Objectionable Word List

[Mike Chappell]

We are tinkering with objectionable content filtering for the student email.  The administration does not want to block anything at this point and they only want to monitor the student OU.  I have created a rule on the Student OU that contains a few test words.  The rule flags the message and sends a BCC to a group I created called 'overseer' which our Vice Principals are members of.  We do not modify the subject line and we do not block the message.  

This appears to work just fine and produces a result (record of offensive/possibly offensive content) that the administration is looking for.

I am wondering if anyone else has worked with this and if you have a sample word list that we can borrow from.  Our focus is mainly on cyber-bullying and drug use.  

Thanks

[Ginny Wexler]

I will be out of the office until Wednesday, December 14, 2011, and will be returning messages at that time.  If you have an urgent issue, please call the help desk at x1035.

[Matthew A. Peskay]

I used this list to seed ours - we actually block the message right when the student tries to send it so that it doesn't actually ever get sent.  

This needed a good amount of 'weeding' but it was a good start.  Probably not necessary but NSFW!

http://urbanoalvarez.es/blog/2008/04/04/bad-words-list/

All my best,

Matthew

--
You received this message because you are subscribed to the Google
Groups "Google Apps K12 Technical Forum" group.
To post to this group, send email to k12ap...@googlegroups.com
To unsubscribe from this group, send email to
k12appstech...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/k12appstech?hl=en?hl=en

[Joel Lowsky]

Hi Mike,

Attached is a list of around 450 offensive words.  I built this list based on a few lists I found online ... I have no idea what some of the words mean.

Joel

PS: I don't understand why GAFE makes you create your own list.  Seems like such an obvious (and easy) service to provide.

--

You received this message because you are subscribed to the Google
Groups "Google Apps K12 Technical Forum" group.
To post to this group, send email to k12ap...@googlegroups.com
To unsubscribe from this group, send email to
k12appstech...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/k12appstech?hl=en?hl=en

--
Joel Lowsky
Director of Technology
North Country School and Camp Treetops
(518) 523-9329, ext. 5406
4382 Cascade Road
Lake Placid, NY, 12946
www.ncstreetops.org

[Kevin Johnson]

Did you add these words to the objectionable content rule via a comma separated list?

[Joel Lowsky]

It's been a while since I added these words to our Objectionable Content List.  I think I exported it as CSV, then just copied and pasted the whole blob into the Add Words area.  I don't recall there being an Upload CSV option or the like.

Joel

PS: I'm writing this from our schools new test Chromebook.  So far so good.

On Tue, Dec 13, 2011 at 5:34 PM, Kevin Johnson <kjoh...@fifeschools.com> wrote:

Did you add these words to the objectionable content rule via a comma separated list?

--

You received this message because you are subscribed to the Google
Groups "Google Apps K12 Technical Forum" group.
To post to this group, send email to k12ap...@googlegroups.com
To unsubscribe from this group, send email to
k12appstech...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/k12appstech?hl=en?hl=en

[Matthew A. Peskay]

I just copied from the txt file (you can just grab the column in the spreadsheet) and pasted right into the list editor in the admin console.  

All my best,

Matthew

[melissa benson]

A few items...

1.) https://docs.google.com/document/d/1zQa_UDR3e0NRIS1glbOSUzCRp_LxKRBrcEkqmEDmUwo/edit?hl=en_US is a list of regular expressions that we currently use in postini. They were broken up by character limits that postini has. Our plan is to use these regexs and use the compliance type filter not the objectionable words. I believe the second one lets you do regexs. I'm still unsure what the difference is between the two - does anyone know? I've already read what each does..they seem like they are the same.

2.) Unless I'm missing something I do not see a way to block messages, or redirect them rather, into a quarantine where I can review them and send them along if they are false positives, send a warning via email to the student if it's a swear word or if it is a more major issue like bullying we will take appropriate action.

This is key for us because we do not want students receiving these emails, especially if it's a bullying situation. We do not want the student being bullied to even see the email. Also, the students seeing their email get bounced back and us sending a warning will squash anymore thoughts of sending these types of emails. Word spreads fast.

So it looks like there are two options block and no one ever sees the email or approve and I can BCC a copy. I don't know which is better now, no email being sent that's great but now we can't see what's going on and lots of times there are false positives. We could see what's going on and send the email but that will make students feel like hey no one can see and that will, in my opinion, make students more likely to send those types of messages. I suppose we could still send out warning emails but that's not ideal.

