IF Database details
Carl Muckenhoupt
of content in the proposed IF database.
This is not a formal spec. Rather, this is a summary of what
information will be available from the database, taking
a Game as a starting point. A Game is a conceptual
entity, not a technical one. Adventureland for the TRS-80
is conceptually the same game as Adventureland for the
Z-machine. The original release of Curses is the same
game as the latest release of Curses. MacWesleyan is the
same game as PC University.
I'm using the word "Implementation" for specific instances
of a game. Each game can have many implementations. Each
implementation exists in only one file at the Archive,
although a file may contain multiple implementations
(usually of different games).
Please suggest anything you see lacking. That's why I'm
posting it.
Details of how this information will be represented in the
database are deliberately glossed over.
I. Objective data
A. Title(s)
Note: There may be multiple alternate titles for a game; "Adventure",
"Original Adventure", and "Colossal Cave" all refer to the same thing.
One title should be chosen as the canonical title. I'm going to
ignore the fact that titles can be specific to implementations.
Note: There may be multiple games with the same title. Graham
Nelson's "The Tempest" and the older AGT "The Tempest" are definitely
different games.
1. Subtitle?
B. Authors
1. First name
2. Middle name
3. Last name
4. Honorific?
5. Email address
Note: This should only be filled in with the author's permission.
6. Web site URL
7. Role
Each author will have exactly one of the following
attributes wrt a particular game. Where no
information is available, it will be assumed that all
authors are Authors.
a. Author
worked on design and implementation of game
b. Designer
non-programming author
c. Implementor
non-designing programmer
C. Companies
This is primarily for games that were originally
published commercially. A particular game might
have several companies: producers, publisher,
distributor, whatever.
1. Name
2. Web site URL
3. Fan site URLs?
D. Licensing status
Posible values are:
1. Free of charge
This does not necessarily mean "freeware" or "public domain".
It just means that no one is asking for money.
2. Shareware
Freely redistributable, but the author requests payment.
3. Former shareware
The author is no longer requesting money. This is distinguished
from "Free of charge" to avoid confusion; implementations may
contain obsolete requests for registration.
4. Commercial
Games whose license does not grant the right to redistribute.
There shouldn't be anything like this at the Archive, but there
probably is.
5. Former commercial
Same idea as "Former shareware".
E. Series
For games that are episodes of a larger whole, like the
Zork trilogy or the Unnkulian series.
1. Series name
2. Index in series
For example, the index of Zork II would be 2. Note that
series indices generally follow the chronology of the fiction,
which does not necessarily correspond to order of
release date. (Example: Unnkulia Zero.) Also note that
games in a shared universe don't necessarily have an ordering.
F. Chronology
This may be deemed unnecessary, because of the dates
on the implementations (see below). However, in many
cases the original release is not available at the Archive.
1. Date of first release
2. Date of most recent release?
G. Genre
The mapping between games and genres is most decidedly NOT
one-to-one.
For a possible list of genres, consult
http://www.wurb.com/if/data/items/genres.
Actually, I'd like to make the genre system more hierarchical,
with subgenres like "Science Fiction:Time Travel" and the like.
H. Attributes
See addendum
I. Specific implementations
1. Filename, relative to archive root
2. Authors
See above for more details about Authors. The authors
listed for the game need not be listed here.
Rather, this space exists to give credit to others who
modified the game after its initial release.
The possible roles for an author of an implementation are:
a. Contributor
made significant additions to game content, as in the
various extended versions of Colossal Cave.
b. Porter
made game work on different platform
c. Translator
translated game text into a different (human) language
3. Version
4. Chronology
a. Date of release
b. Date added to the Archive
For new games, these two dates will usually be the same.
For games that predate the Archive, they certainly won't be.
6. Development system/programming language
a. Location (relative to Archive root) of compilers?
b. Related web sites?
7. Platform
Runtime systems like the TADS runtime and the Z-machine
are considered to be platforms. "Source code only" should
probably also be considered a platform.
a. Required verion of runtime
b. Location (relative to Archive root) of interpreters/emulators?
c. Related web sites?
8. Language
Human language, that is. Not programming language.
9. Attributes
See addendum
10. Comment
A string containing arbitrary information that doesn't fit
elsewhere. Intended for things like "Requires VBRUN300.DLL"
or "Crucial documentation contained in separate file".
J. Related files
Anything at the Archive relating to the game that does not
actually contain an implementation. This will largely be
hint files.
1. Filename, relative to archive root
2. Description
"Solution", "Map", etc. Should this be an arbitrary string
or a selection from a list?
K. Related web sites?
II. Subjective data
A. Baf's Guide
Perhaps the reviews on Baf's Guide should be treated like any
other external reviews. But then again, perhaps not; after all,
the database is largely intended to provide the same functionality
as Baf's Guide (only enhanced), and my intention is to generate the
next generation of the Guide from the database.
1. Rating
2. Review
Possibly in some cases too big to easily stuff into a database?
Or maybe not. If all else fails, this field can hold a filename.
B. External reviews
1. Site
a. Name
b. URL
c. Currently active?
(In case a site goes down, but is expected to come back
up at some point.)
2. URL, relative to site URL
C. Collective ratings?
The idea here would be to report the average rating given by
users. I'm not sure I like this idea, though. It has all of the
Comp's problems magnified.
Addendum: Attributes
There will be a collection of attribute tags that can be
tacked onto various games or implementations. Users should be
able to do searches that require or forbid specific attributes.
Attributes will also be reported with game descriptions in the Guide.
Here are some suggested attributes. Not all of them will
necessarily be used, and some may be folded into the Game or
Implementation structure.
Database record incomplete
(Should be applied to any game that has not actually been
downloaded and played by someone contributing to the database.)
Non-IF
Not Interactive Fiction at all
(Example: Freefall, Robots, Lists and Lists)
Interactive fiction in some sense, but not a "text adventure"
(Example: The Space under the Window)
Borderline IF: Is this a text adventure or a board game?
(Example: cRiMe)
Substandard parser
Two-word parser
Less-than-two-word parser
(Example: Caverns of Chaos.)
Graphics
(The primary concern here is to warn players using text-to-speech
software.)
Pixel graphics are crucial to gameplay
Features optional pixel graphics
Character graphics are crucial to gameplay
Features non-crucial character graphics
Sound
Uses sound, which can be disabled
Uses sound, which cannot be disabled
Features continuing background music
Mazes
(There are people who hate mazes vehemently. However, there are
several possible mitigating factors, so it may be worthwhile
to split this into several attributes.)
Traditional maze
(The type found in Adventure and Zork)
Optional maze
(A traditional maze that the player never actually needs to
enter in oder to complete the game)
Guided maze
(Some trick lets you get through it without mapping it)
Grid maze
(Rooms are indistinguishable, but directions are consistent
and the whole thing can be mapped out on a grid. Almost
not a maze, but mazelike enough to be worth noting.)
Pseudomaze
(Something appears to be a maze, but isn't really a maze.
Only listed here so maze-haters won't be angry.
Example: Photopia)
Other pet peeves
(Please suggest more of these, but try to avoid subjective
qualities like "unfair puzzles".)
Time limits
Sliding blocks
(Like the 15 game. There are people who really loathe this.)
Bidirectional staircases
(I have this blindness: When a room description says that a
staircase goes both up and down, I often fail to notice. I
go in one direction and mark the staircase as "explored" in my
mind, only noticing my mistake long afterwards when I'm forced
to resort to hints. At least one other person has admitted to
this foible. I don't really mind two-way stairs, but I'd feel
better if I were warned about them.)
Runtimes?
(Some games are distributed as platform-independent data files
bundled with the runtime for a particular system. Nearly all
AGT games packaged like this, as are a few TADS games. We need
some way to signal this, and I can think of no better place than
in the Attributes.)
Bundled with runtime for MS-DOS
Bundled with runtime for Mac OS
Unregistered version
Incomplete game; demo contains mere fraction of registered version
Complete game, but cannot be won without registering
Winnable, but lacks hints/goodies available in registered version
Hints included
External hint file bundled with game
Room hints
(Not common today, but many older games have a hint text
specific to each room, available via the "help" command.
Scott Adams did it, and AGT at least includes the
possibility.)
Hint menu
Adaptive hints
Source code included
Grammatical oddities
First person
Third person
Past tense
Randomized combat
T. S Eliot scene
(Something of an in-joke.)
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Doeadeer3
>From: Carl Muckenhoupt <ca...@wurb.com>
>Date: 4/18/99 7:41 AM Pacific Daylight Time
As far as I can tell, you left out sexual content. I realize there are not that
many games at the archive with that, but it should be mentioned for those that
have it.
Since, I, in particular usually find those games, sexist.
Doe :-) Flame away. But in any other categorizing system it would be
mentioned.
Doe doea...@aol.com (formerly known as FemaleDeer)
****************************************************************************
"In all matters of opinion, our adversaries are insane." Mark Twain
TenthStone
wrote:
> 1. Subtitle?
Check. (yes)
> B. Authors
> 1. First name
> 2. Middle name
> 3. Last name
> 4. Honorific?
Possibly.
Nickname, or tag, or whatever. Possibly, name on ifMUD.
> 7. Role
Contributor; that is, contributing artist. For graphical adventures,
or games with music not written by the author.
> C. Companies
> 3. Fan site URLs?
Maybe those specific to this game.
> E. Series
> 2. Index in series
> For example, the index of Zork II would be 2. Note that
> series indices generally follow the chronology of the fiction,
> which does not necessarily correspond to order of
> release date. (Example: Unnkulia Zero.) Also note that
> games in a shared universe don't necessarily have an ordering.
Possibly order in series in terms of release?
> G. Genre
> The mapping between games and genres is most decidedly NOT
> one-to-one.
> For a possible list of genres, consult
> http://www.wurb.com/if/data/items/genres.
> Actually, I'd like to make the genre system more hierarchical,
> with subgenres like "Science Fiction:Time Travel" and the like.
Being able to specify a "minor" genre might be helpful.
> 6. Development system/programming language
> a. Location (relative to Archive root) of compilers?
> b. Related web sites?
These are maybes.
> 7. Platform
> Runtime systems like the TADS runtime and the Z-machine
> are considered to be platforms. "Source code only" should
> probably also be considered a platform.
> a. Required verion of runtime
> b. Location (relative to Archive root) of interpreters/emulators?
> c. Related web sites?
What you could do is insert a link to a page on the Guide
which would be devoted to that language.
> 8. Language
> Human language, that is. Not programming language.
Maybe an attribute for "atrocious mastery of language"?
Something that should only be given in the most extreme cases,
when it might cause a serious problem with gameplay.
> J. Related files
> 2. Description
> "Solution", "Map", etc. Should this be an arbitrary string
> or a selection from a list?
A list should work, but an "other" field might be a good idea.
Solution, Map, Hints, ...?
Source code?
> C. Collective ratings?
> The idea here would be to report the average rating given by
> users. I'm not sure I like this idea, though. It has all of the
> Comp's problems magnified.
Maybe a mean, but also a stem plot? I don't remember how Baf's
does its ratings, but possibly it could report how many ones, twos,
threes, etc. (or how many 0-3, 4-6, 7-10).
>Non-IF
> Interactive fiction in some sense, but not a "text adventure"
> (Example: The Space under the Window)
Wait. How would a graphical adventure fit into this? (I don't know
if Baf's includes graphical adventures now, but it should probably
be prepared to do so someday).
>Graphics
> (The primary concern here is to warn players using text-to-speech
> software.)
Good idea.
>Sound
> Uses sound, which can be disabled
> Uses sound, which cannot be disabled
> Features continuing background music
Uses "beep"?
>Other pet peeves
> Time limits
> Sliding blocks
> (Like the 15 game. There are people who really loathe this.)
Maybe in general "Contains classic puzzles". Okay, that might be
misleading, but the idea is there: anything such as sliding blocks,
Towers of Hanoi, etc.
Weird Beard
Doeadeer3 <doea...@aol.com> wrote in message
news:19990418141150...@ng-ce1.aol.com...
> As far as I can tell, you left out sexual content. I realize there are not
that
> many games at the archive with that, but it should be mentioned for those
that
> have it.
>
> Since, I, in particular usually find those games, sexist.
>
That would, hopefully, be covered in section G if it mattered. And 99% of
the time. it won't matter.
David Glasser
> As far as I can tell, you left out sexual content. I realize there are not
> that many games at the archive with that, but it should be mentioned for
> those that have it.
>
> Since, I, in particular usually find those games, sexist.
