having solved all the puzzles, and thus 
without having tied up all the loose plot ends. This is a result of the 
integration of the narrative and the puzzles, and gives the game more 
replayability than most interactive fiction games out there.

Even with its great puzzles and well-done story, "All Hope Abandon" would be a 
mediocre game if it was riddled with bugs and grammatical errors. This is not 
the case; it is clear that the game was carefully tested. Everything works as 
expected, and it does not appear possible to get the game into a state that was 
not anticipated by the author. I must report that I did find one accidental 
typo; however, but if you can get over such things it will not lessen the 
quality of the game for you.

In conclusion, "All Hope Abandon" by Eric Eve is an extremely well-done work. I 
would definitely commend it to your "to play" list if you have not played it 
yet. If you never play it, you should know that you are missing out on one of 
the best games to be released last year.

From: Valentine Kopteltsev (uux SP@G

I know, I know -- a decent review has to start with some smart preamble on a
more or less abstract topic; unfortunately, I couldn't think of any for All
Hope Abandon. So please excuse me just for this once, and let's move on to

The theme of this game is probably best defined as "subconscious jorney".
However, if the author was trying (please note the conditional mood) to
create a dreamlike atmosphere that suggests itself for a work of that kind,
he didn't succeed too well. Anyway, no matter what the author's intentions
were -- the game more than makes up for it by providing a stunning cocktail
of adventure, theology, and romance, all that spiced with a good shot of
irony. Several ways to victory are laid through it; and although the
denouement is rather predictable at the end of the day -- the rich,
intentionally anachronistic setting full of gadgets to fiddle with, and the
considerable variety of the paths provided make it worth replaying the game
several times to try out each of them.

Another aspect of All Hope Abandon, which was especially pleasant personally
for me: it doesn't act overly symbolic, although, again, it'd be very much in
the tradition of the genre. You know, symbolism is just not my mug of beer,
and games relying on it too much often leave me puzzled. On the other hand,
the fact AHA sets forth its main ideas clearly enough doesn't mean it's as
uncomplicated as a game as I am as a person;) -- I'm sure that players more
skilled at interpreting symbolic links than myself will be able to enjoy the
game on additional levels that remained inaccessible to me.

From the technical point of view, the game is faultless. In comparison with
the previous version, TADS 3 added several interface enhancements (for
instance, let's mention the menu-based built-in hints, and the topic
suggestions mechanism of the conversation system) on its own. All Hope
Abandon not only makes extensive use of these facilities, but introduces, in
its turn, a few more. The most interesting ones are the THINK ABOUT command,
and the ability of the player to look in a specific direction. The first
feature is one I've been looking forward to for a long time. Particularly in
this game, it seems all the more appropriate since the protagonist has some
"specialist knowledge that most players will probably not share"; besides,
it's smartly used in one of the puzzles. The second one probably wasn't as
challenging to implement from the technical point of view, but means A LOT
additional work for the game author (ten extra descriptions -- eight for
compass directions, and two for LOOK UP/DOWN -- in each room); even taking
into account they're more terse than the "main" room description, and that a
few of them are similar -- it's still a feat worthy of esteem.

At this point, I've taken a pause and looked at what I had written so far.
You can bet on it -- I'm utterly disappointed with the results: All Hope
Abandon is a great game that deserves an outstanding, or at least a memorable
review, not the generic stuff I sullied the (virtual) paper with. Of course,
any reviewer will tell you it's much easier to write a memorable review for
a flawed game; sure enough, All Hope Abandon doesn't offer much in this
respect. At best, one could complain about the puzzles being too easy, which 
isn't much of a drawback as times go, and the most sceptical among the
players would probably point out that the whole romantic plotline is a bit
unrealistic, considering the protagonist and his beloved barely knew each
other; well, I consider myself a cynical person, but not cynical enough for 
not believing in love at first sight, so this was OK for me. Still, the
unsufficient flaws of the game are a pretty lame excuse for my review being
so insipid.

I think the problem is, All Hope Abandon just arrived at a wrong point in my
life. The effect is like paying a visit to the British Museum at the very end
of an exhaustive sightseeing day trip through London: the overstrained
tourist feels there's a lot of things to admire, but the emotions just aren't
there. Thus, in spite of my review probably not sounding too enthusiastic,
let me assure you -- this game is a great work suitable practically for any
players, ranging from novices to the versed ones, and represents a glorious
showcase for the opportunities the new version of TADS offers.

SNATS (Score Not Affecting The Scoreboard):

PLOT: Predictable, but still gripping (1.3)
ATMOSPHERE: Motley mix (1.5)
WRITING: Manifold, and splendid in its every manifestation (1.5)
GAMEPLAY: Relaxed trip for the most part (1.4)
BONUSES: Rich setting, fine irony, the THINK ABOUT command (1.4)
TOTAL: 7.1
CHARACTERS: Memorable enough (1.3)
PUZZLES: Well-clued and logical (1.2)
DIFFICULTY: On the easy side (4 out of 10)


From: Mike Snyder (wyndo SP@G

TITLE: The Baron
AUTHOR: Victor Gijsbers
EMAIL: victor SP@G
DATE: March 31, 2006
PARSER: Inform 6
SUPPORTS: Z-code (Inform .Z8) interpreters
VERSION: Release 1

The Baron deserves a spoiler-free review.

The difficulty is it's a game that can barely be discussed at any length 
*without* spoilers. I think this one is as spoiler-free as you'll find, but 
*anything* said about The Baron might be too much. In other words, reader 

The author's introductory text describes the story's theme as disturbing, 
shocking, and tragic. On the surface, it's about a missing girl and the father 
determined to save her from her captor, the evil Baron. I use the word 
"evil" because the game does ("x photo" in your bedroom, near the 
beginning). After that, it's left up to the reader. What kind of monster is 
the Baron? Can he be redeemed, or should he die? Is he a monster at all?

The story (if ever a work of Interactive Fiction wasn't a game, this is it) 
begins in a cave. You must slay the dragon, because nobody else will. I found 
no way to achieve this, but later events make it clear that you don't have to. 
After this, the main quest begins. Along the way - and it's a journey that 
feels much longer than it actually is - you encounter three obstacles. These 
are decision points, not puzzles. Each obstacle can be overcome in numerous 
ways. Not every way is obvious in a first play-through, and some of the 
multiple-choice decisions won't even make sense the first time. It should 
really be played at least twice. The second time, your decisions are likely to 
be wildly different - not because you're poking around for changes to the 
story, but because you will understand the story in an entirely different way.

