|
and old acc is one, that a user has forgotten, so he registered a new one.
an unused acc is one, that was not accessed for e.g. 3 months.
the benefit is, that we have less users, so the database gets smaller!
Don't try it, just do it! 
|
|
|
|
|
Alexander M. wrote:
and old acc is one, that a user has forgotten, so he registered a new one.
And we know that someone has created a replacement account how?
Alexander M. wrote:
an unused acc is one, that was not accessed for e.g. 3 months.
So someone takes a sabbatical or extended holiday something, and we kick their account off the site? Doesn't seem right to me.
Alexander M. wrote:
the benefit is, that we have less users, so the database gets smaller!
In this day and age, I don't think database size is really an issue 
Anyway, if the database is set up correctly, then deleting the user would require deleting the posts and articles they may have put here too (it's called referential integrity). So would you have the articles and posts deleted, just because someone hadn't accessed the site recently?
--
Ian Darling
"The moral of the story is that with a contrived example, you can prove anything." - Joel Spolsky
|
|
|
|
|
Ian Darling wrote:
So someone takes a sabbatical or extended holiday something, and we kick their account off the site? Doesn't seem right to me
Thank GOD, CP didnt do that.... I was out of CP for such a period!!!
Ian Darling wrote:
called referential integrity
 ;P
I was born intelligent
Education ruined me!.
|
|
|
|
|
This is the attachment of SpamAssassin to a newsletter:
Spam detection software, running on the system "xxx.yyy.net", has
identified this incoming email as possible spam. The original message
has been attached to this so you can view it (if it isn't spam) or block
similar future email. If you have any questions, see
postmaster@yyy.net for details.
Content preview: The Code Project Newsletter
URI:http://www.codeproject.com/
URI:http://www.codeproject.com/images/newsletter_header.gif For
comments or enquiries please contact
URI:mailto:webmaster@codeproject.com webmaster@codeproject.com Monday
8th December, 2003 [...]
Content analysis details: (8.4 points, 5.0 required)
pts rule name description
---- ---------------------- --------------------------------------------------
0.3 NO_REAL_NAME From: does not include a real name
0.8 FOR_FREE BODY: No such thing as a free lunch (1)
5.4 BAYES_99 BODY: Bayesian spam probability is 99 to 100%
[score: 1.0000]
1.2 MIME_HTML_MOSTLY BODY: Multipart message mostly text/html MIME
0.5 HTML_50_60 BODY: Message is 50% to 60% HTML
0.1 HTML_MESSAGE BODY: HTML included in message
0.1 HTML_LINK_CLICK_HERE BODY: HTML link text says "click here"
0.0 CLICK_BELOW Asks you to click below
Is this avoidable?
--
- Free Windows-based CMS: www.zeta-software.de/enu/producer/freeware/download.html
- See me: www.magerquark.de
|
|
|
|
|
Well, the HTML thing is avoidable by getting the newsletter in plaintext (what I do, because HTML Mail == Evil). As for the Bayesian analysis, it might be worth retraining SpamAssassin (if you can do that, I don't use it, but SpamBayes can be retrained), marking the CP letter as Ham.
Then ask Chris to sound less Spammy 
--
Ian Darling
"The moral of the story is that with a contrived example, you can prove anything." - Joel Spolsky
|
|
|
|
|
Looks mostly like the newsletter got accidentally added as spam to your SA Bayes database. Just gather up a few CP newsletters and send them to sa-learn as ham and you should be fine.
The default autolearn threshold for spam is 12.0, so I doubt the newsletters were auto-learned as spam (the newsletters slip right through SA on my mail server).
- Mike
|
|
|
|
|
I suggest to put an on-line description of the CodeProject's article rating system.
Or did I miss it?
The article rating system doesn't seem to be an average of all votes. I saw an article that had
8 votes and its rating was 4.8 or so. The next day the number of votes was 9 and the rating was about 3.9. How could one vote out of 9 change the rating by a whole point? What is the algorithm? I hope the CP editor aren't playing some tricks under the covers?
Has someone else noticed anything strange with this?
|
|
|
|
|
Tidbits from different parts of the site, (w/ emphasis added):
Rating is the sum of all votes divided by the number of votes. Voting can only be done by registered members and a persons vote is weighted by their [membership] level *.
Popularity is a measure of how much interest an article generates once it's been read. It's calculated as Rating x Log10(# Votes)
* Membership level calculation:
1-4 articles posted = Bronze. 5-14 articles = Silver. 15-24 articles = Gold. 25+ articles = Platinum. Each 500 messages posted adds one level, and if a user has posted a message then each year of membership adds 1 level. Bonus levels not valid for Gold members. Bronze is awarded at the beginning of the first 500 messages posted instead of the end.
|
|
|
|
|
Thanks for the info.
It be nice to link the rating scores
to a page with such description.
BTW: How exactly a vote is weighted
by the membership level? What is the formula?
|
|
|
|
|
|
I'd like to make another suggestion.
According to your description if someone wrote a bunch of low rated articles, her/his
vote will weight as much as a vote of who wrote
the same number of high rated articles.
It seems that the rating system is fundamentally
flawed.
I suggest to count the rating of voter's articles
when calculating the weights.
|
|
|
|
|
The rating system should also take into account people who vote an article low. I am saying this because it happened to me the article I wrote gets a couple of good ratings and it rates about 4 or higher then someone comes and knocks it down to a '1' for no other reason because they want to.
Don't get me wrong I don't mind if someone votes my article low but leave a response to why. Is that asking much? If you vote it low tell me why you did it so I can either correct the article or take notes for my next article I write.