This, and violence language drugs etc, is a good idea on one hand.
On the other hand, for some games this might be a spoiler. Or even an
anti-spoiler: Photopia starts out with protagonists cursing and
drinking, but changes without you expecting it.
--
David Glasser: gla...@uscom.com | http://www.uscom.com/~glasser/
DGlasser@ifMUD:orange.res.cmu.edu 4001 | raif FAQ http://come.to/raiffaq
'No, GLK is spelled "G L K". What is this Java you speak of?'
--Joe.Mason on that portable thing on rec.arts.int-fiction
Carl Muckenhoupt
doea...@aol.com (Doeadeer3) wrote:
> As far as I can tell, you left out sexual content. I realize there are not
that
> many games at the archive with that, but it should be mentioned for those that
> have it.
Ooh, yes. Good call. This is exactly what the Attribute tags are for.
Graphic violence as well. And strong language. Maybe even drugs.
There are people out there who are downloading games for their children.
(And then there are the folks who would select *for* these qualities...)
Carl Muckenhoupt
ca...@wurb.com
Lelah Conrad
>As far as I can tell, you left out sexual content. I realize there are not that
>many games at the archive with that, but it should be mentioned for those that
>have it.
I would like to see some sort of tag like for explicit sexual content
and also for violence (I'm not talking about "kill troll with sword"
-- I'm really talking about a game in which you as the player
character would have to participate in or read lot of graphical
violence), especially if a game contains a high degree of either.
(Incidentally, this might help teachers and others help sort out games
suitable for young people.) Perhaps a simple tag like "adult themes"
could be enough, since that seems to be a current codeword for such
content (although some people may only want to be warned away from one
and not the other).
Lelah
Lelah Conrad
wrote:
Whew. Excellent and comprehensive system. As a former cataloger of
many years, I salute your thoroughness. Maintenance of data is always
a big job in databases (witness the big technical staffs in research
libraries) but if the archive were to stay stable it might be doable.
This would be an interesting and valuable tool.
Lelah
Andrew Plotkin
> G. Genre
> The mapping between games and genres is most decidedly NOT
> one-to-one.
This is *not* an objective item. It's as subjective as "rating" or "sexual
content".
It's also sometimes a spoiler.
> For a possible list of genres, consult
> http://www.wurb.com/if/data/items/genres.
> Actually, I'd like to make the genre system more hierarchical,
> with subgenres like "Science Fiction:Time Travel" and the like.
(Goob, sitting next to me, remarked "That's fine. Genre Zarf; Subgenre
Zarfian.")
> Addendum: Attributes
> There will be a collection of attribute tags that can be
> tacked onto various games or implementations. Users should be
> able to do searches that require or forbid specific attributes.
Some of these are reasonably objective (non-IF, graphics, etc.)
Others are subjective: mazes, other pet peeves.
For all subjective data, it may be convenient to allow many people to
supply their own versions. You've already considered this for "ratings"
and "reviews", but it makes just as much sense for the rest. Be able to
filter on "Fred thinks this game has adult-only content". Or sort by
Samantha's genre categorization, as opposed to yours.
This may sound overly featurely, but arguments about genre can get
*fierce*. Ditto for what's a sexual situation. Really.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."
Adam J. Thornton
Andrew Plotkin <erky...@netcom.com> wrote:
>This is *not* an objective item. It's as subjective as "rating" or "sexual
>content".
>> There will be a collection of attribute tags that can be
>> tacked onto various games or implementations. Users should be
>> able to do searches that require or forbid specific attributes.
>
>Others are subjective: mazes, other pet peeves.
If there's a "sexual content" tag, I want a "dragons" tag.
Show me all the perverse sexual behaviors you want, I'm still not going to
be nearly as angered as I will be by a $&#*(#&@(ing dragon.
Adam
--
ad...@princeton.edu
"There's a border to somewhere waiting, and a tank full of time." - J. Steinman
Papanele
Andrew Plotkin
> In article <erkyrathF...@netcom.com>,
> Andrew Plotkin <erky...@netcom.com> wrote:
> >Others are subjective: mazes, other pet peeves.
> If there's a "sexual content" tag, I want a "dragons" tag.
> Show me all the perverse sexual behaviors you want, I'm still not going to
> be nearly as angered as I will be by a $&#*(#&@(ing dragon.
That'd be both, wouldn't it?
--Z (oh, look, it's Anne McCaffrey)
White Knight
David Glasser <gla...@uscom.com> wrote
> On the other hand, for some games this might be a spoiler. Or even an
> anti-spoiler: Photopia starts out with protagonists cursing and
> drinking, but changes without you expecting it.
I think calling "contains sex" or "contains violence" a spoiler is
taking
the idea of "spoiler" to an extreme. Most movies have ratings, and this
is essentially the same thing. Also, if a game has adult language, then
I
do not think it should be an exception to the rule just because it also
happens to be a good piece of IF -- for example, Photopia. While I
consider Photopia to be an excellent work of IF, its use of adult
language,
and the very nature of the story, makes it a story that I would reserve
for
teens or older. Not for young children.
knight37
Carl Muckenhoupt
L. Awards
1. Award name (XYZZY Awards, Softworks AGT Contest, etc.)
2. Year
3. Rank (where applicable)
4. Category name (where applicable)
Carl Muckenhoupt
ca...@wurb.com
Doeadeer3
>From: erky...@netcom.com (Andrew Plotkin)
>Date: 4/18/99 9:30 PM Pacific Daylight Time
>Message-id:
>Carl Muckenhoupt (ca...@wurb.com) wrote:
>> G. Genre
>> The mapping between games and genres is most decidedly NOT
>> one-to-one.
>
>This is *not* an objective item. It's as subjective as "rating" or "sexual
>content".
>
>It's also sometimes a spoiler.
>
Well, so are reviews. Sometimes people want to know ahead of time what they are
getting into. Some of us don't have a lot of time to waste and would rather
focus on games we think we might enjoy rather than waste a few hours (or more)
on something we won't (until we find out we won't).
I agree genre can sometimes be a spoiler, thinking of one game I am working on,
on first appearance it is a mystery. On second appearance, well...
One could stick with giving the OBVIOUS genre.
For instance, what would I call, "So Far"? Surrealism. But it could actually be
tagged as "fantasy".
Doe :-)
Lucian Paul Smith
: As promised in my previous post, here is a possible outline
: of content in the proposed IF database.
Yay! This is amazingly complete. Getting all the games *into* this
system is going to be rough, but very worth it.
A few suggestions:
: 4. Honorific?
This is kind of a tricky way of including gender information. Perhaps we
should just include gender directly.
: D. Licensing status
: Posible values are:
: 1. Free of charge
: This does not necessarily mean "freeware" or "public domain".
: It just means that no one is asking for money.
: 2. Shareware
: Freely redistributable, but the author requests payment.
: 3. Former shareware
: The author is no longer requesting money. This is distinguished
: from "Free of charge" to avoid confusion; implementations may
: contain obsolete requests for registration.
: 4. Commercial
: Games whose license does not grant the right to redistribute.
: There shouldn't be anything like this at the Archive, but there
: probably is.
: 5. Former commercial
: Same idea as "Former shareware".
The other important information needed here is 'distribution allowances'.
Whether the game can be distributed freely, whether you should contact the
author first, whether one implementation can be distributed but not
another, etc.
: (Example: Unnkulia Zero.) Also note that
: games in a shared universe don't necessarily have an ordering.
Just be sure games like 'Balances' and 'Spiritwrak' can be included in the
'Zork/Enchanter Universe' without having to have a particular 'number'.
[Also, not that it matters for the if-archive, but what about games that
are a series in their own right, but also part of a shared universe? Like
the 'Enchanter' series.]
: F. Chronology
: This may be deemed unnecessary, because of the dates
: on the implementations (see below). However, in many
: cases the original release is not available at the Archive.
However, all the comp games have both versions there, so this could be
important.
: G. Genre
: The mapping between games and genres is most decidedly NOT
: one-to-one.
You mean that any given genre can have more than one game written in it?
;-)
But yeah, I'd like to see an 'inclusive' genre approach, so things could
fall in more than one category. It might be nice to include some
IF-specific genres, too, like "One-room", "Non-human protagonist", and
"Suburban House"
: "Solution", "Map", etc. Should this be an arbitrary string
: or a selection from a list?
I like a list. Be sure to include 'auxillary file' for some DOS games
(in particular) where the files are required for the program to run.
: A. Baf's Guide
: Perhaps the reviews on Baf's Guide should be treated like any
: other external reviews. But then again, perhaps not; after all,
: the database is largely intended to provide the same functionality
: as Baf's Guide (only enhanced), and my intention is to generate the
: next generation of the Guide from the database.
I think Baf's Guide should be treated like any other external review.
: 1. Rating
: 2. Review
: Possibly in some cases too big to easily stuff into a database?
: Or maybe not. If all else fails, this field can hold a filename.
Perhaps include the first paragraph of the review? If that was all there
was, great, and if not, put [Truncated: see ______] at the end.
: C. Collective ratings?
: The idea here would be to report the average rating given by
: users. I'm not sure I like this idea, though. It has all of the
: Comp's problems magnified.
Perhaps the best you could do here would be to put 'positive', 'negative',
or 'mixed'. And maybe select phrases from the reviews? I'm thinking of
something like the restaurant reviews the Rice GSA puts out (examples at
http://www.ruf.rice.edu/~gsa/YP//res-mex.html)
: Other pet peeves
I'd lump Mazes in with the rest of the 'pet peeves'.
: Hints included
: External hint file bundled with game
: Room hints
: (Not common today, but many older games have a hint text
: specific to each room, available via the "help" command.
: Scott Adams did it, and AGT at least includes the
: possibility.)
: Hint menu
: Adaptive hints
Another possibility is in-game hints, like the devil in Curses or (ahem)
the etchings in Edifice.
: Source code included
I'd just list that in the 'other files' bit, I think. Another reason to
have a list there instead of a string. (There also exist examples of
partial source code available. I did this with 'Nalian.inf'; others have
released modules that they've used in their games. Perhaps, if someone
mentioned that they used the 'hints.h' file in the
Inform/library/contributions directory, it could be included on the same
list? This would mean that some files would be included in several games,
but could be nice for people playing a game who think, "Hey, I want to
have that effect in my game.")
Looks good! If this gets going, it'd be a *great* resource.
-Lucian
Carl Muckenhoupt
lps...@rice.edu (Lucian Paul Smith) wrote:
> Carl Muckenhoupt (ca...@wurb.com) wrote:
> : As promised in my previous post, here is a possible outline
> : of content in the proposed IF database.
>
> Yay! This is amazingly complete. Getting all the games *into* this
> system is going to be rough, but very worth it.
Ah. Then I can count on your participation? :)
I'm reading everyone's comments, but I'm only going to reply to
the aspects that I actually have something to say about (beyond
"good idea" or whatever).
> A few suggestions:
>
> : 4. Honorific?
>
> This is kind of a tricky way of including gender information. Perhaps we
> should just include gender directly.
That isn't actually what I had in mind. The Honorific would only be
used if it's part of the author's signature. For example, there is now
a game at the Archive written under the pseudonym "Mr. Wigglebutt".
(At least I hope it's a pseudonym.) When the system puts together a
page that includes that author's name, "Mr. Wigglebutt" is how it should
appear. And yet I don't feel quite right about claiming that his first
name is "Mr."
In addition, I've seen people around here going under titles like
"Admiral" and "Cardinal". Surely such people deserve to have their
exalted ranks recognized.
> The other important information needed here is 'distribution allowances'.
> Whether the game can be distributed freely, whether you should contact the
> author first, whether one implementation can be distributed but not
> another, etc.
Well, generally speaking, things that can't be distributed freely
shouldn't be in the Archive at all.
> : Source code included
>
> I'd just list that in the 'other files' bit, I think.
What about cases where a single zip file at the Archive contains both
executable and source? Wouldn't it look a little strange to list it
twice (once as an implementation, once as a related file)?
But not disastrous, I suppose. One thing to point out: A listing for
source code needs to indicate what language it's in. If source code
is treated as an "other file", this argues in favor of the file type
being an arbitrary string.
Stephen van Egmond
David Glasser <gla...@uscom.com> wrote:
>This, and violence language drugs etc, is a good idea on one hand.
>
>anti-spoiler: Photopia starts out with protagonists cursing and
>drinking, but changes without you expecting it.
