 In the last video, I said I was going to cover the case of a specific individual. Because most people understand where I'm coming from with this. If it's public, I mean cover it fine. I was talking to this individual, and he requested that I do not, and that's fine. I understand his reasonings. We'll just leave it be. However, I would still like to cover the content that I was going to cover. So we're just going to go over that in a more generalized way. This will be the next in the series of my issues with Edakor. And we're going to specifically be delving into how they approach and deal with the community. Like I had said in the last video, I and I think most people understand why if you're paying for commercial support, you get priority support. That just makes sense. That's what you're paying for. The compiler is already free. You could argue that it's partially licensing things, but considering you can get it with the GM GPL exception for the runtime and avoid that by going through the FSF, then it's not really an issue. So you're strictly paying for that commercial support. And that's fine. Where we run into problems is what seems to be thinking your community isn't really important or is somehow like a burden. And I don't think they intend to come off that way. I don't think that is their actual beliefs, but it comes off that way. You see, it's not your intent that matters as much as how things are perceived. That's why normally you have people that are actually trained in PR to deal with these kinds of things. Not using your developers for that purpose. Your developers are developers. Let them develop. I mean, we are dealing with people, myself included, we tend to have some social oddities, whether it's just being a little different in opinion or something that you actually see a lot where they're very critical but in a often intending to be helpful way. There's a name for that. I don't remember specifically what Adam Grant does. Is that his name? No, that's somebody else. No, maybe that is. No, because I think I'm mixing Adam Grant and Adam Ward up. No, I think his name is Adam Grant. He has a great talk on this and he doesn't delve specifically into how it's often developers that are one of them, but he has a great characterization of them and the problems that they tend to encounter and it's like an exact match to a lot of developers. I'll be adding in the screen caps from various GitHub issues from their stuff. It's been quite nice since they've put all their stuff up on GitHub so that you can actually see these interactions because before you were just kind of reliant on hoping people save their emails, which doesn't usually happen. A lot of the time, it's just part of the typical personality, part of that stereotype, developers typically aren't going to make a big social thing of it when it happens. But we have these things being more public now and transparency is good, allows us to hold things accountable and that's pretty much what I'm trying to do here. One of these is actually, no, we have several of these. I'll be my best to find a few of them, but there's several of these instances where Edakor will get a bug report and they'll say that they haven't fixed it but are looking into it but will close the issue. Now the thing with that is ideally you want a community that supports each other. We are talking about open-source software here. You want to take advantage of that. That's the whole point of really the free software movement specifically and we are dealing with stuff that came out of the FSF and Edakor is basically just providing commercial support for that stuff. When you take one of those issues where it's still an issue, I'm breaking this up in post a little bit because I found something that really warrants more than just a simple little snippet of text down at the bottom. There's quite a bit more I want to say about it and honestly I don't know where to put it in the video so we're just inserting it with the other ones. This one was really disheartening to see for a lot of reasons. Simon Wright happens to be an absolutely wonderful member of the community. He is remarkably helpful and goes way out of his way to actually try to help people. He goes above and beyond what I would really expect out of anybody when it comes to help. He's an all-around wonderful guy, a very knowledgeable individual. If you've ever interacted at all on Stack Overflow asking Edakor questions, you've come across him. He's regularly helping out. This one's a particular problem because not only is it again a case of them just up and closing the problem, not having fixed it. But it also demonstrates something that I really, I mean it doesn't surprise me honestly given the quality of a lot of things but I don't really know what to say so I'll just cover exactly what the issue is here. Look at the difference between Seton's and Wright's understanding of GDB and it's particularly interesting to see Wright respond that way because usually he's not that blunt. Usually he's a lot more tactful so that's surprising and I think sort of says a lot. But yeah, GDB is kind of a very important part of your tooling and honestly the fact that you're not very familiar with it really goes a lot to why your stuff is so buggy. Your debugger should be something that you are intimately familiar with because you should be using it regularly in fixing your code and why are you not especially something so tightly integrated with the IDE just dumbfounded. Somebody looking to help out, somebody going you know I'd really like to contribute to this product because I'm passionate about it, I'm passionate about the community and they go through and look at those GitHub issues to see something they can contribute to. When you've closed that despite it still being an issue, they're not going to see that and they're not going to be able to fix that for you. Alright, I need to add something in post again so hopefully I find a spot that isn't too interruptive. When I was looking for cases of things that were prematurely closed and really shouldn't have been because there are things that you're really going to want community contributions to when possible. Oh boy. I was really hoping when I had addressed this before that it would still be an issue but it is. We have the case of Joe Marino, I believe is his name, I've seen him quite a few