SUMO community discussions

Some findings and further investigation around "WMP in Firefox" article

  1. Hi Team!


    tl, dr; WMP article readers have malware installed in their computer and we want to use these users to improve SUMO and Firefox. We will like to have more brains behind the analysis that we just started.


    It's well known that our Windows Media Player article is by far the most popular one with more than 25.000 pageviews a day.

    On a hallway chat yesterday we started to dig more into the details of this traffic. I was aware that more than 95% is coming from several "radio station pages". I was half correct. The truth is that the traffic comes from a radio widget bundled with several crapware/toolbars. Specifically Babylon and Montiera are the main drivers. We want to understand why do we have this traffic and how can we make SUMO and Firefox better by understanding it.

    Based in some surveys, we learn that a good amount of users miss the ability to translate pages offered by the Google Toolbar so we think that this could be a driver to install these type of toolbars. The other theory is that they are just bundleware and the majority of users are forced to install it without any intention to use it.

    The idea is to use the visitors to this Windows Media Player article to learn more about the situation. We want to learn if they are really looking to visit the article to learn more about the WMP or simply they want to remove the toolbar or they landed in the article unintentionally.

    Based on that we drafted a quick action plan. It will be great to enrich it with any other idea, theory or task that you think is valuable and you want to contribute with.

    For now we have:

    • Verdi = To install, test and investigate how these toolbars work, are installed and why users will land in the article. You can see some of that in these screenshots:

    The toolbar has a nice link to our article: http://people.mozilla.org/~mverdi/screenshots/babylon-20120124-174437.jpg

    And it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg

    Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg

    • Ibai = To push for a survey that will run for a day in that article to learn more about the reasons of the visit, installation of the toolbar, etc. It's going to be a little bit annoying but the users are probably annoyed already. You can check the bug here: https://bugzilla.mozilla.org/show_bug.cgi?id=720870

    One of the things that we are not covering is that possibly, the users visiting the article have just installed the toolbar and they are not yet annoyed by the different hijacks. It will be great if we can find a way to survey them after a week or so to understand if they still want the toolbar or not.

    Based on the results we can try to push different actions:

    • Change the article to explain how to remove the crapware.
    • Block this toolbars.
    • Push for a translation feature?
    • Any other idea?
    Hi Team! ---- tl, dr; WMP article readers have malware installed in their computer and we want to use these users to improve SUMO and Firefox. We will like to have more brains behind the analysis that we just started. ----- It's well known that our Windows Media Player article is by far the most popular one with more than 25.000 pageviews a day. On a hallway chat yesterday we started to dig more into the details of this traffic. I was aware that more than 95% is coming from several "radio station pages". I was half correct. The truth is that the traffic comes from a radio widget bundled with several crapware/toolbars. Specifically Babylon and Montiera are the main drivers. We want to understand why do we have this traffic and how can we make SUMO and Firefox better by understanding it. Based in some surveys, we learn that a good amount of users miss the ability to translate pages offered by the Google Toolbar so we think that this could be a driver to install these type of toolbars. The other theory is that they are just bundleware and the majority of users are forced to install it without any intention to use it. The idea is to use the visitors to this Windows Media Player article to learn more about the situation. We want to learn if they are really looking to visit the article to learn more about the WMP or simply they want to remove the toolbar or they landed in the article unintentionally. Based on that we drafted a quick action plan. It will be great to enrich it with any other idea, theory or task that you think is valuable and you want to contribute with. For now we have: * Verdi = To install, test and investigate how these toolbars work, are installed and why users will land in the article. You can see some of that in these screenshots: The toolbar has a nice link to our article: http://people.mozilla.org/~mverdi/screenshots/babylon-20120124-174437.jpg And it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg * Ibai = To push for a survey that will run for a day in that article to learn more about the reasons of the visit, installation of the toolbar, etc. It's going to be a little bit annoying but the users are probably annoyed already. You can check the bug here: https://bugzilla.mozilla.org/show_bug.cgi?id=720870 One of the things that we are not covering is that possibly, the users visiting the article have just installed the toolbar and they are not yet annoyed by the different hijacks. It will be great if we can find a way to survey them after a week or so to understand if they still want the toolbar or not. Based on the results we can try to push different actions: * Change the article to explain how to remove the crapware. * Block this toolbars. * Push for a translation feature? * Any other idea?

    Modified by Ibai on

  2. I filed this bug asking to block the Babylon toolbar. And I also filed a bug about some crazy behavior I saw from the Ask toolbar while looking for things that bundled the Babylon toolbar.