Before setting out toward the Baron's castle, look around the house first. At 
the Baron's castle, it also pays to poke around. Even though the story lacks 
puzzles, it features bonus material for the observant reader. A torture 
chamber, found through a hatch under a rock at the castle, hints that things 
aren't exactly as they seem. Well, not so much that, but it's a good 
indication that the author is relying on symbolism to enhance the story.

In relating what has happened at the *end* of the story, the PC mentions 
nothing of a dragon. It stands to reason that the story's first scene was 
someone else's experience. If this is the case, it might have made more sense 
for the dragon to approach from a southern lair, while the PC stands firm. 
When it ends, the story offers no congratulations. You haven't won. You 
haven't lost. The final choices allow the player to affirm his or her 
convictions. The story doesn't tell *you* what's right and what isn't. *You* 
tell the story.

What I expected from The Baron wasn't what I got. In his introductory text, 
Gijsbers does a good job of preparing the player. Actions should be taken 
because they're meaningful in the situation, not because they "solve a 
puzzle". My first reaction was "sure - I've heard this before." I can't 
help but treat IF as a game - even when the author tells me not to - because 
every decision affects the outcome. In The Baron, that's not the case. Some 
decisions affect the PC's dialogue at the end, but none of it affects the 
experience of the *reader* except to the extent that the decisions themselves 
are part of the experience. So, even though the author warned me that it 
wasn't a game, I tried to play it like a game. I expected something dark and 
sinister. I expected torture, helplessness, suffering, and perhaps victory in 
the end. The story delivers these things, but in an unconventional way... in a 
disturbing, shocking, and tragic way.

If all of this leaves you wondering just what you might be getting into if you 
try The Baron, by all means read a spoilery review. Even though this could 
soften the punch of experiencing it for yourself, you might be doing yourself 
a favor. You may say to yourself "bah - I can handle blood and gore and text-
rendered pain." If that's what The Baron actually had in store for you, a 
disclaimer would be unnecessary.

It's difficult to say if The Baron hits the mark, without knowing what the 
mark was. The final choices in the walkthrough included with the Spring Thing 
version (available from the HELP menu) might be how the author imagines it. 
Most of us won't be able to feel compassion or empathy for the Baron, though - 
let alone identify (thank goodness) with the story itself. So, are these final 
decisions meaningful to us, as readers?

With precious little else to be said without delving into spoilers, some 
discussion of the design and craft is fitting. The story file is in .Z8 
format, written in Inform. The English translation of the Dutch original (also 
included) is surprisingly good. Aside from a few typos, not much in the 
translation detracts from the experience. Even with a second play-through (or 
read-through) of some of the story, I found it easy to complete in an hour and 
a half. Certain bits - especially the dialogue - are presented in multiple 
choice lists. The rest of it, however, manages to maintain the traditional IF-
style command system. You move around a map. You get, drop, and examine 
things. You open doors. You take an active part, just as IF is meant to be.

It's hard to describe The Baron as a *good* story, in the way a game can be a 
*good* game. It's an *effective* story. Appreciating it doesn't mean *liking* 
it. Even so, I can imagine the opinions of various readers will vary wildly. 
Some may say it was emotional. Some may say it wasn't. Some may say it was 
purposely manipulative. Some may say it was an honest and heart-rending story. 
Some may resent becoming an unwitting participant as the story unfolds. Some 
may describe it as grim. Some may feel entirely detached from it. Some may say 
it will receive accolades it doesn't deserve, while others may believe it to 
be unfairly criticized. Some may even say it's a story that didn't *need* 

I say... nothing, except that it was an interesting experiment. In the context 
of the Spring Thing competition, it's far too short (even adding a replay or 
two). I was moved (I'm a parent - how could I not be moved?), but this alone 
doesn't make it a clear winner when this year's competition features three 
other very good games. Scoring it is even harder than reviewing it. After some 
thought, I have settled on a middle-of-the-road score. It succeeds as 
Interactive Fiction, and it doesn't pretend to be a game. It fails as 
entertainment (for me), even though it's more like art for the sake of 
emotion. In another context, it might be a "9" or a "10". It should prove 
to be one of the most memorable works of 2006, regardless.

My Spring Thing score: "6"


From: José Manuel García-Patos (josemanuelinform SP@G

TITLE: Damnatio Memoriae
AUTHOR: Emily Short
EMAIL: emshort SP@G
DATE: March 1, 2006
PARSER: Inform 7
SUPPORTS: Z-code (Inform .Z5) interpreters with blorb support
VERSION: Release 3

Once upon a time there was this guy in Greece -- ancient Greece, that is -- who
set fire to some very important temple (Athena, maybe?). The reason why he did 
so was simple: He wanted to be famous. What did the other Greeks do? They forbid
any mention of the incendiary's name. They condemned him to oblivion. That's 
what the title of the game, Damnatio Memoriae, means. And that's what the player 
must avoid in it: To be forgotten.

No, that's not true. The game's real goal is not to get caught by your enemies,
as in any other game. Posterity is secondary. Let's say it's a plus. I don't
think that's a flaw, but I would've liked a *** You have died ***-like message
every time you had managed to survive but not to save your memory from
destruction. I would've liked to see a game where your life wasn't the most
important thing at stake. (Has a game like that ever been written? One where the
last message would be *** You have died, but you have won ***?) Also, I would've
liked to read some thoughts on the subject of fame and posterity, because the
story really had potential for that.

Damnatio Memoriae shares a similar design pattern with Galatea. In the latter,
the reactions of the NPC depended on several variables, and the ending was the
verbal expression of the final values of those variables. Here you have your 
life and your fame. And the ending depends on the final values of both. This 
always reminds me of this Japanese show, Takeshi's Castle, where participants 
were put inside a giant ball and thrown down through what looked like a 
Tartaglia's triangle pinball. The outcome depended on whether they fell on safe 
or death spots. With this system, it doesn't really matter if you choose A or B, 
because the result is the same. Only the value of the variables change. Also, 
you can end up taking the B path even if you had chosen A earlier.

My own storytelling concept is quite different. I think of interactive stories 
as trees. And trees are nothing but lists with a common root. So, interactive 
stories are a bunch of different linear stories that, depending on the player's 
choices can lead to lots of different endings. I find games designed like this 
easier to write and to plan, and more human, because it lets the story evolve 
almost without constraints. Doing things the other way seems confusing to me.