places though and he's bringing up the fact that Lancet is for Python 2.7 only which is quite old by this point, you know the Python 3.0 rollout was a while ago and he's bringing up back in July of 2017 that Python 2.7 is nearly end of life and you know in his opinion no-do development should be done using Python 2.7 which I agree with, if it's a legacy thing that you know isn't really going to change it may be fine to leave it there but you should really consider migrating to the newer stuff just because it's end of life. It's not getting any more bug fixes and that's the big issue. You're not upgrading for features, you are upgrading so that you can still get bug fixes and that's actually really significant. And I'll kind of just let the screenshots speak for themselves but I've said a few times now that just sort of the Aida vendors in general, the whole ARG really treats Aida as more of a hardcore legacy compatibility language not a high reliability language and you're really seeing some of that creep through here where they just, they clearly believe that this is okay. And they're trying to argue that they're going to get this out before 2020 and we're got slightly over a month left and it's not in sight. So yeah, Python EOL is still 2020, that deadline's coming real soon. Yeah just how sure of himself that there's going to be another extension and that this isn't something that you need to worry about but we are talking about dead deprecated software that is getting no more fixes. No more bug fixes. A company that claims to produce highly reliable systems that are highly reliable software stuff to develop highly reliable systems that they stress that that they emphasize the living hell out of that and yet not alarming to them. Oh my god. Now I'm under no disillusionment about how this stuff works overwhelmingly. Your open source projects are developed by a handful of people, sometimes even just one or two people even when it's sponsored by a major company. Your contributions don't happen that much. But when you set things up this way, they're not going to happen. Furthermore it sets up sort of a false sense of quality. Your closing an issue and making it look like there are less current issues going on than there really are. And I hope that's not the intention here. I certainly can't prove that that's the intention. But you are making it look a little like that's what's going on. I hope it's not. But that's what it looks like. Well I gotta add in even more. This is getting just kind of ridiculous at this point. I'll say this now. This was supposed to be like a 18 minute long video that should have taken about an hour, hour and a half of editing and then just upload. Many times much closer to 30 minutes now because of things I've needed to add. And the total amount of work I've put into it is right about 40 hours at this point because there is so much of just a rabbit hole and I'd like to cover this entire issue at once. But oh boy. I wound up originally having a place to put this in the video and due to the things I've needed to add, otherwise I sort of lost the place for this. So this edition largely serves that purpose. As I was talking about, there are quite a few cases of Edakor just closing the issue when it's not fixed yet. And where I'm adding this, you would have just seen two cases of issues being closed and sure as shit. The issue is still there. This is yet another reason why you don't do that. See it's not just your community members who are not going to be aware of the issues. It's your own employees who are not going to be aware of those issues. And it does seem like you use some internal tracker, but you are still introducing an opportunity for error and that error is still obviously happening. I'm not sure what you're using internally, but speaking of somebody who does use a different internal system from GitHub, once a project receives a certain amount of attention, I do start to put those issues up on GitHub instead for like enhancement requests or whatever that I have planned. Just because since it's getting attention, that makes sense then, but some of the stuff will remain on my internal system. And what I do is I have, I go through Azure DevOps. They implement this. I know a few others implement this, but I don't know what Edakor is using internally. Then I have an internal thing. I can link to those GitHub issues and it will track them along with the internal one. I'd really suggest doing that and if your system doesn't migrate to something that does, please, again, these are not just issues that affect community members. They are issues that affect everybody using these products. You're clearly forgetting about them. You're not letting community members who may be willing and able to help fix them. Which makes things easier on you guys. You're not following up. It's just nothing. I just get closed and abandoned, but they're still issues. And lastly, I want to add that while I've been going through these, I have been finding these cases of closed issues that they either said they were fixed or just didn't follow up on at all. And I'll test them on the newer version of NAT Community Edition and I've replicated quite a few of them. Come on. Another issue I see happen quite often and I really wish it didn't is somebody will report an issue and for whatever reason, and I'll say some stuff about this in a second, there just isn't much in the way of diagnostics or error logs or ways to reproduce it. And so what Iacore will do is just outright close the issue. There's a number of problems with that. And I tackle this by bringing up the worst case scenario and just, well, you'll figure out where this is going as I go along. You want new developers. You want to spread the language. That's not just a tech evangelism thing. That's also a business thing. More customer, more experienced ITA developers means more potential customers. Not just directly through them, but also you have more businesses that are willing to rely on ITA because they can find those developers. And what happens at least in a few cases is you have somebody who is a newcomer, a newbie, whatever you want to call them and they're new to this. They really don't know about how to make a good error report yet. They don't know the kind of information that you need to help the debugging along. Ask them for that information. If