Allen Bierbaum | 4 Jan 20:54 2007

[Trac-dev] Status of SpamFilter and captcha


Sorry to bother the list with this again, but what is the current status 
of the SpamFilter?  Are people using this successfully and do you use it 
with a captcha plugin of some type?

I am asking because one of the trac sites I manage is still getting 
hammered with spam while running the SpamFilter plugin from early 
December.  I am hoping there is some good news on the horizon.  It is 
just taking to much effort to keep removing spam every day.

Thanks,
Allen

--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups "Trac Development" group.
To post to this group, send email to trac-dev <at> googlegroups.com
To unsubscribe from this group, send email to trac-dev-unsubscribe <at> googlegroups.com
For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Matt Good | 4 Jan 21:48 2007
Picon

[Trac-dev] Re: Status of SpamFilter and captcha


Allen Bierbaum wrote:
> Sorry to bother the list with this again, but what is the current status
> of the SpamFilter?  Are people using this successfully and do you use it
> with a captcha plugin of some type?

The SpamFilter is stable and has been extremely effective on the
Edgewall Trac site after the introduction of the SpamBayes filter and
karma system.  There's no captcha plugin, but based on the
effectiveness of the current filters there's no need to bug users with
this.  The SpamBayes filter is the most important.  With a little
training at the beginning it catches almost all spam.  The IPBlacklist
filter has also been helpful.  The Akismet service seems to be working
again, though it's been rather flakey in the past, so if you enable it
I'd only recommend giving it a small karma weighting.

BTW, this is probably more of a Users topic than a Dev one.

-- Matt Good

--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups "Trac Development" group.
To post to this group, send email to trac-dev <at> googlegroups.com
To unsubscribe from this group, send email to trac-dev-unsubscribe <at> googlegroups.com
For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Chris Forsythe | 4 Jan 21:35 2007

[Trac-dev] Re: Status of SpamFilter and captcha


We had this problem on Adium's trac for a while, moved to account 
manager and haven't had any problems since.

Allen Bierbaum wrote:

>
> Sorry to bother the list with this again, but what is the current 
> status of the SpamFilter?  Are people using this successfully and do 
> you use it with a captcha plugin of some type?
>
> I am asking because one of the trac sites I manage is still getting 
> hammered with spam while running the SpamFilter plugin from early 
> December.  I am hoping there is some good news on the horizon.  It is 
> just taking to much effort to keep removing spam every day.
>
> Thanks,
> Allen
>

--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups "Trac Development" group.
To post to this group, send email to trac-dev <at> googlegroups.com
To unsubscribe from this group, send email to trac-dev-unsubscribe <at> googlegroups.com
For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Allen Bierbaum | 4 Jan 22:57 2007

[Trac-dev] Re: Status of SpamFilter and captcha


Chris Forsythe wrote:
>
> We had this problem on Adium's trac for a while, moved to account 
> manager and haven't had any problems since.

We have account manager running and that helped for a while.  I have the 
SpamFilter setup so it deny's most commits to the wiki unless the user 
has registered and editing their account information.  This worked very 
well for a while.  Then the spam started coming in again.  I need to 
take a look and make sure I have the most recent version of everything.

It is good to know that there is nothing major that I am missing though.

-Allen

>
> Allen Bierbaum wrote:
>
>>
>> Sorry to bother the list with this again, but what is the current 
>> status of the SpamFilter?  Are people using this successfully and do 
>> you use it with a captcha plugin of some type?
>>
>> I am asking because one of the trac sites I manage is still getting 
>> hammered with spam while running the SpamFilter plugin from early 
>> December.  I am hoping there is some good news on the horizon.  It is 
>> just taking to much effort to keep removing spam every day.
>>
>> Thanks,
(Continue reading)

Allen Bierbaum | 4 Jan 23:02 2007

[Trac-dev] Re: Status of SpamFilter and captcha


Matt Good wrote:
>
> Allen Bierbaum wrote:
>> Sorry to bother the list with this again, but what is the current status
>> of the SpamFilter?  Are people using this successfully and do you use it
>> with a captcha plugin of some type?
>
> The SpamFilter is stable and has been extremely effective on the
> Edgewall Trac site after the introduction of the SpamBayes filter and
> karma system.  There's no captcha plugin, but based on the
> effectiveness of the current filters there's no need to bug users with
> this.  
I was thinking about the version of the plugin that only bugged users 
with a captcha when they tried to edit their user data.  This would 
effectively allow the user to prove that they are human once.  Then the 
SpamFilter code something it could check to know if the user is human.  
I think (but could be wrong) that this is already used by the SpamFilter 
code to allow known users to post even if Akismet thinks the content is 
spam.

Once again though, I haven't followed this discussion for quite a while 
so I could be completely wrong.
> The SpamBayes filter is the most important.  With a little
> training at the beginning it catches almost all spam.  The IPBlacklist
> filter has also been helpful.  The Akismet service seems to be working
> again, though it's been rather flakey in the past, so if you enable it
> I'd only recommend giving it a small karma weighting.
Ok.  I will give this a try.