And now that I'm dealing with theoretical aspects, let me mention a detail that 
I consider a wrong way of doing things. If you have already played Damnatio 
Memoriae and the ending ever caught you in the other location, you'll have 
noticed that you're brought back to the original room and offered again a full 
description of it. The reason why I think this is wrong is because it breaks the 
rhythym of the narration. I'm not talking about this particular game or about 
the author, but about the system that produced it. Why are Inform games 
organized by location instead of being organized by sequence? Remember the 
opening sequence of Touch of Evil? Why can't IF games do things like that? Let's 
suppose you have this PC who's talking with a friend while walking down the 
street. Are you going to interrupt the conversation every time they
turn a corner? Let's suppose you have a PC who's being chased by the police. Are 
you going to interrupt the chase every time the player types E or S to offer a 
boring description that nobody would care about at that particular moment? If 
games were organized by sequence, only each one's opening location's description 
would be offered and the rest could then be simply asked for. To me this makes a 
lot more sense, and also leaves freedom to the author to write a narrative text 
and make the story advance instead of offering a description that's always more 
or less the same.

But, finally, the good news. I liked the game. Really. I did. It was short, so 
it didn't get boring. Actually, when I reached the ending for the first time I 
thought: Hey,this is it? I want more. Also, it was easier than most of her other 
games. By easier, I mean the interaction was easier. For example, in Savoir 
Faire the non-standard verbs drove me crazy. It was like watching one of those 
pretentious B/W indie movies. It's not me the one who has to adapt to your 
style, it's you to mine, stupid filmmaker! But in this case, I liked the story 
(I love the Romans) and I could take the time to learn how the special commands 
worked, because the learning gave almost immediate results. Also, I think it is 
Emily's best written game.

[Final note: Oh. In case anyone was thinking: You uneducated freak! It was the 
temple of...! And the guy's name was...! I know perfectly well who the temple 
was dedicated to, and who the incendiary was, and even the exact date when that 
happened, but I'm not doing him a favour even by giving you hints about him.]


From: David Wanaselja (wanaselja SP@G

TITLE: Dracula Part I: The First Night
AUTHOR: El Clerigo Urbatain
EMAIL: urbatain SP@G
DATE: September 5, 2005
PARSER: Inform
SUPPORTS: Glulx interpreters
AVAILABILITY: freeware; IF Archive
VERSION: Release 6
Occasionally I get the urge to play some gothic horror game, featuring vampires, 
werewolves, and all the other sorts of unnamed horrors that lurk below the 
surface of the night.  Thankfully, Interactive Fiction has its fair share of 
this genre, some excellent, some average, and some just plain awful.  While 
Dracula Part I: The First Night falls between the average and just plain awful 
categories, it does have some redeeming qualities that make it worth playing for 
the 10 minutes it takes to complete.

Dracula Part I: The First Night is a remake of the Rod Pike game Dracula, 
released in 1986 on the Commodore 64.  Actually, it's a remake of the first part 
of Rod Pike's Dracula game.  The authors have promised the second and third 
parts are forthcoming, and I can't imagine what is taking them so long, as this 
first part literally takes less than 45 minutes to see all it has to offer.  
[As of this publication, the second part of the trilogy -- Dracula: The 
Arrival -- has just been released. --ed.]  Throw in the fact that it is a near 
identical game to the original, and I'm left wondering why they didn't just 
remake Pike's original trilogy in full.  The text is ripped directly from the 
original, and although the graphics have been redone, almost everything else is 
the same, aside from some minor additions.  The differences are few, but using 
the Inform parser is by far the greatest improvement.  The original was clunky 
and hard to manipulate, but this version is far easier to handle by comparison.  

The story is of John Harker, who is on his way to meet his "client," no doubt 
Dracula.  Harker has arrived at the Golden Krone Hotel in the Carpathian 
mountains for a brief respite before continuing his journey.  It is told from 
the first person perspective, with the parser constantly demanding the player 
"tell me what to do."  This first person narrative is fairly entertaining, but 
is punctuated by far too many exclamations and instances of bad grammar and a 
few misspelled words.  Seems understandable when you realize that the first 
language of the author is Spanish, but unforgivable when you think about the 
fact that the text is nearly identical to the 1986 version.

The main puzzles hinge on what you have to eat and drink for dinner.  Some are 
more difficult to manage than the others, but all are fairly simplistic.  There 
are some interesting graphics that pop on screen for these interludes, probably 
the most improved aspect from the original aside from the parser.  Depending on 
the puzzle and how you die (if you fail) you get a different picture.  They also 
threw in some familiar music to add ambience at this point, so turn up those 

By far the most annoying part of the game is the fact that it makes assumptions 
about what the player knows.  Have you examined everything before you've tried a 
particular course of action?  If not, you may find yourself in a bind at one 
point or another, which will lead to your death or failing to complete the game. 
 It's not a big loss, as you'll always be able to "undo" your last move or just 
play through again since it's so short.  However, it is a tad annoying.

Overall, Dracula Part I: The First Night suffers greatly from its short length 
and questionable prose.  If the game was less remake and more makeover, it would 
work far better and be a much more successful game.  The fact that parts 2 and 3 
are not yet available (excepting the Commodore 64 originals) also puts a damper 
on the enjoyability of this title.  As it stands right now, Dracula: The First 
Night is a below average game that offers almost no real reasons to play through 
it, aside from the pretty graphics, and the fact that you can finish it while 
waiting for your wife to get out of the bathroom.  Once parts 2 and 3 become 
available, it will be well worth it.  Until then, it feels like a text adventure 
from 1986 that needs to be reworked badly.

From: Mike Tulloch (Poster SP@G

To be fair from the onset, Dracula the First Night (DTFN) is a remake and a 
translation of an older game, and the initial splash screen captures well the 
spirit of the old EGA graphics while conveying a hint of mystery. Still, this 
screen reads, "Remaked by..." and unfortunately, that signifies what lies in 

Immediately you're given the option of choosing maroon text on black or gold 
text on blue. (Ugh.) The EMPHASIZED WORDS IN ALL CAPS grate the nerves as does 
the chat-like sentence construction...a bunch of short phrases and half-
sentences...arranged like this. Spelling errors are legion; the writing tone is 
utterly bombastic and overwrought; exclamation marks deluge the player! Yes, 
this is the familiar topography of the Penny Dreadful and the 1950's B-movie. 
Was this what the author wanted to achieve? I suspect so.

Most of the puzzles are simple and appropriate, except for one. It's a timed 
puzzle that you'd never know was timed until you lost the game. To solve it, you 
have to do something non-obvious a priori. (Argh.) Some puzzles involve rather 
improbable scenarios, such as waking up carrying everything that you were 
holding before you went to sleep. This detracts from the atmosphere but then 
again, that may have been an intentional nod to, or spoof of the horror genre.