they're not willing to provide it, if they just kind of be a little bit of an ass, I would still say leave it open. You didn't fix the problem. It was still a problem. Would look into it. Maybe create a special GitHub tag for, you know, this needs further investigation. We don't know what's going on here. We don't know if this is a real problem or not. But at the very least, it was reported. There are people, and I'll say the type of people first and then I'll explain, but there are people who don't necessarily have a whole lot of programming knowledge but have a great amount of knowledge when it comes to investigating why these problems happen. And they've been able to do it to immensely larger systems than just an IDE or just a compiler. There are people who don't even need the source code. They'll go through it with the bytes, not assembly with the bytes and will find the problem. You're seeing this a lot and there's some controversy over it so you can definitely find some stories. But you're seeing this a lot with some big games right now. The big one that you see the controversy with is Fallout 76, where people have been going through with cheat engine and finding, because it's cheat engine, it's not a disassembler. It's literally just a memory analyzer that shows you the byte at that address. They're still finding how these bugs and exploits work. The controversy comes from Bethesda banning them instead of treating them like the helpers that they really are. But the point there is that if you have this stuff up, if you have it so people know about these problems, you can get some help. Maybe it's not a programmer, but it's an experienced, I guess we'll call them debuggers, because that's really all they do. I'm not trying to be mean there. You are actually doing some amazing work, those of you who do that. I'm seriously impressed by your ability to figure things out without the source code. But they can help you. And when you just close the issue, you're saying that it's no longer an issue. And again, they don't know that it's still a problem and they can't help you. Let them help you. So as I had said, it is a problem that they just close the issue and don't ask. If we are dealing with, and I know at least a few of these cases are newcomers, they don't know what kind of information needs to be provided. When you're providing good technical support, you want to guide them in how to get that information. If you ask and they say they don't really know how to provide that stuff or they don't really know what you're talking about, then direct them with how to get error logs or how to have reproduction steps. If nothing else, how to take a screenshot of the problem. And sometimes that's enough. You know, I've been able to fix bugs and things that I've developed for other people literally just based on a very rudimentary explanation of what the problem was, no steps to reproduce it at all. And you can honestly usually find where the bug is and still get that fixed. Sometimes it might take a bit of work. Sometimes it might take actually being present there when things are going on and trying to get that idea. But you really can still figure it out in a lot of cases. Don't just straight up close the issue. Why this is a particular problem though is again when we go back to these newcomers. I've seen two big things really happen. One and it's not as bad as the other, but it's still not what you want to see happen because you want to grow the community, right? As they look at it as boy, this is the most popular item vendor and this is the quality of their products and this is the quality of their customer support. I definitely don't want to buy from them. And even though if you bought from them, you would get better customer support. Think about that image. It doesn't build a whole lot of confidence. Where else we get into this as a problem is, and I've seen this a few cases, not as much, but is when they come to a slightly different conclusion. And it's not necessarily a wrong conclusion. This software is developed in Ida and you're saying that Ida is meant for highly reliable systems and just because of how it's designed makes bugs almost impossible. I mean, that's touted a lot by the community. So if this is so buggy, then isn't that a bit of a marketing lie and how much of all Ida code is just horribly buggy? No, I say that's not necessarily the wrong assumption because your code quality ultimately comes down to two things. Your overall engineering approach and how you as an organization or individual approach and handle bugs. And I think the real issue here is just a slight adjustment of that, of what these disenfranchised newcomers are coming to. There's not that all Ida code is bad because it's definitely not, but that an organization that repeatedly has a bad bug handling, it's not really surprising that their code is buggy. I'm going to finish this up by just saying that your community is a lot more important than you're treating them. I don't want to assume what you think here, but a lot more important than you're treating them. You're not treating bug reports that come from these individuals seriously. And the thing is, we're all using the same software. These bugs that people encounter are not just affecting them, and they're certainly not just affecting non-paying community addition people. They are bugs in the software. By not taking them seriously just because who they are coming from, you're leaving these bugs in your commercial products as well. Remember, the big selling point of open sourcing your products is that you can have the community help out. And that's not just being little code monkeys for you, which is unfortunately what I think a lot of these companies sort of treat it as a want out of it. They're going to be people that help you with diagnostics, that help you with feature suggestions. Maybe not the implementation of them, but at least with the ideas. And they're going to be people that report bugs. That's not an attack on your product. They wouldn't report the bugs if they didn't want to help you. If it wasn't an attack on your product. Well, they'd talk about how bad it is, but they wouldn't report the bugs. They wouldn't try to get these fixed. Your community matters. And while I'm not sure whether you respect them or not, it doesn't look that way. It doesn't look that way.