Home Page Forums Thread Like Summary - BFG Scorpion 6 Skeleton

I noticed the Scorpion 6 skeleton PR ( https://github.com/mamedev/mame/pull/11121 ) has been closed but it doesn't seem to be merged ?
Just wondered why.

OK, after I review a few more pull requests, I’m going to try and come up with some more written documentation.

I appreciate it, but I want to be clear, there are multiple issues, not all of which will be assuaged by documentation:
- Prototypes should be un-defaced, but it needs to be understood that someone typically isn't going to submit one unless there's no other options available.
- A lot of people who submit pull requests are doing so on a lark. If there's any major feedback beyond "please tidy up this or that thing", the person is as likely to walk away as anything. For softlist contributions, if it falls within the 3-year rule but is being offered at no charge on a public website, then it needs to either be cordially explained that clarification is needed from the person whose software is being added to the list, or someone needs to go off and obtain that clarification on their own volition.

If you're looking for someone to volunteer to do that, fuck it, I volunteer as tribute. If I can track down a developer who went radio-silence for 24 years and actually get reasonable statements out of him after an initial brick-wall response, I'm pretty confident that I can wheel and deal with people making retro homebrew in a pay-what-you-like scenario.

Why do you care if Haze calls you a cunt or a fascist? Haze and I have absolutely been at each others' throats so many times before, it should give a moment's pause to consider why he and I seem to be aligned on this particular topic. Sure, you're a cunt - a good cunt, in the Aussie sense - and the term "fascist" is thrown around so often these days that you can pretty much write it off as someone mistakenly using it in place of the word "the".

As for putting words in your mouth, the problem circles back to a lack of concretely-stated policy or documentation on the MAME team's part. To quote Talking Heads: Same as it ever was.


This gives more information than most of the comments I've seen on pull requests or other things, let alone the official website. It at least gives something to discuss.

You state:


Your post earlier in the thread, that's how. At this point we're veering away from the other issue of defaced dumps, but I'm going to respond to that in another post, so here we go with the commercialization thing.

There's an ongoing assumption I've seen over the past 15-20 years that we've been communicating, where you assume people have the same grounding of information that you do, and the reality is: Most people don't.

The appropriate solution for this situation would be for whoever is interested - which is, Haze - to reach out to whoever has this pay-whatever site and ask them how they feel about their title being added to MAME's softlists. Additionally, for that person (Haze) to also ask if they have archives of any previous versions of the software.

If the person says it's not cool, then hey, so be it. I have a dump from the DPB-7000 that I can't include just yet out of courtesy to the person who dumped it. That's how life goes.

But if the person says it is cool, then that's a reasonable justification for waiving the 3-year rule, and in such case I would hope that you'd approve its inclusion in the softlist.

What I take issue with is that you find yourself being upset because people get angry about various objections to PRs, but they themselves can't phrase it very well, so there's this inevitable circle of anger that arises. I'm quite possibly the least-qualified person in the universe to talk about how someone communicates with other people, but you either need to appoint someone who will uphold your standards while being able to communicate with contributors in a better way, or you need to reduce your surrounding workload in order to be in a better mind-frame when dealing with potentially-charged pull requests. People aren't idiots, but they also don't have the decades of experience and context that you or I have.

What I'm furious about right now is how this is tracking so very closely with Aaron's departure. The cavalier hand-waving of a valued and long-term contributor. Someone who wasn't perfect, but someone who may have actually stuck around and contributed if the time were taken to educate, rather than castigate.

It's like clockwork: Nobody cares about cleanup; I step up and do what I can. Lamentations about how pull requests seem to constantly run afoul of various guidelines; I write up a guideline document in order to make that easier.

I'm growing weary of having to hear how stupid people are on Reddit when it's just as easy to not participate in that hellscape of a site. I'm growing weary of trying to draw people back to the project, who have done nothing but show love and care for it, after they've been burnt. Phil had for all intents and purposes peaced out, and I did my best to build a bridge back up during the Polygonet finalization, and I genuinely worry that that bridge is going to have fire set to it once it comes time for the Loopy merge.

Do everyone a favour and get yourself under control, because for as much grief as people want to pin on me, there seems to be plenty of acrimony to go around that appears to be coming from within. I believe with every fiber of my body that you're a kind, good, and well-meaning person, and that you want what is objectively the best for MAME, but the way you're going about it, the terse communication and expectations of a psychic uplink, is just harming things. Reduce your workload, take a break beyond getting releases out, or start delegating part of the release processes, so that you have time to just de-stress. You're in exactly the same rut with MAME that I was at my day job, and I beg of you, help everyone by helping yourself to just chill.

Not for reasons unknown, just reasons unasked.

Upon asking, it was a "straw breaking the camel's back" situation: He perceives there to be a push to effectively scrub all cracked/introed dumps or images from softlists in favour of only documenting "clean" dumps.

If true, I tend to agree with his level of frustration; if MAME is supposed to be about documenting history, then the crack/intro scene is history, point-blank. The No-Intro crew have even begun to re-think their attitude towards pirate dumps, whereas MAME appears to be skewing in entirely the opposite direction.

