How about an article subpage for comments by unregistered visitors ?

[Arne Eickenberg]

I haven't read the replies, so this goes directly to OP Milton: I had the same idea a few years ago, and I'm all for it. A specialist actually contacted me on one of my articles, but it was quite hard for him, because he had to switch to the article history, find my name, navigate to my user page, then google for my name, search for the e-mail address etc.

I'm open to debate the system for visitor comments. It shouldn't be a subpage or a second discussion page. This would be vandalized sooner or later, and it would divert energy for maintenance. So it could rather be an online comment/email form incl. captcha to send a message to an author, maybe via a CZ "constable office". There needs to be a spam filter nonetheless, and if the visitors actually enter the topic/article in the subject heading, the message is likely to reach the right author via a constable. Maybe we can also leave out the constable completely and find a fully automated system, like a pop-up menu to select authors/articles to contact/comment on. Such an automated system should also be able to check, if the email address entered in the online comment form is valid. If it's not, the message would be treated as spam.

[Arne Eickenberg]

I haven't read the replies, so this goes directly to OP Milton: I had the same idea a few years ago, and I'm all for it. A specialist actually contacted me on one of my articles, but it was quite hard for him, because he had to switch to the article history, find my name, navigate to my user page, then google for my name, search for the e-mail address etc.

I'm open to debate the system for visitor comments. It shouldn't be a subpage or a second discussion page. This would be vandalized sooner or later, and it would divert energy for maintenance. So it could rather be an online comment/email form incl. captcha to send a message to an author, maybe via a CZ "constable office". There needs to be a spam filter nonetheless, and if the visitors actually enter the topic/article in the subject heading, the message is likely to reach the right author via a constable. Maybe we can also leave out the constable completely and find a fully automated system, like a pop-up menu to select authors/articles to contact/comment on. Such an automated system should also be able to check, if the email address entered in the online comment form is valid. If it's not, the message would be treated as spam.

Suggestion for a comment system:

(1) there is a "Comment" link in the left CZ column, e.g. under "Permanent link".
(2) Clicking it leads to a special CZ page, with an online email/comment form, containing:
(a) Subject heading 1: automatically generated from the referral, e.g. "RE: CZ article 'Biology'"
(b) Recipient 1: automatically generated, i.e. send one copy of the message to the CZ editor in charge of the article
(c) Recipient 2: one of the article's authors, to be selected by the anonymous commenter from a pop-up menu that only contains the authors that actually worked on the article, sorted by the number and/or the scope of their edits, most active first
(d) Subject heading 2: topic of the comment to be typed-in by the commenter
(e) Field where the commenter needs to enter his e-mail address
(f) Field to repeat the e-mail address
(g) Optional field for the commenter to enter his real name
(h) Text field for the message
(i) CAPTCHA

Send button. A copy of the message will be sent to the supplied commenter's email address. If the message returns with permanent errors, the CZ system will not forward the message to the editor and the selected author.

[Daniel Mietchen]

I haven't read the replies, so this goes directly to OP Milton: I had the same idea a few years ago, and I'm all for it. A specialist actually contacted me on one of my articles, but it was quite hard for him, because he had to switch to the article history, find my name, navigate to my user page, then google for my name, search for the e-mail address etc.

I'm open to debate the system for visitor comments. It shouldn't be a subpage or a second discussion page. This would be vandalized sooner or later, and it would divert energy for maintenance. So it could rather be an online comment/email form incl. captcha to send a message to an author, maybe via a CZ "constable office". There needs to be a spam filter nonetheless, and if the visitors actually enter the topic/article in the subject heading, the message is likely to reach the right author via a constable. Maybe we can also leave out the constable completely and find a fully automated system, like a pop-up menu to select authors/articles to contact/comment on. Such an automated system should also be able to check, if the email address entered in the online comment form is valid. If it's not, the message would be treated as spam.

Suggestion for a comment system:

(1) there is a "Comment" link in the left CZ column, e.g. under "Permanent link".
(2) Clicking it leads to a special CZ page, with an online email/comment form, containing:
(a) Subject heading 1: automatically generated from the referral, e.g. "RE: CZ article 'Biology'"
(b) Recipient 1: automatically generated, i.e. send one copy of the message to the CZ editor in charge of the article
(c) Recipient 2: one of the article's authors, to be selected by the anonymous commenter from a pop-up menu that only contains the authors that actually worked on the article, sorted by the number and/or the scope of their edits, most active first
(d) Subject heading 2: topic of the comment to be typed-in by the commenter
(e) Field where the commenter needs to enter his e-mail address
(f) Field to repeat the e-mail address
(g) Optional field for the commenter to enter his real name
(h) Text field for the message
(i) CAPTCHA

Send button. A copy of the message will be sent to the supplied commenter's email address. If the message returns with permanent errors, the CZ system will not forward the message to the editor and the selected author.

