Inform 7 v10.1.0 is now open-source

This is probably the same detection as the Windows Defender one that Drew noted above: both he and I submitted a false positive report on that one to Microsoft.

The last time I looked into it, getting a code signing certificate as an individual developer was a nightmare, and I don’t intend to create a company just so I can get a certificate.

6 Likes

This is another why Linux, FreeBSD and all other opensource OS are the Right Choice…

Best regards from Italy,
dott. Piergiorgio.

6 Likes

Digression to share some experience of code-signing:

The main author of PuTTY, Simon Tatham, has acquired a code-signing certificate as an individual (it was not possible to get one as an informal non-company team), but as I recall it was indeed quite a faff both on initial application and renewal (involving notaries and such), even with pretty good support from the certificate provider (who had reached out to donate the certificate). (Not to mention all the code-signing tooling, which I understand is pretty wretched to use, although PuTTY has its own peculiar requirements in that regard.)

Anyway, I mainly came here to say that code signing did not stop the false positives from AV software at all, so I don’t recommend going to the trouble of code-signing just for that reason. (There may be other good reasons.)

6 Likes

IFTF is in the process of getting code-signing certificates to use for the Twine application. When we get that sorted out we should be able to offer them for use for the Inform developers as well.

(It’s a nuisance on the Mac side too, although for different reasons. Toby Nelson and I both have individual-developer Mac certs.)

8 Likes

Ok, a more serious bug found in the Windows IDE. When I open an extension, the examples are truncated to one or two lines per example, and thus are rendered useless.

The IDE otherwise works very well.

2 Likes

I reported it to Malwarebytes and it’ll be fixed in the next update.

2 Likes

I think this is a probably the same problem that ADRIFT suffers. I’ve had Windows defender delete the Adrift exe.

I can guess what’s wrong here, clicking on the purple oval should open the example fully, but I can see why that is probably not working.

1 Like

Yeah, it just leads to a blank page.

Yeah, Jira can be tricky if you haven’t used it before. Actually, I have used it before, for years, and I still find it tricky :smile: .

I am having similar problems taming the (typically) right side of the Mac IDE. To that end, I wanted to provide an update to your issue report that has the issue number in it, so that others in our situation can find it more easily.

Appreciate the report.

2 Likes

Yup my bug report is number I7-2187. Hopefully you can add a comment in the bug system below that with your update.

It took me ages in the website to find how to report an issue. Tried umpteen pages. Not sure how I finally managed it. Not sure I’d ever manage it again if I tried :stuck_out_tongue_winking_eye:

Whoops, I wrote that post wrong. It should read “To that end, I wanted to provide an updated link to your issue report that has the issue number in it…”.

Your bug report doesn’t need an update :slight_smile:. I will watch it and see where it goes. Thanks.

1 Like

Inform 7 10.1.2 released along with an updated Windows IDE.

8 Likes

Not only that, but v10.1.2 now includes the first release build for Linux! (Which I’ve been waiting for.)
I guess this wasn’t the case when you posted, since the Linux files are dated 6 hours ago as I write.

I guess it might not be quite ready for announcement, since inform7.com hasn’t yet got around to linking to v10.1.2, and I don’t see specific release notes for the Linux build – I guess the inform7-ide 2.0.0 tag covers these builds. (Thanks to @ptomato for all your work.)
The release candidate thread has install instructions and some talk about distro compatibility and such. Unless you’re comfortable with Linux (or have been following the RCs) it might be worth waiting a bit for guidance, but here’s my impatient look at the release files anyway (not that I have run any of them yet):

  • two files (.flatpak and .flatpakref) which I assume together constitute a Flatpak release (which the RC thread says is likely to be the recommended install method); but I imagine it’d be preferable to install from Flathub once it’s updated, and the wheels look to be in motion for that.
    (Disclaimer: I don’t know much about Flatpak.)
  • .deb – experimentally, definitely needs an OS newer than Debian buster (current oldstable) (which seems fair enough, I should really upgrade)
  • .rpm – with “fc35” in the name (same as the RCs), which I assume denotes “Fedora Core 35” and tells you something about compatibility
4 Likes

Note: This protocol and “helpers” will install a working command line version of the current Inform 7.10

Hopefully distros such as Debian, Raspbian, and Ubuntu will pick up the IDE version.

Here is a link to an installation protocol that works very well for the command line:

https://www.hoowl.se/getting_started_writing_interactive_fiction_with_inform7.html

I have been using it since the initial beta release. It continues to work well.

Zeds i7 helpers are also very nice to use:

It’s really unfortunate that GitHub releases don’t leave any room for installation instructions or notes for the individual files. I was looking for a good place to put these notes yesterday, didn’t find one, gave up and figured I’d deal with it later.

The .flatpakref file just contains a link that should install the app from Flathub when you open it with your software manager. (It’s the same file that gets downloaded when you click “Install” on Flathub—An app store and build service for Linux) The .flatpak file is a standalone package of the app which you can use if you prefer not to download it from Flathub. You only need one or the other, not both.

The .fc35 indicates the RPM package was built using Fedora 35, it will probably work on later versions and may work on earlier ones. It also may work on other Linux flavours that use RPM packages.

The DEB package was built on Ubuntu 20.04 because that’s the default environment that GitHub provides. Same thing goes here, it will probably work on later versions and may work on earlier ones. It also may work on other Linux flavours that use DEB packages.

It’s correct that I would prefer that people get the app on Flathub if they can! Building all the packages is a hassle and takes time that I’d prefer to spend on actually improving the app. In that light I wonder if in future releases we should instead turn the tables, put a link to Flathub on the official Inform release page, and put all the packages in a still-discoverable but less official-looking place like the ptomato/inform7-ide release page.

3 Likes

I have a silly question I’ve been wondering for a while. Maybe only @GrahamNelson could say, but maybe someone else here knows.

The original Inform 7 release announcement in 2006 called it a “Public Beta”. Writing with Inform never stopped saying (the generated version varying per platform):

{OSX:}This Public Beta of Inform 7 runs on Mac OS X through the graphical user interface created by Andrew Hunter and Toby Nelson.
{Windows:}This Public Beta of Inform 7 runs on Windows through the graphical user interface created by David Kinder.
{Linux:}This Public Beta of Inform 7 runs on Linux through the text-only interface created by Adam Thornton.
{GNOME:}This Public Beta of Inform 7 runs on Linux through the graphical user interface created by Philip Chimento.

so… is Inform 7 to this day still defined as a Beta release?

There’s a joke in the British tv show “Yes Minister” where Sir Humphrey, the civil servant, insists that reform of the Honours system cannot be contemplated because we have to give the present system a fair chance. Jim Hacker, the minister, replies that the Order of the Garter was founded in 1348 by Edward III, so perhaps it’s getting towards the end of its trial period now.

Anyway, no, Inform is not now considered a beta (or not by me, anyway - your lived experience may vary), so those documentation references should go.

13 Likes

/me *coughs*

4 Likes

Well, in defense of Graham, I must point that art.16 of the Bill of player’s rights don’t apply, because is a developer documentation :wink:

Best regards from Italy,
dott. Piergiorgio.

2 Likes