Board index FlightGear The FlightGear project

FlightGear Development Push

Questions about the FlightGear organisation, website, wiki etc.

FlightGear Development Push

Postby Sebulba » Thu Nov 19, 2015 12:50 pm

Hello to developers and users alike.
This post contains a suggestion that is addressed to the first for consideration, and to the latter for their opinion submission. This is why I didn't write it on the development mailing list, but here.

The suggestion is simple. Form a team that will undertake the task of raising money for the acceleration of the development of the project. This could be done in the following ways:

1.) Start a crowdfunding campaign - 0 A.D. did this on Indiegogo (https://www.indiegogo.com/projects/supp ... egy-game#/) and they raised $33250. In my opinion, FG user base is much larger and a crowdfunding campaign could possibly yield even more.

2.) Set up a Patreon account - I can imagine many FG users who would happily donate at least $1 per month for their favorite open source project. (https://www.patreon.com/)

3.) Put a Flattr micro-donation button on the end of each blog post - I am sure that at least some of the visitors of this website use Flattr (https://flattr.com/)

The purpose of all this would probably be to hire full-time developers in some places that the projects needs them, or even pay those who are currently developing for FG at their free time to work more on the project.
Right now, I believe that the project has formed a huge community of users from around the globe, but very few of them can (have the skills to) contribute to something as complicated as FG coding etc.

I would love to hear your thoughts on this everyone.
Cheers :)
Sebulba
 
Posts: 117
Joined: Wed Nov 04, 2009 2:10 pm
Location: Greece

Re: FlightGear Development Push

Postby Hooray » Thu Nov 19, 2015 1:25 pm

This has been previously suggested/discussed (do a search for "funding" or "non-profit" in the archives).

In summary, FlightGear tends to attract hackers, not business folks, and certainly not lawyers/tax experts.
Curt mentioned a few times that the website, and it using AdSense to make revenue, is already causing some irritation.

If you want to literally boost development, you'd need to ensure that patches/merge requests don't end up being unreviewed/uncommitted for months, and that core developers actually follow-up on seeing contributions they originally encouraged.

However, that is currently not possible because of the lack of manpower - which is to say, that you'd need to establish a procedure to ensure that there is a minimum number of active committers, who can help triage/review bug reports and commit bug fixes.

For the time being, the situation is extremely unfortunate from a manpower perspective, and two of the most active, and most experienced, core developers are working on heavily overlapping GUI functionality, using completely separate technology stacks that are not commonly used elsewhere in FlightGear, or even just in standard code paths, (Qt5, as well as HTML5/JavaScript).

So this is adding to the overall situation, i.e. lack of active core developers who can help with usual duties, such as reviewing merge requests, patches and bug reports.

You only need to look at the devel list to see that all core developers agree that they're fundamentally overstretched in terms what they can handle, and that they need more active core developers involved - for instance, you could look at the recent interview on sourceforge, the stated conclusion also was that more developers are needed.

And ideally, those new developers would be students, or even teenagers, not people with jobs families, because that is exactly how most of the remaining/active core developers ended up being extremely crucial to the project, i.e. when they were young and had plenty of time on their hands, while this is now no longer the case, and turning out to be the bottleneck because other things are taking precedence obviously.

Alternatively, you would need to get more professional users/developers involved, i.e. those who have a track record of doing related FG work, such as poweroftwo (osgEarth).

Absent that, FlightGear is inevitably going to be forked sooner or later - no matter if that is going to be successful or not (in fact, I have been approached on more than one occasion to see if I would be willing to help with such a fork).

Core development is a very real bottleneck, and if nothing is done to welcome/mentor new developers, those whose contributions/patches are not incorporated will move on and create their own "CommunityGear" at some point.

However, that does not necessarily have to be a bad thing, gcc faced the same situation that FlightGear is now facing, see the GCC/egcs split - so while such a development may be unfortunate, it may help the FlightGear project to get our act together and deal with some of the more pressing issues.
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby Sebulba » Thu Nov 19, 2015 1:32 pm

Ok thanks for the reply, however I don't see why you've written all that you've written.
There is literally nothing that money can't buy, so if you need more core developers who are students like you suggest, wouldn't it be nice to pay them in order to lure the best for interviews etc?
Do you need a lawyer for this? If I go to a lawyer and explain the purpose, the size of the user base, and the estimated financial gains, I am sure he would make a fair bidding. If we make 50k on a crowdfunding campaign, we could pay a lawyer too, so where's the problem? The adsense that you suggest are a problem because they don't generate enough money to pay a lawyer, so the income from them is somewhat illegal. I can understand that, but I am proposing something way bigger here in terms of financial support.
Sebulba
 
Posts: 117
Joined: Wed Nov 04, 2009 2:10 pm
Location: Greece

Re: FlightGear Development Push

Postby Hooray » Thu Nov 19, 2015 1:45 pm

I suggest you do a little research first by referring to the archives, using the search queries I mentioned - refer to postings made by Curt and other core developers, and you will see that this is something that they are not interested in pursuing.

