Board index Other Forum

Forum communication

Questions about the forum itself, suggestions or issues with the forum software.

Re: Forum communication

Postby radi » Tue Apr 22, 2014 7:11 am

I took one of Thorsten's posts and drafted a short wiki article. It is meant to be given to, umm, "unwilling" users. I think it's a good idea to keep it rather short.
OSM buildings for LOWI, EDDC
Custom scenery for VHXX YMML
Edit .stg via the FG Object Placement Tool
radi
 
Posts: 643
Joined: Mon Aug 25, 2008 4:24 pm
Location: YMML, EDDC

Re: Forum communication

Postby Thorsten » Tue Apr 22, 2014 9:14 am

Since forum moderation apparently does not subscribe to a), b) or d) and a three strikes approach does not seem viable in practice, I will stop dealing with bug and problem reports in the forum till a reasonable mode to address those can be found.
Thorsten
 
Posts: 11381
Joined: Mon Nov 02, 2009 8:33 am

Re: Forum communication

Postby KL-666 » Tue Apr 22, 2014 10:29 am

The posts of Thorsten are always good food for thought. They make me stand still and reflect about what we are actually doing, in this case in the forum.

I would like to propose an idea that might be helpful for principle a) Respect for knowledge: Create a group in the forum so that it can be easily seen that someone has deep knowledge of fg. Maybe by colouring the name differently.

This will not enforce respect, but it can help like in:

If just anyone suggests to me to try something "drastic" like reinstalling, i will not be happy if it turns out not to solve the problem. So if i have counter-indications against reinstalling, i would discuss them until i am satisfied they are not relevant.

If someone with good knowledge tells me the same, i more readily just try it, because he is probably right and my counter-indications probably irrelevant.

But it took me a while on the forum to know who i can trust almost blindly for their good knowledge. It would have saved me and the people that tried to help me some time if i had seen immediately that someone is a "top developer" or "serious helper" or something of the like.

It also works for entering discussions (or not). If i see some people i have recognized to have good knowledge in a discussion, i am not likely to enter it, because what i have to say from my little knowledge of fg is likely already evident to them.

Of course it will be hard to define criteria for such a group. But i am confident they can be found. The criteria should prove deep knowledge of fg. Possible components could be:
- having developed something that requires a lot of knowledge of fg
- having helped out people with difficult problems on the forum

Well i'm sure others with more (forum) experience can think of better criteria.

Kind regards, Vincent
KL-666
 
Posts: 784
Joined: Sat Jan 19, 2013 1:32 pm

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 3:16 pm

Thorsten wrote in Tue Apr 22, 2014 9:14 am:I will stop dealing with bug and problem reports in the forum till a reasonable mode to address those can be found.


That's basically what I am doing by laying out my requirements for "proper" support through some separate wiki article, if people are willing to work through such an article and provide the requested feedback, I am all game, and will extend the article over time (usually, inviting them to help with this) - if not, I'll ignore them, simple as that - and it works kinda well for me to "filter" users/topics.

And I've not seen a single instance where using the wiki for this, caused irritation in the sense of flame wars - the usual complaint is that those articles are "too technical" and not user-friendly, but that is something that can be solved over time. Speaking from experience, most "conventional" support threads may require more than 30-45 minutes and 5-6 exchanges until things are "solved".

PS: radi, thanks for doing this - and I'd like to invite everybody else to help improve this article, so that we can use it to explain to newcomers how they should ideally act around here to ensure that their questions are answered in a timely manner: http://wiki.flightgear.org/Forum_communication
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby Jabberwocky » Tue Apr 22, 2014 4:59 pm

Well, not everyone is willing or able to work with a debugger, so the logical consequence is, avoiding to ask questions about whether something is a user problem (wrong button, wrong settings) or an error. Which is totally okay, if FlightGear is intended to run anyway only for a little group of enthusiasts. If not, the question remains, how to inform and get background information about plane and program behavior without mentioning that kind of problems to avoid being jumped for "you want help, so you have to produce all the debugger information and otherwise I call you rude."

P.S. And if you decide now, to kick me from the board for pointing out a basic communication problem, so be it.
Jabberwocky
Retired
 
Posts: 1319
Joined: Sat Mar 22, 2014 7:36 pm
Callsign: JWOCKY
Version: 3.0.0
OS: Ubuntu 14.04

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 5:27 pm

Jabberwocky wrote in Tue Apr 22, 2014 4:59 pm:Well, not everyone is willing or able to work with a debugger, so the logical consequence is, avoiding to ask questions about whether something is a user problem (wrong button, wrong settings) or an error. Which is totally okay, if FlightGear is intended to run anyway only for a little group of enthusiasts. If not, the question remains, how to inform and get background information about plane and program behavior without mentioning that kind of problems to avoid being jumped for "you want help, so you have to produce all the debugger information and otherwise I call you rude."