(Continue reading)

Eli Carter | 4 Jan 23:23 2007

[Trac-dev] Re: Status of SpamFilter and captcha


On Thursday 04 January 2007 2:48 pm, Matt Good wrote:
> The SpamFilter is stable and has been extremely effective on the
> Edgewall Trac site after the introduction of the SpamBayes filter and
> karma system.

Do you track false-positives?  I had one the other day trying to note a 
duplicate...

Eli

--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups "Trac Development" group.
To post to this group, send email to trac-dev <at> googlegroups.com
To unsubscribe from this group, send email to trac-dev-unsubscribe <at> googlegroups.com
For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Alec Thomas | 5 Jan 01:56 2007

[Trac-dev] Re: Status of SpamFilter and captcha


On Thu, Jan 04, 2007 at 04:02:34PM -0600, Allen Bierbaum wrote:
> I was thinking about the version of the plugin that only bugged users 
> with a captcha when they tried to edit their user data.  This would 
> effectively allow the user to prove that they are human once.  Then the 
> SpamFilter code something it could check to know if the user is human.  
> I think (but could be wrong) that this is already used by the SpamFilter 
> code to allow known users to post even if Akismet thinks the content is 
> spam.
> 
> Once again though, I haven't followed this discussion for quite a while 
> so I could be completely wrong.

You are not mistaken, you're thinking of this:

    http://trac.edgewall.org/browser/sandbox/spam-filter-captcha

Which is SpamFilter with fallback to a captcha verification if the
other filters believe a submission as spam. ie. it provides a mechanism
for users to confirm that their submission is not spam.

--~--~---------~--~----~------------~-------~--~----~
 You received this message because you are subscribed to the Google Groups "Trac Development" group.
To post to this group, send email to trac-dev <at> googlegroups.com
To unsubscribe from this group, send email to trac-dev-unsubscribe <at> googlegroups.com
For more options, visit this group at http://groups.google.com/group/trac-dev?hl=en
-~----------~----~----~----~------~----~------~--~---

Christian Boos | 8 Jan 15:05 2007
Picon

[Trac-dev] Re: Reviving the Blame Branch


Christian Boos wrote:
> Tim Hatch wrote:
>> I notice that there's a cmlenz-dev/blame branch that hasn't been
>> touched in about a year.  This would be pretty much impossible to merge
>> with trunk at this point, so what does everyone think about creating a
>> new one under sandbox for collaborative work[1]?
>>   
>
> Ok, to get the ball rolling, I've committed a first shot at the blame 
> support, see:
>

Hello everybody, and happy new year!

I've enhanced the blame support quite a bit, and now I think it looks 
really good:

http://trac.edgewall.org/attachment/wiki/ChristianBoos/blame-screenshot2.png

Please try out the blame branch (svn co 
https://svn.edgewall.org/repos/trac/sandbox/blame).

Tim is also working on doing our own blame algorithm, which will be 
useful in order to annotate other versioned content (wiki pages, ticket 
description, ...). To that end, I'm thinking of adding a resource API, 
in connection with the WikiContext API and the work Alec started in the 
security branch. The idea would be to get to the versioned content 
starting from the context:

(Continue reading)

W Craig Trader | 8 Jan 17:26 2007
Picon

[Trac-dev] Re: Reviving the Blame Branch


Christian ...

Very pretty!  How soon before you think you could merge it into the trunk?

- Craig -

Christian Boos wrote:
>
> Christian Boos wrote:
>> Tim Hatch wrote:
>>> I notice that there's a cmlenz-dev/blame branch that hasn't been
>>> touched in about a year.  This would be pretty much impossible to merge
>>> with trunk at this point, so what does everyone think about creating a
>>> new one under sandbox for collaborative work[1]?
>>>   
>>
>> Ok, to get the ball rolling, I've committed a first shot at the blame 
>> support, see:
>>
>
> Hello everybody, and happy new year!
>
> I've enhanced the blame support quite a bit, and now I think it looks 
> really good:
>
> http://trac.edgewall.org/attachment/wiki/ChristianBoos/blame-screenshot2.png 
>
>
> Please try out the blame branch (svn co 
(Continue reading)

Jonas Borgström | 8 Jan 20:35 2007

[Trac-dev] Re: Reviving the Blame Branch


Christian Boos wrote:
> Christian Boos wrote:
>> Tim Hatch wrote:
>>> I notice that there's a cmlenz-dev/blame branch that hasn't been
>>> touched in about a year.  This would be pretty much impossible to merge
>>> with trunk at this point, so what does everyone think about creating a
>>> new one under sandbox for collaborative work[1]?
>>>   
>> Ok, to get the ball rolling, I've committed a first shot at the blame 
>> support, see:
>>
> 
> Hello everybody, and happy new year!
> 
> I've enhanced the blame support quite a bit, and now I think it looks 
> really good:
> 
> http://trac.edgewall.org/attachment/wiki/ChristianBoos/blame-screenshot2.png
> 

Cool, looks really nice!

Since I guess these pages are pretty expensive to generate, would it be
a good idea to add "<meta name="ROBOTS" content="NOINDEX"/>"-tags to the
template?

(To stop search engines like Google from killing public trac sites)

Cheers,
(Continue reading)


Gmane