Anyone else experiencing these question marks?

[Curts, Eric]

We use a "bad word" list in an "Objectionable content" filter of GAES.  Any student email with words from the list will not be delivered.  We used to have a list of about 200 words, but got way too many false positives.  There are just so many reasons why a word that might be profane, could also be acceptable. So when we moved from Postini to the new GAES, we took some time to cut the list down to words that are very clearly profane.  The list is at 112 words now, and I have linked it below as a Google Doc:

https://docs.google.com/document/d/1spBBqOTpSSo_ZYd6lt5H9LKiM3SaJl1sUawCLYNZMiQ/edit

However, in the end I do not think it is realistic to think you are going to block profanity. Students will simply use variations of words, even if you come up with regex rules to try to catch the variants (plus you will increase the likelihood of false positives).  Yes, it is good to have a bad word filter (which we do), but it is more important to have a good Student Acceptable Use Policy.  If a student sends something inappropriate and another student informs you, you have the digital trail right there with the email and can address the misbehavior as described in your AUP.

Unfortunately the current version of GAES does not allow for blocking the message and sending a copy to someone else for review.  I imagine they will add this feature eventually, because it is part of Postini, and Google did say they would replicate all the features.

Eric

* Eric Curts

* Technology Director, North Canton City Schools

* 525 - 7th St NE, North Canton, OH 44720

* te...@northcantonschools.org

* (330) 497-5600 x377, FAX (330) 497-5618

*  Google+ Profile - Click here

*  Follow me on Twitter - http://twitter.com/ericcurts

*  Find me on Facebook - http://www.facebook.com/ericcurts

* http://www.ericcurts.com

* http://www.northcantonschools.org

* http://www.ohiotreasurechest.org

* http://www.appsusergroup.org

* http://www.thestateoftech.org

On Tue, Dec 13, 2011 at 5:02 PM, Mike Chappell <mcha...@abschools.org> wrote:

--

[Melissa Benson]

Good points thanks for the info. And yeah I do remember seeing them say they would have all of postini features eventually in the dashboard.

sent from mobile

[Mike Chappell]

First off, thanks for all the word lists.  I'm sure they will be helpful.

I toyed with the idea of blocking the email altogether, but we felt it better that the students be aware that someone was actually watching and paying attention so the flagged emails would go through with a modified subject line letting the students know that the message may contain objectionable material.  

There are two big problems though.  The first is the amount of false positives.  I am dealing with high school kids and flagging profanity will quickly eat up a ton of administrative time weeding through kids dropping an F-bomb in their email.  Another unforeseen issue came when I added the "N" word to the list and discovered the English class that was analyzing "To Kill A Mockingbird" which made for a pretty huge load of flagged emails and concerned students.

The other problem are the "pranksters" who found out that there is a word list and immediately began tossing random words in their emails with the intent of setting off the filter.

To deal with this I have removed the subject line modification.  While at first glance this may appear to be counter-productive, it lets the email go through without notifying the student that the email has been flagged but still forwards a blind copy to our Email Overseer group.  This should curtail the amount of prank/'testing the system' emails and remove the scare factor from legitimate emails that trigger the filter (i.e. "To Kill A Mockingbird") while still maintaining a copy of the email for review by the Administration.  

I have removed the moderation of messages from my Email Overseer group so all messages that trigger the filter are posted for review by the Disciplinary group of Vice Principals.  They are responsible for monitoring the messages for any potential bullying/drug use references. In general, I think the intention is to use this information for supporting evidence in the event of other issues.  Once we have a comprehensive word list in place I think this should meet my Administration's requirements.

[melissa benson]

Mike, We will/are running into the same issues.

I toyed with the idea of blocking the email altogether, but we felt it better that the students be aware that someone was actually watching and paying attention so the flagged emails would go through with a modified subject line letting the students know that the message may contain objectionable material. 

I agree it's good that they know someone is watching. When you say you edited the subject line, is that the only place you can edit it? I was thinking of doing the same thing. Does the recipient only see it or does the sender see it as well?

There are two big problems though.  The first is the amount of false positives.  I am dealing with high school kids and flagging profanity will quickly eat up a ton of administrative time weeding through kids dropping an F-bomb in their email.  Another unforeseen issue came when I added the "N" word to the list and discovered the English class that was analyzing "To Kill A Mockingbird" which made for a pretty huge load of flagged emails and concerned students.