As for mechanics, the parser seemed lacking. Examining anything but an object 
with an Examine routine returns the result for the coded object. There are a few 
doors, but you can't knock on them because the parser doesn't recognize that 
word. The same goes for answer, table, and many others. However, because the 
puzzles are simple, you don't have to play "guess the verb" very often.

DTFN does exploit other features of Glulx, such as full-window graphics and 
music. These graphics are presented after long sections of text as a way of 
visually enhancing the effect. The last graphic presents a well done panorama of 
your destination. The music consists of a simple synth organ which begins when a 
graphic appears and pauses only when a new graphic appears. (Ow.) Though well-
played, the classical MIDI melody oozes cliche`s.

The verdict? I salute the author for creating a game in Glulx. I've tried and 
found the going awfully rough. The game includes random elements that make each 
time through slightly different but not overly so. That's creative. And the 
graphics, though sparse, do effectively enhance the atmosphere. The plot works 
well for a game of this length -- fifty turns once you know the way. However, 
the parser, the writing, the grammar, and the colors prove very trying.

If you are expecting a modern Glulx game, I'm afraid that DTFN isn't it. The 
author isn't a native English speaker but he deserves some credit for his work; 
that's why I give this game a solid 3. If you're a fan of the old-school EGA 
games and/or IF with graphic touches, of course, you might enjoy Dracula: The 
First Night much more. An hour or so of retro diversion awaits you.


From: Mike Snyder (wyndo SP@G

TITLE: Pantomime
AUTHOR: Robb Sherwin
EMAIL: beaver SP@G
DATE: March 31, 2006
SUPPORTS: Any Hugo Interpreter or GLK Hugo
VERSION: Release 1.00

In the early morning minutes of October 1st, 2005, a list of IFComp entries (at 
the official IFComp website) showed 50 or so potential games. As titles were 
knocked off the list (probably because the authors never completed and uploaded 
them), the list dwindled down to 36. For those of us online at the time, it was 
an interesting thing to watch. Among those original intents, though, was Robb 
Sherwin's Pantomime.

Now released for Spring Thing 2006, it weighs in on the light side (considering 
the competition's focus on medium- to long-sized games). As an intended IFComp 
entry, this makes sense. From chatting with Robb at his Hugo forum, I know that 
it took much effort and some sleepless nights even to finish for the Spring 
Thing deadline. Even short and perhaps rushed, Pantomime is a solid game with an 
entertaining story.

In Pantomime, you live on Phobos, currently a moon of Mars, but soon to break 
apart into an orbiting ring of debris. It's the last place anyone would choose 
to be, even before the crisis. It's a world where cloning is commonplace, robots 
named after Unix commands are a man's best friend, and being good at chess is 
cause for embarrassment. This is a vision of the near future.

I've only played the beginning of one or two Robb Sherwin games prior to this. 
That's probably why my reactions bounced between "whoa. did he really say that?" 
and "wow. what's it going to be *next*?" Somehow, even if he's holding back, it 
doesn't *feel* that way. Aside from a few typos - probably the result of the 
hurried effort to meet the deadline - the writing is great. It flows better 
because it's more casual. It's not just *how* Sherwin writes - it's also *what* 
he writes: the insults between characters, the one-off jokes, the clever 
descriptions and bits of back-story. I usually cringe at coarse passages and 
lowbrow humor in a game, but that's part of what makes Pantomime so interesting. 
Sherwin seems to write it in a convincing, honest way.

Pantomime is what an episode of Futurama might be, if the script came from 
Cartoon Network's Williams Street crew and it aired on HBO after hours. The 
little censor that lives inside Robb Sherwin's mind has a freedom not given most 
other IF authors, save maybe Adam Thornton. I mean, if a wacked-out robot needs 
to sport a cloned copy of a male porn star's money-maker, Sherwin will work it 
into the story. And it'll be *funny*.

The game *is* meant to be funny. I think. It's sometimes tongue-in-cheek humor. 
It's *definitely* black humor, where the absurd and the macabre come together. 
It might be an allegory for some of today's issues, but if so, I didn't really 
get that. More likely, it's just a strange but fascinating story.

The puzzles aren't difficult (generally just a matter of figuring out what 
action to take to move things along), and inventory is almost non-existent. 
antomime is very story-driven. The most difficult bit may have been passing the 
spiked gate, but even that obstacle yields to some creative but simple reasoning 
(okay, okay - I solved it by blind luck and experimentation, but it made sense 
afterwards). Even the second-to-last confrontation doesn't require anything more 
complicated than following instructions and listening to the bad guy's diatribe. 
This should be particularly appealing to anyone who prefers IF to be more 
*fiction* than *game*.

A few minor bugs remain in the competition release. They range from typos to an 
odd exit back to Kangaroo's Club - nothing game-killing. This seems to happen 
more toward the end. What's most likely to work against Pantomime, though, is 
that it doesn't seem long enough for a Spring Thing game. It also glosses over 
the additional detail in most places, when it comes to interacting with (even if 
only to "look at") scenery objects. Knowing Hugo, I think this could be fixed 
easily, even without real objects. Just add an "extra_scenery" property to each 
room, with lists of keywords that will cause a different reference message. It 
means the difference between something "not there" when it really is, and simply 
being "unimportant".

A few plot points left me confused. Who sent me the vial? What was the purpose 
of the seemingly unnecessary gate code? And how drunk was Sherwin when he came 
up with the interaction that helps the PC escape the airlock? Other than that, 
everything is wrapped up tidily at the end, where a couple of fitting plot 
twists are thrown in.

I enjoyed Pantomime, and I recommend it - especially if an update comes after 
the competition. Without hints, I finished in two and a half hours (plus some 
re-play of select earlier bits). It's definitely a game that wouldn't have been 
out of place in the annual IFComp, but even snack-sized by Spring Thing 
standards, it's a worthy entry.

My Spring Thing score: "7"


From: Mike Snyder (wyndo SP@G

TITLE: The Potter and the Mould
AUTHOR: Robert Street
DATE: March 31, 2006
PARSER: Adrift
SUPPORTS: Adrift Runner and GLK Adrift
VERSION: Release 1

The superhero genre isn't my favorite. I've never been a big fan of larger-
than-life, hard-to-believe super powers. Sure, I enjoy the comic-based movie 
from time to time - Superman with his flying, strength, and x-ray vision; 
Spiderman with his spider-sense and web-slinging; Batman with his. super wealth; 
X-Men, of course. I've been known to read a superhero comic, although not 
recently. I've played superhero IF once or twice. 

