New NOT_WORKING machines #9881
New NOT_WORKING machines #9881
Conversation
------------------------------ Tom's Adventures [Museo del Recreativo] Ice Cold Beer (ICE) [Museo del Recreativo]
| okim6295_device &oki2(OKIM6295(config, "oki2", 12.288_MHz_XTAL/16, okim6295_device::PIN7_HIGH)); // Clock frequency & pin 7 not verified | ||
| oki2.add_route(ALL_OUTPUTS, "lspeaker", 0.45); // Guess | ||
| oki2.add_route(ALL_OUTPUTS, "rspeaker", 0.45); // Guessgit checkout |
There was a problem hiding this comment.
@clawgrip what did I say about checking your work before opening pull requests? And @ajrhacker, haven’t I said enough times that people need to actually review pull requests before merging them?
There was a problem hiding this comment.
You're talking about the git command text on the comment that is obviously an unadverted error?
You think I haven't checked it two or three times? I could have checked it a hundred times and still fail to see it.
I drove 800 Km to dump those machines, had to stay at night leaving my wife and baby alone at home, but hey, there's a typo on a comment! If you want to discourage contributions, you're doing a perfect job.
I have some dyslexia and probably a problem with details, buy never been a problem on my whole career (more than 20 years) as a Java developer on big software companies (Sun Microsystems, Atos, Accenture, etc.)... Never but here, where I really feel afraid to contribute.
On every other projects every colleage is allways supportive, friendly, willing to help. Here I feel like my work is not appreciated at all, and that it's better to stop.
I'm like I am, if you feel like there's no place for me on this project, just tell me.
On the last PR there was an obvious typo on the PR description... You edited the PR to remove the name of the person who was all night checking against dozens of PCBs looking for errors to fix on MAME, but did not edited it for fixing the typo.
What was the point? To make sure this guy will never contribute again by removing his name (I'm talking about the PR description, not the whatsnew!!!) and tell everyone I make typos??
There was a problem hiding this comment.
You're talking about the git command text on the comment that is obviously an unadverted error?
You think I haven't checked it two or three times? I could have checked it a hundred times and still fail to see it.
OK, so if it would be missed after checking hundreds of times, why did I see it straight away? It stands out visually, it’s highlighted in a spell-cheking editor, and if you simply read the comments in the file, surely you’d notice it.
I drove 800 Km to dump those machines, had to stay at night leaving my wife and baby alone at home, but hey, there's a typo on a comment! If you want to discourage contributions, you're doing a perfect job.
Family comes first. If this is taking you away from your family and causing a problem, then adjust your priorities, Dumping PCBs isn’t more important than family.
I have some dyslexia and probably a problem with details, buy never been a problem on my whole career (more than 20 years) as a Java developer on big software companies (Sun Microsystems, Atos, Accenture, etc.)... Never but here, where I really feel afraid to contribute.
On every other projects every colleage is allways supportive, friendly, willing to help. Here I feel like my work is not appreciated at all, and that it's better to stop.
Look at this from my point of view. I put a massive amount of time and effort into this project. Consider how much time I spent reviewing over 1,600 files for #9788 – I haven’t just skimmed the changes, I’ve looked for logic errors and potential pitfalls in every case. Look at the feedback cycle on pull request like #9841 and #9773, or #9762 – note that I don’t just say “this won’t work” and close it, I have to explain why it won’t work each time. While this is going on, I’m still trying to implement usability improvements (e.g. #9783), fix bugs (e.g. #9812), give other developers a hand when they ask (e.g. #9748), slowly chip away at our big pile of technical debt (e.g. #9884), and occasionally actually work on emulation (e.g. #9699). On top of that, I try to help with issue triage over at MAME Testers, and helping out users with questions on reddit.
I’m feeling very burned out at the moment, it’s 7:45AM and I haven’t slept all night, but I’m still trying to catch up on what’s going on in MAME because if I don’t, I’m only making a rod for my own back later.
What does frustrate me is seeing the same pattern of errors from the same people. I get frustrated at seeing pull requests where the code won’t build or validate, or where it obviously won’t work. I get frustrated at pull requests being merged with apparently no review at all. I get frustrated when people call absolutely any error a “typo” when they’re clearly not typographical errors. I’m not asking for contributors to be perfect. I’m just asking people to slow down, put a fair effort in and not make the same mistakes repeatedly. Quality is everyone’s responsibility.
On the last PR there was an obvious typo on the PR description... You edited the PR to remove the name of the person who was all night checking against dozens of PCBs looking for errors to fix on MAME, but did not edited it for fixing the typo.
