Redefine your "Definition of Done"

[nautic 40]

This game is great and i love it. I wish it would continue to be a great success.

 

Yet, from an outside perspective i got the impression that your dod (=definition of done) is far too sloppy. The amount of issues and the kind of issues that you have, leave and introduce while continuously adding more and more content is too large in my opinion. Many of the bugs are so trivial to find, that i wonder if the developers even tested it, or if you just intended to ship with bugs. Yes this is alpha, and we get to see what the QA gets to see. But in my opinion, in a professional software project, QA should barely find anything. Missing content is fine, but broken functionality is not. And i'm really worried, that by adding more and more content, without getting rid of all these bugs, but introducing new problems, they will become too much of a burden. Harming your image, your progress and your success. I'd love you to introduce better developer testing, maybe a Test-First approach, and to increase the iteration-lifetime, if you had to. A company that releases early access alpha to the public should have a pretty high quality software development process in order to keep their image clean, and their success safe. Well, that's my opinion at least.

 

Maybe you think i'm too harsh, and i'm alright with that. This is not about me, because i like and enjoy the game at the current state despite of all it's issues. I'm more worried about people that aren't as forgiving and patient as me, and i'm also worried that your projects gets into trouble because of the continuously growing burdon of problems, and technical debt you introduce.

 

I wish you best of luck and a great success.

 

A Dayz addict.

 

Ps: I'm friendly, don't shoot!

[ChristmasLunch 52]

Well actually, the two have virtually nothing to do with each other.

There is a team of people at Bohemia who's objective is to work on new content in one area, and there's another team whos' job is to fix existing content.

There's a team that work on modelling and texturing bicycles and there's a team working on optimizing code for smoother game-play and more zombies.

The two are 'independent' of each other, in a sense, and therefore release updates of separate things at different times, and are sometimes co-ordinated.

 

Don't worry, not all the firepower is going toward adding new things. An equal amount of power is going into fixing things too.

 

Well, at least, that's what I understand.

[Geckofrog7 1168]

I'll keep it short and sweet:

 

They can't fix bugs without first introducing the items.

Everytime they fixed a bug, it'd just reappear after they added something new.

[quantum2k6 348]

imaging, they fix all bugs right now but remove alot of this "fixed" content before release because they had better ideas...

 

so they dont spend alot of time for useless bugfixes...

 

i think its ok for alpha imo. we just testing content atm.

 

love this game too :)

[Awruk 216]

YOU are a tester. They just implent new features and release. You actually paid 29$ to test an unreliable, clunky alpha game, enjoy.

[SFRGaming 718]

This game is great and i love it. I wish it would continue to be a great success.

 

Yet, from an outside perspective i got the impression that your dod (=definition of done) is far too sloppy. The amount of issues and the kind of issues that you have, leave and introduce while continuously adding more and more content is too large in my opinion. Many of the bugs are so trivial to find, that i wonder if the developers even tested it, or if you just intended to ship with bugs. Yes this is alpha, and we get to see what the QA gets to see. But in my opinion, in a professional software project, QA should barely find anything. Missing content is fine, but broken functionality is not. And i'm really worried, that by adding more and more content, without getting rid of all these bugs, but introducing new problems, they will become too much of a burden. Harming your image, your progress and your success. I'd love you to introduce better developer testing, maybe a Test-First approach, and to increase the iteration-lifetime, if you had to. A company that releases early access alpha to the public should have a pretty high quality software development process in order to keep their image clean, and their success safe. Well, that's my opinion at least.

 

Maybe you think i'm too harsh, and i'm alright with that. This is not about me, because i like and enjoy the game at the current state despite of all it's issues. I'm more worried about people that aren't as forgiving and patient as me, and i'm also worried that your projects gets into trouble because of the continuously growing burdon of problems, and technical debt you introduce.

 

I wish you best of luck and a great success.

 

A Dayz addict.

 

Ps: I'm friendly, don't shoot!

With this game being open alpha, you're not going to have a smooth game whenever you get it. People who do internal testing for larger game developing companies see all the bugs and glitches. What you get is a (hopefully) smooth and ironed-out game. But DayZ has an open alpha, so the community gets to see all the things wrong with the game. Plus, games like Battlefield have a specialty engine built just for the battlefield franchise and is engineered to do what the game needs to do. In DayZ, we have taken an engine from another game that had superior rendering capabilities, and made a 225 km2 large zombie survival simulator. When doing things like that, you're always going to come across problems. In my honest opinion, I actually would have expected more bugs from this game. A lot more.

