Inform: new build (5Z71), new website, new blog

[Graham Nelson]

Inform now has a larger and more resourceful website, at a new URL:

http://inform7.com

We will shortly redirect inform-fiction.org to this, but for a few
days the two sites will coexist. Among the new material are a
screencast for new users (by Aaron Reed), a news blog and a new
section of resources for education users. The site is now running off
a bespoke content management system, built for us on top of Django by
Liza Daly of Threepress Consulting, and with the aim of making it much
easier to expand and maintain what is now quite a large website. The
downloads area now maintains an archive of all builds for all
platforms, back to 4S08 (the first really stable build, two years
ago), and it handles a wider range of platforms and architectures: Mac
OS X, Windows, Debian, Fedora, Arch, Ubuntu, XO, Solaris, and so
forth. The new extensions area (organised by Christopher Armstrong)
now includes documentation and the source text for online browsing, so
that it's possible to see exactly what an extension is and does before
downloading it. Two more of the underlying Inform programs have been
published, moved to an open-source licence (the Artistic License 2.0),
and presented in literate source; another purpose of the new site is
to provide a framework to hold the Inform source code as we publish
it, which remains an ongoing process.

Build 5Z71, the first new release in six months, is also out today and
available from the site. Miscellaneous benefits come out of a
substantial internal reform in logic and the handling of kinds of
value. Many of those lift niggling restrictions - see the full change
log for details. Build 5Z71 uses around a third as much memory as
5U92, does much less file-handling, is a little faster, and can handle
projects unlimited in size. Dimensional checking became much more
flexible, allowing Inform authors to model the world with quantitative
physics, and a new built-in extension, "Metric Units by Graham
Nelson", defines all the kinds of value needed. Equations, imitating
those in books and papers, were added to the language. The release
mechanism was also rewritten, allowing much better CSS-friendly
websites to be produced by the Release button. The Recipe Book was
expanded and rewritten in its coverage of actions and commands. Lastly
but most visibly, the Index was redesigned and the Problem messages
relaid to incorporate links to relevant documentation. All 132 issues
arising from bug reports since build 5U92 were closed out.

We are also publishing a new design document, responding to many
suggestions for improvements in Inform, along the lines of the January
2007 document: see http://inform7.com/learn/papers/

This release has been something of an epic business - we began the
preparations before Christmas, but it has now become the traditional
"birthday" release (though Inform's sixteenth anniversary is strictly
speaking not until next week). Many, many thanks to all of those who
helped.

[Dannii]

> 2007 document: seehttp://inform7.com/learn/papers/

>
> This release has been something of an epic business - we began the
> preparations before Christmas, but it has now become the traditional
> "birthday" release (though Inform's sixteenth anniversary is strictly
> speaking not until next week). Many, many thanks to all of those who
> helped.

Good to see more of I7 being open sourced :)

Would be good if the source code could actually be browsed online,
rather than through PDFs.

[S. John Ross]

Many thankings!

[peter...@hotmail.com]

Great work! However, I installed it and tried to compile my WIP with
it, and got a lot of errors... which point to Standard Rules and such.

In Part SR3 - Activities, Section SR2/13 - Locale descriptions -
Unindexed:

Problem. You wrote 'otherwise change the locale description priority
entry to N' : but when I expanded that command using its inline
definition, I ran into an expansion written in braces { and } which
was not one of the names of tokens from its specification, so I don't
know how to compile this. What you've defined here is not a phrase
which will compile to an I6 routine, but rather a phrase which will
compile 'inline' directly into the body of other routines: this
happened because it was a 1-command definition, and that command was
an I6 insertion. The ability to write inline phrases is really
intended only for the Standard Rules and other low-level system
extensions, and it is (intentionally) limited in its flexibility. A
good rule of thumb is: if you can define a phrase without using I6
insertions, always do so.

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

Problem. You wrote 'change the locale description priority entry to
N' : again, when I expanded that command using its inline definition,
I ran into an expansion written in braces { and } which was not one of
the names of tokens from its specification.

In Section 1:

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

Problem. You wrote 'change used entry to turn count' : again, when I
expanded that command using its inline definition, I ran into an
expansion written in braces { and } which was not one of the names of
tokens from its specification.

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

Problem. You wrote 'change used entry to turn count' : again, when I
expanded that command using its inline definition, I ran into an
expansion written in braces { and } which was not one of the names of
tokens from its specification.