This brings to mind the "plot keywords" used by IMDB for movies. For
instance, Trainspotting offers (spoilers)
black-comedy time-lapse vulgarity addiction aids based-on-novel betrayal
drugs heroin realtors relationship revenge schoolgirl-uniform surreal
terminal-illness
(realtors?)
These are pretty neutral; there is an external (to IMDB) review service
called "Screen It!" which catalogues everything someone might find
annoying about a movie - not at all neutral.
http://www.screenit.com/movies/1996/trainspotting.html
(only 141 times?)
Do these strike people as being particularly useful?
/Steve
Yes, I'm back
--
,,,
(. .)
+--ooO-(_)-Ooo------------ --- -- - - - -
| Stephen van Egmond http://bang.ml.org/
Magnus Olsson
Andrew Plotkin <erky...@netcom.com> wrote:
>Adam J. Thornton (ad...@princeton.edu) wrote:
>> If there's a "sexual content" tag, I want a "dragons" tag.
>
>> Show me all the perverse sexual behaviors you want, I'm still not going to
>> be nearly as angered as I will be by a $&#*(#&@(ing dragon.
>
>That'd be both, wouldn't it?
What's so perverse about two dragons $&#*(#&@(ing? I mean, they've
got to continue the species, don't they?
Provided, of course, that all the parties involved are dragons.
>--Z (oh, look, it's Anne McCaffrey)
I see. That explains it.
:-)
--
Magnus Olsson (m...@df.lth.se, zeb...@pobox.com)
------ http://www.pobox.com/~zebulon ------
Carl Muckenhoupt
ad...@princeton.edu (Adam J. Thornton) wrote:
> If there's a "sexual content" tag, I want a "dragons" tag.
Well, I suppose it's no more unreasonable than tagging
bidirectional staircases...
Magnus Olsson
Anson Turner <anson@DELETE_THISpobox.com> wrote:
>In article <7ffssk$r6l$1...@joe.rice.edu>, lps...@rice.edu (Lucian Paul
>Smith) wrote:
>
>:
>:This is kind of a tricky way of including gender information. Perhaps we
>:should just include gender directly.
>
>*girl*!
I can't see why we should include gender information at all - frankly,
it smacks of sexism.
Besides, most IF authors have names or pseudonyms that are rather clearly
gender-specific. Those who haven't may not be very keen on having their
gender disclosed.
But back to the question of honorifics. I don't think Carl meant to
put "Mr" or "Ms" in that field. (Is "Mr" a honorific at all, by the
way?). But how many IF authors have titles? There are a few PhD's
around, and one Cardinal :-). I can only speak for myself, of course,
but I really don't see a need to put a "Dr" in my entries in the data
base, and a "Mr" would seem rather ridiculously formal.
Magnus Olsson
White Knight <knig...@flash.NOSPAM.net> wrote:
>
>David Glasser <gla...@uscom.com> wrote
>
>> On the other hand, for some games this might be a spoiler. Or even an
>> anti-spoiler: Photopia starts out with protagonists cursing and
>> drinking, but changes without you expecting it.
>
>taking
>the idea of "spoiler" to an extreme.
I can't think of any game off-hand for which this would be a spoiler.
>Most movies have ratings, and this
>is essentially the same thing.
I don't know how the American system of movie ratings works, but is a
film automatically R-rated (or whatever) if it contains just one
four-letter word? If taken literally, the characterization "contains
adult language" means just that. And what about "contains nudity"? I-0
would certainyl merit such a label :-), but would Anchorhead (where
the protagonist has to undress completely to take a bath)?
My point is that we must be careful with such categories, to avoid
having them backfire. I mean, if all games where the (female) player
can do "take off shirt" are labeled as "contains nudity", then all
those games will necessarily be lumped together in the minds of many
people - even though the category would contain everything from
totally innocuous games (a children's game where you have to change
clothes in the privacy of your own room) to "Stiffy Makane".
Adult language is a little more clear-cut, perhaps, but "sexual
content" is also a difficult one. And "violence" - does whacking
trolls with your Elvish sword count?
>Also, if a game has adult language, then I
>do not think it should be an exception to the rule just because it also
>happens to be a good piece of IF -- for example, Photopia.
I think this depends a lot on what people are going to use the ratings
for. If a parent doesn't want her eight-year-old daighter to be
exposed to dirty language, she may want to filter out *all* games that
contain such words in any context. Fine - that's her decision (though
I'm afraid it's rather futile - children tend to pick up that kind of
language anyway).
But my concern is that putting a big orange sticker "WARNING: Contains
adult language and sexual references" on Photopia would give a lot
of people an entirely wrong idea of what the game is about.
And, of course, there's the nagging suspicion that characterizing
works in this way just paves the way for the censors and self-appointed
morality watchers.
Lucian Paul Smith
: I can't see why we should include gender information at all - frankly,
: it smacks of sexism.
I must, respectfully, strongly disagree.
Information is just information. It can be used in a sexist way or not.
I could scan through the archive looking at gender ratios and post, "Hey,
boyz are better than girlz, cuz lookit all the great games by males!
Also, they have cooties!", or I could post, "Hey--good news: over the
past three years the number of archived games by women has increased by
about 15% a year." To exclude this information because it could be
misused offends my sensibilities.
Besides, the principle utility I see in including this information is to
eliminate minor embarrassing gaffes. Remember someone's review of
'Everybody Loves A Parade" where the reviewer consistently referred to
Cody as a woman? And especially when authors are from other countries, I
often don't have a good idea of whether the name is male or female. And,
should I want to write a review of the game or e-mail them directly, it's
embarrassing to have to ask, "Hey, are you male or female?"
: Besides, most IF authors have names or pseudonyms that are rather clearly
: gender-specific. Those who haven't may not be very keen on having their
: gender disclosed.
If any author objects to *any* of the information presented in the archive
about them, they should have the right to ask for it to be removed.
: But back to the question of honorifics. I don't think Carl meant to
: put "Mr" or "Ms" in that field. (Is "Mr" a honorific at all, by the
: way?). But how many IF authors have titles? There are a few PhD's
: around, and one Cardinal :-). I can only speak for myself, of course,
: but I really don't see a need to put a "Dr" in my entries in the data
: base, and a "Mr" would seem rather ridiculously formal.
Here I agree, and is one of the reasons I objected to the honorific idea
in the first place (although I did indeed feel he meant the 'Mr' and 'Ms'
thing.).
-Lucian
Carl Muckenhoupt
not giving too much away, but the plot hinges on incest. I'd wouldn't call
the game "sexually explicit" - although it's made clear what's happened, it's
never depicted. And I certainly wouldn't call the game sexist, which is what
Doe was concerned about. But I would want to warn our hypothetical concerned
parent. What's do be done?
I'm tempted to agree with Zarf and leave these tags out on the basis that they
cannot be defined clearly. But there seems to be a demand for them. We'll
probably just have to muddle through as best we can.
Magnus Olsson
Lucian Paul Smith <lps...@rice.edu> wrote:
>Magnus Olsson (m...@bartlet.df.lth.se) wrote:
>
>: I can't see why we should include gender information at all - frankly,
>: it smacks of sexism.
>
>I must, respectfully, strongly disagree.
>
>Information is just information. It can be used in a sexist way or not.
>I could scan through the archive looking at gender ratios and post, "Hey,
>boyz are better than girlz, cuz lookit all the great games by males!
>Also, they have cooties!", or I could post, "Hey--good news: over the
>past three years the number of archived games by women has increased by
>about 15% a year." To exclude this information because it could be
>misused offends my sensibilities.
Well, yes, I see your point, and I think I overstated my point a bit.
WHat I object to is not really providing the gender information, but
the idea of classifying works of art by the artist's sex. "The top
ten IF games written by women" - a list like that would make me
wonder a little about the intentions of its compiler.
A similar classification based on the author's race or religion would
be rather controversial, wouldn't it?
On the other hand, I can think of many situationbs where it could be
intersting to know, say, the sex of the author of a specific work. Or
even the religion: "This game has such an interesting take on
reincarnation. It would be interesting to know if the author is a
Hindu".
I'm in two minds about whether such information should be included in
a *database*, though. Perhaps it should be provided at the author's
initiative? Perhaps each author in the database should be allow to
supply biographical details, including - at his or her discretion -
things like gender, ethnicity, religion etc. That way, an author who
would like to be identified as a female Inuit Buddhist could expose
those facts, while another author may choose to let his works speak
for him.
>Besides, the principle utility I see in including this information is to
>eliminate minor embarrassing gaffes. Remember someone's review of
>'Everybody Loves A Parade" where the reviewer consistently referred to
>Cody as a woman? And especially when authors are from other countries, I
>often don't have a good idea of whether the name is male or female. And,
>should I want to write a review of the game or e-mail them directly, it's
>embarrassing to have to ask, "Hey, are you male or female?"
That's a point, and I remembered the Cody incident after I posted my
previous article, but again I think that it's not a given that such
facts are - or should be - relevant for a review.
>: Besides, most IF authors have names or pseudonyms that are rather clearly
>: gender-specific. Those who haven't may not be very keen on having their
>: gender disclosed.
>
>If any author objects to *any* of the information presented in the archive
>about them, they should have the right to ask for it to be removed.
Definitely.
And in many countries, such a database would be outright illegal
without prior consent from all the people represented.
Doeadeer3
>From: lps...@rice.edu (Lucian Paul Smith)
>Date: 4/20/99 7:43 AM Pacific Daylight Time
>Besides, the principle utility I see in including this information is to
>eliminate minor embarrassing gaffes. Remember someone's review of
>'Everybody Loves A Parade" where the reviewer consistently referred to
>Cody as a woman?
So???
How about those women who are using a male pseudonym or those men using a
female one?
Frankly, I don't care much either way, but I think one must allow for an author
who does not want to disclose his/her gender.
Doe :-) I also would not want an "honorific" before my name.
Second April
> WHat I object to is not really providing the gender information, but
> the idea of classifying works of art by the artist's sex. "The top
> ten IF games written by women" - a list like that would make me
> wonder a little about the intentions of its compiler.
>
> A similar classification based on the author's race or religion would
> be rather controversial, wouldn't it?
>
> On the other hand, I can think of many situationbs where it could be
> intersting to know, say, the sex of the author of a specific work. Or
> even the religion: "This game has such an interesting take on
> reincarnation. It would be interesting to know if the author is a
> Hindu".
>
> I'm in two minds about whether such information should be included in
> a *database*, though. Perhaps it should be provided at the author's
> initiative? Perhaps each author in the database should be allow to
> supply biographical details, including - at his or her discretion -
> things like gender, ethnicity, religion etc. That way, an author who
> would like to be identified as a female Inuit Buddhist could expose
> those facts, while another author may choose to let his works speak
> for him.
Given that many biographical details aren't apparent at a glance, the
author would kinda have to supply them. It's also important to remember
that some authors may _not_ want to be evaluated in those terms; they
might not want anyone to know they're female, Japanese, Muslim, etc., for
fear of being pigeonholed.
> >Besides, the principle utility I see in including this information is to
> >eliminate minor embarrassing gaffes. Remember someone's review of
> >'Everybody Loves A Parade" where the reviewer consistently referred to
> >Cody as a woman? And especially when authors are from other countries, I
> >often don't have a good idea of whether the name is male or female. And,
> >should I want to write a review of the game or e-mail them directly, it's
> >embarrassing to have to ask, "Hey, are you male or female?"
That was me. On the other hand, there wasn't much skin off anyone's
nose--Cody was a little bemused and most folks had a good chuckle. I
didn't go out and beat my head against a wall. It wasn't relevant to
that particular review (well, that review did bring up some gender
issues while avoiding spoilers, but I didn't make the author's supposed
gender an issue). I'd rather have a little embarrassment than forced
disclosure, and I think the author should be able to supply biographical
info at his/her discretion and leisure.
> >: Besides, most IF authors have names or pseudonyms that are rather clearly
> >: gender-specific. Those who haven't may not be very keen on having their
> >: gender disclosed.
> >
> >If any author objects to *any* of the information presented in the archive
> >about them, they should have the right to ask for it to be removed.
Of course. But not all authors will necessarily be hip to the ways of the
tagging system, and you might get situations where biographical info
(maybe even incorrect info) gets supplied by someone else and attached to
a game without the author's realizing it.
Duncan Stevens
d-st...@nwu.edu
773-728-9721
Love in the open hand, no thing but that,
Ungemmed, unhidden, wishing not to hurt,
As one should bring you cowslips in a hat
Swung from her hand, or apples in her skirt,
I bring you, calling out as children do,
"Look what I have!--And these are all for you."