In essence, I'm not the ideal fan for The Potter and the Mould. This is a story 
for superhero fans, and it comes complete with all the trappings. Origin story? 
Check. Mutant-like powers? Definitely. An imperfect, self-doubting protagonist? 
Uh huh. The mentor/student relationship? You bet. An evil but misunderstood 
villain? Sure. Motivation by revenge? Of course. It's everything you'd expect 
from a superhero story.

What I like best about The Potter and the Mould is that it keeps moving. It will 
appeal most to superhero fans, but it's a frenetic and fast-paced adventure that 
kept me enthralled for the four hours it took to complete. It *feels* more 
difficult than it really is, which is a credit to the author's talent. The 
puzzles aren't hard enough to impede the action, yet they leave a sense of 
accomplishment in their wake. My longest sticking point involved a machine-room 
and a clay dog. After solving it - which was easier than I tried to make it - I 
realized that the puzzles were simple and understated. They work to keep the 
story moving, not to work against it, and that's probably the best kind.

In most cases, Robert avoids text-dumps by way of action prompting. This 
typically comes as a nudge from an NPC, beginning with the rescue scene at the 
beginning, through Waterfall's revelations in the mall, to the hurried trip to 
the Potter's inner sanctuary. It works very well, and keeps things interactive.

The story has some surprises and twists. It strays from the predictable formula 
particularly near the end. Even though I appreciate the change-up, the victory 
felt less than satisfying. It was like facing down Dr. Evil, and getting a back
-stabbing Mini-Me instead. What follows is an exciting bit, but not what I 

The writing in this, Robert Street's latest game, left me puzzled. Even though 
I've commented about this in prior reviews, it seemed more noticeable here; 
equally hard to pin down, but more prevalent. It could just be his style, but 
I'm not convinced that's it. Too many sentences (even some in close proximity) 
ended awkwardly with the word "though". There was a glaring lack of contractions 
throughout the text - not always, but in certain passages - making it more 
awkward. This alone isn't a sticking point, but coupled with some awkward 
phrasings in general, it just didn't always read *right*.

Take the following - just one random example from hundreds of lines of text:

"The shops in this corner seem to be trying to outdo each other with silly 

The words "seem to be trying to" are the awkward bit. I might write this as:

"The shops in this corner strive to outdo each other with silly displays."

These kinds of things weave themselves throughout the text. The removed bit was 
clunky and passive. The replacement flows better, and it's more active. I 
usually qualify these kinds of comments with a reminder that I'm no expert. I 
have to look two or three times to figure out just what it is about a sentence 
that bothers me. Even in the example above, I might change "each other" to "one 
another". At times, the text in The Potter and the Mould felt like a first 
draft, as though it had been written once and then left alone. At other times, 
the text felt too heavily edited, as if the smooth flow and original expression 
had been lost under the weight of so much revision. Which the case may be, I 
don't know. Like I said, it might just be the author's unique style.

Robert didn't skimp on details. Even though this is an ever-moving game, the 
extra effort shows in the responses, from looking around to trying various other 
actions. This isn't always the case in story-heavy works, where the only 
important thing is doing exactly what advances the plot. I liked that The Potter 
and the Mould stood up to some prodding.

The hero's premise, now that I've come to it, is that he can "mould" his shape 
into various things. Robert implemented this in a logical, user-friendly way. 
Mould options are task-specific and presented in a list. This maintains the 
illusion that you're really able to morph into anything, where free-form input 
(John Evans's games come to mind) makes this very difficult. It also gives the 
PC some say-so as to how the player proceeds. In other words, the PC dreams up 
these forms so the player doesn't have to. More than that, it eliminates the "I 
tried to become a diamond-tipped drill, but it didn't work" complaints, at the 
expense of limiting the player's options. Occasionally, this let me figure out a 
puzzle where I wouldn't otherwise have had all the facts, but all in all, it was 
a good design.

Next comes the obligatory Adrift discussion - but I'll keep it short. I'm a fan 
of Adrift's auto-mapping, even if I found some necessary exits unmarked on it. 
It can be a crutch sometimes, and I probably shouldn't have tried to rely on the 
map as much. I'm not a fan, however, of Adrift's pick-apart parsing. I call it 
that, because I don't know if it has a real name. This is most problematic when 
commenting a transcript. If you type something that has the word "undo" anywhere 
in it, Adrift believes you intend to undo your last move. If you comment about 
"he" or "she", Adrift starts matching pronouns, leading to some interesting 
responses. Knowing nothing about its inner workings, I still get the impression 
that grammar isn't built the way it is in Hugo and other IF languages. Adrift 
makes plenty of assumptions, and it doesn't conform to any grammar rules. If it 
finds what appears to be a verb, and what appears to be a noun referencing a 
known object, it reacts. At times, such as when it reacts on an NPC you haven't 
encountered yet, it can even be a little spoilery.

Even though I'm not a superhero fan, I enjoyed The Potter and the Mould. It was 
trippy, but it was fun. It's a solid game, and a credit to the author's 

My Spring Thing score: "8"


From: Mike Harris (M.Harris SP@G

TITLE: Threnody
AUTHOR: John Schiff
EMAIL: john SP@G
DATE: March 13, 2005
SUPPORTS: HTML TADS interpreters
VERSION: Release 1.0b

Threnody bills itself as "a lighthearted, puzzle-rich fantasy game" and as such 
hits the mark.  The backstory is a pedestrian "castle full of treasures" meme, 
with the mildly interesting twist that in addition to treasure “points” 
accumulated, each of the NPC’s has a "Catena" which must be located and 
destroyed to "release" the NPC, a tally of which scores in a parallel point 

The player must choose at the beginning whether the PC will be a warrior, wizard 
or thief, as well as the gender of the PC.  While the former affects what 
options the PC has available to solve portions of the game, the purpose of the 
gender choice seems only for fleshing out the story thread.  The verbiage is 
standard "fantasy 101" with the exception that objects’ and sometimes NPC’s 
descriptions contain horrible puns.  The title NPC is a talking cat and is 
generally helpful as a guide but which purpose seems more to further the story 
than the game.  I delight in both bad puns and cats; those not so enamored might 
find the game less charming.

I prefer to be very methodical when playing IF, searching and examining 
everything in a room and solving as many puzzles as possible before moving on to 
the next room.  This approach works poorly with Threnody.  Those who prefer a 
brief visit to every accessible room before beginning any puzzle solving in 
earnest might find Threnody less frustrating.  Firstly, objects that might aid 
in solving puzzles are often available; I did a lot of guessing.  Secondly, many 
rooms and locations are empty, there as "place holders."  Lastly, the game is 
full of objects that seem to be nothing more than red herrings.