Code4Food
----
"There is no try; only do or do not"
-Yoda
|
|
|
|
|
Change my picture ?
I uploaded the new one but seems didn't affected at all 
How bout if you add with some short of "erase button" ?
Thanks for your comment.
"Courage choose who will follow, Fate choose who will lead" - Lord Gunner, Septerra Core
"Press any key to continue, where's the ANY key ?" - Homer Simpsons
Drinking gives me amazing powers of insight. I can solve all the worlds problems when drunk, but can never remember the solutions in the morning. - Michael P Butler to Paul Watson on 12/08/03
|
|
|
|
|
go to your bio, hit CTRL+F5
Shog9
---
You'd better turn back, before the frost sets in.
These desert nights are for weathered men,
The ones who've already given in...
|
|
|
|
|
Shog9 wrote:
go to your bio, hit CTRL+F5
Coooolll, many thanks....
"Courage choose who will follow, Fate choose who will lead" - Lord Gunner, Septerra Core
"Press any key to continue, where's the ANY key ?" - Homer Simpsons
Drinking gives me amazing powers of insight. I can solve all the worlds problems when drunk, but can never remember the solutions in the morning. - Michael P Butler to Paul Watson on 12/08/03
|
|
|
|
|
hey CP team,
why are unregistered users allowed to write into the forums?
they mostly ask silly questions, or they repeat their messages the whole time...
is it possible to change that?
greez,
alex
Don't try it, just do it!
|
|
|
|
|
CodeProject likes to be friendly to newcomers. If you have problem, you might be put off having to go through a registration process. (I had a similiar issue with the Post Office site, where I just wanted to look up one post-code but had to register first; it was annoying)
Yeah, I know we get some rubbish questions posted by unregistered users, but heck we get crap questions from registered users too.
It is best just to ignore the silly questions and repeat messages if they annoy you - that's what most of us here do.
Michael
Blue canary in the outlet by the light switch
Who watches over you
Make a little birdhouse in your soul - They Might Be Giants
|
|
|
|
|
Besides what Michael said, I also think this just mucks up the user count. Call me a data-nut, but I'd like to see a count of actual registered users rather than entities that just want to plug a product or post something really stupid.
-----BEGIN GEEK CODE BLOCK-----
Version: 3.21
GCS/G/MU d- s: a- C++++ UL@ P++(+++) L+(--) E--- W+++ N++ o+ K? w++++ O- M(+) V? PS-- PE Y++ PGP++ t++@ 5 X+++ R+@ tv+ b(-)>b++ DI++++ D+ G e++>+++ h---* r+++ y+++
-----END GEEK CODE BLOCK-----
|
|
|
|
|
I just did a search of the Lounge for "awasu" but almost all the hits are on Taka Muraoka's sig.  It would be really swell if the search ignored sigs since they aren't really what a person is looking for when doing a search. Could you mark the sig text with a tag (maybe <sig>...</sig> ) which would be invisible on the board, but which the search could use to know what text to ignore?
I know it would be impossible to do this for existing posts, but maybe for the future...?
--Mike--
Ericahist | CP SearchBar v2.0.2 | Homepage | RightClick-Encrypt | 1ClickPicGrabber
Actual sign at the laundromat I go to: "No tinting or dying."
|
|
|
|
|
The sigs and messages aren't stored separately so it's impossible at the moment to search one and not the other. Sorry
cheers,
Chris Maunder
|
|
|
|
|
I know they're not stored separately, that's why I said "I know it would be impossible to do this for existing posts". But the sig is in a separate textarea, and I assume it's submitted separately from the message text. (If not, then never mind) If that's the case, then there must be a line somewhere before the message goes into the database that looks like
ContentText = ContentText + '\n' + Sig
Could you add this just before that line:
Sig = '<sig>' + Sig + '</sig>'
so the search code could know where the sig is, and ignore it?
--Mike--
Ericahist | CP SearchBar v2.0.2 | Homepage | RightClick-Encrypt | 1ClickPicGrabber
|
|
|
|
|
Michael Dunn wrote:
I know they're not stored separately, that's why I said "I know it would be impossible to do this for existing posts".
OK - assume I've not had a great deal of sleep lately.
I could definitely wrap the sig in the XML tags but comments search is done using the Full Text Indexing in SQL which, AFAIK, can't be told to index only a certain section of a text column.
cheers,
Chris Maunder
|
|
|
|
|
Chris Maunder wrote:
OK - assume I've not had a great deal of sleep lately.
Sounds like sig material.
|
|
|
|
|
Not if you want to be able to search for it though.
Chris Meech
It's much easier to get rich telling people what they want to hear. Chistopher Duncan
I can't help getting older, but I refuse to grow up. Roger Wright
I've been meaning to change my sig. Thanks! Alvaro Mendez
We're more like a hobbiest in a Home Depot drooling at all the shiny power tools, rather than a craftsman that makes the chair to an exacting level of comfort by measuring the customer's butt. Marc Clifton
|
|
|
|
|
Chris Maunder wrote:
the Full Text Indexing in SQL which, AFAIK, can't be told to index only a certain section of a text column.
Come on Chris, are you going to let M$ and BillG tell you what you can and can't do? Don't let the Borg keep you down.
Or... better yet, go get some sleep
--Mike--
Ericahist | CP SearchBar v2.0.2 | Homepage | RightClick-Encrypt | 1ClickPicGrabber
Kosh reminded me of some of the prima-donna programmers I've worked with. Knew everything but when you asked them a question; never gave you a straight answer.
-- Michael P. Butler in the Lounge
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