In Volume 2 - Locks and Keys, Book 3 - The Bolting Action, Part 2 -
Rules for Bolting and Unbolting, Chapter 1 - Unbolting Rules:

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

Problem. I attempted to compile 'Before unbolting something lockable
when the noun provides the property matching key and the known-key of
the noun is not no-key and the known-key of the noun is touchable
( this is the convert unbolt to unlock with rule )' using its inline
definition, but this contained the invalid annotation 'aorp'.

In Part SR3 - Activities, Section SR2/13 - Locale descriptions -
Unindexed:

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

Problem. I attempted to compile 'if the item provides the property
initial appearance and the item is not handled begin' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Part SR4 - Actions, Section SR4/2 - Actions concerning the actor's
possessions:

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

Problem. I attempted to compile 'if H provides the property carrying
capacity begin' using its inline definition, but this contained the
invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the second noun provides the
property carrying capacity begin' using its inline definition, but
this contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the second noun provides the
property carrying capacity begin' using its inline definition, but
this contained the invalid annotation 'aorp'.

In Volume 2 - Locks and Keys, Book 3 - The Bolting Action, Part 2 -
Rules for Bolting and Unbolting, Chapter 3 - Rules for Using a Key on
Something that Doesn't Need One:

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

Problem. I attempted to compile 'Check locking something lockable with
something when the noun provides the property matching key and the
matching key of the noun is no-key ( this is the object doesn't need a
key to lock rule )' using its inline definition, but this contained
the invalid annotation 'aorp'.

In Part SR4 - Actions, Section SR4/5 - Actions which change the state
of things:

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

Problem. I attempted to compile 'if the noun provides the property
lockable and the noun is lockable, continue the action' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Volume 2 - Locks and Keys, Book 4 - The Keyring Kind:

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

Problem. I attempted to compile 'if ( the holder of the second noun is
not the actor and the holder of the second noun is not a keyring
carried by the actor ) or the noun does not provide the property
matching key or the matching key of the noun is not the second noun,
stop the action with library message locking it with action number 4
for the second noun' using its inline definition, but this contained
the invalid annotation 'aorp'.

In Book 3 - The Bolting Action, Part 2 - Rules for Bolting and
Unbolting, Chapter 3 - Rules for Using a Key on Something that Doesn't
Need One:

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

Problem. I attempted to compile 'Check unlocking something lockable
with something when the noun provides the property matching key and
the matching key of the noun is no-key ( this is the object doesn't
need a key to unlock rule )' using its inline definition, but this
contained the invalid annotation 'aorp'.

In Part SR4 - Actions, Section SR4/5 - Actions which change the state
of things:

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

Problem. I attempted to compile 'if the noun provides the property
lockable and the noun is lockable, continue the action' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Volume 2 - Locks and Keys, Book 4 - The Keyring Kind:

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

Problem. I attempted to compile 'if ( the holder of the second noun is
not the actor and the holder of the second noun is not a keyring
carried by the actor ) or the noun does not provide the property
matching key or the matching key of the noun is not the second noun,
stop the action with library message unlocking it with action number 3
for the second noun' using its inline definition, but this contained
the invalid annotation 'aorp'.

In Part SR4 - Actions, Section SR4/5 - Actions which change the state
of things:

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

Problem. I attempted to compile 'if the noun provides the property
switched on, continue the action' using its inline definition, but
this contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the noun provides the property
switched on, continue the action' using its inline definition, but
this contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the noun provides the property
openable and the noun is openable, continue the action' using its
inline definition, but this contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the noun provides the property
lockable and the noun is locked, stop the action with library message
opening action number 2 for the noun' using its inline definition,
but this contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the noun provides the property
openable and the noun is openable, continue the action' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Volume 2 - Locks and Keys, Book 3 - The Bolting Action, Part 2 -
Rules for Bolting and Unbolting, Chapter 2 - Bolting Rules:

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

Problem. I attempted to compile 'Precondition for bolting an open
lockable thing when the noun provides the property matching key and
the matching key of the noun is no-key' using its inline definition,
but this contained the invalid annotation 'aorp'.

In Volume 1 - Implicit Action Framework, Book 3 - Precondition Rules
Using Implicit Actions:

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

Problem. I attempted to compile 'Precondition for opening something
locked when the noun is lockable and the noun provides the property
matching key and the matching key of the noun is the known-key of the
noun ( this is the unlock before opening rule )' using its inline
definition, but this contained the invalid annotation 'aorp'.