I like this suggestion.

[Milton Beychok]

I like Arne's idea ... very much.

Milt Beychok

[Dan Nessett]

Suggestion for a comment system:

(1) there is a "Comment" link in the left CZ column, e.g. under "Permanent link".
(2) Clicking it leads to a special CZ page, with an online email/comment form, containing:
(a) Subject heading 1: automatically generated from the referral, e.g. "RE: CZ article 'Biology'"
(b) Recipient 1: automatically generated, i.e. send one copy of the message to the CZ editor in charge of the article
(c) Recipient 2: one of the article's authors, to be selected by the anonymous commenter from a pop-up menu that only contains the authors that actually worked on the article, sorted by the number and/or the scope of their edits, most active first
(d) Subject heading 2: topic of the comment to be typed-in by the commenter
(e) Field where the commenter needs to enter his e-mail address
(f) Field to repeat the e-mail address
(g) Optional field for the commenter to enter his real name
(h) Text field for the message
(i) CAPTCHA

Send button. A copy of the message will be sent to the supplied commenter's email address. If the message returns with permanent errors, the CZ system will not forward the message to the editor and the selected author.

If we decide to do this, I think we should give authors (and perhaps editors) the ability to opt-out. Not everyone wishes to receive anonymous comments. Speaking from experience (not at CZ, but elsewhere), such comments generally are low-grade ore. Also, we need to evaluate the technical feasibility of this proposal. Right now unregistered users cannot send email to registered users. (Try it. Log out and then go to someone's personal page. You will see that the "Email this user" link in the Toolbox has disappeared.)

[Matt Innis]

Considering all of the above, my net feeling is that our goal should be to get anyone who wants to make a comment to join in order to do that.  Real names discourages bad behavior so making sure that their account is tied to them encourages only constructive criticism.  Our challenge is to make the joining process easy and quick enough to make a quick comment, but still require verification. 

We should work to make account creation as close to immediate as we possibly can.

[Howard C. Berkowitz]

Considering all of the above, my net feeling is that our goal should be to get anyone who wants to make a comment to join in order to do that.  Real names discourages bad behavior so making sure that their account is tied to them encourages only constructive criticism.  Our challenge is to make the joining process easy and quick enough to make a quick comment, but still require verification. 

We should work to make account creation as close to immediate as we possibly can.

I agree with the points that anonymous comments don't fit the culture here, and that account creation should be fast.

I continue to advocate reputation/karma systems as a more effective means of avoiding problems than extensively bureaucratic verification. Among the worst security disasters in governments came from people that passed extensive security clearances, but then raised red behavioral flags that were ignored.

[Arne Eickenberg]

I'm always in favor of perfecting the account creation process.

[Matt Innis]

Considering all of the above, my net feeling is that our goal should be to get anyone who wants to make a comment to join in order to do that.  Real names discourages bad behavior so making sure that their account is tied to them encourages only constructive criticism.  Our challenge is to make the joining process easy and quick enough to make a quick comment, but still require verification. 

We should work to make account creation as close to immediate as we possibly can.

I agree with the points that anonymous comments don't fit the culture here, and that account creation should be fast.

I continue to advocate reputation/karma systems as a more effective means of avoiding problems than extensively bureaucratic verification. Among the worst security disasters in governments came from people that passed extensive security clearances, but then raised red behavioral flags that were ignored.

So... Can we require an short account creation process that can autoconfirm a non-free email account and then allow immediate use, but limit that use to a particular type page (or namespace) until their account can be verified or "karma'ed"?

[Arne Eickenberg]

So... Can we require an short account creation process that can autoconfirm a non-free email account and then allow immediate use, but limit that use to a particular type page (or namespace) until their account can be verified or "karma'ed"?

2-step process. Good. Agreed.

