Philosophical Discussion on Bugs vs. Releases

[Jonathan Fong]

On 8/4/2016 at 6:00 PM, patrickbc said:

At this time Id choose the buggy, and have the bugs fixed in hotfixes and updates 

I'm tired of waiting

Keep in mind I'm talking about major bugs, and by major bugs, I mean things that will ruin your experience with the plane. Things like installers that corrupt your FSX/P3D installs, forcing a complete reinstall of the simulator and all installed addons, or bugs in the plane causing it to, say, climb uncontrollably upon descent without any hope of pilot recovery bar reloading the sim and hoping for the best.

Guess what? These are all bugs found in release versions of several flight sim addons. In fact, the latter one was a bug that occurred with FSLabs's own Concorde X v1.3 update, and you know how much people pressured FSLabs to release that quicker.

Would you still pick the buggy release?

[João Silva]

Keep in mind I'm talking about major bugs, and by major bugs, I mean things that will ruin your experience with the plane. Things like installers that corrupt your FSX/P3D installs, forcing a complete reinstall of the simulator and all installed addons, or bugs in the plane causing it to, say, climb uncontrollably upon descent without any hope of pilot recovery bar reloading the sim and hoping for the best.

Guess what? These are all bugs found in release versions of several flight sim addons. In fact, the latter one was a bug that occurred with FSLabs's own Concorde X v1.3 update, and you know how much people pressured FSLabs to release that quicker.

Would you still pick the buggy release?

Indeed, what's the logig of having something that doesn't work properly. We don't want another airbus crashing after some minutes.

Enviado do meu 7043K através de Tapatalk

[peter kelberg]

just to add my 2  cents worth can some name  a addon  that's been released  that hasn't  had  some  bugs and  had  to be  fixed by updates  cant  see  any difference to   this when its released

[Jonathan Fong]

2 hours ago, pete_auau said:

just to add my 2  cents worth can some name  a addon  that's been released  that hasn't  had  some  bugs and  had  to be  fixed by updates  cant  see  any difference to   this when its released

It's practically impossible to release with no bugs whatsoever, but tell me, what's the difference between having a plane that you can't use due to major bugs installed on your computer and having no plane installed at all? In both cases, you can't use the plane, and you'll be pestering FSLabs one way or another - either to update the software to fix the bugs, or to release it already.

[David Porrett]

2 minutes ago, fcisuperguy said:

It's practically impossible to release with no bugs whatsoever, but tell me, what's the difference between having a plane that you can't use due to major bugs installed on your computer and having no plane installed at all? In both cases, you can't use the plane, and you'll be pestering FSLabs one way or another - either to update the software to fix the bugs, or to release it already.

Isn't beta finished though?

[Jonathan Fong]

Just now, DavidP said:

Isn't beta finished though?

Don't ask me, I'm not a developer. My point still stands, though. Think about it - why would a major addon developer hold the release of an addon for so long if there were no major bugs left, especially considering the amount of hype surrounding the release? Surely they must be doing something, and the only 'something' they could possibly be doing at this point is fixing bugs. Remember, bugs can exist anywhere and everywhere, from the plane's flight dynamics to the installer and uninstaller. They may have finished the beta period of the plane itself, but there may be issues with external programs that interact with the plane in some way.

[patrickbc]

9 hours ago, fcisuperguy said:

Keep in mind I'm talking about major bugs, and by major bugs, I mean things that will ruin your experience with the plane. Things like installers that corrupt your FSX/P3D installs, forcing a complete reinstall of the simulator and all installed addons, or bugs in the plane causing it to, say, climb uncontrollably upon descent without any hope of pilot recovery bar reloading the sim and hoping for the best.

Guess what? These are all bugs found in release versions of several flight sim addons. In fact, the latter one was a bug that occurred with FSLabs's own Concorde X v1.3 update, and you know how much people pressured FSLabs to release that quicker.

Would you still pick the buggy release?

Indeed major bugs is not something I want. But major bugs would probably be easy to find (because theyre well, major) and with a beta which has taken many many months, they should have fixed any major bug long ago.

[Jonathan Fong]

23 minutes ago, patrickbc said:

Indeed major bugs is not something I want. But major bugs would probably be easy to find (because theyre well, major) and with a beta which has taken many many months, they should have fixed any major bug long ago.

See, this is another programming assumption which causes such unrealistic expectations.

Considering the complexity and realism which FSLabs claims the A320X to have, surely they must have implemented a massive amount of variables to simulate the aircraft in such a way that it mirrors real life enough for it to be used as a training method. Therefore, to TEST for bugs, the FSLabs team would have to run tests on most, if not all, combinations of said variables to weed out bugs, and considering how many variables there actually are, that would take some time, no?

You say major bugs should be easy to find. In that case, why are there so many cases of programs being released with bugs that rendered the software in question unusable? 'Major' bugs only implies that the bug causes an issue so severe that the software can't be used at all, not necessarily that it can be found and replicated easily.

Once again, I'll use the example of the Concorde X v1.3 update - the issue where the aircraft would suddenly begin a steep uncommanded climb at a certain point in the descent without any chance of pilot recovery was experienced by many users of the software on many different systems; yet, despite the fact that the replication steps were so simple and straightforward (during a normal flight, the issue occured just after reaching subsonic speeds), the bug wasn't identified and remained undetected until pointed out by several users.