In Volume 2 - Locks and Keys, Book 3 - The Bolting Action, Part 2 -
Rules for Bolting and Unbolting, Chapter 2 - Bolting Rules:

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

Problem. I attempted to compile 'Check an actor bolting something
lockable when the noun provides the property matching key and the
matching key of the noun is not no-key ( this is the can't bolt
without key rule )' using its inline definition, but this contained
the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'Check an actor bolting something
lockable when the noun does not provide the property matching key
( this is the can't bolt without matching key rule )' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Chapter 1 - Unbolting Rules:

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

Problem. I attempted to compile 'Check an actor unbolting something
lockable when the noun provides the property matching key and the
matching key of the noun is not no-key ( this is the can't unbolt
without a key rule )' using its inline definition, but this
contained the invalid annotation 'aorp'.

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

Problem. I attempted to compile 'Check an actor unbolting something
lockable when the noun does not provide the property matching key
( this is the can't unbolt without matching key rule )' using its
inline definition, but this contained the invalid annotation 'aorp'.

In Part SR4 - Actions, Section SR4/5 - Actions which change the state
of things:

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

Problem. I attempted to compile 'if the holder of the second noun is
not the actor or the noun does not provide the property matching key
or the matching key of the noun is not the second noun, stop the
action with library message locking it with action number 4 for the
second noun' using its inline definition, but this contained the
invalid annotation 'aorp'.

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

Problem. I attempted to compile 'if the holder of the second noun is
not the actor or the noun does not provide the property matching key
or the matching key of the noun is not the second noun, stop the
action with library message unlocking it with action number 3 for the
second noun' using its inline definition, but this contained the
invalid annotation 'aorp'.

[peter...@hotmail.com]

Hmmm, that happened on my ZCode WIP. On my Glulxe WIP (a different
project) I didn't get any of that. I did, however, get these two:

"In Chapter 1 - Initialisations, windows and values, section -
definitions of properties and values:

Problem. You wrote 'g-present property translates into I6 as
"light"' : but this property does not exist, so cannot be
translated."

This is in Jon Ingold's Flexible Windows extension. Does this mean the
extension is broken with the new Inform?

Problem. You wrote 'Taking inventory, going, looking, examining,
waiting, swearing mildly, swearing obscenely, sleeping, listening and
thinking is innocuous behavior' : but an action can't be the same as
a thing, so my guess is that this is an attempt to categorise an
action which went wrong because there was already something of that
name in existence. For instance, 'Taking something is theft' would
fail if 'theft' was already a value. (But it can also happen with a
sentence which tries to set several actions at once to a named kind of
action, like 'Taking and dropping are manipulation.' - only one can be
named at a time.)

I had "Taking inventory, going, looking, examining, waiting, swearing
mildly, swearing obscenely, sleeping, listening and thinking is
innocuous behavior." What does it mean, "only one can be named at a
time"? Do I have to set each action to "innocuous behaviour", one at a
time? Isn't this a bit of a regression?

Also, I can't tell whether the previous errors *didn't* occur - maybe
they would show up after these two were corrected? But I can't just
comment out the existence of Ingold's extension, because I have
sooooooo many references to it spread throughout my source text...

[Emily Short]

On Apr 23, 5:27 am, peter_pe...@hotmail.com wrote:
> Also, I can't tell whether the previous errors *didn't* occur - maybe
> they would show up after these two were corrected? But I can't just
> comment out the existence of Ingold's extension, because I have
> sooooooo many references to it spread throughout my source text...

I can confirm that Flexible Windows does fail under 5Z71, and does so
in a way that causes I6 errors that affect the compilation of
apparently-unrelated parts of the source. I think this is, as you saw,
due to its attempts to save space by double-naming a built-in I6
property.

I have hacked my own version of Flexible Windows so that it does
compile, but I'm not totally certain that this hasn't resulted in some
other performance changes; what you probably want to do is contact Jon
and see whether he can release an update to accommodate the new build.
(Sorry; there is always a little fallout of this kind from a new
build.)

[Jon Ingold]

Hey -

I'll look into this as soon as I can, and get it posted once I
understand it!

cheers
jon

[David C.]

