|
Delete his questions and move on.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Reputation History is a great new feature. It seems to work well, however I do have some suggestions:
1. the datetime format is pretty weird, I'd never put time in front of date. I personally would use "dd-MMM-yyyy HH:mm", or alternatively "yyyy-MMM-dd HH:mm".
[EDIT] The order has been fixed, now the hour format is still weird, it has both 24 hours and AM/PM! [EDIT]
2. I would put the Reputation column to the right of the Points column (they belong together), and right-align the Points column; perhaps even merge both columns.
3. I would appreciate the forum name be mentioned when the item is "Forum Message..."
4. Maybe you could add an active/passive column or a background code to differentiate own activities (e.g. post message) and other's activities (e.g. message got up-voted).
There is one bug (or at least a very bad feature): when not viewing page 1, and altering the search criteria, the page number is not reset to 1, causing an unexpected and possibly empty result. BTW: the same behavior is present in all your search facilities AFAICR.
And a wish list would include more search criteria, such as searching a specific rep category, negative points only, etc.
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles] Nil Volentibus Arduum
Please use <PRE> tags for code snippets, they preserve indentation, improve readability, and make me actually look at the code. modified on Monday, December 20, 2010 11:16 AM
|
|
|
|
|
Rep history has a massive, ugly bug and was not meant to be public just yet. You also spotted this a couple of days ago. I've left the page up ther since you know about it and if you think it's useful in it's current form then great - but I am planning on fixing it properly this weekend and then releasing.
On a broader topic I've been planning to rerun the rep calculations to take into account further things like points for bookmarking but I'm loathe to do it 2 weeks before I announce MVPs, but I think it's important we have MVPs for 2011 be based on what the rep calculations should be, not what they are.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: points for bookmarking
You know, this post you've just made seems important. I believe I'll bookmark it. 
|
|
|
|
|
Yes I think it could be useful, especially for (a) putting some minds at ease, and (b) spotting rep bugs if and when they occur. I would even suggest to offer the same info through the web service to be (there may be an identification problem to be dealt with first).
Not sure why you resent recalculating rep points if there is a need; all I would suggest is to organize it in such a way that nobody notices, i.e. have distinct accumulator fields while calculating, and just copy the results when all is done (I have more ideas on the subject). That way it is only some disk space and a lot of server load, no continuous visible change in rep values and colors and privileges.
Personally I don't expect big changes (AFAIK my total is off by 10 points), except indeed I suspect you only have captured recent bookmarking. A few people of the top-10 may be juggled a bit (up to 3K is my estimate) as they have lots of articles and bookmarks, so be it (one could argue it is unfair you change the rules after the facts, I won't). JSOP and POH might swap places once more.
Judging by the dates this year's MVP awards got published, you still have 4 weeks, however sooner would indeed be better. I'm also not expecting a strict relationship between rep points and MVPship; there may still be some relevant factors that escape the rep system.
IMO you should bite the bullet.
[EDIT]
Estimate was way off: by articles being bookmarked alone Sacha may gain 70K author rep, Marc 50K, Nish 37K, CG 15K, John 10K, Pete 5K, etc.
I'm in favor of articles getting rewarded more, as you well know.
[/EDIT]
Luc Pattyn [Forum Guidelines] [Why QA sucks] [My Articles] Nil Volentibus Arduum
Please use <PRE> tags for code snippets, they preserve indentation, improve readability, and make me actually look at the code. modified on Friday, December 17, 2010 9:02 PM
|
|
|
|
|
Or... OR, he could assign MVP's based on both the old AND the new rep points. The more the merrier.
|
|
|
|
|
I'm afraid you've been reading too many Vilmos sigs.
|
|
|
|
|
You misread me: I certainly don't resent recalculting. It's just a button click
What I worry about is the endless round of "but where did my points go????".
However, points should all go up, not down.