This leads us to bugs. 

It is possible to locate and destroy the Catenas of some NPC’s before 
encountering the NPC’s themselves.  While the disappearance/destruction of the 
NPC should mean its absence when its location is eventually found, this is not 
always the case.  Taking the Catenas to unusual locations to be destroyed will 
sometimes trigger the same bug.  In any case, this bug allows the game to be put 
in an unwinnable state as the NPC can not be vanquished.  I also stumbled upon a 
sequence of moves that "recreated" the Catena of one particular NPC, which I 
could then destroy repeatedly for points.

In some cases there is more than one way to "beat" the NPC, but I found another 
bug involving engaging a NPC in a game of chance using an object found 
elsewhere.  While defeating the NPC in the game of chance should achieve 
success, and the NPC states that you can take the (blank) because you’ve won, 
any attempt to actually do so results in the pre-defeat response.

I found the "red herrings" to be frustrating as well.  While they may not be 
true "red herrings" but have some purpose in solving puzzles when the PC is in a 
different iteration, their sheer number gets tedious – "open (blank) with 
(blank)"; "put (blank) on (blank)" is fun for the first thirty or forty tries 
but does start to pall once one gets into triple digits.  Furthermore, while the 
PC’s carrying capacity is large it’s not unlimited.  Three quarters of the way 
through the game I found that I was no longer able to pick up objects and had to 
decide which objects in my vast inventory to jettison, hoping that I would not 
rid myself of anything truly useful in the process.

While some puzzles are ludicrously simple, some are so difficult that I had no 
recourse but to refer to the hints or the walkthrough.  There is some trouble of 
the “guess the verb” variety but in many cases there seems to be no way to solve 
the puzzle except through brute force, trying every possible combination hoping 
that one works.

Some minor squawks – default responses sometimes come up when the story would 
dictate otherwise.  For example, upon entering a room there are often lavish 
descriptions of furnishings, floor etc., further examination of which elicits 
"you see nothing special about the (blank)" type responses, or directly 
contradictory responses – "The floor is made of large slabs of a dark, rough 
stone" after the description states something very different.  The story 
emphasizes the importance of the Catenas to the NPC’s, but (with the exception 
of the title NPC) showing the catena to the NPC elicits a default "(NPC) is not 
impressed."  And speaking of "guess the verb," during play I could often not 
recall the term "Catena" or names of NPC’s and more generic descriptions such as 
"man," "crystal" or "blue" were ineffective.

In summary, while Threnody is a mildly entertaining diversion exactly as 
advertised, its flaws and generic storyline made it less than compelling.  The 
optional graphics are very much like the game itself – workmanlike, well 
executed but ultimately unremarkable, and doing little to enhance the storyline. 
Once played through as one PC I had no desire to play more than a small amount 
with the PC in a different iteration, just enough to get a feel for that 
particular PC and some of the differences in puzzle solving.  On a scale of 1 to 
10 I would rate it a 5 for difficulty and an overall rating of 5.5 to 6.


From: Mike Snyder (wyndo SP@G

TITLE: The Warlord, The Princess & The Bulldog
AUTHOR: David Whyld
EMAIL: dwhyld SP@G
DATE: March 31, 2006
PARSER: Adrift
SUPPORTS: Adrift Runner and GLK Adrift
VERSION: Version 1

It's not the next chapter in The Chronicles of Narnia (and if no other reviewer 
makes the same joke, I'll be surprised). David Whyld's Spring Thing 2006 entry, 
written in Adrift, reprises the exploits of mercenary bad-ass Stavros "The 
Bulldog" McGrogan in a sequel to his earlier A Spot of Bother. It's up to The 
Bulldog to sneak, fight, grunt, and puzzle-solve his way to victory against the 
evil Warlord, Baron Grishtak.

At times, this is a contradiction. From the start, the goal is clear. I don't 
mean the goal of the story (which is also clear), but the goal of the game 
itself. Finish the three primary objectives with full health for a score boost, 
and pick up more points for solving puzzles rather than pushing past them with 
brute force. This opens the game to a variety of play styles, but that "best 
score" objective is the carrot dangling just beyond reach. The Bulldog loses 
life points when he fights, and without a clear idea of how to gain them back 
(let alone how many can *be* regained), my inclination was to avoid fights and 
slink about the castle solving puzzles, preserving every point of health 
possible. So much for being a bad-ass.

Even though I enjoyed the game using this strategy, I might have enjoyed it 
more if I hadn't been aiming for a perfect game. In the end, it didn't matter. 
I didn't complete one of the three objectives, and I won with a score of only 
80 and health of 90. The ending - and the death ending too, when I purposely 
let The Bulldog get pounced by The Tiger - was still satisfying.

Things take a bit longer when you play for points. Instead of beating up the 
bad guys, I lured them into traps, tricked them into leaving, or simply avoided 
them entirely. When I stumbled into traps or lost health in unexpected ways, I 
opted to "undo" or "load" a prior save, so I could try another approach. This 
made the game tougher. To get it all right the first time, I would have needed 
to read the author's mind. The interesting thing is that this was just another 
way of playing the game. With a different objective - let The Bulldog fight 
enemies and bully his way past the tough parts - it doesn't require psychic 
powers. It's a fair system which rewards do-overs without making do-overs 
essential to win.

I've mentioned "health points" several times. If you have visions of RPG stats 
and random dice-rolls - especially if you don't *like* those things - take 
heart. That's not how WPB works. Think of it as the antithesis of a scoring 
system. When you earn "score" points, it's for completing a task, reaching a 
milestone, or hitting some score-worthy trigger. These are things built into the 
game, and the points are set. If you play much IF, you've probably seen this in 
action. WPB has this *in addition to* its health point system. Points come off 
by making mistakes, or in other predetermined ways that involve alternate puzzle 
solutions. Sometimes, these mistakes (especially in facing enemies) can be 
repeated, but on the whole it's more like a credit system. The Bulldog is 
extended so many of these "mistake" points, and he spends them as necessary.

