Release level

[Marco van Herwaarden]

To calculate the release levels a lot more factors are considered then only the number of releases. (and no we will not disclose the full calculation).

I won't reply to the second half of your post as those questions are all "what if's" based on the assumption that our system is flawed. There are currently no flaws that we are aware of.

[EnIgMa1234]

Originally Posted by Shelley_c

Another question paul. I've looked at princetons release amount (which is 31)
or there abouts, now I've got a release amount of 60+ so how can this warrant such a high level release for princeton and why is is system looking like it's flawed?

If the release levels system is a flaw and crawling with inconsistencies what's to say the board/hack of the month system isn't riddled in flaws. I'm sure there may be reasons why some people have higher release levels for such a small contribution count but if the release level system is flawed and/or being tampered with to benefit the few privileged then why have a release level system in place and shouldn't these issues/inconsistencies be addressed?

A lot more than the release amount is considered. For example you have 700+ installs. That would also play a factor.

[King Kovifor]

Originally Posted by Shelley_c

Another question paul. I've looked at princetons release amount (which is 31)
or there abouts, now I've got a release amount of 60+ so how can this warrant such a high level release for princeton and why is is system looking like it's flawed?

If the release levels system is a flaw and crawling with inconsistencies what's to say the board/hack of the month system isn't riddled in flaws. I'm sure there may be reasons why some people have higher release levels for such a small contribution count but if the release level system is flawed and/or being tampered with to benefit the few privileged then why have a release level system in place and shouldn't these issues/inconsistencies be addressed?

The favoritism was a joke, but lets look at the facts:

You have 61 releases, yes, but only 700 installs. Princeton has 31 releases, you beat him in that section. But he has over 2200 installs.

It's not just based upon releases.

[Paul M]

The system is working fine, if you want to believe otherwise that's up to you. I have other things to work on.

[Attilitus]

And no system is perfect... Even if this system was "crawling with inconsistencies" it wouldn't be that big of a deal... at the end of the day it is just the number that appears next to Release Level when you mouse over the small-texted usertitle of a user on a message board.

I doubt that the fabric of space-time will come undone if the system is not absolutely perfect.

[hambil]

For what it's worth, from someone who has been here a while, what we have now is a subjective system. This is not news. We where told back when the old system was being retired that the new system would be much more subjective. I'm not thrilled with it, but it is better than the objective system we had before, which generated some truly ridiculous results.

[Marco van Herwaarden]

Hmmm i think you are mixing up objective and subjective.

The old system implicated a "value" (ie. Senior coder, Master coder) to the quality of the work of a coder, only based on numbers like number of installs or number of releases. The problem with this was that the number of releases does not say anything about the quality of the work.

With the new system we have been trying to present an objective number that does not implicate anything about the quality of the work.

[hambil]

Originally Posted by Marco van Herwaarden

Hmmm i think you are mixing up objective and subjective.

The old system implicated a "value" (ie. Senior coder, Master coder) to the quality of the work of a coder, only based on numbers like number of installs or number of releases. The problem with this was that the number of releases does not say anything about the quality of the work.

With the new system we have been trying to present an objective number that does not implicate anything about the quality of the work.

No, I think I got it right. Take the simple example of a question:
"Is this text right aligned?" is an objective question, with an objective answer.
"Is this text any good?" is a subjective question, with a subjective answer.

The system you have now uses subjective questions such as 'how good are these releases', 'how supportive is the author', 'how valuable to the community are these releases'. The old system only used objective questions such as 'how many installs does this author have' and 'how many releases do they have'.

[Marco van Herwaarden]

I see this different:

The old system gave a subjective value: Master Coder (is this member a better coder then a Junior coder????)

The current system gives only a number, no value attached to that. A higher release level does not implicate a better coder.

[hambil]

Originally Posted by Marco van Herwaarden

I see this different:

The old system gave a subjective value: Master Coder (is this member a better coder then a Junior coder????)

The current system gives only a number, no value attached to that. A higher release level does not implicate a better coder.

Perhaps you are applying a negative connotation to the word 'subjective' that doesn't exist. Whatever the result (which I've already agreed is improved from the old system) the means you use to achieve that result are subjective. It's a good thing. It's also fact, and you're too smart for me to go quoting a dictionary on you

I think you are, perhaps, referring to the fact that release level is far more language neutral than the previous system. That's a good thing in this case, and a smart move, but not related to the discussion at hand. Are you saying that a higher release level is not better than a lower release level? Because that's certainly not the impression given by staff comments about it, or by the results seen in the release level list, which shows a strong correspondence between older, more experienced, more successful hackers, and higher release levels.