--Edna St. Vincent Millay
Greg Kuperberg
Carl Muckenhoupt <ca...@wurb.com> wrote:
>As promised in my previous post, here is a possible outline
>of content in the proposed IF database.
Here is a concrete proposal. The beginning of your outline sounded
very strong, but the list of information that you plan to include could
perhaps be too complicated. I think there is a way to do what you want
that will keep things as simple as possible. Keeping things
simple requires considering the implementation, unfortunately.
Each database record for a game might look something like this:
--------------------------
ID: minster
Title: Christminster
Author: Gareth Rees
Year: 1995
Genres: College; Mystery
Comments: Female protagonist
File: review.baf
Type: Review (*****)
Author: Carl Muckenhoupt
File: v1.z5
Type: Inform
Date: February 29, 1995
File: v1.tar.gz
Type: Inform source
Date: whatever
File: sol.1
Type: Walk-through
...
--------------------------
In this arrangement, the database only needs to know two kinds
of "entities": A game and a file associated to a game. The first
paragraph of the database record describes the game; each subsequent
paragraph describes the associated files. The game paragraph has just
six types of fields (ID, Title, Authors, Year, Genres, Comments);
each file paragraph has five (File, Type, Date, Author, Commments).
"File" can be a URL or a file internal to the database.
Of course I'm not suggesting that you follow this exactly; this is just
an example of what might work. My main point is that I think it could
work, but if you do something much more complicated, then it probably
wouldn't work. There is no need to commit yourself to supporting a
lot of peripheral information at the same level as the essentials.
For example, gender information can be interesting, but it does not
deserve as much support as the title and author of the game. It's the
kind of annotation that can and should be relegated to the Comments field.
Likewise it is usually not interesting whether an author is a designer
or an implementor. If such information does happen to be interesting
for some particular game, you can put it in parentheses without adding
the complication of an entirely new attribute:
Authors: Douglas Adams (designer), Joe Ghostwriter (implementer)
Of course, at least initially the database must be a "parasite" of the
ftp archive. However, there may be no compelling reason to point to
the ftp archive when you can deliver the same files yourself. This is
especially true if you can offer better identifiers than the ftp archive
does. In this case you could either automatically or semi-automatically
get all of the games from ftp.gmd.de and put them in the right place at
your web site.
What would that right place be? The example above suggests having a
directory called minster with a file called v1.z5, so that minster/v1.z5
would be Chistminster version 1, Inform platform. Perhaps the data file
that I mentioned could be minster.data in the same directory as minster.
But but but but there is a point here that many people missed in the
discussion in October, which is that the stable identifier for a game
does not have to be the same as its internal location in the database.
Identifier = filename is an old equation from the days of ftp which
should now be followed much more loosely, if at all.
In practice, one or more of the files associated to a game might be
"default files" for the game. These may include the best version for
each platform (among versions), the Baf review (among reviews), etc.
These could be marked with an asterisk in the records, for example:
File*: review.baf
You might have especially simple URLs to deliver these, for example:
http://ifdb.org/game/minster.z5
http://ifdb.org/review/minster
I have a technical comment about these URLs to reinforce the conceptual
point above: Apache lets you process the tail of a URLs in the same
way as query arguments. If you have a script called "game" in the right
place, Apache would give it the entire URI "/game/minster.z5", which
it could process as an argument list. In other words, just because it
looks like a directory path, it doesn't mean that it is one.
Another technical point is the use of ASCII records. You might think
that having colon-labelled fields (as in mail headers) that can be read
and edited directly is hack that is only suitable for small databases.
It is a hack, but the advantages of human readability and editability do
not go away as the database gets more sophisticated. In practice even
some huge databases have "raw forms" that look much like mail headers.
The raw form can be compiled into something fancier, like an SQL database,
for use by CGI scripts.
--
/\ Greg Kuperberg (UC Davis)
/ \
\ / Visit the Math Archive Front at http://front.math.ucdavis.edu/
\/ * Mathematics should be selected and not censored *
Adam Cadre
> WHat I object to is not really providing the gender information, but
> the idea of classifying works of art by the artist's sex. "The top
> ten IF games written by women" - a list like that would make me
> wonder a little about the intentions of its compiler.
Left Bank Books here in Seattle does, in fact, arrange its shelves
so that books by women are along one wall and books by men are along
the opposite wall.
I didn't see any books co-authored by male and female writers --
I guess they'd just pile those on the floor or something.
-----
Adam Cadre, Seattle, WA
http://adamcadre.ac
Magnus Olsson
Carl Muckenhoupt <ca...@wurb.com> wrote:
>Anchorhead has a good deal more than innocent bather's nudity.
Of course. Sorry if I gave any other impression, but I didn't want to
put in too much spoilers.
Anyway, I was referring *only* to the context of the protagonist's
nudity in the game, a context which is almost entirely non-sexual. The
game itself is certainly *not* innocent - there's the incest motif (an
update on Lovecraft, who - true to his time - seems to have been more
concerned with inbreeding than with incest itself), some other sexual
allusions, and lots of generally nasty things.
OK, it was a bad example, but my point was that although the game
requires the protagonist to get naked to advance the plot, that nudity
has nothing whatsoever to do with the "adult" themes of the game.
Michael Straight
> G. Genre
[...]
> Mazes
[...]
> Other pet peeves
> Time limits
> Sliding blocks
> Bidirectional staircases [!?!]
> Grammatical oddities
> Randomized combat
I'd suggest setting up the front end so that information like this that
might be spoilers is only if the user asks for it.
I still haven't gotten around to finishing _Spider and Web_, but I'm sure
glad I played the intro before reading any reviews.
SMTIRCAHIAGEHLT
Doeadeer3
>From: gr...@manifold.math.ucdavis.edu (Greg Kuperberg)
>Date: 4/20/99 12:37 PM Pacific Daylight Time
>that will keep things as simple as possible. Keeping things
>simple requires considering the implementation, unfortunately.
>
>Each database record for a game might look something like this:
>
>--------------------------
>ID: minster
>Title: Christminster
>Author: Gareth Rees
>Year: 1995
>Genres: College; Mystery
>Comments: Female protagonist
>
>File: review.baf
>Type: Review (*****)
>Author: Carl Muckenhoupt
I didn't agree with a lot of what you said last year, but I do agree with the
above. The simplier the better (usually known as KISS :-) ).
Actually this is different, a web page maintained by someone else that points
to the archive. Not a revamping of the archive itself.
I have a bit of a problem with listing pet peeves, for instance.
We all have different ones. For some a dragon would be an attraction, not an
irritation. Maybe, ditto, a maze.
I would tend to put such things in a LONG comment field that would only be
readable with another click (another URL, subpage), so that who don't want
spoilers can avoid them. Actually, comments on mazes, dragons, etc. really
fall under the classification of reviews, anyway.
Telling what the protagonist is, whether there is sexual content and/or strong
language or the historical period it is set in could be in a short comment
field.
Setting: Zork Universe, Underground
Protagonist: Usual Brave Adventurer
Setting: Surburan House
Protagonist: Stuffed Teddy Bear
Setting: Turn of the Century
Protagonist: Aging Reverand
That would be enough. The rest could be covered in reviews or longer "click on"
comment fields. (One room, cute -- could be for children if are old enough to
understand how to play IF, more emphasis on character interaction than
puzzles.)
The way Amazon.com describes a book might be a good way to go. Split page,
basic info at top, comments and reviews at the bottom. One previewing games to
play could skip scrolling down to see the bottom if they don't want spoilers.
Of course, that would be a great deal of web pages and a large enough ftp site
to store them.
Doe :-)
Lucian Paul Smith
: In article <7fi3qr$cd3$1...@joe.rice.edu>,: >
: >Information is just information. It can be used in a sexist way or not.
<snip>
: Well, yes, I see your point, and I think I overstated my point a bit.
: WHat I object to is not really providing the gender information, but
: the idea of classifying works of art by the artist's sex. "The top
: ten IF games written by women" - a list like that would make me
: wonder a little about the intentions of its compiler.
OK. But again, this is just the database, not a classification scheme.
: A similar classification based on the author's race or religion would
: be rather controversial, wouldn't it?
Hrm, good point. I don't think the analogy extends *quite* that far, but
I can see what you're getting at.
: I'm in two minds about whether such information should be included in
: a *database*, though. Perhaps it should be provided at the author's
: initiative? Perhaps each author in the database should be allow to
: supply biographical details, including - at his or her discretion -
: things like gender, ethnicity, religion etc. That way, an author who
: would like to be identified as a female Inuit Buddhist could expose
: those facts, while another author may choose to let his works speak
: for him.
Ah, OK. I'm much more comfortable excluding information on a privacy
issue than I am on an "it's sexist" issue.
I suppose it comes down to what the 'default' should be. I'm inclined to
say, "Include as much information in the db as comes with the game
itself." Since, usually, the inclusion of the author's given name also
indicates gender, I'd be for going ahead and filling in that field in the
db for gender-obvious names. For those names that are ambiguous, I'd say
that the author should be contacted before the information is included.
Does this sound fair?
I guess this is my answer to the race/religion thing. It's not
immediately clear from the information in the game what race or religion
the author is, while in the majority of cases, gender *is* known. It's a
weak argument, but it's the only one I got ;-) Presumably, those authors
who wished to conceal their gender wrote their games anonymously or
pseudonymously.
...which brings up another point: How to classify 'Meteor' and
'Photopia', both games whose authors entered 'disguised' as women, but who
later revealed their actual identity? Leaving the field blank would seem
suspicious; filling it in as male would be spoiler-y, and filling it in as
female would be misleading (though perhaps only equally misleading as the
pseudonym.) We're fortunate here in that we can ask the authors
personally, but what if we couldn't? My vote: fill it in female. But I
can also see the situation as an argument for leaving it blank by default.
<snip>
: I remembered the Cody incident after I posted my
: previous article, but again I think that it's not a given that such
: facts are - or should be - relevant for a review.
No, but I wasn't saying that the review was worse for getting Cody's
gender wrong. I was just saying it was embarrassing for the reviewer.
And I also think it's not a given that such facts are *not* or should not
be relevant for a review.
: >: Besides, most IF authors have names or pseudonyms that are rather clearly
: >: gender-specific. Those who haven't may not be very keen on having their
: >: gender disclosed.
: >
: >If any author objects to *any* of the information presented in the archive
: >about them, they should have the right to ask for it to be removed.
: Definitely.
: And in many countries, such a database would be outright illegal
: without prior consent from all the people represented.
Which is another reason to lean towards the 'information present within
the game' idea.
The advantage I see in including this information is that it could clear
up ambiguity in situations where ambiguity was not intended. Since some
of the ambiguity is cultural (Americans looking at Finnish names, for
example, or (I'd expect) visa-versa), it'd be nice to simply include this
information by default. For authors with gender-ambiguous names who
don't intend ambiguity it'd be nice to provide an avenue for them to
clarify the issue without being somehow singled out.
I also think it *would* be interesting to see if female IF authorship has
increased in the past few years. My impression is that it has, but I like
a statistic. ;-)
-Lucian
Greg Kuperberg
Carl Muckenhoupt <ca...@wurb.com> wrote:
>I've stated this elsewhere, but this thread seems to be ignoring it:
>The "honorific" field, if included in the database, would only be
>used in those few cases where author signs the work with an honorific.
...
>Whether these few cases justify an extra field is debatable - hence the
>"?" in the outline.
Of course it doesn't justify an extra field! You don't need more than
one field for the author's name, which can be taken as a single string,
either "Carl Muckenhoupt" or "Mr. Carl Muckenhoupt" or however the author
signs his/her/its games. For that matter, whatever bio information
you have about authors should be in a separate table from the games,
since it depends only on the author and not the game.
My advice (short version) is to build up the database from its
mission-critical parts. Getting the essentials right is hard enough;
it will be a long time before you have to worry about peripheral details.
Greg Kuperberg
Carl Muckenhoupt <ca...@wurb.com> wrote:
>How would people feel about replacing the "genres" system with
>something more like the IMDB keywords?
I like the genres; you shouldn't dilute them in a more generic
keyword system.
David Glasser
> This brings to mind the "plot keywords" used by IMDB for movies. For
> instance, Trainspotting offers (spoilers)
>
> black-comedy time-lapse vulgarity addiction aids based-on-novel betrayal
> drugs heroin realtors relationship revenge schoolgirl-uniform surreal
> terminal-illness
>
> (realtors?)