Edited February 24, 2014 by Shadow134

[Wolfguarde 108]

Yes this is alpha, and we get to see what the QA gets to see. But-

 

Not trying to be an ass in doing this, but that's the thing... you (and a great many other players thinking along the same lines) are probably used to beta testing for other games - where most of the features are in and integrated smoothly, with most of the critical bugs ironed out. Alpha is completely different - they're exposing a stage of the game's development most developers would not dare to show to the public, in order to speed up testing drastically and get some kickstart funding to keep the ball rolling. The features are being added in so that a basic framework can be established before the whole system begins the massive clusterfuck of debugging; they want everything on the table before they start sorting the mess out, so to speak. We're all used to beta/early release testing... not a lot of games have had open alphas, so bear in mind that it's a very different stage of development, and you're being shown the building progress that happens at a bare-bones level. And remember that because they have opened the game to the public during its alpha phase, they're going to be working their asses off constantly to fix/manage critical bugs in order to keep the game playable for the people testing it.

Edited February 24, 2014 by Wolfguarde

[nautic 40]

With this game being open alpha, you're not going to have a smooth game whenever you get it. People who do internal testing for larger game developing companies see all the bugs and glitches. What you get is a (hopefully) smooth and ironed-out game. But DayZ has an open alpha, so the community gets to see all the things wrong with the game. Plus, games like Battlefield have a specialty engine built just for the battlefield franchise and is engineered to do what the game needs to do. In DayZ, we have taken an engine from another game that had superior rendering capabilities, and made a 225 km2 large zombie survival simulator. When doing things like that, you're always going to come across problems. In my honest opinion, I actually would have expected more bugs from this game. A lot more.

 

This is also what i thought. I think that the fact, that bohemia is building on top of the arma 2.5 engine, which is probably a big ball of mud / legacy code with lots of issues, it's sometimes extremely hard to make it "fine". I can really imagine that this is very hard. But other than that i think that there are many other trivial issues that are not only related to the engine. I don't know. that's just how it appears to me. I also mean little things like: Splitting ruined bullets only leaves one ruined bullet. Ruined can opener opens cans. Duplicate my gun when i log off while having two guns. Can run around with free hands after logging in with 2 guns. Can't put something in my hands if it wasnt in the inventory first. Can't put a 4slot item into 4 free lots sometimes? Can move while relogging. These are just a few for example. And i am not talking about glitch issues, like glitch through wall here, which i am quite sure are engine related. I'm talking about the small things.

[nautic 40]

Not trying to be an ass in doing this, but that's the thing... you (and a great many other players thinking along the same lines) are probably used to beta testing for other games - where most of the features are in and integrated smoothly, with most of the critical bugs ironed out. Alpha is completely different - they're exposing a stage of the game's development most developers would not dare to show to the public, in order to speed up testing drastically and get some kickstart funding to keep the ball rolling. The features are being added in so that a basic framework can be established before the whole system begins the massive clusterfuck of debugging; they want everything on the table before they start sorting the mess out, so to speak. We're all used to beta/early release testing... not a lot of games have had open alphas, so bear in mind that it's a very different stage of development, and you're being shown the building progress that happens at a bare-bones level. And remember that because they have opened the game to the public during its alpha phase, they're going to be working their asses off constantly to fix/manage critical bugs in order to keep the game playable for the people testing it.

 

Yes, i see truth in your statement. Nevertheless it makes me worried to see the kind of issues i'm seeing in this particular Alpha, because i believe many of them are avoidable, and should be avoided, even in alpha. I've experienced other early access alpha programs (not games in particular) that weren't nearly as unstable as this.

[Korsbaek 1778]

I'm going to have to tell you something. This game is not done. Not yet. 

 

I don't think it will ever be. The devs will continue development even after full release, and then they'll abandon it at some point.

[nautic 40]

I'm going to have to tell you something. This game is not done. Not yet. 

 

I don't think it will ever be. The devs will continue development even after full release, and then they'll abandon it at some point.

 

Thats the nature of software. It will never be done.

 

Yet, we try to define when something is done for every iteration, in order to increase it's overall quality. We do this by defining requirements and positive, but also negative test-cases to avoid communication errors.

https://www.scrum.org/Resources/Scrum-Glossary/Definition-of-Done

[entspeak 374]

They don't need to redefine their definition of done; they fully acknowledge that it's NOT done. What you're describe would best be defined as a beta - that seems to be what you want them to give you. So, I think rather than BI redefining their definition of done, you need to accept the definition of alpha.

Experiencing bugs, crashes, bugs, bad performance and... oh yeah, big huge game crashing bugs are part and parcel of testing an alpha - which is, essentially, what you are doing when you play this early-access game.

Public alphas are a relatively new phenomenon, so it's understandable that people have a misconception about what the experience should be; but, you can't reasonably ask that an alpha be anything other than what it is.

Edited February 24, 2014 by entspeak

[nautic 40]

They don't need to redefine their definition of done; they fully acknowledge that it's NOT done. What you're describe would best be defined as a beta - that seems to be what you want them to give you. So, I think rather than BI redefining their definition of done, you need to accept the definition of alpha.

 

I also acknowledge that it's not done. i'm just worried of the amount, and the kind of halfhearted features i get to see. There is no real definition of alpha, but an alpha usually is the first version of a program that is to be tested by the QA. So, it usually is what the programmers hand over to the testers when they think they are done with all their tasks of the current iteration. And i think that they could drastically improve this 'done'. 