On Apr 23, 3:12 am, Graham Nelson <gra...@gnelson.demon.co.uk> wrote:
> Inform now has a larger and more resourceful website, at a new URL:
>
> http://inform7.com
>

Bravo Graham, Emily, Andrew, David, and all involved.

David C.

[mikegentry]

> 2007 document: seehttp://inform7.com/learn/papers/

>
> This release has been something of an epic business - we began the
> preparations before Christmas, but it has now become the traditional
> "birthday" release (though Inform's sixteenth anniversary is strictly
> speaking not until next week). Many, many thanks to all of those who
> helped.

SO COOL.

[Adam Thornton]

In article <6f1e8087-802e-41b1...@r37g2000yqn.googlegroups.com>,

<peter...@hotmail.com> wrote:
>Great work! However, I installed it and tried to compile my WIP with
>it, and got a lot of errors... which point to Standard Rules and such.

When I got the aorp errors it was because I had an old Standard Rules.

Adam

[Andrew Plotkin]

Here, S. John Ross <sj...@io.com> wrote:
>
> Many thankings!

And felicitations.

Let me point out something that folks might not have registered:
the first release of the Inform 6 compiler under an open-source
license. (Previous versions of Inform 6 were released as source code,
with a "use this but don't change it" restriction.)

This means I6 will be more fully glomped into the Linux/etc software
packaging ecosystem. Which probably doesn't thrill most of you, but it
is a good step forward, and one I have been anticipating for a while
now.

So, thanks.

--Z

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

[James Jolley]

I'm curious folks. Is this on Windows only? I ask because the new build
is working superbly here. Love the new formula creation tools.

[Jim Aikin]

Fantastic! Many thanks, Graham.

One suggestion for the new website: Add a link on the Downloads page
to the Extensions page. Currently the Extensions (which one would
download) are tucked away under the Write tab.

--JA

[George Oliver]

The new site really looks excellent! Nice work guys.

[QF]

Andrew Plotkin <erky...@eblong.com> wrote:
[...]

> Let me point out something that folks might not have registered:
> the first release of the Inform 6 compiler under an open-source
> license. (Previous versions of Inform 6 were released as source code,
> with a "use this but don't change it" restriction.)
>
> This means I6 will be more fully glomped into the Linux/etc software
> packaging ecosystem. Which probably doesn't thrill most of you, but it
> is a good step forward, and one I have been anticipating for a while
> now.

[...]

It certainly thrills me! (Even though TADS 3 is currently still my IF
platform of choice.) Many Linux users are just the kind of people who
might be more inclined to play/write IF, including myself.

Congrats on the release!

--QF

[Otto Grimwald]

> Andrew Plotkin <erky...@eblong.com> wrote:

>> the first release of the Inform 6 compiler under an open-source

>> license. /.../

>> This means I6 will be more fully glomped into the Linux/etc software
>> packaging ecosystem.

That's great, and I see the I7 libraries are under the same license as well:

http://inform7.com/sources/webs/

And how about the 6/11 libraries, such as
http://www.inform-fiction.org/source/library/index.html ?

[Andrew Plotkin]

I think the exact situation is that the library files are under the
same license as the (previous) I6 compiler. But any derivative work
compiled from the library files is entirely yours and can be licensed
however you like, as long as you include the compiler/library version
info (which is in the default "version" banner). The DM4 copyright
page explains this.

It would be good to explicitly post all the 5/ and 6/ libraries under
the same dual license (Artistic and historic Inform) as the new I6
compiler.

[peter...@hotmail.com]

On 23 Abr, 16:17, a...@fileserver.fsf.net (Adam Thornton) wrote:
> In article <6f1e8087-802e-41b1-8f2e-c20c9cb3c...@r37g2000yqn.googlegroups.com>,

>
>  <peter_pe...@hotmail.com> wrote:
> >Great work! However, I installed it and tried to compile my WIP with
> >it, and got a lot of errors... which point to Standard Rules and such.
>
> When I got the aorp errors it was because I had an old Standard Rules.
>
> Adam

I say, well spotted! I used to have all my extensions in "My
Documents", because apparently that was where things went, or
wherever, and so I moved the built-in extentions to there. And there
they still were, being used instead of the new, built-in ones!

And thank you Emily for your reply, and thank you Jon for looking into
it.

Now the only question is, do I really now have to define all those
actions as "innocuous behaviour" at a time? How come? It's so unwieldy.