Surreal realtors in schoolgirl-uniforms, apparently.
> These are pretty neutral; there is an external (to IMDB) review service
> called "Screen It!" which catalogues everything someone might find
> annoying about a movie - not at all neutral.
>
> http://www.screenit.com/movies/1996/trainspotting.html
>
> (only 141 times?)
>
> Do these strike people as being particularly useful?
The IMDB's keywords are great. Not for cross-linking as much as just
seeing about a movie. Screenit has frequently been mocked on ifMUD, on
the other hand.
> /Steve
> Yes, I'm back
Yay! I take it that bang.dhs.org is the new URL for your site?
--
David Glasser: gla...@uscom.com | http://www.uscom.com/~glasser/
DGlasser@ifMUD:orange.res.cmu.edu 4001 | raif FAQ http://come.to/raiffaq
"Well, it's interesting. We're willing to kill in the name of David
Glasser." --Steven Marsh on rec.arts.int-fiction
Greg Kuperberg
Dan Shiovitz <d...@cs.wisc.edu> wrote:
>In article <7fjb9f$ee5$1...@manifold.math.ucdavis.edu>,
>Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>>I like the genres; you shouldn't dilute them in a more generic
>>keyword system.
>the game. How much does "fantasy" tell you?
Of course it doesn't say much about a specific game. If I want that,
I'll read the review. But the genre does help organize the list of games.
I might be in the mood for a college game, or I might want to know how
many college games have been written, etc.
In practice, such classifications only work if they aren't too
complicated. If you have 10-30 genres, it works well. If you have a
nebulous keyword system with 800 possible keywords, it doesn't work.
Greg Kuperberg
Andrew Plotkin <erky...@netcom.com> wrote:
>> In practice, such classifications only work if they aren't too
>> complicated. If you have 10-30 genres, it works well. If you have a
>> nebulous keyword system with 800 possible keywords, it doesn't work.
>
Work for the goal of helping me find the games I want, and to understand
the list of games as a whole.
I have used a number of different on-line databases of different sizes
over the past ten years. My nearly universal experience is that keywords
are very hard to implement properly, and in fact they are destined for
failure if they are author-provided. Even when they do work well, they
are not a substitute for a coarser classification on the leverl of genres.
Two examples are the "META" tags in HTML documents and keywords in the
Internet Movie Database (IMDB). META tags are so widely misused, both
unintentionally and intentionally, that many search engines ignore them.
IMDB is one of the rare exceptions with good keywords, but it also
classifies movies separately by genre.
Another reason that keywords are not that important is that the 10-line
Baf reviews would already provide pretty good keywords if they were
searchable. This is a lesson of e-print archives: The author-provided
abstracts are a pretty good source of keywords, while author-provided
keyword lists would be useless, and maintainer-provided keywords would
be prohibitively time-consuming.
In other words, the genre classification works pretty well and should
not be replaced by something that might or might not work at all.
Adam J. Thornton
Not even Luminary Doe?
Adam
--
ad...@princeton.edu
"There's a border to somewhere waiting, and a tank full of time." - J. Steinman
Carl Muckenhoupt
The "honorific" field, if included in the database, would only be
used in those few cases where author signs the work with an honorific.
revealed one Mr., one Cardinal, and one Admiral (surprisingly enough,
not Jota.) There are probably a few more that aren't mentioned in the
master index, and maybe some honorifics that I didn't think to look
for. (Is "The" an honorific?)
Whether these few cases justify an extra field is debatable - hence the
"?" in the outline. But I certainly didn't think the suggestion would
be nearly this controversial...
As for whether or not to include gender information: Does anyone feel
strongly that we should? I'm inclined to leave it out; it seems like
more trouble than it's worth. For example, there are many authors for
whom this information is simply unavailable. But if people want it,
I'm willing to include it. More generally, I think that the best way
to include information about the authors (beyond which games they wrote)
is through a URL to the authors' home pages, where they can tell you
exactly as much about themselves as they wish you to know. The only
problem with this is that you can't do searches on the info in external
web pages.
> >If any author objects to *any* of the information presented in the archive
> >about them, they should have the right to ask for it to be removed.
Indeed, under the current proposal, authors wouldn't even have to ask.
They could just edit the database directly. :)
> Definitely.
>
> And in many countries, such a database would be outright illegal
> without prior consent from all the people represented.
Hmm. This is a concern. Does it make a difference if all the
information presented is already available to the public?
Carl Muckenhoupt
svane...@bang.dhs.org (Stephen van Egmond) wrote:
> In article <1dqhdon.t6...@usol-209-186-16-213.uscom.com>,
> David Glasser <gla...@uscom.com> wrote:
> >This, and violence language drugs etc, is a good idea on one hand.
> >
> >anti-spoiler: Photopia starts out with protagonists cursing and
> >drinking, but changes without you expecting it.
Perhaps it would help if some attributes weren't reported with the
review. They could be used solely as search criteria, and otherwise
invisible.
This wouldn't be necessary for all of the proposed attributes, however
- "source code included" is hardly a spoiler - and in some cases it
would actually be detrimental. But we could make some tags visible
and others invisible.
> This brings to mind the "plot keywords" used by IMDB for movies.
How would people feel about making the proposed "tag" system more like
the IMDB's keyword system?
How would people feel about replacing the "genres" system with
something more like the IMDB keywords?
Dan Shiovitz
Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>>something more like the IMDB keywords?
>
>keyword system.
In practice, I find that genres provide very little information about
the game. How much does "fantasy" tell you? Even the more specific
ones like "horror" don't do much -- compare "The Zuni Doll" and
"Anchorhead". I could be wrong, I guess. Does anyone out there only
play IF games of a certain genre? Plus, of course, there's the
psychological issue that providing a specified genre list encourages
new authors to pigeonhole, and that's a Bad Thing.
> /\ Greg Kuperberg (UC Davis)
--
Dan Shiovitz || d...@cs.wisc.edu || http://www.cs.wisc.edu/~dbs
"...Incensed by some crack he had made about modern enlightened
thought, modern enlightened thought being practically a personal buddy
of hers, Florence gave him the swift heave-ho and--much against my
will, but she seemed to wish it--became betrothed to me." - PGW, J.a.t.F.S.
Andrew Plotkin
> In article <7fjf0k$m...@spool.cs.wisc.edu>,
> Dan Shiovitz <d...@cs.wisc.edu> wrote:
> >Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
> >>I like the genres; you shouldn't dilute them in a more generic
> >>keyword system.
> >In practice, I find that genres provide very little information about
> >the game. How much does "fantasy" tell you?
> Of course it doesn't say much about a specific game. If I want that,
> I'll read the review. But the genre does help organize the list of games.
> I might be in the mood for a college game, or I might want to know how
> many college games have been written, etc.
> In practice, such classifications only work if they aren't too
> complicated. If you have 10-30 genres, it works well. If you have a
> nebulous keyword system with 800 possible keywords, it doesn't work.
I dunno. "Work" for what goal?
I like the idea of an indefinite nebula of keywords, with five or ten
assigned to each game. Glancing at that doesn't take any more effort than
glancing at a single "genre" tag.
Filtering by nebula is harder than filtering by genre, but the problem may
prove fruitful to consider. Start with the obvious approach, where
"fantasy", "sf", "college", "horror", etc *are* keywords whose scope is
just what you think of as genre.
--Z
--
"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."
Palfalvi Tamas
> Setting: Zork Universe, Underground
> Protagonist: Usual Brave Adventurer
>
> Setting: Surburan House
> Protagonist: Stuffed Teddy Bear
>
> Setting: Turn of the Century
> Protagonist: Aging Reverand
Possible spoilers again. How would you describe the protagonist in Photopia?
Changing all the time, sometimes different person, sometimes the same at
different age... And you can kill the maze part by trying to precisely define
the character...
Void
Carl Muckenhoupt
gr...@manifold.math.ucdavis.edu (Greg Kuperberg) wrote:
> >Whether these few cases justify an extra field is debatable - hence the
> >"?" in the outline.
>
> one field for the author's name, which can be taken as a single string,
> either "Carl Muckenhoupt" or "Mr. Carl Muckenhoupt" or however the author
> signs his/her/its games.
I agree that there doesn't seem to be any good reason to split up the
name to the degree in the proposal. But I do want to keep the last name
separate, because I want to be able to generate a list of authors,
sorted by last name. So if I don't keep the last name in its own field,
I'll have to parse it out at runtime. And then I'll have to worry about
special cases like Neil de Mause. (The Guide currently deals with this
by distorting him to "Neil deMause". I'd like to avoid altering people's
names if possible.)
> For that matter, whatever bio information
> you have about authors should be in a separate table from the games,
> since it depends only on the author and not the game.
Obviously.
> My advice (short version) is to build up the database from its
> mission-critical parts. Getting the essentials right is hard enough;
> it will be a long time before you have to worry about peripheral details.
Frankly, I'm still trying to figure out what the mission-critical
parts are...
Carl Muckenhoupt
ca...@wurb.com
Michael Straight
> What would that right place be? The example above suggests having a
> directory called minster with a file called v1.z5, so that minster/v1.z5
> would be Chistminster version 1, Inform platform. Perhaps the data file
> that I mentioned could be minster.data in the same directory as minster.
That would make things more unfriendly for the person doing the
downloading. If I tried to downloaded several such games, I'd get a bunch
of files that look like this:
v1.z5
v2.gam
v1-1.z5
v1.z5 [This file already exists. Replace existing file?]
Unless I go to the trouble of renaming files as I download them, or
making a whole bunch of directories to put them in.
SMTIRCAHIAGEHLT
L. Ross Raszewski
Greg Kuperberg
Carl Muckenhoupt <ca...@wurb.com> wrote:
>I agree that there doesn't seem to be any good reason to split up the
>name to the degree in the proposal. But I do want to keep the last name
>separate, because I want to be able to generate a list of authors,
>sorted by last name. So if I don't keep the last name in its own field,
>I'll have to parse it out at runtime. And then I'll have to worry about
>special cases like Neil de Mause.
There are several ways to do this. The most reliable one is
to store the names in inverted order:
de Mause, Neil
It is then trivial in Perl to convert to the more attractive First
M. Last form for display purposes:
$name =~ s/(.*),\s*(.*)/$2 $1/;
Note that the Front (http://front.math.ucdavis.edu/) extracts last names
from the author fields from another database which does not explicitly
"know" this information. It assumes that the last name is the last
word plus all previous words that match a heuristic. The heuristic is
lower-case words plus special cases like De and Von. There is another
heuristic for trailing Sr. and Jr. Of course this is more complicated
than what you need to do, since you have the privilege of designing the
raw database yourself.
The real advantage of these tricks is that you can make do
with a single author field for multiple-author games:
Author: Smith, Joe; Doe, Jane
Of course you then have to split on semicolons (split(/\s*;\s*/) as they
say in Perl) to identify individual authors.
The more general issue here is: What details should be supported
explicitly by the database structure, and what should be handled by
post-processing when you read the database? My experience is that even
though relational databases are very powerful, Perl is also very powerful
and it is less cumbersome. It often saves both design time and database
input time if you simplify the database itself and compensate with more
intelligent Perl post-processing. (Note that you can also use Perl to
generate secondary parts of the database.)
>> My advice (short version) is to build up the database from its
>> mission-critical parts.
>Frankly, I'm still trying to figure out what the mission-critical
>parts are...
Right. My suggestion before was that the only essential
attributes for each game are the following:
Identifier
Title
Authors
Year
Genre
Rating
Comments
and in my example I hid the "Rating" attribute in the separate record for
the game review. (It could reasonably go in either place.) I thought
it over and I can suggest one more: a difficulty rating from 1 to 5.
IMO that is really the only essential attribute missing from the Guide.
All of the others that have been proposed are fun to discuss but no fun
to implement; they should be relegated to "Comments" or to the records
for individual files supporting a game.
--
/\ Greg Kuperberg (UC Davis)
Greg Kuperberg
Michael Straight <stra...@email.unc.edu> wrote:
>> What would that right place be? The example above suggests having a
>> directory called minster with a file called v1.z5, so that minster/v1.z5
>> would be Chistminster version 1, Inform platform. Perhaps the data file
>> that I mentioned could be minster.data in the same directory as minster.
>downloading.
No it wouldn't, because as I already said, the users do not have to see
the internal file structure of the database. On the Web, where I put my
files does not have to equal where you find them. It's a simple point,
but a subtle one.