(1) Verify email > initial activation > limited account > limited e.g. to editing personal namespace and contacting other authors
(2) Verify person, name, credentials etc. > full account

[Matt Innis]

So... Can we require an short account creation process that can autoconfirm a non-free email account and then allow immediate use, but limit that use to a particular type page (or namespace) until their account can be verified or "karma'ed"?

2-step process. Good. Agreed.

(1) Verify email > initial activation > limited account > limited e.g. to editing personal namespace and contacting other authors
(2) Verify person, name, credentials etc. > full account

Can it be done programatically with what we have?

[Dan Nessett]

So... Can we require an short account creation process that can autoconfirm a non-free email account and then allow immediate use, but limit that use to a particular type page (or namespace) until their account can be verified or "karma'ed"?

2-step process. Good. Agreed.

(1) Verify email > initial activation > limited account > limited e.g. to editing personal namespace and contacting other authors
(2) Verify person, name, credentials etc. > full account

I agree that this is a good idea and support it. Except. Right now we have a lot of ideas that require technical development of the wiki software. I have elsewhere described the resources we currently have for doing such technical development. In summary they are extremely limited. I don't want to get the reputation as a Scrooge who is always bah-humbugging good ideas, but people really need to understand we are technical developer poor.

I am trying to create some infrastructure that will help train those without software development experience how to work on the wiki code, but it is going slower than I had hoped (The last email Greg sent me was on Nov. 3 and he hasn't been responding to me since). If we can get a software version control repository configured (Greg has installed the necessary software, but not enabled access to it), we will be in a position where some of us with the interest and determination can start learning the code. However, everyone should understand this is going to take time. There are 460,000 lines of PHP code in the modified mediawiki software that runs CZ's wiki (that does not include comments or blank lines in the file). The core of the code (where most of the execution occurs) is about 125,000 lines. That is a big code to learn and even the experts don't understand all of it.

So, by all means come up with these ideas. When there is some convergence on what it is people want to do, file an enhancement request in CZ Bugzilla. That will guarentee the idea doesn't get lost in the warp and woof of fora threads. But, please understand having a good idea is about 10% of the solution. The other 90% is the sweat of actually implementing it.

[Matt Innis]

Thanks, Dan   Of course, no-one knows what you're saying better then me... I think I was the one that first burst your bubble

I appreciate your difficult position and will count on you to be there to advise our production staff when the time *is* right!  Team work!

[Howard C. Berkowitz]

Considering all of the above, my net feeling is that our goal should be to get anyone who wants to make a comment to join in order to do that.  Real names discourages bad behavior so making sure that their account is tied to them encourages only constructive criticism.  Our challenge is to make the joining process easy and quick enough to make a quick comment, but still require verification. 

We should work to make account creation as close to immediate as we possibly can.

I agree with the points that anonymous comments don't fit the culture here, and that account creation should be fast.

I continue to advocate reputation/karma systems as a more effective means of avoiding problems than extensively bureaucratic verification. Among the worst security disasters in governments came from people that passed extensive security clearances, but then raised red behavioral flags that were ignored.

So... Can we require an short account creation process that can autoconfirm a non-free email account and then allow immediate use, but limit that use to a particular type page (or namespace) until their account can be verified or "karma'ed"?

If there is a karma process, why worry about free email accounts?

Surely a broadband operator email account (e.g., comcast) is only minimally more exclusive than hotmail, and it can be very inexpensive to set up a domain from which one wants to do mischief.

[Hayford Peirce]

Surely a broadband operator email account (e.g., comcast) is only minimally more exclusive than hotmail, and it can be very inexpensive to set up a domain from which one wants to do mischief.

Well, I myself use Comcast as my Internet Service Provider and it costs about $45 per month.  Which is a lot of money to pay for a teenage vandal who wants to do mischief at CZ, let's say.  I just don't believe that very many of them would pay $500 a year for the privelege of harassing us.  As for exclusivity, so what if Comcast has 10 million subscribers?  All we're concerned about is *verifying* the identity of Citizens.  If Jean-Pierre de Gaulle sends us an email from JPdeGaulle@comcast.net, I think we can be somewhat more certain that he's a real person than if the same request to open an account came from JPdeGaulle@gmail.com.

And setting up a domain is something else again -- once again, sure, it's possible for someone to create some sort of domain that would fool us poor Dumb Kops into accepting someone under a phony name.  But here's the key question: Why would anyone bother?  It's not as if we have anything valuable here worth stealing....