|
Luc Pattyn wrote: It has ten thousands
Sorry Luc, but we had over 10 million unique visitors last year based on IP. I know when you say "user" you mean an author or someone who posts a question (so yes, your estimates are closer for that) but for me, everyone who visits, or bookmarks an item, or downloads a zip file, or just comes and uses the site in whatever fashion suits them, is a user.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
My estimate was for unique physical people with at least one CP account, used for whatever purpose they choose; I suspect most users create their first account in order to download something, and I expect the majority returns to download one or a few files a year, and create an account each time, rather than remembering it somehow. Gmail becoming slightly harder to create new accounts may have a dampening effect on this; other e-mail providers may have compensated for it, I don't know.
As you know IP address counts may be misleading, with dynamic IP and all. I once observed I got hundreds of different addresses myself in a week's time, when my ISP decided to break connections every 30 minutes in an attempt to make do with the limited range he had available; he has added a couple of ranges since, so now the switch is much less frequent, however the choice is much larger (he now holds well over 100K IP adr).
|
|
|
|
|
10 million users may be about right, but what about the number of members?
|
|
|
|
|
7,497,076
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Ah, so then how many accounts?
|
|
|
|
|
If my members have split personality disorder then it's polite not to disciminate against them
In all seriousness, though, the number of dupe accounts isn't super high. My figure of over 10M visitors last year is based on a system that aggressively recombines multiple user records into single records based on IP address and member ID.
While it's true a single member may visit from multiple IPs, they do not typically employ a ton of different email addresses with which to sign up, and logs show that our "send my password" page is heavily used, meaning there's a great deal of people recovering their accounts.
As you are aware there is also a bigger issue: Corporate firewalls. Single home users may have multiple IPs, but you can have thousands of users all appearing as if they are from a single IP.
All we can accurately say is that 7.5M unique email addresses have been used to sign up, and based on a combination of email address, IP address, and cookies, we saw 10M+ people wander through our messy halls last year.
The simplest measure of our membership is "lots".
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
Chris Maunder wrote: it's polite not to disciminate
Then my optical drives need to learn some manners... they disc emanate all the time.
Thanks for explaining all that. I'd say your figures are fairly trustworthy.
|
|
|
|
|
I'm typing one hand on a laptop keyboard and the other hand is working on my main PC's keyboard.
Spelling comes a distant 3rd in all of this.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
I'm gonna have to start a tip/trick entitled "The Maunder Dictionary".
|
|
|
|
|
Interesting. Thanks.
You might want (or not want) to gather and publish activity histograms: downloading, voting, publishing.
PS: and you may want to update this page on some quiet day: http://www.codeproject.com/info/MediaKit.aspx[^]
|
|
|
|
|
Chris Maunder wrote: The simplest measure of our membership is "lots".
Or, "Not Enough" as the case may be!
------------------------------------
I will never again mention that I was the poster of the One Millionth Lounge Post, nor that it was complete drivel. Dalek Dave
CCC League Table Link
CCC Link[ ^]
|
|
|
|
|
We used to cutoff to 20, but I doubled that to allow separate calculations for authors and those who answer questions in the fora.
Thanks for the bug: I'll add it to the list.
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
You should have chosen 21 each Chris for a total of...!
|
|
|
|
|
I would like to suggest that article approval (on "pending" articles) require multiple approvals, say five. If it is not approved (garnering five approval votes) within a set time frame (say, one week), it is automatically deleted, and an email is sent to the author stating that not enough people approved it for publication.
The article page would show how many approval votes have been accumulated so that everyone can see what the current approval status is.
EDIT ===========
A variation of multiple votes would be maybe a points value that each article was required to meet before it was approved, with each vote being weighed by the user's reputation (like the vote weights are done now). That way, it might take 10 votes from low-rep users, but maybe only 2 or 3 from high-rep users to give the article enough points to be published.
".45 ACP - because shooting twice is just silly" - JSOP, 2010 ----- You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010 ----- "Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass." - Dale Earnhardt, 1997
modified on Tuesday, January 4, 2011 5:56 PM
|
|
|
|
|
Is there any accountability to article approvals? I'm wondering if more senior members of the site, perhaps including those with approval privileges, can see who approves the article. There can then be accountability among those who know what's going on and acceptable.
I wasn't, now I am, then I won't be anymore.
|
|
|
|
|
I don't think knowing who approved a given article is that important.
".45 ACP - because shooting twice is just silly" - JSOP, 2010 ----- You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010 ----- "Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass." - Dale Earnhardt, 1997
|
|
|
|
|
To clarify this - only articles that are Pending, so authors have a chance to finish their articles when they are at composing. As a mentor, I get to see articles that aren't ready for publish and they shouldn't follow the same criteria.
|
|
|
|
|
Right - only pending articles (articles that would otherwise need to be approved but that haven't been completed shouldn't even be allowed to be approved).
".45 ACP - because shooting twice is just silly" - JSOP, 2010 ----- You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010 ----- "Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass." - Dale Earnhardt, 1997
|
|
|
|
|
What about a multiple disapprovals? Say two or three? (reports included)
Thanks,
Sean Ewington
The Code Project
|
|
|
|
|
Well, in my eye, enough disapprovals to delete the pending article would not give the author a chance to fix things that were (hopefully) cited by the people doing the approving, especially if we were to use a weighted system.
".45 ACP - because shooting twice is just silly" - JSOP, 2010 ----- You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010 ----- "Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass." - Dale Earnhardt, 1997
|
|
|
|
|
John Simmons / outlaw programmer wrote: enough disapprovals to delete the pending article would not give the author a chance to fix things that were (hopefully) cited by the people doing the approving
Agree 100%
cheers,
Chris Maunder
The Code Project | Co-founder
Microsoft C++ MVP
|
|
|
|
|
in the points accumulation scheme I proposed (see a message below), I think the accumulated value should be reduced when the article gets updated:
- I would divide by 2 when positive, so a change pulls it down as those points do not belong to the revised version;
- I hesitate what to do with negative accumulator values; in theory I would divide by 2 too, however that may entice the author to just start a completely new article, which may not be what CP wants. Maybe resetting to zero *and* sending an e-mail to earlier disapprovers would be better.
PS: I also think earlier comments should remain visible to their author, and the article's author, no matter what. If I offer a comment, I want to be able to check it against the revised article, and so should the article's author.
|
|
|
|
|
Maybe that Lounge thread is what triggered you to start this one, I'll quote from my message there[^]:
I can't but repeat my long standing comment on the approval process; I find it wrong that a single person can approve an article (while it takes several people to report it and get it removed). In my view, approval would need a number of points (say 32), where each reader can vote from -2 to +2, which gets multiplied by the voter's author+authority weight (take the "largest" color, and give it say 8 for platinum, 4 for gold, and 2 for silver if applicable). So it would take at least two platinum's full support votes to pass, and each negative vote would make it harder to pass.
ADDED: I probably would *not* display the accumulator's value (except to the author himself), so a reader does not get influenced by the points given earlier.
|
|
|
|
|
I thought of the points thing and edited my message as you were adding this.
".45 ACP - because shooting twice is just silly" - JSOP, 2010 ----- You can never have too much ammo - unless you're swimming, or on fire. - JSOP, 2010 ----- "Why don't you tie a kerosene-soaked rag around your ankles so the ants won't climb up and eat your candy ass." - Dale Earnhardt, 1997
|
|
|
|
|
which is a bit of a surprise, as I know you're not particularly fond of down voting, and this scheme takes it all the way!
|
|
|
|
|