What is true is that many browsers assume a filename from the tail of the
URL, and therefore this tail deserves some extra attention. A carefully
chosen URL for Christminster, version 1, might be one of these:
http://ifdb.org/games/v1/minster.z5
http://ifdb.org/games/minster.v1.z5
If he wanted to, Carl could honor either one or both of these URLs,
or twenty different ones including these two, regardless of the actual
names of his files.
Stephen Granade
> And then I'll have to worry about
> special cases like Neil de Mause. (The Guide currently deals with this
> by distorting him to "Neil deMause". I'd like to avoid altering people's
> names if possible.)
A minor correction in Neil's case: his last name is actually 'deMause'
rather than 'de Mause'.
Stephen
--
Stephen Granade | Interested in adventure games?
sgra...@phy.duke.edu | Visit Mining Co.'s IF Page
Duke University, Physics Dept | http://interactfiction.miningco.com
Stephen van Egmond
David Glasser <gla...@uscom.com> wrote:
>The IMDB's keywords are great. Not for cross-linking as much as just
>seeing about a movie. Screenit has frequently been mocked on ifMUD, on
>the other hand.
Sorry I missed it...
>Yay! I take it that bang.dhs.org is the new URL for your site?
Yes, the site has been renamed, nothing has moved. bang.ml.org/foo now
redirects to bang.dhs.org/foo. I love Apache. Here's to hoping that DHS
lasts longer.
/Steve
--
,,,
(. .)
+--ooO-(_)-Ooo------------ --- -- - - - -
| Stephen van Egmond http://bang.ml.org/
Jacob Solomon Weinstein
> Do these strike people as being particularly useful?
Actually, the IMDB keywords have been quite useful to me as an aspiring
screenwriter. There have been a number of times when I've wanted to see
if somebody has made a movie similar to one I'm about to write, in order
to learn from their mistakes and successes, and avoid unintentional
duplications.
Most of the objections I've seen so far to a specific feature of the
proposed database have been, "That can't be useful, because I personally
wouldn't find it useful." It seems to me that one of the advantages of
an electronic database with a sufficiently well designed front end is
that you can make of it what you want. If you think keywords would be
useless to you, or gender information would be offensive, than it ought
to be a simple matter to ask the database not to display that
information when you call up an entry.
So, if you say, "This information is inherently unworthy of being
included", you're setting yourself up for a not-very-useful debate. If
you must object to a feature, it is more useful to say, "Information A
is more worthy of being included than Information B, and both are more
worthy than Information C." Then the implementors of the database can
determine on their own, based on how much time they have to type the
damn thing in, whether to include (A) or (A + B) or (A+B+C). Of course,
it looks like Baf is willing to enter (A+B+C+...+Z). Remember, he
prefaced his lengthy list of planned inclusions by asking what was
LACKING, not what he ought to delete...
Best,
Jacob Weinstein
--
------------------------------------
To reply to me by e-mail, swap the last two letters of my e-mail
address.
J. Robinson Wheeler
> There are several ways to do this. The most reliable one is
> to store the names in inverted order:
>
> de Mause, Neil
>
> It is then trivial in Perl to convert to the more attractive First
> M. Last form for display purposes:
First M. Last is a troublesome annoyance to those of us who use
F. Middle Last. Let this be the one form I fill out that doesn't
force me to use First M. Last.
--
J. Robinson Wheeler
whe...@jump.net http://www.jump.net/~wheeler/jrw/home.html
Greg Kuperberg
J. Robinson Wheeler <whe...@jump.net> wrote:
>Greg Kuperberg wrote:
>> There are several ways to do this. The most reliable one is
>> to store the names in inverted order:
>>
>> de Mause, Neil
>>
>> It is then trivial in Perl to convert to the more attractive First
>First M. Last is a troublesome annoyance to those of us who use
>F. Middle Last.
I didn't mean it so literally; I just meant converting from last name
at the beginning to last name at the end.
If the issue is not the internal representation in the database but how
authors should enter their names in an on-line form, you can have an
input-processing Perl script assume that the last name is a single word.
The maintainer can intervene in the few cases in which it is not true.
As I explained before, if the assumption is corrected with heuristics
to identify surname prefixes, it can be >99% reliable.
David Glasser
> I've stated this elsewhere, but this thread seems to be ignoring it:
> The "honorific" field, if included in the database, would only be
> used in those few cases where author signs the work with an honorific.
> This is a rare condition; a quick grep of the Archive's master index
> revealed one Mr., one Cardinal, and one Admiral (surprisingly enough,
> not Jota.) There are probably a few more that aren't mentioned in the
> master index, and maybe some honorifics that I didn't think to look
> for. (Is "The" an honorific?)
Also Jr/Sr/III/whatever, I'd guess.
If this actually gets finished (not that I'm doubting, just that I'm not
jumping straight in), would you want to try to merge the ifHOS in
somehow? (The Interactive Fiction Hall of Shame is a bunch of pictures
of ifMUDders, though it may eventually become a people database or
something.)
--
David Glasser: gla...@uscom.com | raif FAQ: http://come.to/raiffaq/
"It's good to explore the G.U.E. caves / It's good to explore the G.U.E.
caves / You can count all the leaves / You can KILL TROLL WITH SWORD /
You'll get stuck but you won't be bored"-Joe.Mason, rec.arts.int-fiction
Magnus Olsson
Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>In article <7fkv7o$bmb$1...@nnrp1.dejanews.com>,
>Carl Muckenhoupt <ca...@wurb.com> wrote:
>>I agree that there doesn't seem to be any good reason to split up the
>>name to the degree in the proposal. But I do want to keep the last name
>>separate, because I want to be able to generate a list of authors,
>>sorted by last name. So if I don't keep the last name in its own field,
>>special cases like Neil de Mause.
>
>to store the names in inverted order:
>
>de Mause, Neil
This, of course, assumes that "de Mause" is to be alphabetized under "d".
But many European names of this form should be alphabetized under the
main word (just as American book titles starting with "The" aren't
usually filed under "t"). So if this were a music database, Beethoven
should be entered as
Beethoven, Ludwig van
rather than
van Beethoven, Ludwig
but this is a bit suboptimal, since his last name really was
"van Beethoven"; "van" is *not* his middle name.
--
Magnus Olsson (m...@df.lth.se, zeb...@pobox.com)
------ http://www.pobox.com/~zebulon ------
Greg Kuperberg
Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>>There are several ways to do this. The most reliable one is
>>to store the names in inverted order:
>>
>>de Mause, Neil
>
>This, of course, assumes that "de Mause" is to be alphabetized under "d".
>But many European names of this form should be alphabetized under the
>main word (just as American book titles starting with "The" aren't
>usually filed under "t"). So if this were a music database, Beethoven
>should be entered as
>
>Beethoven, Ludwig van
>
>rather than
>
>van Beethoven, Ludwig
>
>but this is a bit suboptimal, since his last name really was
>"van Beethoven"; "van" is *not* his middle name.
(a) This obscure distinction has been thoroughly mangled by the
authorities and it is not clear that there is any sense in remembering it.
Even the Grove Dictionary of Music lists Beethoven's name as "Beethoven,
Ludwig van". And certainly you'll never find Van Halen under "H"!
(b) If the users never see the inverted names that the database keeps
internally, it doesn't matter anyway.
Paul O'Brian
> But I would want to warn our hypothetical concerned
> parent. What's do be done?
How about including a generic "Warning" or "Parental Advisory" field, for
which authors could supply content if they so chose?
--
Paul O'Brian obr...@colorado.edu http://ucsu.colorado.edu/~obrian
"I don't find it fantastic or think it absurd
When the gun in the first act goes off in the third." -- Aimee Mann
Palfalvi Tamas
> Perhaps a simple AFGNCAAP [yes] [no] selection?
Really? And what is *that*?
Void
Magnus Olsson
Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>In article <7fmm8s$3sv$1...@bartlet.df.lth.se>,
>Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>>>There are several ways to do this. The most reliable one is
>>>to store the names in inverted order:
>>>
>>>de Mause, Neil
>>
>>This, of course, assumes that "de Mause" is to be alphabetized under "d".
>>But many European names of this form should be alphabetized under the
>>main word (just as American book titles starting with "The" aren't
>>usually filed under "t"). So if this were a music database, Beethoven
>>should be entered as
>>
>>Beethoven, Ludwig van
>>
>>rather than
>>
>>van Beethoven, Ludwig
>>
>>but this is a bit suboptimal, since his last name really was
>>"van Beethoven"; "van" is *not* his middle name.
>
>(a) This obscure distinction has been thoroughly mangled by the
>authorities and it is not clear that there is any sense in remembering it.
>Even the Grove Dictionary of Music lists Beethoven's name as "Beethoven,
>Ludwig van".
The Grove dictionary of Music is American, right?
> And certainly you'll never find Van Halen under "H"!
Non sequitur. There are different "Vans" in names. It could be a first
name, or a middle name that can be shortened to an initial, or a
capitalized part of a surname that should be alphabetized under V, or
a non-capitalized part of a surname that shouldn't.
Serious, I don't think it's an earth-shattering issue, as long as you
keep in mind that the IF community is multi-national and some damn
furriners with funny-looking names may take offence at having their
names mangled by American "standards".
>(b) If the users never see the inverted names that the database keeps
>internally, it doesn't matter anyway.
Of course. The database can mangle the names any way it likes as long as
1) People should be able to find what they're look for
and
2) Names (and other info) shouldn't have to be output (or even worse
input) in a mangled form just because it's mangled inside the
database.
Greg Kuperberg
Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>In article <7fn9gs$i0n$1...@manifold.math.ucdavis.edu>,
>Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>>>but this is a bit suboptimal, since his last name really was
>>>"van Beethoven"; "van" is *not* his middle name.
>>
>>(a) This obscure distinction has been thoroughly mangled by the
>>authorities and it is not clear that there is any sense in remembering it.
>>Even the Grove Dictionary of Music lists Beethoven's name as "Beethoven,
>>Ludwig van".
>The Grove dictionary of Music is American, right?
"Users of the dictionary are asked to write to the editorial office, c/o
Macmillan, Little Essex Street, London WC2R 3LF, to notify us of errors
or omissions so that they may be recorded and set right in the future.
- Stanley Sadie, London, 1979"
"In its authorship and its editing, this dictionary is as much American
as British. But its traditions are primarily British, and it accordingly
preserves British terminological usage."
>>[Van Halen example]
>Non sequitur. There are different "Vans" in names. It could be a first
>name, or a middle name that can be shortened to an initial, or a
>capitalized part of a surname that should be alphabetized under V, or
>a non-capitalized part of a surname that shouldn't.
The section of the Grove Dictionary on alphabetization explicitly explains
that it does use capitalized prefixes to determine alphabetical order,
which implies that there is no consensus on this point.
>Serious, I don't think it's an earth-shattering issue, as long as you
>keep in mind that the IF community is multi-national and some damn
>furriners with funny-looking names may take offence at having their
>names mangled by American "standards".
If you want to carp about American leadership, it would be hard to find
a tinier or less accurate nit to pick than this. Maybe Kosovo is a more
important case?
Magnus Olsson
Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>In article <7fnehr$2il$1...@bartlet.df.lth.se>,
>Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>>In article <7fn9gs$i0n$1...@manifold.math.ucdavis.edu>,
>>Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>>>>but this is a bit suboptimal, since his last name really was
>>>>"van Beethoven"; "van" is *not* his middle name.
>>>
>>>(a) This obscure distinction has been thoroughly mangled by the
>>>authorities and it is not clear that there is any sense in remembering it.
>>>Even the Grove Dictionary of Music lists Beethoven's name as "Beethoven,
>>>Ludwig van".
>>The Grove dictionary of Music is American, right?
(...)
>"In its authorship and its editing, this dictionary is as much American
>as British. But its traditions are primarily British, and it accordingly
>preserves British terminological usage."
OK. My point remains, though, since van Beethoven is a German, not an
English name, while the dictionary is written in English by English-
speaking people.
Of course, a German dictionary will have similar problems with
non-German names. *All* dictionaries have problems with names that are
spelled with diacritics that don't exist in the dictionary's language
- even if the editors can find a font with the correct diacritics
(which they usually can, nowadays), they'll still have to use the
collation order of the target language.
This is really not a problem specific to American dictionaries, or
libraries, or databases. It's not a problem specific to English, or
German, or Swedish, or Chinese dictionaries or libraries or databases
either.
It's a problem that occurs as soon as you want to put things like
names (which vary a lot from culture to culture) int a databse whose
design is based on how names (in this case) work in one culture.