    I filed [https://bugzilla.mozilla.org/show_bug.cgi?id=721264 this bug] asking to block the Babylon toolbar. And I also filed [https://bugzilla.mozilla.org/show_bug.cgi?id=721258 a bug] about some crazy behavior I saw from the Ask toolbar while looking for things that bundled the Babylon toolbar.
  3. I note the bug already has a comment c1 suggesting this should not be blocked.

    If blocking this does not meet current criteria then maybe the criteria could do with a rewording or clarification. I am sure many users are not really aware of

    • what they are agreeing to install
    • the potential problems this installation may cause

    The blocklist page ( https://wiki.mozilla.org/Blocklisting#A_High_Bar ) does say regarding its stringent conditions Note: The above conditions assume users have chosen to install the add-on or software in question. IMO users of this have not given an informed consent, especially as you point out it takes over the UI.

    I hope you have success in either blocking this or; maybe because of such a threat; having the providers of this permanently modify the behaviour of the add-ons.

    I note the bug already has a [https://bugzilla.mozilla.org/show_bug.cgi?id=721264#c1 comment c1] suggesting this should not be blocked. If blocking this does not meet current criteria then maybe the criteria could do with a rewording or clarification. I am sure many users are not really aware of * what they are agreeing to install * the potential problems this installation may cause The blocklist page ( https://wiki.mozilla.org/Blocklisting#A_High_Bar ) does say regarding its stringent conditions ''Note: The above conditions assume users have chosen to install the add-on or software in question.'' IMO users of this have not given an '''informed consent''', especially as you point out it takes over the UI. I hope you have success in either blocking this or; maybe because of such a threat; having the providers of this permanently modify the behaviour of the add-ons.
  4. Assuming that the Babylon Toolbar isn't going to be added to the Why does Mozilla disable some add-ons from running in Firefox? (at least not any time soon) the best we can do is improve the Fix problems with your home page or search article (Michael Verdi has a revision pending) and maybe rename and revise it to include general info on dealing with browser hijacks. Related discussion: https://support.mozilla.org/en-US/kb/removing-babylon-searchqu-or-mystart/discuss/2297

    Assuming that the Babylon Toolbar isn't going to be added to the [[Add-ons Blocklist]] (at least not any time soon) the best we can do is improve the [[Removing Babylon, SearchQU or MyStart]] article (Michael Verdi has a revision pending) and maybe rename and revise it to include general info on dealing with browser hijacks. Related discussion: https://support.mozilla.org/en-US/kb/removing-babylon-searchqu-or-mystart/discuss/2297
  5. more options

    This situation is almost worse now than it was in Firefox 0.9 before the "whitelist" of "allowed sites" or trusted websites was added to prevent drive-by extension installation with no warning or "Allow" button. And when that started happening an extension was made within a day or two to add the "whitelist" feature, and a new 0.9.1 or 0.9.2 (I can't recall exactly which version had the fix) version was created within a week or so.

    https://bugzilla.mozilla.org/show_bug.cgi?id=721264#c1
    Clearly he hasn't searched the forums for "Babylon" threads and read a dozen or so of those threads to see what Firefox users are going through.

    Either that or there might be an assumption that the user's are suffering "buyer's remorse" over selecting an option to install it, or the user's are just plain stupid and don't remember installing it, or they are lying and won't admit to selecting it as an option.

    If Babylon isn't going to be added to the Blocklist due to "technicalities" with the rules for the Blocklist, maybe what is needed is a 2nd list? A "questionable practices list" or "reported annoyances list" which would have a lower threshold for inclusion than the Blocklist seems to have. Then add a pref in the Security panel that the user can select to "Warn me when I am about to allow an add-on to be installed that I will probably regret very shortly." And give a Mozilla support person the ability to determine what is added to that list based on feedback from support contributors and Firefox users.
    And instead of just having an "Allow" button, have a two stage warning with a 5 second delay.
    1 - Are you really sure you want to install this? It has been reported by users like you as having some undesirable features.
    2- Are you really, really sure?