This is quite hard, since there is a communication gap between designers and developers. designers think the obvious, but the obvious is not as obvious in a programming world. This gap could be closed by acceptance tests, for example. Also, the developers could be way more strict themselves in terms of 'when to hand over a feature'. Of a good programmer, i expect a feature to be rock solid when he hands it over to the QA. It might contain some bugs, but they should be really hard to find. QA should be more about defining when something is done in the first place. Well, this is a basic issue that the development sector is facing atm.

 

I believe they knew how bad it was, but they had to take the risk to release the alpha in order to get the money to continue. But now that they have sold so many copies, i think it's time to step up.

Edited February 24, 2014 by nautic

[ploulaf 89]

the thing is they add stuff on broken foundations.

At one point they will have to start it all up again to fix some major issues that still aren't fixed yet. And they will have a hard time doing that because they kept adding and adding more to it instead on focusing ALL the efforts on having a stable ground.

The engine itself is flawed. Once they'll change it they will have to start all over again.

[dallas 5195]

If you're looking for done, you're looking the wrong place.

 

I never see BI games as done, to me they are nothing more than user content templates and while BI delivers the platform, the users develop the content. DayZ is the ultimate user content, so popular that it became it's own standalone game, still like all other custom content, it's not a static thing, it's something in motion and constant development. If you're looking for a finished product, you're going to be unhappy with ArmA/DayZ. If you're looking for a developing experience, you're at the right place.

[fing24 36]

Like all things it is lack of communication. We report the bugs and some are big ones yet after a few updates they are still there.  We don't know if they are fixing it or ignoring it for the moment.  I except there are going to be bugs in the game I have no problem with that, but it would be nice to see more of them being ironed out so when things are added later it might be easier for them to fix things. The game was very good at the start and had the odd few issues, but as time has gone by more bugs are creeping in and it seams that only a few are being fixed. Take loot spawns at the military camps, they know about it, but have not fixed it as of yet.

[entspeak 374]

Some bugs are harder to fix than others. Because it's a public alpha, they have to balance holding off an update for a particular big fix with the constant call for more updates from players. The way it seems to be working right now - they seem to address bugs that kill the ability to play before things like loot spawns in some buildings. Also, sometimes bugs seem fixed in internal testing, but once it goes to experimental, they crop up again. It is a balance of priority, how long it will take to fix the problem and how badly the bug impacts gameplay.

[entspeak 374]

I believe they knew how bad it was, but they had to take the risk to release the alpha in order to get the money to continue. But now that they have sold so many copies, i think it's time to step up.

Yes, they knew exactly what the state of the alpha was. That's why Rocket handled the release the way he did. That's why there's a warning that pretty much let's everyone playing know the state of the game. You can speculate about the funds BI may or may not have until the cows come home. But it was felt that the there was a game there to play - even if it was incomplete.

Not all software is the same and public alphas for games are a completely different beast. What you're describing about the testing process is exactly what happens. The game is in development, they put features in and explain that some are placeholders, some don't work properly... it's the stuff they don't mention that you come across that are the bugs. That's how development testing works. This isn't the type of alpha where they completely finish the game and hand it to you to see if it all works. That's not what early-access is about.

"Time to step up..." I don't know what you're talking about. You want a beta, that's what you're asking for. This isn't ready to be a beta. It was made clear to everyone where this game was and they've been pretty good about keeping people apprised of progress via changelogs. My suggestion to you would be, if the progress they are making doesn't work for you, wait til later to play. Progress for the development shouldn't be based on the number of people willing to play the game in its current state. Nobody's been misled. Caveat emptor.

[Wolfguarde 108]

I also acknowledge that it's not done. i'm just worried of the amount, and the kind of halfhearted features i get to see. There is no real definition of alpha, but an alpha usually is the first version of a program that is to be tested by the QA. So, it usually is what the programmers hand over to the testers when they think they are done with all their tasks of the current iteration. And i think that they could drastically improve this 'done'. 

There is that. But the thing is, as I see it, they've pushed for an early release when they weren't ready for it - and are now having to pump out features to an audience that's just playing the game like it's an open release and expecting the same sort of release pattern you would expect in a finished game. They're responding to the pressure of demand to keep their testers interested. I agree entirely that it's not the best way to handle a release - they shouldn't have released it at all until they'd gotten most of the alpha stage done, in my opinion - but they set a release date that they could not meet and were running a long way overdue. It was probably a calculated risk to release it as early as they did, and now they're running themselves ragged trying to keep up with the pace of development we expect from them.

It could as easily - or also - be a lack of experience handling an alpha release, since it's not really something you're going to be ready for without having done it before. In my mind, I can easily see it as being like standing in front of oncoming traffic. You can research it, learn as much as you possibly can about what is going to happen, but nothing's going to prepare you for the moment of impact.

Probably a really macabre analogy, but you get the drift of what I'm saying.

Edited February 25, 2014 by Wolfguarde