P.S. And if you decide now, to kick me from the board for pointing out a basic communication problem, so be it.


I don't think that you are getting "kicked" for posting constructive things :D

I am not sure why you are referring to the debugger issue - but we don't necessarily ask everybody to use a debugger, but once someone mentions being 1) a programmer, 2) being a Linux user, asking if running a gdb session would be possible isn't all that far-fetched. There's a ton of information that can be gathered even without using any "developer tools".

I also don't think that you are getting called "rude" here for not being able to provide certain information - but you are indeed usually requested to comply with certain requests, and state clearly if you're unable. There's only so much we can do after all, without things getting fairly technical. And people have been working on features to provide backtraces to developers without end-users having to be aware of the whole process, as in the whole crashrpt thing that Zakalawe implemented - but all this is still in its infancy unfortunately.

Just keep in mind that FlightGear users are not dealing with a finished or "polished" product - things are very much in flux, and people can be part of the evolutionary process here.

FlightGear as a project and community really has a ton of problems, usability certainly among them - but friendliness only gets us so far - we are seeing more support requests than we can deal with, so we really need to "filter" things to see who's even capable of processing our requests. We could spend a whole release cycle providing topnotch support, but we wouldn't be getting anything else done then unfortunately. And others are usually unwilling or unable to provide support up to a certain extent/quality. While many end-users may consider it unfriendly not to have personal 1:1 support, the wiki -and dedicated troubleshooting articles- are the only way to deal with such things at a wider scale.

I am responsible for some of the most technical "troubleshooting" guides that you can find on the wiki, still those are seeing thousands of views within a few months - and frankly, I'd rather deal with people who can actually work with that info, than having to deal with people who barely know how to switch on their computer, or how to run a program via a shell/terminal. It's simple as that - we only have a finite amount of time that we can spend each month contributing here, and we need to determine how to spend it properly, and how we want to spend it.

This is not a forum to teach people basic computer skills, and it also isn't a programmers forum - but it seems natural that "programmers" are easier to deal with for us, and that they're more vital to the project than end-users who want to edit some livery/XML/camera file but fail to work through the docs...
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby Jabberwocky » Tue Apr 22, 2014 6:08 pm

Oookay, lets clear up a few misunderstandings:

Not every Linux user is a hardcore system program on the metal guy. Ubuntu for example has managed to come up with a version also a normal garden variety user can use without all those wierd glitches Windows produces sometimes.

Second, as you may have read already on my homepage, I was a programmer, I was a project manager, I was then on general management level, I earned me a burn out syndrome, a heart attack and a divorve that way and jumped the boat. So, I haven't programmed C++ for example in a while. I program still Java, when I have the time.

Third, I can't, as much as I would like, take the time, bury myself in source codes and the project to get an overview, what's going on in it. I run already another non-profit-project. Admittedly, right now, programming would look far more enjoyable than that one, but there are commitments, sorry.

Fourth, and the is the key point: The problems, FG has are NOT software problems. Thorsten is, according to the original post, frustrated about communication problems. And yes, I can see why. On the software side, FG can compete with any other FlightSim on the market as it looks. So, FG may is not finished and not polished, but it is already quite impressive. And if a guy like me can fly in it, that means a thing or two, because I'm not that good a pilot. Those software glitches are solvable problems. But to build up a working communication structure is often harder than writing a program.

Stating clearly I can't comply to certain requeast? What was the hard part of "Sorry, I don't have more information, very sorry"? Especially, after Thorsten said already, he thinks, it was a scenery not loaded problem and I had already agreed, I had the same suspicion?

Bottom line is, before we can talk about communication and organization, and again, that is the problem here, we need to figure out one answer to a very basic question: What is the goal of FG? A niche software only for hardcore enthusiasts? Or do you want to go out and attract users? Bacause if the intetion is hardcore group and leave the world out, you just need a "phonebook". Not a physical book, just a little database, say behind the FlightGear site, to bring people with different skills together, for example a 3D artist and a cockpit programmer and so on.
If you want to go big, you need to level your organization. Someone like Thorsten, aside of all willingness to help, is too valuable to deal with "what is the button for" problems. That would be, in any commercial software project a question caught by first line support, not a developer. FG has no first line support. If FG wants to go bigger, FG needs one. That's a task for enthusiastic pilots, not for developers. Developers come in when it is clear it is a bug and when it is clear, in which program part, at least aboutish.
A bug shouldn't even reach the developers if it is sure, it is a bug (actually, when I wrote the post that led to so much trouble, I wasn't even sure, it was a bug. Could have been, from my point of view something, I did wrong). And in many cases, debug traces are not only hard to get but also unnecessary. The discussion about those reliable crashes at KSFO for example. It happens only in MP with many others around and the checkbox "load additional objects" not checked. So without any debugging, the logical question is, what happens if an additional object is requested but can't be loaded? To look right there would probably save hours of reading some thousand line backtracks and debugging sessions all over the place. Most crashes and errors are on that level.
The next thing is testing. Developers are not good testers. They do things automatically the right way, by all means, they wrote the stuff. So, developers need test monkeys. Not to get full traces in most cases, but to hear first about the existence of a glitch. That is the first step, everything else follows later. But first, someone has to figure there is something rthat needs to be looked at in the first place.