Well, first of all, I don’t know what PR you’re even talking about. The most recent PR I edited the title of was #9887 to make it more descriptive. I didn’t remove any names. I often edit pull request titles to make them more descriptive or to better match the usual format. I don’t frequently edit pull request descriptions at all.
When you’re writing a pull request title, consider that some developer is going to be seeing it in a list of search results months or possibly years later when trying to track something down, most likely while tired and frustrated. Pull request titles should be terse and descriptive so someone has a good chance of seeing what it does without having to read the description, or resort to looking at the diff itself. If a pull request title doesn’t follow the format developers expect on a project, it takes longer to interpret, also increasing the workload. It’s not a place for stuffing names for street cred, and while using joke in PR titles might seem like a good idea at the time, it’s unhelpful. It’s completely redundant to include your own name in a pull request title, as the author’s name is already part of the pull request metadata.
We have a significant problem with pull request titles that aren’t descriptive, and cases of only describing the pull request in a comment, so even expanding the description isn’t enough to work out what it’s supposed to do (remember web browsers aren‘t the only way people view data – GitHub has APIs that work with other applications, too). This is an example of a pull request title and description that isn’t particularly useful – it just says “New NOT_WORKING machines”. When you see that in a list, it doesn’t give any clues about what the machines are, and even the PR description doesn’t mention that it adds a skeleton driver.
What was the point? To make sure this guy will never contribute again by removing his name (I'm talking about the PR description, not the whatsnew!!!) and tell everyone I make typos??
I still don’t know what you’re actually referring to. If you’re referring to the summarised pull request titles that appear in the release announcement, there’s a strict 132-character line length limit. I need to aggressively edit the titles to make them terse enough to fit, as well as coming up with descriptive titles for pull requests that were submitted with unhelpful titles. The pull request author is always included (that would be you if it was a pull request you opened), but other names may be omitted just to save space.
If I missed a typo in the summarised pull request titles, I’m sorry. Every month I need to do summarise a large number of bug reports and pull requests, as well as sorting the additions and promotions, and it’s always under time pressure given how much goes into each release. I do check it multiple times, and I try to get it into public version control at least 24 hours before a release in case other people want to review it (@0kmg pointed out a couple of editing errors for 0.244).
|
Oh are you referring to the commit message for #9846 that appears in master? I didn’t remove any names from the first-line description. Because the branch for that pull request had a single commit, the default commit message when it’s merged is copied directly from the commit message you used on your branch.
|
|
And if you’re asking why I didn’t edit the description of #9846 to fix your “typo”, I honestly didn’t know what you were trying to say. I’m damned if I do, and damned if I don’t – if I made assumptions and edited it to something that didn’t reflect your intent, you’d be angry at me for screwing it up; I asked for clarification, and you think I’m trying to mock you for making “typos”. Do you understand the unwinnable position you put me in here? |
|
Let's be real here, nobody's a saint in this sad and petty exchange. The feedback on an obvious mis-keying could have been phrased in a far better way. If something like proofreading is a consistent problem for a person, @cuavas, you're not going to get a great response from people by rubbing their noses in it like a dog. And you're certainly not going to engender much motivation for people to even bother merging pull requests by castigating them for overlooking something that, like it or not, is easy to miss if you're doing a relatively fast scan. When I still had direct commit access, I often gave a proofreading pass over the release announcement, yet it wasn't altogether infrequent that I would miss this or that spelling or grammar error. That's just life. I understand your frustration here - the piling on of tech debt - but let's not blow out of proportion what amounts to an honest mistake. It's not poorly-structured code, it's what amounts to having "fat-fingered" a Git command prior to a window-change taking effect. At the same time, let's at least attempt to approach things from the standpoint that everyone has their own personal life, their own frame in which they exist that nobody else knows about, but we all give a damn about MAME. It's not like people who are skilled enough to contribute to MAME are necessarily guaranteed to have the happiest personal life. That goes for everyone, from me, to @clawgrip, to @cuavas , to all of us who choose to use our little spare time towards at least wanting to contribute towards digital preservation. And for the people who do manage to walk that fine line of having a happy family life while also contributing, so much the better, I'm not directing this at you right now. Lacking info - info that nobody is obligated to give - it's easy enough to interpret an angry remark, a lashing-out, as a direct and personal attack. Most of the time, it isn't - it's that person at the end of their mental rope when it comes to being able to deal with stress. It's not okay, but it's not a reason to go off the deep end, either. Take a moment, a few deep breaths, and assume positive intent, despite