[David Kinder]

peter...@hotmail.com wrote:
> I say, well spotted! I used to have all my extensions in "My
> Documents", because apparently that was where things went, or
> wherever, and so I moved the built-in extentions to there. And there
> they still were, being used instead of the new, built-in ones!

That's definitely not a good idea - the built-in extensions live with Inform
7, and any additional ones you download go in "My Documents". Mixing these
two together will have (as you found) unpleasant results :(

David

[peter...@hotmail.com]

On 23 Abr, 21:09, David Kinder <da...@david.david> wrote:

Yeah, as I found. Pity. I like to keep my stuff together, and I
reckoned, if those built-in extensions get updated I'll want to place
them along the others - otherwise I may forget they came built-in with
Inform and end up having both of them around, never knowing which I7
will pick.

Except now I know: I7 will pick the one in My Documents.

Incidently, I still have an extension-related problem - the
documentation doesn't get opened. I get a 404 error, or server error,
or what not - it's in portuguese, so it wouldn't help any even if I
pasted it here. The documentation is in Os Meus Documentos\Inform
\Documentation\Extensions, and if I go there directly I am able to
read them with Firefox. The rest of the manual is properly readable.

[David Kinder]

peter...@hotmail.com wrote:
> Incidently, I still have an extension-related problem - the
> documentation doesn't get opened. I get a 404 error, or server error,

I've just had another report of this, and I can partly reproduce the problem
here: I can click on the documentation and it opens, but if I go back and
then forward, I get the 404. Not sure what's gone wrong there, but I'll look
into it.

David

[Mike]

On 23 Apr, 22:15, David Kinder <da...@david.david> wrote:

Many thanks to Graham and all involved in the new release.

In relation to the documentation for extensions, I have experienced
this as well with Windows XP. I am trying this on Windows 7 but at
the moment, IE8 will not let me download the exe file. It reports the
file size as about 358 KB and then refuses to run it (even if I save
the exe file first before running it).

[Mike]

Interesting. I have finally managed to download and install using
windows 7 and IE8 (internet connection is really ropey at the moment)
and extension documentation works correctly. Using Windows XP and
IE6, extension documentation produces the errors referred to above.

[Ron Newcomb]

> 2007 document: seehttp://inform7.com/learn/papers/

>
> This release has been something of an epic business - we began the
> preparations before Christmas, but it has now become the traditional
> "birthday" release (though Inform's sixteenth anniversary is strictly
> speaking not until next week). Many, many thanks to all of those who
> helped.

Yay! Xmas in April!

Also, are we out of beta? With the launch of the new website and the
changes to the splash screen, it feels like it.

Also: Yay!

[ChrisBlue]

Huzzah! And our German extension still works (after some minor
changes)!

http://www.pageturner.de/adventure/GermanBeta.zip

Thanks for your great work!

[JJ Sonick]

On Apr 23, 1:12 am, Graham Nelson <gra...@gnelson.demon.co.uk> wrote:

> 2007 document: seehttp://inform7.com/learn/papers/

>
> This release has been something of an epic business - we began the
> preparations before Christmas, but it has now become the traditional
> "birthday" release (though Inform's sixteenth anniversary is strictly
> speaking not until next week). Many, many thanks to all of those who
> helped.

Fantastic news! Nate Cull's Planner still works fine with it too
(hurrah!).

Josh

[Adam Thornton]

In article <75bm5mF...@mid.individual.net>,

James Jolley <jrjo...@me.com> wrote:
>On 2009-04-23 16:17:05 +0100, ad...@fileserver.fsf.net (Adam Thornton) said:
>> When I got the aorp errors it was because I had an old Standard Rules.

>I'm curious folks. Is this on Windows only? I ask because the new build
>is working superbly here. Love the new formula creation tools.

Er, no. I got the dreaded aorp errors when regression-testing the Linux
build. But it was because I had, for some reason, my own local copy of
the Standard Rules in my Extensions directory. Don't do that.

Adam

[dott.Piergiorgio]

Graham Nelson ha scritto:

> Two more of the underlying Inform programs have been
> published, moved to an open-source licence (the Artistic License 2.0),
> and presented in literate source;

Whose is a good thing, because reduces the issues around including it in
Linux distros (albeit I don't known how the Debian mantainers consider
the Artistic 2.0 in their strict canon of Open Source)

Personally, keeping the I7 closed source was (and is ?) a good thing, if
we look at the long discussions on NL & syntax....

