FreeOrion

I have a real concern about the future of this project. I really want it to succeed, but I am slowly coming to the realization that it is dying. Here are the things that concern me:

The level of capability evident in the code I have seen so far is somewhat low. There are two problems with this. First, the number of errors that we currently have in the code and the small amount of code that we have at present suggest that this project is doomed from a technical standpoint; when enough code like this gets put into a project, it becomes harder and harder to add new code, to the point of impossibility. Second, the general bugginess of the current code would not necessarily be a problem by itself, but I also feel that those with fewer skills are essentially unwilling to listen when their mistakes are brought to light. Those that do appear to listen do not seem to follow through by changing their code. I hope I am just getting the wrong impression because all our communication is being done through text, and occassional text at that.

Related to this is another problem: I have been unable to add new code to the game many times because the code that exists either does not do what the documentation says it does at all, or does it incorrectly. I have tried to get around this by asking people to fix things, and then have inevitably ended up rewriting some things that other people have written to fix them, because I could not reasonably wait any longer. I do not like to do this, both because it makes it hard for people to maintain their own code when it has been changed without their involvement, and because it breeds resentment.

This brings me to my next point. The number of commits is slowing to a crawl. We are now in what I consider the fun stuff. The ground work has been laid, and the major design decisions have been made. Now we get to add actual functionality to the game, and get to sit back and say "Gee whiz, it works." The fact that no one is doing much under these conditions suggests the worst. It really appears to me as though people are losing interest in the actual coding. This spells out the demise of this project.

Another problem that is somewhat related to the second one is that I feel that much of the code in the repository has been committed with little or no testing and verification. It is inappropriate in a project like this to just write some code that probably works, then commit it. This leads to the kind of problems I have had, in which I assume that existing code is functional and try to add new code that uses it, only to find that whoever committed the code clearly did not verify that it works.

Here are some things that we could definitely improve on:

If we change someone else's code, we should at least notify them of what changed and exactly why. Preferably, we should notify them before hand to allow them time to respond; we may be changing something vital that works differently than we think it does.

If we are uncertain about our design choices and/or that our code works, we should ask someone. I am always available to help with this, and I usually respond the same day.

We should only be committing code that we are reasonably certain is usable and works as documented; we should wait until we are reasonably certain and have had time to test the code before committing it. Testing it means using it in the way it will be used in the larger project.

We should document all our code according to the published standard, and make sure that the documentation changes as soon as our code does.


This said, I am seriously considering leaving the project. However, as some of you know from reading my past posts, my mind is easily unmade when I see a persuasive argument to the contrary. I will stick around for the next few weeks, and if things start to look up, I'll stay.

I want everyone to understand that this is not an ulitmatum. I'm not saying "If you don't do this, I'm outta here!" What I am saying is that I have serious concerns about this project, but also a great deal of investment in it. The only thing that will drive me away from it is if I feel that it cannot succeed and that I am therefore wasting my time, which is very valuable to me.

Please let me know what you all think.

I'm not a coder, and thus don't know about the fundamental needs of programmers, still maybe one reason for the slow down could be summer time.

Yeah, believe it or not, I heard that even computer nerds sometimes get out of their climatized rooms, trying to get a sniff of fresh air, and maybe, yes maybe even meet someone of the opposite sex outside of a chat room .


As much as I can understand your sorrows tzlaine, I have major hope in the weather. When it gets grey outside, and the skirts become longer again, you'll also face more code (good or bad) flowing in.

Well, for my part, I apologize for not commiting lately. I've been slow to pick up with the project lately and finish the fleet window, and I'm going to start putting more time into it from now on.

I'll post some thoughts after I've had a chance to go over the code and get an idea of what our current state is. Many of us (myself included) have indeed been shirking the unit testing at one point or another, and it's really going to take a concerted effort by those involved to get around that problem.

_________________
Empire Team Lead

I agree tzlaine re: the lack of code being submitted, and I'm to blame also. Half my problem is not knowing what I should be working on, and half laziness.

