I'm glad to see this thread has triggered a civil pissing match. It is a good thing, I think.
Angelox, I am a bit hurt about some of the things you said. EQEmu is my first priority, with database/quest development being next. I've often been prone to taking the less wise path in an effort to keep both projects moving, whether I look bad in the process or not. But, I know you and know you don't hold any malice towards anybody who doesn't deserve it so I am going to chalk this up as a miscommunication and call it done with.
Back on to topic about the devels...
I have worked with a few of them pretty closely and I do want to say some of the flames toward them is a bit unfair.
KLS - Grab the full changelog and do a search for KLS. That about sums it up. Other than answering development posts KLS keeps themself pretty quiet on forums and such. I have no problem with that at all, and don't see why others do. The work gets done and it's quality.
Wildcard - Again, look through the changelog. Add that to the fact that he is a key member of PEQ. In addition to coding, he does database work, event scripting, helps me with misc chores that we find ourselves needing, and handles some of the GM work that I don't particularly care for. He has a full plate that's for sure.
FNW - In the modern day of EQEmu he kept this project going, Atlas style. I still remember back in the day his MASSIVE changes to the changelog. Getting new code from him almost felt like Christmas. In the past 2 months, he corrected a few severe memory leaks that I know were brutal on Grand Creation and I am certain effected other servers as well. I think he and Doodman do more behind the scenes than anybody knows, or gives them credit for. I found with PEQ that running a project isn't just about getting the actual development work done, there are tons of other things on your plate that require attending to, or else there won't be a project left to develop for. I think it should also be noted that when EQEmu was losing its hosting, FNW told me if he had to, he would host the site and login on his PERSONAL box and bandwidth.
Doodman - I haven't spoken with him much, but I have to admit I haven't ever tried, either. I know like FNW, he does a ton behind the scenes much, much more than he is given credit for. On a personal note, there was a time recently when Wild and I were bomboarding with him questions mostly dealing with packet collection and collecting spawn opcodes so we can try to get eqextractor working with eqbuilder again, and he was more than friendly and helpful to provide us the information we needed.
Rogean - I know less about him than any other, I admit but his recent HoTT implementation as well as fix for a certain form of exploit was well received by me. His work seems to be quality, so that's all the counts for me.
Now, I don't know the specifics of how it works so I'm now expert, but I do know that all code that is put into CVS is quality controlled to prevent bogus and/or incompatible code from getting into the repo. I know there have been a few times developers had to rewrite certain pieces of code because while they worked and worked well, they didn't meet the quality EQEmu has tried to maintain in recently time. I think that may be one of the reasons why code from the forums doesn't make it into the CVS often. Here is the next issue I have seen... lack of free time. Developers don't have the time to find a piece of code on the forums, analyze it, then either approve it, or hand it back saying this, that, and this need to be changed because of that guy over there. Finally, submit the approved code to SVN which winds it's way to the CVS, and eventually to the nightly builds.
What I feel EQEmu needs to correct this situation is a sort of Forum Developer. Somebody who will act as a lisaon between the coders on the forums, and the developers with CVS access. They would need to be a good developer, and have a very solid grasp on the EQEmu code. They could find the code on the forums, diff it if it isn't already, find immediate mistakes with it and either correct it themselves, or give it back to the author. Then, pass the mostly polished code up to whoever. If they deny it, the liason would then send the code back to the author and either let them know what's wrong, or have them contact the CVS developer directly. Of course I have no power to make this a reality, it's just an idea really. I think instead of fighting back and forth, that's what we need to do come up with ideas and put them into motion.
|