$50k is "nothing" once you have a family, a house and bills to pay - so it would need to be much more massive than that, you could maybe sponsor 3-4 students using that kind of budget.
But just look at how "GSoC" never worked out for FG due to the same factors.

If you have any kind of serious budget, (think 1M), it would make more sense to start a company and hire a team of experienced C++/OSG developers
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby Sebulba » Thu Nov 19, 2015 1:51 pm

Well, 50k is basically just a number that I pulled out of my ass. 0 A.D. managed to get 35k, so if FG is say three times larger, It should make at least 100k right? Now this may not be much, but it would certainly help speed up development. Maybe enough to at least pay one full-time developer. Or three students. Anything would be welcome, I mean why not? Why not do the best that we can for the benefit of the project we love? You sound like "either bring me millions or I don't care". This rationale sucks mate and I am saying it in a friendly tone. We should put things on the table and try to make them work, instead of rejecting everything before we even discuss it, or putting our minds into thinking a thousand reasons of why things wouldn't work.
Sebulba
 
Posts: 117
Joined: Wed Nov 04, 2009 2:10 pm
Location: Greece

Re: FlightGear Development Push

Postby curt » Thu Nov 19, 2015 2:02 pm

Here is the big challenge as I see it. Even if I had $100k in my pocket right now, how do I find a really good qualified developer? What would they focus on? Most of the really good, qualified, experienced developers who contribute to FlightGear are doing it as an outlet from their stressful day jobs. Don't underestimate the challenges of convincing one of these guys to take on extra duties with hard timeline commitments and potentially lots of extra stress ... in an area of their life that is currently for fun. It's different if you are in your early 20's looking for your first real job, versus in your 30's, 40's, 50's, etc. staring at an accumulation of stress. Hiring a random person for a year would be easy, but may not produce anything useful, may not produce quality code, may not understand the larger vision or direction ... I don't know. We could always get lucky and find the perfect person on the first try and be wildly successful, but for every success in that area, there are probably 100's of less fortunate stories to be told.

Groups that get funding (like through kickstarter etc.) have a long track record of delays and under-delivering ... or out right failure ... because people just starting out don't understand the difficulties of committing to developing something new on a fixed time and dollar budget ... along with managing all the other challenges of a business.

So sure, I'm not opposed to something like this in theory ... but there are many many many challenges to being successful. Getting the money is the easy part in comparison. Delivering is hard.
curt
Administrator
 
Posts: 1174
Joined: Thu Jan 01, 1970 12:00 am
Location: Minneapolis, MN

Re: FlightGear Development Push

Postby Sebulba » Thu Nov 19, 2015 2:10 pm

Ok thanks for this answer. I understand your points and I agree with most of them. However, I would like to point out that
1.) In Indiegogo, you don't have to meet a goal or deliver "anything" at all in order to get the money raised. Kickstarter is a different story.
2.) Even if the campaign failed, it would generate publicity for the project, and more young developers could hop in the core dev team thanks to this.
3.) Maybe we should focus on finding a person who would do a more "generic" work in order to ensure that his/her contributions would be meaningful to the project no matter what? Something like reviewing of testing reports, bugs, and new contributions?
Sebulba
 
Posts: 117
Joined: Wed Nov 04, 2009 2:10 pm
Location: Greece

Re: FlightGear Development Push

Postby Hooray » Thu Nov 19, 2015 2:13 pm

To the OP: this has been discussed in great detail on both, the forum, and the devel list. So it would seem like a good idea to review the main arguments that have been previously discussed.

Otherwise I do agree with Curt, even though I am not sure that FG as a project, could actually get its act together to start a compelling crowd-funding campaign, which should ideally be very organized and convincing, including concrete plans what is to be done with the money.

Regarding finding/hiring skilled developers: that seems like a no-brainer to me, you only need to look at people in the OSG community who are doing OSG related contract work, there is plenty of manpower over there, and they're looking for jobs - and we have plenty of OSG related rendering work ahead of us (fgviewer, hla, splitting off the renderer, adding compositeviewer support - you name it)

And then, you could obviously look at existing FG contributors who did FG related contract work, e.g. poweroftwo originally implemented FG/osgEarth integration for a corresponding project.