>>>[Van Halen example]
>>Non sequitur. There are different "Vans" in names. It could be a first
>>name, or a middle name that can be shortened to an initial, or a
>>capitalized part of a surname that should be alphabetized under V, or
>>a non-capitalized part of a surname that shouldn't.
>
>The section of the Grove Dictionary on alphabetization explicitly explains
>that it does use capitalized prefixes to determine alphabetical
>order,
That's a common convention. And you have to have conventions. My point
is really that you should be careful when selecting those
conventions. For example, you can't assume that if a name breaks up
into three space-spearated parts, then those parts are first, middle
and last name, respectively.
>>Serious, I don't think it's an earth-shattering issue, as long as you
>>keep in mind that the IF community is multi-national and some damn
>>furriners with funny-looking names may take offence at having their
>>names mangled by American "standards".
>
>If you want to carp about American leadership, it would be hard to find
>a tinier or less accurate nit to pick than this.
For crying out loud, I do *not* want to "carp about American
leadership". I want to carp about one particular individual (you), who
happen to be an American, who seems to believe that people of
different ethnicities won't object to having their names mangled to
conform to *your* standards. That you're American is totally
irrelevant. And being insensitive and patronizing is not my definition
of "leadership".
Doeadeer3
>From: ad...@princeton.edu (Adam J. Thornton)
>Date: 4/20/99 12:01 PM Pacific Daylight Time
>>Doe :-) I also would not want an "honorific" before my name.
>
>Not even Luminary Doe?
>
>Adam
It's been a long time and I am basically a pacifist, but...
>thwap Adam<
Doe :-) Hehehe.
Doe doea...@aol.com (formerly known as FemaleDeer)
****************************************************************************
"In all matters of opinion, our adversaries are insane." Mark Twain
Carl Muckenhoupt
Stephen Granade <sgra...@lepton.phy.duke.edu> wrote:
> A minor correction in Neil's case: his last name is actually 'deMause'
> rather than 'de Mause'.
Whoops. My apologies, Neil. Still, you see my point.
And as far as I can tell, storing names internally as
"surname, rest of name" is an adequate solution.
Also, since it's been brought up: Alphabetization of
titles. I've been assuming that other countries also
conventially ignore leading articles. Thus, for
example, the game "Die Karawane der siebten Dynastie"
is filed under K. If this assumption is in error,
someone should tell me so.
Carl Muckenhoupt
gla...@uscom.com (David Glasser) wrote:
> If this actually gets finished (not that I'm doubting, just that I'm not
> jumping straight in), would you want to try to merge the ifHOS in
> somehow? (The Interactive Fiction Hall of Shame is a bunch of pictures
> of ifMUDders, though it may eventually become a people database or
> something.)
I'd be happy to provide links to the ifHOS entries in the author
details, if such a thing is desired by whoever's in charge of the
ifHOS.
Does the ifHOS in fact exist right now? All the links to it that
I've found seem to be broken.
Carl Muckenhoupt
jac...@alumni.princeton.eud (Jacob Solomon Weinstein) wrote:
> Of course,
> it looks like Baf is willing to enter (A+B+C+...+Z). Remember, he
> prefaced his lengthy list of planned inclusions by asking what was
> LACKING, not what he ought to delete...
Mind you, that doesn't mean that everything that anyone suggests
is going to be included. I'm treating this as a brainstorming
session. The idea behind brainstorming is that you collect as many
ideas as possible, then pare the list down afterwards.
Carl Muckenhoupt
It's a term used in "Zork: Grand Inquisitor". It stands for
"Ageless, Faceless, Gender-Neutral, Culturally-Ambiguous
Adventure Person".
But you can get pretty close to being AFGNCAAP without
fulfilling that description completely. I can think of a
few games in which the main character is referred to using
male pronouns, but is otherwise left blank.
Magnus Olsson
Carl Muckenhoupt <ca...@wurb.com> wrote:
>In article <jdk8v5a...@lepton.phy.duke.edu>,
> Stephen Granade <sgra...@lepton.phy.duke.edu> wrote:
>> A minor correction in Neil's case: his last name is actually 'deMause'
>> rather than 'de Mause'.
>
>Whoops. My apologies, Neil. Still, you see my point.
>And as far as I can tell, storing names internally as
>"surname, rest of name" is an adequate solution.
Since my recent replies to Greg's articles may perhaps give a
different impression, let me say that I think so, too, with the
qualification that for some names, what is actually part of the
surname be put in the "rest of name" field (as in
"Beethoven, Ludwig van").
Doeadeer3
>From: anson@DELETE_THISpobox.com (Anson Turner)
>Date: 4/20/99 5:16 PM Pacific Daylight Time
>doea...@aol.com (Doeadeer3) wrote:
>
>:The way Amazon.com describes a book might be a good way to go.
>
>Hmm.
>
>Title: Everybody Loves a Parade
>Author: Ms. Cody Sandifer
>Year: 1997
>Genres: Parades; Love; Rocks
>Comments: I like a smock.
>
>People who downloaded Everybody Loves a Parade also downloaded: Acid
>Whiplash, A Change in the Weather, the TextFire 12-Pack, and Pick Up the
>Phone Booth and Die.
>
>Player reviews:
>
>Rating: *
>This game sux! Not only was it writen by a girl, but the main caracter is
>a girl too! But you dont find that out until later, accept now Ive told
>you, so nah nah! Plus, its got hell in it. Thats bad! -- Stiffy Makane
>
I guess you are saying the database SHOULD include gender information, so
negative reactions/reviews directed at games "written by girls" can be directed
at only those games ACTUALLY written by women (i.e. those of us with female
gentalia.)
I guess that is what you are saying.
Reminds me of a comment I got from someone once about an action I had in The
Family Legacy. He told me the game was obviously written by a woman because a
male wouldn't perform that action and a male author wouldn't have thought to
include it.
I found that a highly irrelvant comment because as a game player I am always
performing actions I would not in real life.
Fold sheet was the action (one didn't HAVE to fold the sheet). I guess that
means no males ever do their own laundry or housework (if the writer of that
comment was to be believed).
Hmmmm.
Sometimes things like that make me think the author's gender should NOT be
specified in the database. One can tell by the author's name or not.
BTW - Anson I used the examples I did because Greg (or whomever) mentioned the
protagonist of Christminister was female. If the protagonist of Chrisminister
was to be specified then ALL of the protagonists should be specified. However,
that information COULD be skipped in the comment fields and left to reviews.
Doe :-)
Doe :-)
David Glasser
> In article <1dql3cf.mel...@usol-209-186-16-97.uscom.com>,
> gla...@uscom.com (David Glasser) wrote:
> > If this actually gets finished (not that I'm doubting, just that I'm not
> > jumping straight in), would you want to try to merge the ifHOS in
> > somehow? (The Interactive Fiction Hall of Shame is a bunch of pictures
> > of ifMUDders, though it may eventually become a people database or
> > something.)
>
> I'd be happy to provide links to the ifHOS entries in the author
> details, if such a thing is desired by whoever's in charge of the
> ifHOS.
>
> Does the ifHOS in fact exist right now? All the links to it that
> I've found seem to be broken.
It's currently at http://www.uscom.com/~glasser/ifmud/ifhos/
Many pictures there were lost in the crash, so some links on it are
defunct.
--
David Glasser: gla...@uscom.com | http://www.uscom.com/~glasser/
DGlasser@ifMUD:orange.res.cmu.edu 4001 | raif FAQ http://come.to/raiffaq
"Also, if/when is David Glasser v2 coming out, and will it support
HTML-TADS? Version 1 is pretty buggy." --Steven Marsh on raif
Adam J. Thornton
Doeadeer3 <doea...@aol.com> wrote:
>>thwap Adam<
Ahhhhh.
Adam
--
ad...@princeton.edu
"There's a border to somewhere waiting, and a tank full of time." - J. Steinman
cody sandifer
doea...@aol.com (Doeadeer3) wrote:
> I guess you are saying the database SHOULD include gender information, so
> negative reactions/reviews directed at games "written by girls" can be
directed
> at only those games ACTUALLY written by women (i.e. those of us with female
> gentalia.)
You go girl!
cody
Knight37
Magnus Olsson <m...@bartlet.df.lth.se> wrote:
> This is really not a problem specific to American dictionaries, or
> libraries, or databases. It's not a problem specific to English, or
> German, or Swedish, or Chinese dictionaries or libraries or databases
> either.
>
> It's a problem that occurs as soon as you want to put things like
> names (which vary a lot from culture to culture) int a databse whose
> design is based on how names (in this case) work in one culture.
Yeah?
Well what durn language is this web site going to be in, anyway? :)
I loved yer "damn ferners" bit, tho ya did misspelt ferner.
You'll have to excuse my apparent total lack of Politically
Correctness... you see, I was born without any. Oh, and I have been
reading a lot of the "DeadLands" RPG lately. Stuff is startin' ta
grow on me, dagg nabbit.
knight37
Joe Pfeiffer
> Carl Muckenhoupt <ca...@wurb.com> wrote:
> >In article <jdk8v5a...@lepton.phy.duke.edu>,
> > Stephen Granade <sgra...@lepton.phy.duke.edu> wrote:
> >> A minor correction in Neil's case: his last name is actually 'deMause'
> >> rather than 'de Mause'.
> >
> >Whoops. My apologies, Neil. Still, you see my point.
> >And as far as I can tell, storing names internally as
> >"surname, rest of name" is an adequate solution.
Surname, RestOfName is eurocentric. Not all cultures make the family
name be the surname. Not all cultures have a tradition of maintaining
a family name across generations. A strawman suggestion would go
something like
Social title (Mr, Ms, etc) optional
Name[] array of names in author-defined order
Honorific (Esq, MS, etc) optional
Sort order (order of name fields to be used in a sorted list of authors)
Normal order (order of names to be used in normal contexts)
So (if I were an author), an entry for me would be
Social title: Dr.
Name[0]: Joseph
Name[1]: J.
Name[2]: Pfeiffer
Name[3]: \0
Honorific: Jr,, Ph.D.
Sort order: 2, 0, 1
Normal order: 0, 1, 2
When a name is displayed, it is always social title first (if
appropriate), then name fields, then comma, then honorific (if
appropriate). If the name is displayed in normal order, no additional
characters are added. If it is displayed in sort order, and the sort
order is not the same as the normal order, a comma follows the first
displayed name.
In normal order, mine would appear as Joseph J. Pfeiffer, Jr., Ph.D.
In sorted order, it would be Pfeiffer, Joseph J., Jr., Ph.D.
In a mention, it would Dr. Pfeiffer (notice that you don't use social
titles and honorifics at the same time -- though lots of people do).
A Chinese author named Li Peng would appear as Li Peng in both sorted
and normal order.
Can you tell I'm tired of seeing my name appear on mailling labels as
``Pfeiffer Jr., Joseph J.?''
--
Joseph J. Pfeiffer, Jr., Ph.D. Phone -- (505) 646-1605
Department of Computer Science FAX -- (505) 646-1002
New Mexico State University http://www.cs.nmsu.edu/~pfeiffer
Magnus Olsson
Knight37 <knig...@flash.SPAMBLOCKA.net> wrote:
>
>Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>
>> This is really not a problem specific to American dictionaries, or
>> libraries, or databases. It's not a problem specific to English, or
>> German, or Swedish, or Chinese dictionaries or libraries or databases
>> either.
>>
>> It's a problem that occurs as soon as you want to put things like
>> names (which vary a lot from culture to culture) int a databse whose
>> design is based on how names (in this case) work in one culture.
>
>Yeah?
>
>Well what durn language is this web site going to be in, anyway? :)
English, I suppose.
And that means, presumably, that it will use a character set such as
ISO Latin-1 (so that you can't enter names in, say, Cyrillic letters),
and English alphabetization. I don't have a problem with that.
WHat I'm asking for is that the people who'll be implementing the
databse at least *try* to make it flexible enough to handle personal
names from other languages, without trying to squeeze them into
something like a "First name, middle initial, last name" mold (not
that the actual proposal is that rigid, far from it, this was an
example).
Greg Kuperberg
Magnus Olsson <m...@bartlet.df.lth.se> wrote:
>WHat I'm asking for is that the people who'll be implementing the
>databse at least *try* to make it flexible enough to handle personal
>names from other languages, without trying to squeeze them into
>something like a "First name, middle initial, last name" mold (not
>that the actual proposal is that rigid, far from it, this was an
>example).