Mailing lists: Uh-oh, mailing lists. When a project gets big, mailing lists produce thousands of mails, all with nice clean issue numbers in the four and five digit range. Nobody really knows what is in them after a while except for those actually working in this certain corner of the project. The rest is all fog, gloom, oblivion. Because mailing lists don't cartegorize, nor do they tag on multiple levels. So mailing lists are project coordination only till you got yourself something better, for example a little database with tag search and categories that reflect support level and subprojects. That is another point that leads to communication problems, you have four bases, the website, the mailing list, the forum and the Wiki to cover and everything is unsynchronized. No wonder, things get lost, misunderstood or simply ingored. Which kind of closes the circle to parts of Thorsten's original post.

Of course, this is now all a little bit more out of the gut, I would like to see more of the project, the subprojects and most important about the ideas where FG wants to go before I can make suggestions more in detail.

J.
Jabberwocky
Retired
 
Posts: 1319
Joined: Sat Mar 22, 2014 7:36 pm
Callsign: JWOCKY
Version: 3.0.0
OS: Ubuntu 14.04

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 6:38 pm

I don't even disagree with you, and we've been seeing countless of such postings before - some written by Thorsten and myself back when we started out here.
However, requirements are easy to come up with, but implementation strategies are much more valuable - especially in a non-commercial project where there's no "reward" or incentive for people to do certain -unpopular- chores.
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby hvengel » Tue Apr 22, 2014 7:05 pm

To related this to commercial software. Where I work we currently have 9 "engineers" (IE. people doing software development and technical data work). But we have about 20 front line support staff (not counting account managers and sales people who also do some support work). As you can see that is about a 2 to 1 ratio of support people to engineers. The engineers where I work only interact with end users an rare occasions and when this happens the issue has specifically been escalated by the support staffs management and engineering management has reviewed the issue and has determined that an engineer needs to be involved in working with the end user. This happens perhaps once or twice a year in total.

The reasons for this are fairly easy to explain. If we didn't have the support buffer between the end users and the engineers the engineers would be overwhelmed dealing with support issues and would be unable to do what they are paid to do. Basically it would waist a very expensive and limited resource.

One of the things that happens when there is a support staff is that they work with the user gathering information about their issue. And because they are professionals they typically are very "user friendly" in how they work with the end users. The info they gather is basically the same type of info that the wiki pages that Hooray points people at asks for. Support tickets are not forwarded to engineering until all of that info is gathered since there is no reason for an engineer to look at the issue until that info is in hand. In addition if an engineer is assigned a ticket and he/she finds that there is more information needed they document what is needed and assign the ticket back to support for follow up with the end user.

This is not how open source projects typically operate since there is no "support staff" that can provide a buffer between the end users and the engineers. As a result much of the support here is provided by the engineers/programmers. So when Hooray asks a user to go to a wiki page that has trouble shooting steps and info gathering steps he is basically saying that you, the end user, are responsible for at least some of those support staff duties and here is what you need to do. By doing those steps you reduce how much engineering resources are tried up doing support work which means in the long run more actual engineering gets done. I would argue that there is an implied contract between end users and the programmers/engineers of open source projects that the end user has a responsibility to do as much of their own support as possible. Because of this implied contract when Hooray or some other programmer asks someone reporting an issue to gather more information he is not being rude and is in fact basically reminding that user what their responsibility in the process is.
hvengel
Retired
 
Posts: 1128
Joined: Sun Dec 24, 2006 4:35 am
Location: Minden Nevada

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 7:22 pm

indeed, it kinda works like that around here, and "we" (contributors helping maintain such troubleshooting guides, and providing hand-holding in the process of working through them) are even willing to act as "support staff" to some degree - usually, once some issue is sufficiently understood, we'll file an issue at the tracker where we can provide all the info, which is often the end-result of days (or even weeks) of frustrating troubleshooting on the forum.