    This situation is almost worse now than it was in Firefox 0.9 before the "whitelist" of "allowed sites" or trusted websites was added to prevent drive-by extension installation with no warning or "Allow" button. And when that started happening an extension was made within a day or two to add the "whitelist" feature, and a new 0.9.1 or 0.9.2 ''(I can't recall exactly which version had the fix)'' version was created within a week or so. https://bugzilla.mozilla.org/show_bug.cgi?id=721264#c1 <br /> Clearly he hasn't searched the forums for "Babylon" threads and read a dozen or so of those threads to see what Firefox users are going through. Either that or there might be an assumption that the user's are suffering "buyer's remorse" over selecting an option to install it, or the user's are just plain stupid and don't remember installing it, or they are lying and won't admit to selecting it as an option. If Babylon isn't going to be added to the Blocklist due to "technicalities" with the rules for the Blocklist, maybe what is needed is a 2nd list? A "questionable practices list" or "reported annoyances list" which would have a lower threshold for inclusion than the Blocklist seems to have. Then add a pref in the Security panel that the user can select to '''"Warn me when I am about to allow an add-on to be installed that I will probably regret very shortly."''' And give a Mozilla support person the ability to determine what is added to that list based on feedback from support contributors and Firefox users. <br /> And instead of just having an "Allow" button, have a two stage warning with a 5 second delay. <br /> 1 - '''Are you really sure you want to install this? It has been reported by users like you as having some undesirable features.''' <br /> 2- '''Are you really, really sure?'''

    Modified by the-edmeister on

  6. Honestly, from the behaviour described by Verdi I'm not sure why it doesn't qualify. Certainly it wouldn't hurt to add a list of links to forum threads for more background info. Also if there are brave souls willing to try installing this to see what users are actually prompted with and take screenshots that would help as well.

    Honestly, from the behaviour described by Verdi I'm not sure why it doesn't qualify. Certainly it wouldn't hurt to add a list of links to forum threads for more background info. Also if there are brave souls willing to try installing this to see what users are actually prompted with and take screenshots that would help as well.
  7. I note access to the bug is now blocked. I wonder if that is an indication that babylon is being blocklisted or action being taken to prevent its bad behaviour as documented above.

    and it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg
    Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg

    P.S.
    Did any one see results posted anywhere for the toolbar survey http://blog.mozilla.com/sumo/2012/02/27/surveying-our-toolbar-users/

    I note access to the bug is now blocked. I wonder if that is an indication that babylon is being blocklisted or action being taken to prevent its bad behaviour as documented above. <blockquote>and it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg <br/> Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg </blockquote> ------- P.S. <br/> Did any one see results posted anywhere for the toolbar survey http://blog.mozilla.com/sumo/2012/02/27/surveying-our-toolbar-users/

    Modified by John99 on

  8. Hey John, it looks like it has been marked as security sensitive. It's also "assigned" but there hasn't really been any comments on it in about a month. I'll see if I can find out what's planned.

    Hey John, it looks like it has been marked as security sensitive. It's also "assigned" but there hasn't really been any comments on it in about a month. I'll see if I can find out what's planned.
  9. John99 said

    I note access to the bug is now blocked. I wonder if that is an indication that babylon is being blocklisted or action being taken to prevent its bad behaviour as documented above.
    and it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg
    Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg

    P.S.
    Did any one see results posted anywhere for the toolbar survey http://blog.mozilla.com/sumo/2012/02/27/surveying-our-toolbar-users/

    We decided not to post the results in the format that we were planning because of various reasons. But here they are for the contributors consumption: https://etherpad.mozilla.org/babylon-toolbar-survey

    ''John99 [[#post-46086|said]]'' <blockquote> I note access to the bug is now blocked. I wonder if that is an indication that babylon is being blocklisted or action being taken to prevent its bad behaviour as documented above. <blockquote>and it takes over our UI: http://people.mozilla.org/~mverdi/screenshots/babylon2-20120124-175833.jpg <br/> Bad practices all over: http://people.mozilla.org/~mverdi/screenshots/evil-20120124-181413.jpg </blockquote> ------- P.S. <br/> Did any one see results posted anywhere for the toolbar survey http://blog.mozilla.com/sumo/2012/02/27/surveying-our-toolbar-users/ </blockquote> We decided not to post the results in the format that we were planning because of various reasons. But here they are for the contributors consumption: https://etherpad.mozilla.org/babylon-toolbar-survey
  10. Thanks Ibai,
    the results were probably as expected, and I would conclude it is useful for translation but otherwise seems mostly unwanted.

    Verdi,
    I assumed it was blocked for security reasons. No point in asking further questions at present, but thanks for the reply. I note babylon technical support have apparently been posting comments in the support forum recently e.g. /questions/850289#answer-324976 and /questions/812492#answer-324972

    Thanks Ibai, <br/> the results were probably as expected, and I would conclude it is useful for translation but otherwise seems mostly unwanted. Verdi, <br/> I assumed it was blocked for security reasons. No point in asking further questions at present, but thanks for the reply. I note babylon technical support have apparently been posting comments in the support forum recently e.g. [/questions/850289#answer-324976] and [/questions/812492#answer-324972]