The actual proposal as I understand it is to invert the names strictly
internally and strictly for alphabetization purposes.
FWIW, the xxx archive system (http://xxx.lanl.gov/), and consequently my
front end, allow TeX symbols for denoting accents and non-Latin letters.
The users themselves started it; almost everyone using that archive writes
in TeX anyway. For example, Poincare becomes Poincar\'e. This has the
major advantage that the raw database can be in restricted 7-bit ASCII.
It has the minor disadvantages that accents must be translated to HTML,
which is done whenever possible, and that they must be stripped for
indexing purposes.
However, the IF archive is substantially less international than physics
and mathematics research. This is very unfortunate for the IF archive
in my opinion, but it means that many cultural variations can be handled
manually as special cases.
Carl Muckenhoupt
m...@bartlet.df.lth.se (Magnus Olsson) wrote:
> And that means, presumably, that it will use a character set such as
> ISO Latin-1 (so that you can't enter names in, say, Cyrillic letters),
> and English alphabetization. I don't have a problem with that.
Indeed, even if it was trivial to enter names in their native
alphabet, I'd tend to discourage it. After all, when was the last
time you saw an alphabetized list in which some entries are /
> WHat I'm asking for is that the people who'll be implementing the
> databse at least *try* to make it flexible enough to handle personal
> names from other languages, without trying to squeeze them into
> something like a "First name, middle initial, last name" mold (not
> that the actual proposal is that rigid, far from it, this was an
> example).
Definitely. Whatever system we use, Li Peng goes under L. (And
Elizabeth Regina would go under E, I suppose, if it turns out that
she wrote some text adventures.)
Branko Collin
wrote:
>In article <7fng51$i7v$1...@manifold.math.ucdavis.edu>,
>Greg Kuperberg <gr...@manifold.math.ucdavis.edu> wrote:
>>Magnus Olsson <m...@bartlet.df.lth.se> wrote:
.
>>>The Grove dictionary of Music is American, right?
>>
>>as British. But its traditions are primarily British, and it accordingly
>>preserves British terminological usage."
>
>OK. My point remains, though, since van Beethoven is a German, not an
>English name, while the dictionary is written in English by English-
>speaking people.
The name is Dutch/Flemish. Van Beethoven moved to Germany, but was
originally from Belgium (IIRC). Hence, the Dutch/Flemish surname.
Actually, German-language people have a similar preposition for
surnames, namely "von".
The rule for Dutch is fairly simple, but may be confusing. If the
surname is preceded by one or more of the first names, "van" is
written without a first capital. In any other case, "van" is
capitalized.
So:
I met Van Halen yesterday.
I met Alex van Halen yesterday.
("van", by the way, means "of", "from".
--
branko
-- loving me ... is easy cuz i'm beautiful
(shalalalala)
Doeadeer3
>From: anson@DELETE_THISpobox.com (Anson Turner)
>"Doe, how would you define yourself as a woman?"
>"I have female genitalia."
>
>Did you really mean to suggest this? Anyway, when I use the term "gender"
>outside of a grammatical context I mean psychological, not necessarily
>biological, sex.
Actually a biological definition IS the definition of gender, as any transexual
will tell you. ;-)
>It's hard to imagine what significance one's genitalia
>has to any IF they write. (If it does, I doubt that's a game I want to
>play.)
>
Agreed.
>I had in The
>:Family Legacy. He told me the game was obviously written by a woman because
>a
>:male wouldn't perform that action and a male author wouldn't have thought to
>:include it.
>
>I'm sorry that I reminded you of that. Here's an idea: Shoot him. See if
>he says "I know that a man told you to do that because a woman wouldn't
>have thought to perform that action."*
Okay. Good idea.
And I like the fact that I can say a male thought of it.
Doe :-) Hehe.
Remember my tongue is usually in cheek, but you seem to have caught that fact.
David Glasser
> anson@DELETE_THISpobox.com (Anson Turner) :
> >doea...@aol.com (Doeadeer3) wrote:
> >
> >:The way Amazon.com describes a book might be a good way to go.
> >
> >Hmm.
> >
> >Title: Everybody Loves a Parade
> >Author: Ms. Cody Sandifer
> >Year: 1997
> >Genres: Parades; Love; Rocks
> >Comments: I like a smock.
[snip]
> I guess you are saying the database SHOULD include gender information, so
> negative reactions/reviews directed at games "written by girls" can be
> directed at only those games ACTUALLY written by women (i.e. those of us
> with female gentalia.)
>
> I guess that is what you are saying.
Me, I just thought he was making fun of Amazon.com.
--
David Glasser: gla...@uscom.com | http://www.uscom.com/~glasser/
DGlasser@ifMUD:orange.res.cmu.edu 4001 | raif FAQ http://come.to/raiffaq
"Well, it's interesting. We're willing to kill in the name of David
Glasser." --Steven Marsh on rec.arts.int-fiction
Doeadeer3
>Date: 4/23/99 3:22 PM Pacific Daylight Time
>Doeadeer3 <doea...@aol.com> wrote:
>
>> anson@DELETE_THISpobox.com (Anson Turner) :
>> >doea...@aol.com (Doeadeer3) wrote:
>> >
>> >:The way Amazon.com describes a book might be a good way to go.
>> >
>> >Hmm.
>> >
>> >Title: Everybody Loves a Parade
>> >Author: Ms. Cody Sandifer
>> >Year: 1997
>> >Genres: Parades; Love; Rocks
>> >Comments: I like a smock.
>[snip]
>> I guess you are saying the database SHOULD include gender information, so
>> negative reactions/reviews directed at games "written by girls" can be
>> directed at only those games ACTUALLY written by women (i.e. those of us
>> with female gentalia.)
>>
>> I guess that is what you are saying.
>
>Me, I just thought he was making fun of Amazon.com.
Well that too. As Anson said his post had many deep layers of meaning. ;-)
But he also specifically did Cody, who Lucian had pointed out earlier someone
once mistakenly referred to in a review as being a female.
Doe :-)
Go cody go!
Branko Collin
Kuperberg) wrote:
[Greg's simpler proposal for a database]
>ID: minster
>Title: Christminster
>Author: Gareth Rees
>Year: 1995
>Genres: College; Mystery
>Comments: Female protagonist
Why would you choose 'genre' as a special (searchable?) field? It
seems just as arbitrary as, say, 'language', with the added
disadvantage that some authors undoubtedly do not want their game
labelled to belong to some genre.
Would it perhaps be easier to let the game author choose a number of
keywords, or to have a review record in which the reviewer can select
the keywords?
keyword1: Dutch
keyword2: dragon
keyword3: female
keyword4:
keyword5:
Of course, this has its shortcomings. For instance, what does 'female'
stand for here? The protagonist? The author? The intended audience?
And what if the person searching the database is looking for 'woman'
instead?
The latter could be partly solved by offering the author/reviewer a
list of recommended keywords and the person visiting the database a
list of searchable keywords.
Greg Kuperberg
Branko Collin <col...@xs4all.nl> wrote:
>On 20 Apr 1999 12:37:13 -0700, gr...@manifold.math.ucdavis.edu (Greg
>Kuperberg) wrote:
>[Greg's simpler proposal for a database]
>>ID: minster
>>Title: Christminster
>>Author: Gareth Rees
>>Year: 1995
>>Genres: College; Mystery
>>Comments: Female protagonist
>
>Why would you choose 'genre' as a special (searchable?) field?
First, all of the fields should be searchable. The question is which
information should be *browsable*, meaning for which fields X should
there be a table in the database "games organized by X".
In my opinion, it is extremely helpful for the games to be browsable by
genre, for same reason that many bookstores organize fiction by genre.
Of course, unlike at a bookstore, if you don't like to browse by genre
on-line you can browse by some other attribute instead.
>It seems just as arbitrary as, say, 'language', with the added
>disadvantage that some authors undoubtedly do not want their game
>labelled to belong to some genre.
In fact language is also a crucial attribute that would merit a separate
field if there were more non-English games. It's too bad that there
are so few; as a start, someone should work on porting Inform to other
languages. Baf's Guide currently combines language with genre, which
is a reasonable solution for now.
Actually an author could argue against any of the fields; they are all
a priori arbitrary. But that's irrelevant, because the database itself
is not the exclusive property of the authors, even though the games that
it lists are. The usual way to compromise with those who do not like
genres is to first, listen to the community about what genres should
exist, and second, include a generic genre like "None" or "Innovative".
>Would it perhaps be easier to let the game author choose a number of
>keywords, or to have a review record in which the reviewer can select
>the keywords?
No, because there is no way for hundreds of people to independently
choose consistent keywords, even if they wanted to. For example, Usenet
postings have a keyword field, and that field is a failure.
To do keywords well you need editors to provide consistency. If Carl is
willing to do that, then keywords could also be useful. But even then,
they would not make the genre superfluous.
Moreover, simply indexing all of the words in the database record of
the game plus a short review is a cheap way to get keywords. It's not
perfect, but it's not bad. One question is whether the database
should allow author-contributed synopses. These are almost as a good
a source for keywords as mini-reviews.
Dave Gatewood
> Right. My suggestion before was that the only essential
> attributes for each game are the following:
>
> Identifier
> Title
> Authors
> Year
> Genre
> Rating
> Comments
Game Length: There is a tremendous qualitative difference between
"Curses" and "Bear's Night Out" and "Silence of the Lambs". In choosing
a game to play, this would always be the number one criterion I would
search by. Lengths would preferably be chosen off a short list (less
than 10 min, 10-30 min, 30 min - 2 hours, 2-5 hours, more than 5
hours). Yes, this is subjective, but also very necessary.
A couple of "genres" that get discussed periodically: Children's Games
and Educational Tools. Not many of either exist yet, but rgif posts
suggest there are regular searches for these two types.
Carl Muckenhoupt
gr...@manifold.math.ucdavis.edu (Greg Kuperberg) wrote:
> In my opinion, it is extremely helpful for the games to be browsable by
> genre, for same reason that many bookstores organize fiction by genre.
> Of course, unlike at a bookstore, if you don't like to browse by genre
> on-line you can browse by some other attribute instead.
Also, bookstores usually don't put a book under more than one
genre. We're under no such limitation.
> In fact language is also a crucial attribute that would merit a separate
> field if there were more non-English games. It's too bad that there
> are so few; as a start, someone should work on porting Inform to other
> languages. Baf's Guide currently combines language with genre, which
> is a reasonable solution for now.
Actually, Baf's Guide currently has a separate field for Language, with
its own browsable list. Unfortunately, this isn't very useful, because
I only review games that I've played, and I generally don't even
download games that aren't in English. The only non-English games
listed in the Guide are translations of English games, and those
that were in if-archive/games/pc when I initially started the Guide.
This is one of the reasons I want a comprehensive database.
I think it's a good idea to keep the Language field separate, even
if the non-English games form a small minority of the games in the
Archive. It is, after all, a conceptually distinct property of a game.
Besides, for all we know there may be hundreds of text adventures in
Russian that none of us have heard of, but which will all be uploaded
to the Archive one day...
Branko Collin
wrote:
>In article <7fsm6i$r0i$1...@manifold.math.ucdavis.edu>,
> gr...@manifold.math.ucdavis.edu (Greg Kuperberg) wrote:
>> In fact language is also a crucial attribute that would merit a separate
>> field if there were more non-English games. It's too bad that there
>> are so few; as a start, someone should work on porting Inform to other
>> languages. Baf's Guide currently combines language with genre, which
>> is a reasonable solution for now.
>
[...]
>I think it's a good idea to keep the Language field separate, even
>if the non-English games form a small minority of the games in the
>Archive. It is, after all, a conceptually distinct property of a game.
>Besides, for all we know there may be hundreds of text adventures in
>Russian that none of us have heard of, but which will all be uploaded
>to the Archive one day...
I am working on a database listing all known Dutch text adventures.
This thing is going to be very free form. I estimate that there have
never been more than 200 text adventures published in Dutch, so that
it still would be easy to maintain the database by hand.
The reason for making such a database available, is because at some
point in the future I want to help develop a Dutch library for Inform.
Knowing what adventures were published in Dutch and where to find them
or their authors will probably help to settle on a standard language
to use in Dutch text adventures. (For instance, would you use 'laag',
'omlaag', 'beneden', 'daal' or something else for 'down'? And what
would you use for the abbreviation?)
I have set up most of the homepage of this database, I think I will
announce the address in a week or so in both r.a.i-f and r.g.i-f.