For the rest, It's excellent timing, because now the already needed
rebuild of GNOME I7 can be also a major release :)

Best regards from Italy,
Dott. Piergiorgio.

[dott.Piergiorgio]

Andrew Plotkin ha scritto:

> I think the exact situation is that the library files are under the
> same license as the (previous) I6 compiler. But any derivative work
> compiled from the library files is entirely yours and can be licensed
> however you like, as long as you include the compiler/library version
> info (which is in the default "version" banner). The DM4 copyright
> page explains this.
>
> It would be good to explicitly post all the 5/ and 6/ libraries under
> the same dual license (Artistic and historic Inform) as the new I6
> compiler.

Still I prefer "Zarf's license", whose resolve the issue "code vs. story" ;)

[Victor Gijsbers]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

dott.Piergiorgio wrote:

> Whose is a good thing, because reduces the issues around including it in
> Linux distros (albeit I don't known how the Debian mantainers consider
> the Artistic 2.0 in their strict canon of Open Source)

FSF and Debian both accept Artistic 2.0. In fact, Debian also accepts
the original Artistic. Even more in fact, it's one of the three examples
of a free license they give in the Debian Free Software Guidelines:
http://www.debian.org/social_contract#guidelines.

You can't ask for more than that. :)

Regards,
Victor
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAknx6nwACgkQoiOrMwvIZLzRBgCeKBKimuJmEuYofUDWt7w2tJAD
H5MAnRokyhlhPfsFwB0vxZM/JFdbIRjX
=TJ/2
-----END PGP SIGNATURE-----

[dott.Piergiorgio]

Victor Gijsbers ha scritto:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> dott.Piergiorgio wrote:
>
>> Whose is a good thing, because reduces the issues around including it in
>> Linux distros (albeit I don't known how the Debian mantainers consider
>> the Artistic 2.0 in their strict canon of Open Source)
>
> FSF and Debian both accept Artistic 2.0. In fact, Debian also accepts
> the original Artistic. Even more in fact, it's one of the three examples
> of a free license they give in the Debian Free Software Guidelines:
> http://www.debian.org/social_contract#guidelines.
>
> You can't ask for more than that. :)

Tnx ! :)
Eventually you can propose Gnome-Inform7 for inclusion in the distros :)

Best regards formo Italy,
Dott. Piergiorgio.

[leve...@googlemail.com]

On Apr 23, 9:12 am, Graham Nelson <gra...@gnelson.demon.co.uk> wrote:
> [...]

Nice work!

> Two more of the underlying Inform programs have been
> published, moved to an open-source licence (the Artistic License 2.0),

> and presented in literate source; another purpose of the new site is
> to provide a framework to hold the Inform source code as we publish
> it, which remains an ongoing process.

If I may make a small request on this topic, and that of distribution
packages: "packages" in distributions such as Gentoo consist of a
script that downloads the source code (usually, although binaries work
too if that's all that's available) and builds it on the user's
machine. For this to work properly, a URL has to precisely identify
which version of the software it will download, usually simply by
putting the version in the filename. I've been maintaining unofficial
Gentoo packages of I7 for some time, and now that the source code is
being released I'd like to convert them to compile as much of it as
possible instead of relying on the published binaries.¹ Therefore, I
would like to ask that the various source files be made available with
such a versioned URL (the existing ones can of course continue to
exist as aliases). This also makes it trivial to provide older
versions once newer ones are released, as you already do with the
"main" downloads, if you wish to do so.

If you can't do this for some reason, I do have a workaround in mind,
so it won't be the end of the world. But it seems to me that it would
be cleaner to make the change officially.

As always, thanks to the whole team for all their hard work.

[1] Why bother, when the binaries work fine? It's partly
philosophical - Gentoo /is/ a source distribution, for all the good
and bad that that entails, but there are practical benefits too. For
one thing, it works on platforms where no binary is provided, and much
of http://www.gentoo.org/proj/en/java/why-build-from-source.xml also
applies to non-Java software.

[dott.Piergiorgio]

dott.Piergiorgio ha scritto:

>
> Tnx ! :)
> Eventually you can propose Gnome-Inform7 for inclusion in the distros :)
>
> Best regards formo Italy,
> Dott. Piergiorgio.

ah, I'm rather pleased to report that Gnome I7 5Z71-0ubuntu2 runs fine
and w/o issues on my box :)