The beauty is that making these mistakes usually gets The Bulldog past puzzles. 
For instance, there are several ways to pass the landmines near the beginning of 
the game. One way in particular saves The Bulldog from damage entirely. Other 
ways leave him only slightly scathed (or perhaps unharmed, but with the loss of 
something that might be the key to avoiding damage later). Of course, stepping 
into it (with persistence) solves the puzzle too, at the expense of a chunk of 

It's designed to be winnable, no matter how low your health becomes. The more 
damage The Bulldog takes, though, the fewer risks he can endure. Suppose this 
drops to a single remaining point. The game remains winnable, but every 
additional obstacle must be overcome with brains instead of brawn. This can 
become *very* difficult. Health can be recovered, but I never was quite sure 
how much. If I recall, I healed about 30. The Bulldog has suffered some prior 
to the start of the game, beginning with 63 health. Health of 100 is considered 
"full". It may be possible to recover more than 37, making it possible to take 
damage and still finish with full health. I never figured out the max. It's just 
as possible that every method in the game adds up only to a total of 37, meaning 
a perfect win requires a totally unharmed Bulldog. Maybe a better player than I 
- or Whyld himself - will say for sure.

Really, it's a clever design. I can't think of a single puzzle that didn't 
have two or more solutions. The easier the solution, the fewer the points (and 
often, the more damage The Bulldog would take). Because my goal had me going 
after the toughest of each solution, I hit the built-in help often. After only 
a short ways into the game, I was requesting every hint available in every 
room. In a way, this became just another tool, like "undo". Instead of cheating, 
it seemed more like a part of the game. Some hints even felt more like puzzles 
to solve. Even *with* hints, it was often difficult to work out the best (most 
rewarding point-wise) solutions. Without them, though, I never would have.

This all makes it difficult to say just how tough The Warlord, The Princess, and 
The Bulldog is. I solved many of the puzzles with easier solutions at first, 
costing The Bulldog only a few points of health. I would have finished faster - 
and possibly without so much reliance on hints - if I had just pressed forward 
from those points. I suppose it ranges from "challenging but not overly 
difficult" to "one step down from impossible", depending on what approach you 
take. Mine was more on the side of the latter.

Whyld has done an excellent job of anticipating much of what players may try. 
The implementation level alone is amazing. Very little encountered in the game 
lacks first, second, even third-level implementation. If you look at scenery 
that has parts, you can look at those parts. You can often *interact* with those 
parts. If those parts have parts, they're probably implemented too. It pays to 
really inspect what's around. Even though much of it is optional, enough digging 
can bring up the keys to alternate puzzle solutions.

The prose in WPB is dotted with amusing passages. Generally, Whyld isn't trying 
for real comedy - and if so, it probably wouldn't have worked here anyway. It's 
more the "ah ha, that was funny" kind of subdued but cliched humor you'd expect 
from a story in which the hero only grunts yet everybody understands what he 
means. When Baron Grishtak writes a letter to his ace henchman - subsequently 
obtained by The Bulldog - he admits that he "foolishly jotted down the access 
code to the master computer on the bottom of it." He goes on to encourage his 
henchman to destroy the letter after reading it, for that very reason.

As to the presentation, the author held nothing back. My first fifteen minutes 
were spent just reading the introductory material - details about the game, 
additional commands, the intro, etc. The game font size can be adjusted via the 
command prompt. Screen-clearing at each room change can be turned on or off 
(personally, I liked it on - it was easier to quickly scroll up and reread room 
descriptions that way). Around four different fonts were used - one for room 
headers, one for the room description, the default font for most game messages, 
and a script-style font for letters and notes. It may sound like a hodgepodge, 
but it works well (if you're using the Adrift runner and your Windows-based 
computer has those fonts) and it set WPB apart from other games in terms of 

To now, it may seem as though I have no complaints about The Warlord, The 
Princess & The Bulldog. A big game, though, has more room for things to go 
wrong. None of these problems (in my play-through, anyway), were game-killing, 
but they ranged from mildly annoying to completely preventing (or, at times, 
*allowing*) certain solutions. My transcripts note quite a few typos - not 
surprising in a game of this size and complexity, but still minor dents in the 
proverbial finish. Weirder quirks included things like the non-working pendant 
(it worked once, but after a subsequent "undo" or "restore", shaking it didn't 
work even though it still had 3 charges); being able to enter the guards' 
training courtyard in a "they're gone" state, even though they shouldn't have 
been; a reference to a voodoo doll in the hints, which doesn't seem to be in the 
game (Adrift will usually respond to objects it knows, even in other places, and 
it didn't know that one); being able to break the panel in the sleeping quarters 
repeatedly; I didn't realize it at the time, but the "code to the master 
computer" is too long to work in either of the computers found later in the 
game; some available exits were unmarked on the map; some exits described in the 
text didn't work in the game; you can't "undo" to before a hint screen; I 
couldn't get "exit" to work (even though it was supposed to), when trying one of 
the codes; A seven-letter password scattered throughout the castle appears to 
have two fifth letters; it's possible to set the watch before winning, so that 
it goes off during the final scene; a few other miscellaneous quirks.

As the game progressed, these things either became more common or more 
noticeable. Maybe it was the cumulative effect, but my faith in the game's 
internal consistency was shaken. If I felt at all guilty about reliance on 
hints, the feeling passed when I thought that maybe the game was broken just 
enough to *prevent* the solutions I needed for a perfect win. This may not be 
true. From my experience, the bugs that persist after beta testing are usually 
the bugs in sections that *aren't* vital - else they would have been worked out 
already. Nonetheless, it's a reminder: the better the polish, the higher the 

Most of the design works great. The health point system contributes to alternate 
puzzle solutions, and alternate puzzle solutions are abundant. The hints, 
although cryptic at times, are helpful. Even so, a few specific parts left me 
cold. One very early puzzle (the one that avoids a loss of health - an easier 
but damaging alternate does exist) requires waiting a few turns after taking 
action. I was impressed that the game allowed the particular action, but I 
thought I had messed up - so I did an "undo". Speaking of "undo", you can 
unwittingly make a move that disables it, in what I can only describe as a prank 
perpetrated by the author. It's by no means a necessary (or even an obvious) 
move, but some players will try it. I found no way to re-enable it, aside from 
reverting to a prior save (or starting over). One obstacle requires that you 
lose everything in inventory. The hints describe a way to keep most of it, but 
it requires repeating an action (and it's even possible to undo a failure, 
repeat, and succeed the second or third time).

As a Spring Thing entry, WPB is fittingly sized. My play-through - taking most 
puzzles the hard way and relying heavily on the built-in hints - was eight and 
a half hours. Despite the flaws, I enjoyed the time I spent with The Warlord, 
The Princess & The Bulldog. An incredible amount of effort was put into this 
game, and it really shows. A post-competition release could address the 
remaining problems, making it even more recommendable.

