A critical bug has been left festering on Linux for one month and 18 days, which breaks Inform 7 completely. I reported it when it completely broke all operations under Linux inform7.com/mantis/view.php?id=2030
…only to figure out that it was reported back in August, at “high” priority and “critical” severity, and totally ignored by the supposed “maintainers”. inform7.com/mantis/view.php?id=2025
It only requires a recompile of NI.
However,
(1) NI is closed-source, for no obvious reason
(2) NI has debugging symbols stripped, for no obvious reason, so even as an experienced programmer I would never have traced this down
(3) The only people with access to the source have ignored a critical bug report which breaks the entire system for the average Linux user, and is not debuggable by any ordinary user, for 1 month 18 days
(4) Nowhere is it made clear which people have the secret source to ni, so I don’t even know which email addresses to pester.
Which is why I’m asking here.
OK. So it’s clear that just filing a bug report isn’t doing the trick. How can I wake up the maintainers of the secret source and get NI recompiled (or better yet, get the source released – even if not under an open source license – so it can be recompiled whenever this happens, without having to wake them up)?
(And people wonder why I describe Inform 7 as “very powerful, but practically user-hostile”. This sort of stuff is infuriating.)
As the person who’s volunteered to monitor the forum and send an e-mail to the Inform team when a critical issue arises (mentioned here), I’ve sent an e-mail about this issue.
As for getting NI to be open-source, that’s a perennial discussion, and it seems like discussing it right here isn’t going to have any effect on the people who can actually make it happen. (To be clear: I am absolutely not one of those people, and I don’t know them.) The reason ni isn’t open source as zarf puts in there is that “Graham’s answer about this has been consistent: he’ll do an open-source release of I7 when he’s rewritten, cleaned up, and documented the code to his satisfaction.”
There is information about “the people who have the secret source to ni” (or some of them, at least) on the Inform 7 webpages, under the Sources tab. How up-to-date it is I couldn’t say.
The lack of an open source release might be irritating, but I certainly don’t find it inexplicable. Inform 7 is Graham’s project, and he prefers to keep it that way (at least for the time being).
– I suspect that Graham’s standards for clean and documented code are pretty demanding, so this might take a long time.
I don’t think much is gained by abusing the developers. Accusing them of “callous disregard” seems hyperbolic to me.
At any rate, lecturing this forum about the problem changes nothing. I’m pretty sure everyone who posts here wants to see an open-source release sooner rather than later.
Thanks, Andrew. Since Graham did his best to keep his email address off the Inform 7 website, you’re doing what I can’t.
I think an appropriate Christmas present for Graham Nelson would be a copy of “The Cathedral and the Bazaar”. If I had the opportunity I would send it to him.
One of my areas of expertise is cleaning up and documenting code. I also do refactoring and documentation of code quickly.
I personally probably have substantially higher code and documentation standards than he does, since mine are extremely high, and I can point to lots of things in Inform 6 which ought to be cleaned up.
[rant]There are projects I’ve felt very proprietary about and not wanted to release half-finished. But when I find that I’m not checking the bug tracker for a month at a time, I give up control because anything else is a disservice to the users driven by egoism.[/rant]
Given those areas of expertise, have you considered working on other open source projects? For instance, the web version of Hugo has no save capacity right now, and neither does Lectrote’s hugo. There is also an Inform extension called Glimmr that is needed for Counterfeit Monkey and some other game to display images correctly that is completely broken. Any of those projects would be a major help, and could help pass the time before Inform becomes public.
Why would a source release (“open” or not) require the code to be cleaned up or documented? That seems kind of a strange thing to put up as a blocker.
I personally think Graham has set the tone for IF development by keeping Inform 7 so close and not open sourcing it. A lot more collaboration would go a long way to expanding our community, but our cat-herd mentality is unlikely to change any time soon. I wonder if this is the result of how things were in the past, when working on anything alone was “the only way”. Now we have massive collaboration capabilities that are embraced by the wider developer population, but in IF it seems to be hit or miss. Mostly miss. And specifically with Inform 7, there are so many tooling solutions that would improve with an open source repository, including source control, extension management, plug-in capabilities, and more. I can see how anyone in the IT industry that comes across Inform 7’s state is dumb-founded that someone who gives away their software and yet maintains the source privately. It’s such a foreign concept today. I think the primary frustration is that the wider developer community knows that open source can be improved at a much faster pace. More eyes and hands, more solutions…including solutions that the repo owner may not even have considered.
That said, the OP’s tone was out of line. We can disagree with Graham’s process, but there’s no need to get nasty or even impolite.
Emily has got in touch with me to let me know they’re working on it.
As far as this sort of thing goes, I think that posting on the forum and having me get in touch with Emily is probably more effective than trying to pester (in the words of the OP) her or Graham directly, just because she knows to expect things like this from me every so often. If there’s something that it seems like I should be escalating and I don’t seem to have seen it, feel free to PM me to bring it to my attention.
This should allow people to fix their installations for the time being. Replace the existing ni binary in /usr/libexec/gnome-inform7 (location may vary per Linux distribution) with this one. Or, build gnome-inform7 from a git checkout after dropping this ni binary into the src/ni/ folder.
Hopefully there will be a re-release of ni on all supported architectures before too long; I can only provide x86_64, lacking the hardware to provide the others. When that happens, I’ll issue a full re-release of gnome-inform7.
It comes with a few caveats.
One test from Inform’s test suite failed with this recompiled binary. No idea how serious it is. I’m leaving on vacation imminently, so I really don’t have time to dig into it. Obviously before making a re-release we’d have to make sure this isn’t an issue.
If anyone decides to make a re-release of one of the gnome-inform7 packages with this binary (see github.com/ptomato/gnome-inform7/wiki/Packaging) please differentiate it by including something like “unofficial” in the version string.
.)