Hello. I'm sorry I've been away. I had a 'family intervention' and eventually got whisked away with family because I was in such a bad state. I'm sorry that one of the downsides of this project is that I'm the only one working on it right now. I need to step out for an hour or two in a minute here, but I've spent the last hour and a half or so answering PM's. Things seem to be running well now, but I am terribly sorry that people lost coins in the 1.0.4 fork yesterday around noon.
The problem was caused because the exact same numbers were multiplied together in the exact same way but in different locations of code and produced a different result. This kind of thing can happen when you're dealing with the maximum limits that computers can deal with because compilers (the programs that turn code into new programs) like to try to fix these sort of problems for you. In 1.0.4, the GCC compiler didn't fix the problem for me, and that's fine, I can deal with the 64 bit integer boundaries myself. But when I multiplied the same numbers together in a different place it DID fix the problem to a very small degree. Half the bugs I mentioned before just don't exist anymore (You could stake 100m coins and it'd be fine, and there is no .9-1 coin bug, all courtesy of the magic of compilers.)
The reason for all the problems boils down to lack of time to test. I've given myself no time. The fix yesterday took 6 hours from discovery to release, and that's disgustingly fast. The 1.0.4 release was produced in about 4 hours. Both faced limited testing, and 1.0.4 was actually a smooth transition, there was no forking issues on it, but my 1.0.3 test code got out to servers.
I can do this, and I can code well, but coding perfect code this fast is practically impossible. I did test all the code, but I couldn't test it 100% thoroughly against the current network because the difficulty is just so high. Testing takes days at least, and proper testing could take weeks or months, and clearly we didn't have that.
I'm sorry for the troubles that everyone has had. There shouldn't be anymore serious issues throughout growth, and anything that gets fixed in the future will be thoroughly tested over a reasonable period of time, and go through the proper alpha/beta/release candidate stages before final release.
I'm here to stay, and I will continue to work on SEED. I know this is a stain, and I plan to do everything I can to bring SEED up to rise above it. Honestly, even if the coin 'dies' in the market, I'll still work on it. It's taught me a lot of things about the way proof-of-work and proof-of-stake work together and about the limitations of the proof-of-stake as implemented in other coins. There's a lot of things I have been dying to bring to Crypto, but I have dreaded the pressure and expectations that get put on Dev's that get paid little or nothing for their work.
See?
This is a Dev I can get behind and fight tooth and nail for.
You don't come across many people with this amount of drive and passion for their product. It's people like this that get shit done.
Don't sweat it dude - I'll still be fighting your corner. It's just shit that some people expect
so much in a very little space of time. It takes time to make a project work! Rome wasn't built in a day, the Universe wasn't created in 7 and so on
Chillax people, I expect great things from this coin