I'm not on the list and I haven't had any pull requests affected by this, but if taken at face value, then I see no reason to not support cracked or introed dumps in separate softlists. Drivers can support multiple lists, as far as I'm aware. GoodTools had the wrong idea by documenting outright wrong dumps (overdumps, underdumps, corrupted dumps), but ignoring the influence the cracktro scene, and by extension the demoscene, have had on the gaming industry as a whole - again, if true - would be one of the most foolish and short-sighted management decisions in MAME's long history of foolish and short-sighted management decisions.

Off the top of my head I can name EA DICE, Starbreeze Entertainment (and by extension Machine Games), Funcom, and purportedly EA Tiburon that were started by people involved in the demoscene, and pretending that the demoscene didn't arise out of the intro/cracktro scene would be frankly idiotic.

I'm not aware of any issues with cracked stuff as long as it's in its own software list. The major issue with Haze's stuff was new on-sale software. Vas considers "pay any price even zero" to be on sale and thus subject to holding off until it's no longer that way. Haze considers it to be freeware. They're both right.

The demoscene existed long before dodgy dumps for emulators became a thing. Tech Tech by Sodan & Magician 42 released in 1987 was a defining moment in Amiga demos. Elaborate crack intros appeared on the Apple II in the early to mid ’80s, which were something of a precursor to the demoscene. C64 demos had emerged from that by 1986.

Vas is talking about defaced console dumps, and the demoscene existed well before those. The console releases with crack screens that I'm familiar with were made for copiers and predated emulators though. We really need to emulate some SNES copiers sometime because nobody else is going to.

I'm pretty sure when the 3-year rule was instituted, the indie-game scene and the creation of homebrew titles that were widely disseminated online didn't exist. Nor did the myriad niche distribution networks that, unlike more commercially-oriented ventures like PSN, XBL, and whatever storefronts Nintendo are running this week, don't have any real concept of versioning.

There are groups out there dutifully tracking version identifiers for the major console vendors, and for Steam and other major storefronts, but retro-homebrew titles are a black hole. Data is actively getting lost.

When the 3-year rule was instituted, the intent was pretty explicitly to offer a window for companies like SNK that were still releasing titles on systems that MAME could trivially emulate, where you effectively had to pay. That goes out the window with a pay-as-you-like business model.

The fact that you're happy to see a developer who's been working on MAME longer than you or I have, who has been doing more to promote MAME than whatever half-assed effort Stiletto manages to manifest or what the official team itself ever half-heartedly manages to do, should make you feel deep and intense shame. Aaron, now Haze, to say nothing of anyone else. You'll see this entire project burn before you give an inch towards admitting you might be wrong about something.

Another thing: If you're writing documentation in order to be able to hand off certain responsibilities (like PRs) with a high level of confidence that core team members are going to do you proud, you'll need to write a document that gives a "go-to" example for how to do just about anything within the current core.

One of the things that seems to come up with some frequency, and what I suspect has been a major factor in team members falling out of the habit of reviewing PRs, is doing so to the best of their ability, yet still letting low-quality code through simply because nobody involved - either the submitter or the reviewer - knows about the current best practices.

What I absolutely don't want is for you to take some time to relax only to return to a complete fuck-fest. Only you really seem to have the encyclopedic knowledge of the codebase in order to point to a given driver as the best-case-scenario for doing something; transfer that knowledge into a document, if only to increase the bus factor.

100% agree with asking the author on the pay-as-low-as-zero stuff. I'd even offer up front to buy a copy for some larger than normal amount (maybe $50 or $100?) as a thank-you. I don't think the people writing that stuff expect to get rich, but they'd love to see something for it.

Written review guidelines would be fantastic. We could refer contributors to them up front, and PR reviews would have a predictable checklist so people other than Vas could review things sanely.

And yeah, non-historical defaced stuff could be BAD_DUMP. I'd rather have a defaced copy than no copy, and if it's possible to disable the defacement without burning the source that's always an option. (You'd still want it to be BAD_DUMP in that case, of course).

As these things are being discussed now, I'd like to share my 2c that are based on experience for over eleven years here on this Forum and in the MAME ecosystem:

[*] Please tutor and nurse people like Golden Child and others by giving them valuable but manageable tasks within the main project so they're willing, and don't ridicule their baby steps and creative ways they're using MAME for.
[*]Don't lump everyone in with anonymous outside contributors (likely never to be seen again), ROM kiddies, RA crazies or trolls. Give regular long term low tier contributors a niche to allow them to feel attached to the project.
[*]Don't let old hands get away with insulting other people. It should not matter if a first time poster or Al Kossow throws a hissy fit - they're due a talking to, at least in a PM.
[*]Give external contributors some sense of being acknowledged, a few kind words in a thread or a quick skeleton that shows them theyr stuff is valued and has arrived. (I wish I had gotten something else than "oh, another CP/M machine" after being in contact with guys in two or three different countries to procure everything needed for emulation.)
[*]Keep civil. Moogly is having a constant effort to rein in his creative cursing, and so should everyone.

These are not demands, but musings over what would make the MAME experience more predictable and enjoyable.

The issue with MAME right now is that if you asked myself, OG, Kale, hap, and Osso what that pile of general stuff is you'd get 5 different lists. Meanwhile your idea of that list is the one that matters. I like you because you have the cool architectural vision, but I need to know how I should act on it.

We solve that by documenting what that list is. I know it works because we do it at my work. We have wiki pages with the standards for each language we use. Code is written and reviews are done to those specs. Reviewers spot-check each other. And our code architect spends his time doing actual architect things instead of stalking the Perforce depot.