I will recalc, but I will put on a flame proof suit first. And yes, the rep list is designed specifically for you guys to see more clearly what's happening to save confusion and also help out with diagnostics if anything is amiss.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: where did my points go?
There's two aspects: during recalculation and afterwards.
1.
I think the main problem is what happens during recalculation: everyone gets back to zero, some are surprised, all colors and privileges are gone, long-term members can't edit/delete any longer, etc. That is why I recommend you recalculate in the background (using separate counters), while you continue to use the old values for displaying and decision making.
2.
I agree, the end result is bound to rise. And anyhow, it should fit what the FAQ says, however I still feel the FAQ still isn't good enough. It isn't always clear (some descriptions are long and confusing), it does not mention weight factors, etc.
And yes, some points might go down due to the large number of daily limits which seem to have been added recently; I remember that column being almost empty not so long ago; did you really implement all those limits?
If you want to avoid complaints about sudden changes, I recommend you stop applying rule changes retro-actively. You don't want the tax law changes to work retro-actively either, do you?
|
|
|
|
|
Luc Pattyn wrote: You don't want the tax law changes to work retro-actively either, do you
Oh, but they do. What was legal years back can be re-evaluated and found to have a different interpretation.
But anyway, point (1) has been mitigated in that we use a new method of calculating that drastically minimises the time your points are reset to 0.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
let's have it!
|
|
|
|
|
Chris Maunder wrote: It's just a button click
Go on Click It.......get it out of the way!
|
|
|
|
|
Chris Maunder wrote: It's just a button click
What? No need for two senior sys admins, entering 256-character launch codes simultaneously? reckless it is.
|
|
|
|
|
Standards are dropping woefully, I know...
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Luc Pattyn wrote: I'm in favor of articles getting rewarded more, as you well know.
Same here
You can spend a week writing an article, get 10 5s, or get the same rep points by answering two questions and posting "smart/funny" responses that will gain dozens of 5s. So overall the rep scores are not in favor of authors compared to people answering questions. I reckon the 2011 MVPs will reflect this bias (more of them will be people who are active in the forums than article authors).
|
|
|
|
|
Nishant Sivakumar wrote: Same here
I can't imagine why. 
Of course we need both articles and forum discussions.
|
|
|
|
|
Luc Pattyn wrote: Of course we need both articles and forum discussions.
Absolutely agree but I think it should be balanced.
Luc Pattyn wrote: I can't imagine why
Well I can't defend that accusation. Except that if I was hell-bent on getting a higher rep score, I believe I can do that, whether through articles or through forum activity. Just that these days I am not as concerned about such things as some years ago when I was numerically obsessed with this sort of stats
|
|
|
|
|
I'm glad to see your obsession has become less numerical now.
It is understandable though, you're the only sixtuple platinum, so why bother with rep points any longer?
It is too bad you're such a poor enquirer...
|
|
|
|
|
|
Chris Maunder wrote: rerun the rep calculations to take into account further things like points for bookmarking
About time you did that!
|
|
|
|
|
Luc Pattyn wrote: 1. the datetime format is pretty weird, I'd never put time in front of date. I personally would use "dd-MMM-yyyy HH:mm", or alternatively "yyyy-MMM-dd HH:mm".
Unless you make it user configurable and automatically set the default using location I vote for the ISO format. The US and EU formats will both offend large numbers of people who are too stupid to understand a date that's half backwards. Neither has any justification behind their stupidity so you might as well use the option that is most rational in design and offend both groups equally.
3x12=36
2x12=24
1x12=12
0x12=18
|
|
|
|
|
Dan Neely wrote: offend both groups equally
a noble goal.
|
|
|
|
|
Luc Pattyn wrote: Reputation History is a great new feature.
Where is it? I can't find it. Priviledge issue?
|
|
|
|
|
it comes and goes, it still is work in progress.
|
|
|
|
|
Ya, I read it was work in progress. Will wait to see what is that and the discussion around. Thanks.
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
