Posts by Mariner

I suppose everyone's got their preconceived notions about what it means to be open sourced. The term so often evokes charged feelings and socio-political debate these days. The fact of the matter is that like most things, it isn't a matter of absolutes. There is a lot more grey than black or white.

For that matter, I have yet to see an open source project that is a total free-for-all. What makes it an open source model open is that anyone can view the source code--that's it. It most emphatically does not mean that everyone and their uncle can write back to the master at will. Open source does not turn a project into a subway wall with every passerby overwriting their own brand of graffiti.

Also, open source does not necessarily mean just the Free Software Foundation's 'utopian' GPL3 vision. The project leadership is free to choose a license that makes sense independent of the development model (provided they don't violate the Mojang license.) If they want to have a clause in the license that says users can only play IndustrialCraft on Thursdays and only if you have an attractive redhead on your lap, then they have the right to release under that license. (If only they'd provide the redheads...) They can have that license and use an open source development model.

Furthermore, they can also prohibit redistribution under their license, or allow it for non-commercial applications, or whatever they see fit. Alblaka can keep essentially the same license he has now if he chooses.

There is no more risk of clones, bad addons, or other such nonsense than there is now. Anyone with even the skill to code a bad addon, or even just an eighth grade reading level, can easily drop IC2 into MCP.

Unfortunately, without access to the reversion control system, that's what a well meaning community member would have to do if they wanted to help out. To non-programmers, that may seem like no big deal, but since that happens outside of a reversion control system, it is a nightmare (or at least a PITA) for the devs to try to merge those changes into the master, especially since they've been coding on a branch too and their current stuff isn't the same as what we can download. That's what reversion control systems are designed for. That in turn makes devs just toss out community contributions because it feels like more work to merge than to just tackle the issue yourself. You lose out on a valuable resource that way. The community can be a great resource in this department if you let it.

Quote from MagusUnion

Plus, even if you allow for public bug fixing, it raises the issue of 'rampant bug breaking'...

Think of it this way, if a person developed a fix for a bug, then there is bound to be several who develop the same fix. But while doing so, those other fixes cause additional bugs that 'break code' and create new bugs for the community to fix. Thus, the wild rapid fire of 'fix this' 'debug that' would spiral out-of-control, as many would try to fixed already tamed bugs, and possibly reintroduced new ones thru different, multiple changes to the code that IC is unable to handle...

Not an issue if the code repository is set up with proper permissions. I suspect what you're imagining is an anarchy in which the whole world could write changes to the master code, but that's just not how an open source project works. It can easily be set up so that only Alblaka, Player, and whoever else they choose have write access, and anyone else has to submit their changes for review and approval. They effectively end up with a lot of free labor and just have to sign off on the changes after considering the impacts to the project as a whole. This provides a channel for folks like Greg, Raw, CPW, or whoever to submit their changes to the IC2 core.

Quote from selurgniman

I think a lot of people are used to mods just being abandoned in MC and so have become more apt to just move on once it appears evident that the mod is no longer going to be updated. For example, most of my players are in favor of moving on without EE3 because it's obvious that that mod is probably not ever going anywhere. I wrote up a little forge mod to replace the easy bits and we're moving on without EE3. IC2 would be harder to move on from but not impossible, eventually the features from the game updates are going to be compelling enough.

Hm... you might have inadvertently raised another interesting point. As far as I know, Pahimar live-streams development, he tweets progress reports (if you can count <=140 chars as a progress report), [EDIT: and EE3 is on GitHub (https://github.com/pahimar/Equ…Exchange-3/commits/master)] but apparently that is insufficient since some people evidently feel that EE3 is vaporware.

So what's the lesson-learned there that can be applied to IndustrialCraft? How many communication channels is enough, and what would be the preferred channel(s) keeping in mind that time spent communicating is time not spent developing? E-mail subscription list? Tweets? Live-streaming development? Blog post? Git RSS? Chat to YouTube about how the week went? Facebook? G+?

IMO, an occasional blog post is fine if coupled with access to the commit logs in git and a public bug tracker with actual notes in the tickets (not just "fixed in build #9999.")

Quote from selurgniman

Actually I would say the view count has been dropping due to the pace of development. Here we are over two months later with no 1.3.2 compatible build and 1.4 probably only a couple weeks away. Shortly after 1.3.0 dropped I was checking the site and forums multiple times per day for updates, then once per day, then only when my RSS feed of the main site updated, now only when something interesting pops up in the RSS feed.

My participation is about the same. I used to visit the forums almost daily, then weekly, and now I don't visit the forums at all looking for an update--I just have the blog RSS set up in my reader and I eagerly await posts regarding new versions. I see that as a reason that site visit numbers aren't necessarily a good metric with which to gauge interest in IndustrialCraft. I don't tweet or follow Jeb et. al. but that doesn't mean I don't love the crap out of Minecraft and want to see it continue and thrive.

I came here because I saw your call for opinions in the RSS feed.

Yeah, the [video's] delivery was abrasive and inconducive to constructive dialog, but let's put that aside for the time being and focus on the salient points. It seems the main complaints raised are about transparency and development pace, so I'll put in my two cents on those topics.

The communication issue I agree with. Frankly, I doubt anyone will respond to this thread saying they'd have a problem hearing MORE about the current state of development. One solution already proposed is to open up an RSS of your git. Another idea is to expand your team to appoint someone to handle public relations. You could also make your development goals, priorities, and timeline a public document.

Alblaka makes a very valid but concerning point that this is just a hobby. As a hobby, it will receive attention from developers when the whim grabs them. We, the users, cannot expect anything more than that. However, I believe that is a strong argument for moving to an open source model. That way, when the whim isn't catching anyone, or the right combination of people simultaneously, or if interest starts to wane altogether, other developers can pick up the slack and drive the project forward. You can still choose a license that offers some protection (although I doubt you'd want to expend any resources defending that license if someone did violate it.) If you had a public bug tracker and public git, anyone could read the bugs, grab a branch, hack on it, and submit their fixes for approval.

Alblaka also makes a good point about the forums blowing up with 'bug reports'. I see a simple fix. Public bug tracker and a new forum policy, and promote some long-time / trustworthy forum members as forum moderators to delete the offending posts and ban as appropriate.

At the end of the day, the only reward for working on any Minecraft mod is ego (fame, notoriety, 'credit', feel-good, etc.) or profit. That isn't a value judgement, it just is. There's no way the revenue from ad-sponsored downloads and/or donations offset the man-hours put in, so that leaves ego. Ego can be just as fulfilled with an open source model, possibly even more-so than a closed model. At the end of the day, nobody can take away that Alblaka was the father of IndustrialCraft, no matter how many devs hack on it.

A selfish (for Alblaka and team) motive for opening up development is that you don't need to feel any pressure from the community anymore. If you decide you want to take three months off to write a novel or tour southern Africa, you don't even need to tell anyone or fret about coming back to hate-mail. If the concern is maintaining the scope and balance of the mod, you can still maintain a steering committee--you could be the Linus Torvalds of IC. How cushy would it be to not have to code at all and just field people's commits?

Anyway, you asked for it, so there's my opinion. Love your mod and thanks for all you do (and put up with)... it's a pretty thankless task sometimes, huh?