Agreed. But with the F-bombs that aren't too serious we have been sending a "Canned Response" (email lab) to these first time offenders and keep a list in a Spreadsheet. The warning email says they have violated our policy and if it continues they will be sent to admin to deal with. We also send out an email a few times a year saying "What is bloomington apps email for..." and "what it is not used for". "What it's not used for contains signing up for facebook, myspace or personal conversation." We try to make it like your "work" email and teach them to be responsible with it. I have never had any experience with students purposely adding in swear words and if they do we simply document it and go to the appropriate people. Like I said, word travels fast through students.

The second part about the English class I am currently experiencing. Some teachers use turn it in and use Docs for this type of stuff. (I think words in Docs that get shared/emailed get caught as well/). I would say try to use Docs as much as possible and maybe without emailing a copy when sharing. Still trying to figure this one out.

To deal with this I have removed the subject line modification.  While at first glance this may appear to be counter-productive, it lets the email go through without notifying the student that the email has been flagged but still forwards a blind copy to our Email Overseer group.  This should curtail the amount of prank/'testing the system' emails and remove the scare factor from legitimate emails that trigger the filter (i.e. "To Kill A Mockingbird") while still maintaining a copy of the email for review by the Administration. 

I'll keep this in mind but like I said we've never had any "prankster" instances and if a student sends a legitimate email about to kill a mocking bird I would assume older kids will know hey I did nothing wrong.

I have removed the moderation of messages from my Email Overseer group so all messages that trigger the filter are posted for review by the Disciplinary group of Vice Principals.  They are responsible for monitoring the messages for any potential bullying/drug use references. In general, I think the intention is to use this information for supporting evidence in the event of other issues.  Once we have a comprehensive word list in place I think this should meet my Administration's requirements.

Interesting, so your VPs are doing the monitoring? Are they just monitoring the entire district or just their building? How we do it now is that I monitor them, send out canned responses for simple swear word violations, document it. We have 10k students K-12. I believe K-5 email is shut down, middle school internal only and HS everything is allowed. From the beginning of school I think I've send out around 6-7 warnings and 1 isntance of major bullying where we contacted the principal and put that student in a "restricted" org in Apps that had email disabled. By default we leave them there for 30 days (after the Principal says yes go ahead) and they contact me if they want it longer. If not 30 days later I remove them.

Like you mentioned taking up too much time with sorting through false positives and stuff could be an issue but I'm able to currently do it every few days. I'd like to hear how others are doing it as well, anyone else?

[Chris Franzen]

I have found the approach to be better if you put a stop to it and then don't make a big deal about the students don't continue down the same path.  I have it setup so the if certain words are seen the message can't be sent, end of story.  This method seems to make life easier.

Chris Franzen
Technology Coordinator
Nokomis School District
511 Oberle St.
Nokomis, IL  62075

QR

Please consider the environment before printing this email.

--

[melissa benson]

I'm doing a test and transitioning our dev domain. A few questions:

1.) When setting up the content compliance filters does it take awhile for it to take effect? Like 5 minutes? I'm doing trial and error and I'm getting some seriously mixed results. It will work, not work, I'll even delete the filter and change the message and it will still pop up the message saying the old rejected message!

2.) I'm testing out the send/bcc option. I can see how the recipient will see the modified heading and I can modify the header for the BCC person. But will the SENDER be warned or notified that their message was flagged?

3.) How does the "add header" option work? the two fields where it says X- blank field : blank field ... I can't find any documentation on what that actually is or an example of it in use?


I think we have decided on a list of words that are blocked, no questions asked for now until hey add the quarantine option. I also ahve hope for setting up filters that have multiple conditions - like swear words + the phrase "kill" "ugly". Block those, then have common swear words go through but flagged. But for now we're doing the transition and blocking all for now.

[Mike Chappell]

1) The settings take effect immediately.

2) I'm not sure on if the sender is notified when a message is flagged.

3) No clue. I found just as much info on that as you.

[Ginny Wexler]

The School District of Springfield Township is closed 12/23/11 thru 1/2/2012.  I will be returning e-mails when the district reopens.

[Chris Franzen]

Its not near as nice as Postini, until Google incorporates all of the features Postini has.  Right now, I'm also seeing some odd flags as result of my "bad words".  I've received a couple of messages that I can't figure out why they got flagged.  I would very much call it a work in progress.

  
Chris Franzen
Technology Coordinator
Nokomis School District
511 Oberle St.
Nokomis, IL  62075

QR

Please consider the environment before printing this email.

--