On the other hand, what we could try to improve here is adopting a real web-based "knowledge base" system, i.e. dynamic and interactive, asking questions and presenting possible workarounds in an iterative manner. That is something that I would volunteer to help maintain over time (by adding new questions/answers/solutions/suggestions). But it's taken roughly 5 years for a wiki to be adopted, and another 3 years for the move from seedwiki to the self-hosted wiki we're currently using. And we've probably had half a dozen of "issue trackers", too - such things depend mainly on contributors, i.e. people willing to maintain such infrastructure. Some people on the devel list have previously suggested such tools, but there was little to no interest/support. But technically, a Q&A setup like "stackoverflow" would work better for those among us who really only want to answer technical questions ...
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby Jabberwocky » Tue Apr 22, 2014 7:40 pm

@HVengel and Hooray:
What you write is EXACTLY the reason why I suggest somethi9ng like volunteer testers and supporters. To take that part of pressure off of the developers. That open source projects "usually" have no support staff means what? Some twenty years ago there were also "usually" no open source projects. Nowm, as FG proves, there are. If we get a support of volunteers up, who knows, in some years it would be total normal, that open source projects have a support stuff. The thing is, people like to do things, they like to volunteer in the limitations of their skills. So, give them the things they have or can develop the skills for.

And since we are repeating the same mistakes over and over ... I suggested to Hooray in the original "Can you see the airport", to add "Recruitment" to the board index ;-)
Jabberwocky
Retired
 
Posts: 1319
Joined: Sat Mar 22, 2014 7:36 pm
Callsign: JWOCKY
Version: 3.0.0
OS: Ubuntu 14.04

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 8:39 pm

Jabberwocky wrote in Tue Apr 22, 2014 7:40 pm:What you write is EXACTLY the reason why I suggest somethi9ng like volunteer testers and supporters.


Again, the idea isn't new:
http://wiki.flightgear.org/Volunteer#Pr ... se_Testers
Prior to a release, release candidates are provided to the community. By directly providing feedback about your experiences with FlightGear's development code, you will become a crucial part of the development process and you will basically serve as quality control for FlightGear, your experiences will determine whether FlightGear's development code is ready for a next official release or not. See the Release plan to find out when release candidates will be distributed.
Note: If you are interested in actually doing development for FlightGear, make sure to check out the Developer section.


The important bit is implementing a strategy to motivate people to help with this - so far, this has not been particularly popular.

Ideas are a dime a dozen :D
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby Jabberwocky » Tue Apr 22, 2014 8:57 pm

Hooray, that is the point, I'm still working on. But since some details are not mellowed out yet, give me a little time. Like programs, recruiting measures are not done in five minutes at a cup of coffee.
Jabberwocky
Retired
 
Posts: 1319
Joined: Sat Mar 22, 2014 7:36 pm
Callsign: JWOCKY
Version: 3.0.0
OS: Ubuntu 14.04

Re: Forum communication

Postby Hooray » Tue Apr 22, 2014 9:43 pm

I wish you the best of luck, but this is not the first time we're seeing someone trying something like this - there kinda seems to be a multi-stage process that all aspiring contributors are going through, this being among the first 3 stages ;-)

And yeah, I also went through it, too.

Keep in mind that our way of interacting with people and "managing" things is not necessarily "ideal" for the project, but it happens to work for us, and is tolerated by others (another important point)
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 11510
Joined: Tue Mar 25, 2008 8:40 am

Re: Forum communication

Postby Lydiot » Tue Apr 22, 2014 10:56 pm

Hooray wrote in Tue Apr 22, 2014 8:39 pm:The important bit is implementing a strategy to motivate people to help with this - so far, this has not been particularly popular.


I think part of the problem, and I'm speaking as a relatively new user now, is that there doesn't seem to be this one place where all information exists. There's the webpage, there's the forum, there's the wiki. Now, what determines where I go is what information I'm seeking. If I end up going mostly to the webpage only to download new aircraft then I'll likely bookmark that download page, not because I don't want to see the homepage but because the forum is where a lot of the "life" is taking place. So asking for testers on the webpage is probably going to be missed by many that have already passed that stage (i.e. found the webpage, downloaded the app and some planes and then started flying). And in my experience it's difficult to create an impact on forums simply because of the user interface. It's very easy to miss things of more importance than others, because they all look the same on the page. And the wiki page to me at least is a resource when looking to solve a problem or learn something.

So, the short version is that many users are likely completely unaware that testers are needed for various parts of the project.
Lydiot
 
Posts: 987
Joined: Tue Oct 22, 2013 10:50 pm

PreviousNext

Return to Forum

Who is online

Users browsing this forum: No registered users and 0 guests