Without trying to justify my lack of coding, there is something I have been waiting for: Yoghurt promised us a demo of python so we could finalise that question once and for all. It's a question which answer *could* have a huge impact on the project, development time, and the foundation. In other words, something I want to answer early, which should have possibly been done earlier than now. Anyway, I think Yoghurt is near completed and ready with this, I talked to him the last day or two.

But I really do think we are moving too slowly, and I take part of the blame, and need to reverse that now.

Well, personally I would really like to help where I can, but my C++-skills are... hmm... almost non-existent, we just started to switch from C to C++, but at least I know the basics (references & classes)...

hmm, I really dont have a clue what to write besides that... (I tried to ad 1 or 2 more sentences for 10 minutes now, but... maybe I should just go to bed -.- )

_________________
zaba zaba zud zud

This project, like any open source/community project, is a very fragile house of cards of which this team is the most essential piece.

We all have our ups and downs in terms of our enthusiasm for working on FO, but try to keep in mind why we all started this project in the first place. QSI was getting roasted alive by the public and we set about trying to do things right. We got off to a very good start and I would hate to see everything end here.

That said, it's quite reasonable to want to take a break from working on a project like this. I have some weeks where I get very enthusiastic about something and other weeks where I don't even want to read the forums. So please be honest -- both with yourselves regarding your own commitment, and with the other members of your team. If you want more programmers, you're in the best position to go out and find some -- the minute FO announces itself to places like Slashdot, we're going to have a flood of recruits (and ten trolls for every one). But we need to have something to show before we do that, so the best way for you to get a hand is to recruit one directly.

I'll try to get the v0.2 design document done this week to get us some momentum, but that's really all I can do. Please don't let FreeOrion die young -- we are so much more fortunate than most community projects, because once we get to a certain level, we'll have people lining up to help out just because of the Orion games. We have the potential to turn out as well or better than FreeCiv, and that's something that goes on your resume. And it'll be you guys that get the interviews with Slashdot, not us pansy designers.

Best of luck,
SWK / Aquitaine

_________________
Surprise and Terror! I am greeted by the smooth and hostile face of our old enemy, the Hootmans! No... the Huge-glands, no, I remember, the Hunams!

hummm... it appears to me that you guys need more groundwork in the project and more organization. Giving specific tasks to specific team members (and then setting a rough timeline based on what the member who is doing the coding thinks) would be wise at this point. When the project grows and has a working foundation than it is easier for each programer to do whatever he wants to help the project develop. I good idea would be to assign tasks, and organize plans, with the motive of building a working foundation for the game.

I agree that things have been very slow. I am also concerned about the overall health of the project. I know a few of our main coders have been on vacation at different times this month. I'm hoping some of it can be put off to that.

I have had a few things come up that are putting high demands on my personal time over the past few weeks, and have really sidelined me from the project. My interest and enthusiasm are still high, but I have had to de-prioritize it for now. I'm hoping things will get back to normal over the next couple of weeks, but I can't guarantee that. I suspect it will continue , on and off, like this until the end of October. As such, if there is someone from our initial coding team who is interested in taking over the lead spot, I would be amenable to handing it off. Otherwise you'll have to bear with me until I can start putting in more time again.

Some other thoughts:

--We may want to advertise through sourceforge and moo fan sites to try and attract more developers if we haven't currently got the manpower to keep things moving at a decent rate.

--A lot of Zach's suggestions are good, but I'm not sure they will be followed short of an inspection/code-review regime. This was initially part of our proposed process, but given the small and informal nature of the programming team, a lot of the process was allowed to slide. Should we think about bringing that back?

--Lastly, Zach, I won't hestitate for a moment to say that you are our most talented and valuable contributor. However, you do have exacting standards and can be somewhat abrasive at times with your criticism. I'm not at all suggesting that you shouldn't critique other peoples' code and designs, as I've found your comments extremely valuable. Just that sometimes people will feel a bit put off by them, and while the initial reaction may be resistance, I think people have been generally good about accepting advice and criticism.

_________________
Programming Lead

A suggestion relating to the peer-review thing.
Instead of each of u devs doing it (which apparently u arn't anyway ), assign some enthusiastic not-so-good C++ coders to over-look it.
That way all the code is looked over (if only by a semi-professional eye), and they get to learn stuff too.