[patrickbc]

1 hour ago, fcisuperguy said:

See, this is another programming assumption which causes such unrealistic expectations.

Considering the complexity and realism which FSLabs claims the A320X to have, surely they must have implemented a massive amount of variables to simulate the aircraft in such a way that it mirrors real life enough for it to be used as a training method. Therefore, to TEST for bugs, the FSLabs team would have to run tests on most, if not all, combinations of said variables to weed out bugs, and considering how many variables there actually are, that would take some time, no?

You say major bugs should be easy to find. In that case, why are there so many cases of programs being released with bugs that rendered the software in question unusable? 'Major' bugs only implies that the bug causes an issue so severe that the software can't be used at all, not necessarily that it can be found and replicated easily.

Once again, I'll use the example of the Concorde X v1.3 update - the issue where the aircraft would suddenly begin a steep uncommanded climb at a certain point in the descent without any chance of pilot recovery was experienced by many users of the software on many different systems; yet, despite the fact that the replication steps were so simple and straightforward (during a normal flight, the issue occured just after reaching subsonic speeds), the bug wasn't identified and remained undetected until pointed out by several users.

Surely I don't know much about programming, but when I say 'major bug' I mean something:
1: Most people experience it (>30%)
2: Happens during most flights (>40% of flights)
3: Makes the flight impossible to complete

30% x 40% = >12% of any randomly chosen flight should be impossible to complete.

This is in my opinion a major bug, and should be discovered very early in BETA if not in ALPHA. And no matter how complex the aircraft, I doubt I'll never take 'months' to fix one or more 'major bugs', and if that'd be the case, it wouldn't be difficult to describe in an update.

A minor bug (e.g. one which 10% experience, 10% of the time and renders the flight useless) and even some of them, is not something which would ruin the aircraft for me, so in that case, give me the aircraft and fix that later. 

[Jonathan Fong]

Just now, patrickbc said:

Surely I don't know much about programming, but when I say 'major bug' I mean something:
1: Most people experience it (>30%)
2: Happens during most flights (>40% of flights)
3: Makes the flight impossible to complete

30% x 40% = >12% of any randomly chosen flight should be impossible to complete.

This is in my opinion a major bug, and should be discovered very early in BETA if not in ALPHA. And no matter how complex the aircraft, I doubt I'll never take 'months' to fix one or more 'major bugs', and if that'd be the case, it wouldn't be difficult to describe in an update.

A minor bug (e.g. one which 10% experience, 10% of the time and renders the flight useless) and even some of them, is not something which would ruin the aircraft for me, so in that case, give me the aircraft and fix that later. 

Finally, you admit that you don't know much about programming.

First, you assume that bugs should be discovered early in development. That is not always the case in programming. Keep in mind the code of a product is always changing throughout development. The introduction of code in one part of the software (possibly even to fix another bug) can cause a new bug to pop up. That's not the developer's fault - it's just the way developing complex software works. You can catch new bugs anywhere from alpha to late beta or even with the release candidates.

Second, from what I can see, you are operating under the assumption that FSLabs knows about each and every bug currently present in the A320X release candidate software and, by extension, how the A320X interacts under every possible environment. If they did and they could reliably claim that you would only experience a bug, say, 0.01%, of the time, an early release might actually work because the actual likelihood of bugs occuring would be known to be too low to matter. However, they do not have such information - no developer can do such a thing. Their beta team annd developers couldn't possibly have every possible environment that the A320X could be run in available to test in. Therefore, you are asking them to release software that they know is buggy, but that they don't know the full extent of the bugginess. What may seem bug-free (or minimally buggy) to them may be completely flight-destroying to everyone else (once again, the Concorde X v1.3 update comes to mind - they most certainly didn't catch that bug in testing).

[Lefteris Kalamaras]

Gents,

I am in a good mood today (no specific reason <cough cough>), so I split this discussion into its own topic in the General forum - it had less relevance in the A320-X Release Road Map topic.

Feel free to continue in a respectful manner (a few sentences up there were borderline Jonathan) and I won't lock it as irrelevant, because, after all, it is in the General forum.

[Thomas Antonovich]

2 hours ago, Lefteris Kalamaras said:

Gents,

I am in a good mood today (no specific reason <cough cough>), so I split this discussion into its own topic in the General forum - it had less relevance in the A320-X Release Road Map topic.

Feel free to continue in a respectful manner (a few sentences up there were borderline Jonathan) and I won't lock it as irrelevant, because, after all, it is in the General forum.

Sooo.... Does this mean what I think it means? Don't go breakin' my heart Lefteris! ;-)

[Romain Roux]

3 hours ago, Lefteris Kalamaras said:

I am in a good mood today (no specific reason <cough cough>)

Hello Lefteris,

That's mean of you to create allusions where there is no reason! Or it there? (Rethorical off course)! 

[MoeGhaziri]

21 hours ago, Lefteris Kalamaras said:

Gents,

I am in a good mood today (no specific reason <cough cough>),

Have some tea with lemon and ginger Lefteris, it will cure your coughs