My Spring Thing score: "9"

 ___. .___    _    ___.     ___. 
/  _| |   \  / \  / ._|    /  _| 
\  \  | o_/ |   | | |_.    \  \  
.\  \ | |   | o | | | |    .\  \ 
|___/ |_|   |_|_| \___|    |___/ PECIFICS

SPAG Specifics is a small section of SPAG dedicated to providing in-
depth critical analysis of IF games, spoilers most emphatically


The Baron




From: Dan Shiovitz (dans SP@G

TITLE: The Baron
AUTHOR: Victor Gijsbers
EMAIL: victor SP@G
DATE: March 31, 2006
PARSER: Inform 6
SUPPORTS: Z-code (Inform .Z8) interpreters
VERSION: Release 1

In my review of The Baron for the comp, I mostly just said "I'm pretty sure this 
is not like any IF game you've ever played before" and left it at that. But this 
is a SPAG Specifics review, so I'm going to assume everyone's played the game 
all the way through, and will feel free to get into the nitty-gritty details of 
how and whether I think the game works. 

To get some stuff out of the way, I'm not particularly impressed with the 
writing or storyline here. I think it's too obvious early on that weird stuff is 
going on with the daughter, the "the PC is actually a bad guy!" plot has been 
done before (in Bliss, for example), and most of the actual plot elements feel 
more like setpieces than real places or NPCs -- you got your generic dragon cave 
(bone flooring optional), your generic village, your generic forest, your 
generic ruined castle. So given all that, why did the game work so well for me?

The first time the game astonished me was when when I typed >KILL WOLF on the 
young wolf after I killed its mother, and the game *asked me why I was doing 
it*. Man, even now I am still blown away by how cool that is. When this 
happened, the game essentially defined a whole new reason for why IF is 
interesting. We've talked in the past about buzzwords like "exploration" and 
"agency" and "complicity" to name things you get out of IF that you don't get 
out of static fiction, but The Baron defines a new one -- justification. 

There have been hints of this kind of thing in the past. Piece of Mind, for 
instance, has a famous section where the player is challenged by the game's 
protagonist over the rightness of the action the player's typed in. And 
obviously there are plenty of games where the PC has to make a moral decision. 
The thing is, in almost cases the moral decisions involved in those games are 
lame, or not really a decision at all -- the game clearly identifies one of the 
decisions as right and the other as wrong. 

The game-design insight in The Baron is that by shifting the moral judgement 
onto the player, the decision suddenly becomes much more interesting, because 
now the player (and not just the PC!) has direct personal stakes. This is sort 
of similar to how you can make a storyline more involving by presenting it in 
nonlinear order: in that case the player gets involved in the plot because they 
have to piece it together themselves. In this case they're involved in the 
decision because then they're asked to justify it -- not to the game, which
would be trivial, but to themselves. The game establishes that the PC has been 
doing something bad for a long time, and is on the verge of doing it again, and 
nothing more. The question of how bad a person this makes the PC is entirely up 
to the player, as are the punishment for and future consequences of the PC's 

When The Baron gets away from asking for justification, it tends to lose its way 
a little bit. The scenes at the beginning in the dragon cave and at the end in 
the ruined castle are examples of this: the symbolism comes off as heavy-handed, 
and none of the PC's actions matter. Which is an interesting objection, since 
the PC's actions don't really matter elsewhere either -- there's only one place, 
right at the end, where you have the power to alter the storyline in a
significant way. 

The point seems to be that, again, the PC's actions in The Baron are meaningless 
except as they're evaluated by the player: your actions with the wolf don't 
matter except that the gargoyle forces you to discuss them. I'm surprised that 
the initial scene with the dragon never gets a similar kind of discussion -- the 
choice to fight or run, and the question of whether failure is really inevitable 
fit in just fine with the rest of the game, but without discussion they aren't
particularly important. 

So where to from here? I would be totally interested in seeing new games use 
this one as a jumping-off point. The thing I would most like to see is if 
there's a way to restore some agency to the player here -- The Baron focuses so 
tightly on the moral decisions that it makes the PC's actions irrelevant to the 
course of the narrative even if they're relevant to the player. It seems like it 
must be possible to have a game where the PC's choices are significant both
in-game and out of it. But The Baron is still seriously ground-breaking, and I 
hope in a few years we can look back and see it make the same kind of impact on 
IF design that Photopia or A Mind Forever Voyaging did. 

As one final closing note, Victor Gijsbers is involved in the modern RPG scene, 
and it's hard not to think that The Baron is strongly influenced by trends 
there. Like, here's a quote from Ron Edwards defining Narrativist gaming, the 
closest RPG trend to what Gijsbers's game is about:

  Story Now requires that at least one engaging issue or problematic
  feature of human existence be addressed in the process of
  role-playing. "Address" means: 

  Establishing the issue's Explorative expressions in the game-world,
  "fixing" them into imaginary place. 

  Developing the issue as a source of continued conflict, perhaps
  changing any number of things about it, such as which side is being
  taken by a given character, or providing more depth to why the
  antagonistic side of the issue exists at all. 

  Resolving the issue through the decisions of the players of the
  protagonists, as well as various features and constraints of the

Sounds familiar, hunh? This is from -- somebody interested in 
working on a game like The Baron would be well-advised to check it out. 

Even if you're not interested in a game that's as non-standard as The Baron is, 
there's plenty of RPG theory and craft that still applies: how to set up a 
scene, how to cut between scenes, how to build storylines that allow for and 
adapt to the PCs doing surprising things. The modern IF community has gained a 
lot of valuable insight from static fiction and other types of computer games, 
but RPGs have been mostly absent as an influence since Colossal Cave. It's 
probably time the IF community starts looking at another interactive 
storytelling medium and seeing what techniques they have that we can learn from.

SUBMISSION POLICY ---------------------------------------------------------

SPAG is a non-paying fanzine specializing in reviews of text adventure games, a.
k.a. Interactive Fiction. This includes the classic Infocom games and similar
games, but also some graphic adventures where the primary player-game
communication is text based. Any and all text-based games are eligible for
review, though if a game has been reviewed three times in SPAG, no further
reviews of it will be accepted unless they are extraordinarily original and/or
insightful. SPAG reviews should be free of spoilers, with the exception of
reviews submitted to SPAG Specifics, where spoilers are allowed in the service
of in-depth discussion. In addition, reviewers should play a game to completion
before submitting a review. There are some exceptions to this clause --
competition games reviewed after 2 hours, unfinishable games, games with
hundreds of endings, etc. -- if in doubt, ask me first.

Authors retain the rights to use their reviews in other contexts. We accept
submissions that have been previously published elsewhere, although original
reviews are preferred.

For a more detailed version of this policy, see the SPAG FAQ at http://www.


           Thank you for helping to keep text adventures alive!

Click here for a printable, plain text version of this issue.