I'm not sure it's a good idea to have the code reviewed by the less experienced people. It takes an experienced eye to notice mistakes in code, especially when the code was written by an experienced individual.

_________________
Empire Team Lead

I think the project has gone well os far, we are getting into versoin 0.2 now. I think quiting now will kill the project, but continuing will ensure that it is completed. Besides there is no deadline, no pressure and as time goes by, more people will come. In the last week things have slowed down, but maybe that is just miscommunication.

I don't really agree that the project is dying. I feel, as do others, that the project is simply at a low point.

I know personally that I've not contributed anything since the first iteration of the Galaxy map. I've had some significant health issues lately and I'm in a very busy period, both professionally and personally.

Also, keep in mind that summer is over and I'm sure there are many people involved in this project who are returning to school, which will obviously limit their involvement.

As has been said, people's interest in FreeOrion will change over time. For me, it isn't a high priority right now. I am involved with other projects that I all but ignored for some time to work on FreeOrion. I wish that I was able to spend time each day on it, but it just isn't possible.

I am still very interested in this project and I don't want to see it die. I also know that there are many people out there who have heard about the project and are eagerly awaiting some sort of playable release. Yes, things are slow, but they will pick up eventually as long as we don't quit entirely.

---------

As for Zach's comments about the state of the code, he is right on many points. I do agree that if you're going to commit code, make certain that it works under a variety of test conditions, as well as making sure it isn't platform-specific. I am of the opinion that code in the repository should work as-is, so don't commit works in progress.

I also know, as Oceanmachine said, that Zach can be very critical on code style and such. And while his comments are extremely helpful and indispensable, they can sometimes seem a bit insulting. I have almost always accepted (and agreed with) your comments and done my best to abide by them, but keep in mind that not all of us are professionals or even extremely experienced. For some, it is the first time working on a large project as a group, where coding conventions are paramount. The bottom line is, you cannot expect to achieve the same level of standards as you would in the business world, and if we attempt to expect that, we will all be in for a lot of disappointment.

_________________
FreeOrion Programmer

First off, let me say that I am very encouraged by the response to my post. I have faith that though the project may be in a slump, the slump is temporary. This is heartening news. I feel that we do need some slightly more formal methods, or at least some clear guidelines that people can and will stick to. I hope to have a discussion about this either with Joe one-on-one, or preferably in these forums where everyone can participate.


As Joe and Tony have alluded to, my criticisms of other people's work has been abrasive and insulting. This is both accurate and inaccurate. It is accurate in that if anyone was abraded or insulted by my posts, then by definition, my posts were abrasive and insulting. However, it is inaccurate in that I never intended this when I wrote the posts. There are a couple of factors that create a disconnect between what I write and what you all read. First, the medium is hastily written, technical text. Nuance that may exist between the words I write is largely lost. If I were to say the exact words to you that I have written, you probably would not be insulted, because it would be clear from the way I spoke them that that I meant no offense. This is obviously missing from the words alone. Also, I try to write as clearly as I can, since the subject is abstract and technical by its nature; often, things outside of the specific topic I am writing about fall by the wayside. Second, and far more importantly, is that when I write harsh words about someone's code, I am writing about the code they wrote, not their intelligence or value as a person. Everyone makes mistakes. Everyone makes stupid mistakes. No one is born knowing how to write software. None of us are exceptions to this, least of all me. Writing a colossally boneheaded bit of code should not be embarrassing to anyone, because everybody does it. Mistakes, including really stupid ones, are a constant and prevalent presence in programming. Again, I am no exception. So when I say something harsh about your code (or mine, as I have done as well), don't tie your ego to your code. It's just code, it's not you.

I too find that written words come out so offensive, just because words mean so much more in writing then when spoken.

You could consider using smilies/emoticons when writing messages. That way there would be much less room for mis-interpretation.
They work for me. Being british, I use British humour, which generally comes across as very sarcasic to americans... So i have to put lots of smilies in to try not to offend anyone.