how it might be phrased. We all want the same goal - preserving data - it just isn't always clear how best to get there at the end of the day. None of this is intended as an excuse for negative behaviour. In my opinion, were this an office, there would be apologies in order. But I can't force anyone to do that. Hell, I'm an external contributor. What I want is more open communication. Nobody benefits when a person perceives the weight of the world on his shoulders and never tells anyone of any importance, but lashes out as a result of it. While people shouldn't be praised for approving pull requests that need revisions, all that's left without a carrot is the stick. I've noticed over the years that people only ever seem to hear about things when they're going wrong: Users only complain when something is "wrong". The core members who approve pull requests only hear about it when they've done a bad job. Taking the time to carefully feedback on a pull request and arbitrate with the submitter for a higher code quality is something from which we all benefit, but also something that ends up completely ignored and never praised. It's simply wrong to only reach out to people when they err, and not when they do good work. If anything, this is why the project is in the state it's in - a minimal number of people taking the time to reivew pull requests, for fear of drawing ire if they make a mistake, while having no acknowledgement whatsoever if they take the time to provide feedback and encourage the submitter to improve. Everyone here can do better. The feedback process can be improved, and we could all do a bit better in reviewing our own pull requests. Quite a number of things in my own Rather than hammering on each other, use that energy to hammer out a clear-cut set of communication guidelines. |
|
i want to leave a crucial piece of feedback from all of this. it's all bullshit since i can only give uninformed (and probably even hypocritical) opinions at this point, but i don't care, i'll leave this all here anyway. while it is true that everyone "makes mistakes" or whatever, there is no excuse for anything that's happened here. @MooglyGuy nailed exactly what the current situation with MAME is. while MAME contributors often do make significant mistakes in regards to code, it seems @cuavas is lashing out on everything this time around and putting up immense amounts of mental exercsise to call out even the slightest of mistakes. in all honesty, i won't be swayed by any sort of manipulative tactic that gets thrown at me no matter how many excuses and justifications get thrown my way. boo hoo you can't sleep because MAME doesn't have that sweet sweet dough called "quality" that you want so badly. ugh. @cuavas, let me tell you something. the fact that you invested this much effort into calling someone out is honestly impressive, and in fact reminds me of my worst qualities. i have a tendency to write out a wall-of-text that can veer towards the most punishing of words, and that can range from personal attacks to even death threats. i called out a dev for implementing protection code for one of his tools and attracting virus outlets for treating it as a virus and even zeroed in on him for as much as i could. you mention about being "under pressure" to deliver monthly MAME releases. what pressure? where is it coming from? that must've been made up, because i've seen nothing but frustrated comments from other contributors regarding your presence. speaking of "quality", it's not "everyone's responsibility". so let me disabuse you of this notion. so when i see someone using this word seriously, i'm like "what the fuck is wrong with you?" it's cringeworthy, it's despicable, it's horrendous. and i see people actually practicing this kind of mental gymnastics, to uphold nonsense standards based on whatever feeling they might have for literally anything that comes at them like it's nothing. no, seriously, honest question. what quality? this isn't "quality", and i'm not seeing any of it. i argue it never really existed to begin with. none of this would've happened if you just gave actual feedback that doesn't screw people over in an instant. if you don't know how to give one, don't give one instead of flat-out nuke the living hell out of all incoming contributions here. you must be self-taught at this kind of stuff. see any code mistake, that person deserves the worst words possible. for some reason you think this is okay. what i'm saying is, whenever someone contributes some code to MAME, they must fear your presence so they can learn said standards from your mouth. and you're the source of that fear. all you're saying is that they should be anxious until their worst fears are realized for the mere achievement of you showing up and crashing their party over your personal baggage. the mere fact that CONTRIBUTING.md is empty speaks volumes. that file literally symbolizes and in fact represents MAME's current state (including your controversial decisions no less) in a nutshell and just proves my point. i mean, just look at it. what do you think of it? i don't need this god-awful pathological manipulative nonsense let alone someone with a lot of power over everyone's contributions playing the victim card and crying over about how everyone "just puts me down" like, why must you do this? why? just... why? you're literally irredemable at this point. and i'm well aware of how nitpicky you can get so maybe it's about time you can prove yourself worthy of that denomination if you're really willing to tear this comment apart into a thousand pieces. i mean, it's your project, you dictate how people get to contribute, how code gets a pass because it's not broken or something, how compiled binaries get to be released, etc. you basically treat it like it's your life. do your worst. |
|
Enough. |
Tom's Adventures [Museo del Recreativo]
Ice Cold Beer (ICE) [Museo del Recreativo]