So, looking at crowd-funding is nice - but there is certain surrounding overhead (read: paperwork and other obligations) that needs to be managed - in its current form, FG as a project is already extremely bad at leveraging all the manpower/resources it is attracting, especially when it comes to contributors, and core developers.
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby Sebulba » Thu Nov 19, 2015 3:27 pm

Hooray, thanks for all your (comprehensive) answers.
So, to summarize, you say that the development pace could be essentially and significantly increased if we could leverage all of the contributions that the project receives right now. Could a crowdfunding campaign help the project in that part? Could we find and hire one or two persons who could help in that sector? Maybe experts who could review all of the incoming contributions and forward the most important ones to the core developers with their comments? Or maybe people who could implement some of the contributions themselves? Does that make any sense?
Sebulba
 
Posts: 117
Joined: Wed Nov 04, 2009 2:10 pm
Location: Greece

Re: FlightGear Development Push

Postby Hooray » Thu Nov 19, 2015 3:35 pm

that's not going to help - there's a growing backlog of merge requests/patches, that simply don't get reviewed/committed due to lack of manpower among core developers - either something is done about that, or someone is going to get in touch with all the people whose contributions haven't yet made it, and create the equivalent of "CommunityGear", which is likely going to see more activity than FlightGear core development is seeing recently.
(just look at what fgmembers were doing)

You only need to look at the low number of people who were shouldering most of the workload in the last couple of years (even just referring to the commit logs), most of the work was shouldered by Zakalawe alone - with Torsten and ThorstenB being there to assist - but otherwise, there is a clear bottleneck, and direct dependence on certain individuals.

Previously, it was mfranz shouldering all this workload (i.e. doing the unpopular stuff like reviewing bug reports, and committing patches).

I don't think you are going to magically "solve" this problem by throwing money at all - unless it is such an amount of money that long-timers like Torsten, James and others can simply afford to focus entirely on just FG matters.

FlightGear is not a company, so most of its long-time contributors are unlikely to respond too well to the kind of obligations that money tends to bring with it...
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby Thorsten » Thu Nov 19, 2015 3:39 pm

Simple question: Suppose the cash problem to hire a full-time developer is solved - who gets to decide what this developer works on? What if some of us disagree with the way he does a problem?

The project doesn't have anything like the decision-making structures to assign paid people to a task in place.
Thorsten
 
Posts: 11136
Joined: Mon Nov 02, 2009 8:33 am

Re: FlightGear Development Push

Postby Hooray » Thu Nov 19, 2015 3:44 pm

good question - but I guess that it could be handled analogous to how people were coming up with the updated project roadmap - i.e. seniority and track record are most likely to influence decision making.

You could conduct a poll among core developers, who may in turn conduct a poll among fgdata/scenery (middleware) developers to prioritize things accordingly.

For instance, it seems that there is a general consensus that FlightGear is not particularly good at leveraging multicore power, which is what is motivating HLA/RTI - so that this is mentioned on the roadmap, and also being worked on by different people (see the recent devel list discussion initiated by Stuart).

Equally, it seems that people want to get rid of legacy OpenGL code and make better/more optimized use of OSG.

In other words, there seems to be a shared subset of common goals/ideas that most core developers agree on - hopefully, such a consensus would then be based on active involvement, too - or at least on people being willing to mentor related efforts actively.

Otherwise, it is also true you are NOT going to impress people by telling them to work on something that they fundamentally disagree with - e.g. telling Thorsten to do Rembrandt work, or to texture the ogel :mrgreen:
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby Hooray » Fri Nov 20, 2015 4:54 pm

Here's pretty much the same discussion that took place on the osg-users list 7 years ago, you will see that many of the arguments will seem familiar: http://www.mail-archive.com/osg-users%4 ... 19685.html
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Re: FlightGear Development Push

Postby pommesschranke » Fri Nov 20, 2015 8:14 pm

Hello Sebula

I like your idea very much and I hope that the responses that you got so far are not demotivating for you.
Go ahead!
Create accouts at those 3 sites, then do some marketing about it and let's see if and how much people donate.

I do various stuff for FlightGear (see my wiki page).
If I'd get financial support from FlightGear users I would be able to spend more time on FlightGear development.

helijah has a paypal donate button and I donated to him already
http://helijah.free.fr/flightgear/hangaren.htm
pommesschranke
 
Posts: 1104
Joined: Sat Apr 27, 2013 7:58 pm
Location: EDLM & LJCE
Callsign: d-laser
IRC name: laserman
Version: git
OS: Linux Lubuntu 18.04

Re: FlightGear Development Push

Postby Hooray » Fri Nov 20, 2015 8:41 pm

look, I also didn't mean to discourage anybody - but you need to know what you are doing, i.e. you need a plan and a concept to do this, and you will also need community support if you are serious about this - if this is just about a few dollars, you are better off continuing by donating privately to certain contributors specifically.

But if you are looking for large-scale sponsoring, this needs much more thought, or it will inevitably fire back at the people trying this, or even worse, the whole FlightGear project.

I do know about a handful of contributors who did paid contract work for FG, and they stated quite clearly that they didn't feel comfortable disclosing this on the devel list, so there's that, too - you only need to look at Curt's experiences whenever he managed to do a little contract work related to FG, and the discussions that triggered - and keep in mind that you cannot have any "better" standing than Curt probably ...
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: 11354
Joined: Tue Mar 25, 2008 8:40 am

Next

Return to The FlightGear project

Who is online

Users browsing this forum: No registered users and 1 guest